Digital era policy advising: Clouding ministerial perspectives?

Marando, Dylan ; Craft, Jonathan

Introduction

It is late Friday afternoon. Public servants and political staff around the table are grumpy. This is the umpteenth version of this briefing and they are still wrestling between policy options. As the Minister, you need to bring something to Cabinet, and soon. But you are seeing billions in cost, imagining the inevitable angry calls from key stakeholders, and being told that there are no guarantees the program will actually work with the dated technology in place. Two videoconferences with officials and stakeholders from across the country stand between you and your next rubber chicken dinner, and yet another keynote speech. You need an answer, and fast. How will "going digital" get you out of this mess?

Ministers, like the rest of us, are unable to escape the implications inherent in the shift towards an increasingly digital society and governance. Yet Ministers are unique in that they wield democratic power to make decisions on behalf of citizens. Not that long ago ministers would not have to worry about something "going viral" on social media or how well a government program or service works on a smart phone. "Big data," algorithmic analysis, web-enabled consultation, and data visualization, are all examples of disruptions that have yielded new channels for policy advice and produce qualitatively different inputs for policymakers. They do not however absolve ministers of the dilemmas inherent in governing: representation and accountability imperatives, decisions regarding values and resource allocations, and the use of coercive state authority, to name a few (Pierre 1998). Digital era governance (DEG) (1) will not only result in new disruptions and dilemmas as modern technologies displace older ones, but it as likely to intensify as it is to resolve fundamental "legacy" dilemmas and disruptions linked to governance.

As an example, it remains unclear if DEG will help or hinder well-known features of advisory work involving the very capacity--political, managerial, organizational, and analytical--of government to address policy challenges (Howlett, Wellstead, and Craft 2017; Tiernan 2015). For instance, the data deluge that accompanies digitization may enrich and democratize policymaking. It may facilitate broader and deeper consultation, generate new inputs for decision makers, or provide innovative instruments to strengthen inclusiveness and equity. Conversely, it may cloud policy relevant learning dynamics by overwhelming or paralyzing decision makers, or those who are charged with generating and sorting the advice that flows to them. DEG may be used to justify the exclusion of some policy participants or serve to alienate others. It may foster governance battlegrounds dotted by advocacy coalitions and instrument constituencies who clash on the basis of digital or non-digital problem definitions, policy options, or instrumentation and implementation preferences (Howlett, Wellstead, and Craft 2017; Weible 2008; Voss and Simons 2014). We know already that in some cases digital information and communication technologies expedite or slow the very pace of decision-making, altering the nature and place of advisory work as a component of the policymaking process (Hochtl, Parycek, and Schollhammer 2015).

To tackle the disruptions and dilemmas of DEG advisory work we borrow existing theoretical frames on digital governance, but focus on the micro-processes surrounding ministers of the crown. We posit that three concepts are essential for understanding the implications of digital governance for advisory systems: (1) information processing, (2) policy entrepreneurship, and (3) control and coordination. These three concepts feature prominently in the digital governance and policymaking literature (see Clarke and Craft 2017). However, our focus is novel in that is reappraises them from the vantage point of ministers. The article begins with an overview of policy advisory systems and the longstanding dilemmas they embody, in particular what advice has influence within PAS and why, along with whether these systems have the requisite capacity to serve decision makers well. It then reappraises these through the lens of DEG. The concluding section identifies a research agenda and discusses the key theoretical and empirical gaps that currently exist from the vantage point of ministers and DEG. We employ vignettes throughout, like that at the start of the article, with the aim of personalizing the disruptions and dilemmas of policy advising for ministers in DEG.

Policy advice, ministers, and the dilemmas of governance

Policy advice is most often conceived of broadly in terms of government knowledge utilization or more narrowly as part of the policy-making process (Peters and Barker 1993; Scott and Baehler 2010). In the latter, it is traditionally presented as a formulation stage activity, "covering analysis of problems and the proposing of solutions" (Halligan 1995: 139). It informs how policy environments are assessed and what options are generated, and ultimately what is proposed to decision makers. Ministers are central. This remains the case despite contemporary research which finds policy advising consists of a broader set of activities such as "research, data analysis, proposal development, consultation with stakeholders, formulation of advice for decision makers, guiding policy through governmental and parliamentary processes, and the subsequent evaluation of the outcomes of the policy" (Gregory and Lonti 2008: 838; Vesely 2016). As such, policy advice is better understood as the provision of recommendations, guidance, and the articulation of preferences in support of policy work (Craft and Halligan 2017).

A usefully heuristic to make sense of the growing number of potential advisory inputs is to conceive of them as part of a Policy Advisory Systems (PAS). This captures the reality that ministers benefit from an interlocking set of actors and organizations with unique configurations in each sector and jurisdiction that provides support and guidance to policymakers (Halligan 1995). The concept helps capture the plurality of advisory units, types of advice, and advisory practices involved in advisory work. Three leading categories of research have emerged that focus on questions of location of supply and control over it, the content of policy advice, and the dynamics of how PAS evolve. These categories are, of course, not mutually exclusive but each approach features a distinct focus and favored unit of analysis (Craft and Wilder 2017). Initial PAS research centered on questions of location or where advisory supply was generated (e.g. inside or outside of government) and the degree of control government could exert over it. The logic being that the closer the location of supply was to decision makers the greater control decision makers could exert over its congruence with government aims and its use (Halligan 1995; Weaver and Stares 2001). A second content-based approach favors a unit of analysis that focuses on the content of advice itself, emphasizing the substantive and procedural aspects of advisory work along with the proactive (longer-term) and reactive (short-term) nature of that work (Craft and Howlett 2012; Howlett and Lindquist 2004). A third dynamics approach attends to how individual advisory system units and entire systems change over time, with emphasis typically placed on the externalization away from public service supplies to alternatives and trends towards the politicization of PAS(Craft and Howlett 2013; Lindquist and Tiernan 2011).

A set of debates has emerged from this research linked to questions of the optimal distribution and relative influence of various advisory supplies, the capacity of individual units and entire systems to support effective policymaking, and the ability of PASs to evolve in keeping with the norms and exigencies of governance. On a basic level, there are enduring dilemmas about the structuring of these systems, how open or closed PAS should be, and what the implications of changes in the distribution, location, and type of advice circulating within PAS begets for representative democracy and optimal public administration (Pierre 1998; Plowden 1987).

Many of the dilemmas raised above are all well expressed in the shifting role and influence of the public service, vis a vis other supplies, within PAS. From a demand perspective public service advice has gone from the supply to a supply among many for ministers (Prince 2007; Zussman 2015). Its influence has been weakened by the growth in alternative supplies, the opening up of PAS to additional advisory supplies, but also given ministers' concerns regarding the capabilities and willingness of the public services to provide policy advice responsive to the needs of the government of the day (Rhodes, Wanna, and Weller 2010; Zussman 2015; Savoie 1994; 2015). Indeed, various non-public service advisory supplies, like those of private sector consultants or ministerial political staffs, have seen their fortunes improve as Westminster PAS have been restructured and evolved to meet changing demand (Halligan 1995; Craft 2015; Craft and Halligan 2017). Ministers have always faced dilemmas as to whose advice to seek out, use, and when (Savoie 2003; Bakvis 1997). This becomes more problematic when individual units, like the public service or others, are perceived to offer an uneven policy capacity (Howlett, Wellstead, and Craft 2017; Tiernan 2011), and when newer supplies come online, which are not bound by the professional and non-partisan norms of the public service. This is most vividly materialized in debates around evidence based policy-making practices and the increased use of appointed political staffs in Westminster systems (Zussman 2015; Savoie 2015; Weller 2015).

As elected political actors, ministers face dilemmas about how to balance various considerations that are expressed by these different advisory supplies. These are enduring dilemmas, but as is reviewed below from the DEG perspective, the governance context within which they are embedded is undergoing rapid transformation and there is uncertainty about how well the norms, institutions, processes, and players of traditional PAS will adapt (Craft and Howlett 2012; Prince 2007; Politt 2013; Parsons 2004). Finally, all of these PAS dilemmas are compounded by the sober reality that not all ministers are created equal. Some require more or less advisory support, of one kind or another, are adept at consultation or prefer closed advisory processes, or are particularly engaged in policy matters while other ministers are not (Rhodes 2011; Di Francesco 2012).

Crucially, PAS and the dilemmas they engender have not yet been subject to reappraisal from a digital perspective. How will DEG affect the distribution of available supply? Who will ministers go to for digitally informed policy advice and will they have the capacity to respond? What effects does digitization have on substantive characteristics of policy advice? Will it be transformative in terms of the content or only the channels by which policy advice is generated, brokered, and consumed? Finally, how, if at all, will DEG affect PAS dynamics? Will DEG spur or slow politicization, or the de-institutionalization of the public service as primary advisory unit, or engender altogether new dynamics linked to a need for more "agile" government?

Ministerial PAS dynamics in a digital governance era

To this point, we have sought to establish the volatility of the PAS. There will be a plethora of fault lines in the system as DEG disrupts and matures, but it seems worth highlighting three in particular because of their ability to illustrate the disruptions and dilemmas imbedded in PAS. As part of our conceptual descriptions, we have sought to identify particular dilemmas facing practitioners as the result of these themes. In some cases, we have signalled possible remedies.

Information processing dynamics

As the Minister's Chief of Staff, you're the last line of defense between the boss and a political policy failure. You're scrolling through to appendix G of the cabinet deck just sent to you, looking at another table with even more numbers--sensitive data "scraped" by the department regarding users of the program in question. As the tables, data sets, and versions pile up, you become less and less confident in the value of any one slide. You are not sure of what to prioritize. Given how complicated this is, you wonder if it is easier to go over all these numbers with the Minister? How will you synthesize this? It is easiest to fall back on what you and she are good at--engaging with stakeholders and framing the final policy decision? You can always sprinkle some data on top. Do you have any other choice given the tight timelines and the fact that a day earlier you were handed a separate deck from an external consultant with advice that competes with, and seemingly contradicts, the data being presented by officials? Both department officials and the consultant suggested a new round of data will be available next month. You go to brief the minister wondering if that is a data bridge you should cross once it is fully built?

Given our focus on ministers, special attention should be given to the demand-side advisory processing implications raised by the digital governance context. Like our fictional character above, it is easy to imagine a scenario where ministers' become overwhelmed by the realities of managing advisory work in a digital governance context. Of course the cognitive limitations of governing elites are well-established (Jones 2001) and the phenomenon of information overload has received concerted attention (Schwartz 2004). Researchers point to the tendency of decision-makers to rely on heuristics, which are simplistic and contrary to the actual direction of evidence (Tversky and Kahneman 1974; Miller 2009; Walgrave and Dejaeghere 2016). This has important linkages to the literature on information processing dynamics (Workman, Jones, and Jochim 2009).

In the digital governance context however, it is unclear how politicians become any better or worse at paying attention, and what incentives politicians have to deviate from tested heuristics and traditionally trusted interpreters of information. Critically, and based on a norm of accommodation toward ministerial attention deficits, one must wonder if the information interface of politicians changes at all in the digital era. It has long been an important function of senior officials and political staff to reduce the amount of paper that comes across the minister's desk; the two page briefing note is a staple in a well-functioning PAS. Why is it valuable for minsters and legislators to be actively involved in new ways of collecting and processing data? Why can they not merely be the implied beneficiaries of its outputs, while still enjoying the creature comforts of slide-decks and in-the-car updates? And if the interface is worth changing, how do you do it?

Though there has been much enthusiasm about the potential for data visualization to increase civic literacy (Lindquist 2015; Raphael et al. 2010; Janssen, Charalabidis, and Zuiderwijk 2012; Janssen and Helbig 2016), empirical evidence remains limited, especially as it relates to increasing literacy among segments of the population (e.g. ministers) presumed to already have a reasonable level of policy knowledge. Why, in the very unique circumstance of political decision-making, would we assume that new presentations of information are better than the status quo? Even if we assume they are beneficial, is there any sense of how marginally advantageous new presentations need to be in order to offset any costs associated with transitioning to a new channels or content in advisory work?

Disciples of digital governance will rightly argue that the rejection of modernized/digitized ministerial advice-getting merely pushes the problem of increased information processing upstream and applies more pressure on senior officials and advisers to sift, sort, and synthesize new information. This applies to both just "making sense" of these inputs and potentially translating them into "old" or more digestible formats. Some may also reference framing effects and note that an embrace of interface modernity risks putting too much responsibility and power in the hands of advice brokers. Both criticisms have merit, but rather than refuting a skepticism of new forms of information processing, these critiques simply compound them and demonstrate that judgments of ministerial attention have important ripple effects through the PAS.

In a similar vein, digitization raises crucial questions about the real and perceived value of content and those who produce it--that is to say, one should rightly wonder if politicians even desire to engage with new waves of data, or at least assume a wide variability in political appetites for data. Deference to data, on the part of politicians, should not be taken for granted. As some theory signals, growth in data could lead to a cheapening, in a strictly economic sense, of individual units of data (Bates 1990). We might imagine a corresponding demotion of data producers and maybe a promotion of those supplying skills ancillary to content (Shapiro and Varian 1999), interpretive or brokerage functions in the case of the PAS.

Even more starkly, as data become increasingly ubiquitous we may not only value but also proportionally demand it less, perhaps desensitized to its use, or overconfident that data will be available when we need it, or unsure of how to distinguish between good and bad data when novel and shiny data is being hurled from so many sources. Indeed, research within and outside of political science has pointed to the way in which we can overestimate personal knowledge as the result of easier access to information (Fisher, Goddu, and Keil 2015), and how too many options may result in the effective opting-out of deciding (Iyengar and Lepper 2000; Botti and Iyengar 2006; Bawden and Robinson 2009), reminiscent of Simon's (1956) earlier work on satisficing. There is also an established interdisciplinary tradition of recognizing the predominance of rhetoric and narrative in shaping analysis (Majone 1989; Forester 1993) as well as growing consensus that even within so-called evidence-based policy systems, unavoidable analytical trade-offs are made (Greenhalgh and Russell 2009; Davoudi 2006). The concept of data devaluation helps to capture these various risks, highlighting digitization's role in disrupting the data supply and demand equilibrium. It raises implications not just for the channels by which new advisory inputs will flow, but also for the decision making settings for ministers. As noted above, ministers are not a homogenous set of actors and DEG may serve to support and empower some ministers, while distracting, paralyzing, or simply boring others.

Policy entrepreneurship

You get it. The Premier is young and hip. She should be on social media, it serves as a way for her to keep up to date, connect with everyday citizens, and access thinking she might not otherwise have come across. But it seems like her policy preferences are heavily conditioned by those with the loudest social media voices. You are an expert in your field; you have lots of credibility owing to years of experience in the private sector, stints in government, and work at the highest levels of international non-governmental agencies. The Premier should be listening to you, right? You are sixty-five and think a hashtag is something that comes with a breakfast sandwich and coffee. (2) Your overtures for advice to the Premier, pro bono, keep being meet with polite form emails thanking you for your time. If only you knew how to get past the digital gatekeepers. How can you mobilize the constituencies most affected by the policy in question in time for the next round of consultations? Maybe your nephew can help. He is studying computers.

The notion of policy entrepreneurship has been influential since its inception as part of Kingdon's (1995) research on agenda setting. As the concept concerns DEG, one key question is what role ministers will play as entrepreneurs themselves? On the one hand, ministers are uniquely positioned to facilitate or impede DEG transformation both in terms of institutional machinery of government decisions pertinent to PAS structuring, but also as significant contributors to the cultural change associated with digital government and governance. DEG advisory work requires certain capacity, skills, and has potentially transformative implications for policymaking processes including choices of instrumentation, policy goals, and basic policy design activities (Clarke and Craft 2017) Governments naming ministers "responsible for digital" signals the interest, and intent, of political executives to drive DEG transformation (Thornton and Campbell 2017).

From a policy process perspective, entrepreneurship is crucial not only to the "coupling" between policy, problem, and political streams noted by Kingdon but also integral to the activities within the individual streams themselves. How will digitization affect advisory activity in the political, policy, and problem streams, and the function of entrepreneurs in exploiting their intersections at opportune times? On the one hand, it is plausible that as pertinent information is more readily disseminated, existing entrepreneurs will use their first mover advantage to capture that knowledge and lever their status. Perhaps, as the mergers of policy process streams, policy entrepreneurs will not particularly care how fast water is flowing, as long as they have the tools to build the dams. As likely, however, is that the intensity of policy's rushing waters has reached a point where dams of yesterday simply will not be up to code (pun intended). Existing entrepreneurs seem likely to come under siege. More data flowing to more advisory units means more competition, particularly from an emerging class of "techy" entrepreneurs (McNeill 2016). Automation is another threat, with algorithms possibly assuming the defacto role of robot policy entrepreneurs (Just and Latzer 2016). And some long-in-the-tooth entrepreneurs may simply opt-out of the PAS, feeling that they just do not have the energy to keep up with the technological and data fads.

If the technologically literate are those best disposed to heightened levels of policy entrepreneurship, the PAS may well need to respond to a resurgence of technocratic impulses. Those with affinities for number crunching may be coveted, while those more accustomed to the smokey backrooms and backslapping of yesteryear may be shown the back door. Such a phenomenon would resemble what have been termed "instrument constituencies"--policy actor sets dedicated to the articulation and promotion of particular kinds of solutions regardless of problem, or context (Beland and Howlett 2016; Voss and Simons 2014). As new digital policy instruments find their place in the tool boxes of policy designers, important questions arise regarding instrument selection, target population behavior, and broader design questions that are linked to the complementarity or various policy instruments along with digital and non-digital policy goals (Clarke and Craft 2017). For example, a new breed policy entrepreneur could be the crowdsourcing fetishists. This constituency of digital democrats seems to have penetrated fairly deep into the political consciousness and gained an impressive credibility and influence. As Denis Linders (2012) notes, using digital and social media technologies to open-up government processes has increasingly become the norm. Those who lead or mediate these open government initiatives have a significant ability to validate or invalidate policy (Heikka 2015), perhaps independent of content. Capacity to wield social media channels may become more important than capacity to understand or critique technical or normative aspects of policymaking (Zavattaro and Sementelli 2014). But again, this view stands in tension to some of the earlier theorizing related to information processing, pointing to another dilemma. Is a digital PAS going to stress our tendency to seek-out cognitive security blankets, not more sophistication, in the face of information overload? West (2005: 6) observes that in the face of visions of electronic governance:

[M]any government officials are conservative when it comes to change. Rather than rushing to embrace new technology, major political and economic interests slow the pace of technical innovation until they can figure out how to make sure their own vested interests are well-protected. This keeps the danger from new technology as low as possible, and forces technology to accommodate existing power structures rather than the other way around.

West notes the budding view that government internet technology in particular will not transform democracy in the long run, but rather reinforces existing social and political patterns. So perhaps those most inclined to entrepreneurship in the digitized PAS are not the technological literate but the classic policy entrepreneurs, now simply better resourced. All of which raises the spectre of a further restricting of access to the policy process and an entrenchment of elite interests.

Control and coordination

The Prime Minister's Office, working with the Privy Council Office (PCO), has developed a digital dashboard to monitor the progress of high profile policy priorities. Policy files are assigned the colour green (on track), yellow (at risk of running off track), or red (completely off the rails) and managed accordingly. Each week ministers' offices have to provide the PMO with a political update on how files are progressing. Concurrently, departments feed in real-time data that they are collecting along with pertinent data from various external sources like the World Bank. Each week senior PMO and PCO staffs walk the Prime Minister, and her key advisers, through the dashboard to fine-tune strategy. As a Deputy Minister, you know that your department's data paints an increasingly poor picture; especially juxtaposed against the trend lines emerging internationally. You try to brief up your minister but he is already late for cabinet committee and prefers face to face briefings to lengthy memos. The minister says his office will schedule a meeting first thing next week. You reach out to a colleague in PCO to give them a heads up; the minister is not very interested in the file. They confer with others in PCO and then reach out to their PMO policy shop counterpart who says the Prime Minister is already aware. The file is being downgraded by the PMO to yellow. The PMO asks PCO to convene a four corners meeting between PMO, central agencies, ministers, and departments for a course correction.

A third consideration that will be front and center as PAS digitize will be the tension between control and coordination. The erosion of traditional government control over public service advisory and the imperative to coordinate the seemingly endless plurality of advisory supplies and processes has consumed much oxygen in PAS scholarship (c.f. Halligan 1995; Dahlstrom, Peters, and Pierre 2011). The concepts have already been pointed to as salient in the broader literature on digital governance (Margetts and Dunleavy 2013). Yet, little attention has been paid to what the implications are for ministers. As advisory supplies, units, and practices "go digital," the degree of control government may be able to exert (or even want to exert), and the imperative to coordinate, are likely to become acute dilemmas. When should control over the advisory process or content be completely seeded or retained? Digitization is clearly well positioned to disrupt politicians' ability to exercise control, in part due to the creation of new "open" and "transparent" policy and governance processes and arenas, both internal and with non-governmental policy actors (Thornton and Campbell 2017). Paradoxically, digitization may in fact serve to increase the control government is able to exert through its unique ability to set standards and regulate. The amount and the types of control that are sought by governments over the public service advisory units and its regulation of the broader societal parameters around digital platforms and processes represent another dilemma.

Fishenden and Thompson (2013) note how digitization can be "characterized by a centralization/decentralization dialectic, involving, on the one hand, a tight central mandating of standards and interfaces, by a core function that is thus positioned to leverage the innovation and cost advantages of a plural, disaggregated delivery marketplace, on the other" (Fishenden and Thompson 2013: 14). Examining trends in the private sector, particularly as they relate to service delivery, Margetts and Dunleavy expand on Fishden's point, observing that digitization has allowed higher tier decision-makers in the private-sector to "keep tabs on more subordinates, be periodically involved in more decisions, insist on being consulted in real time and intervene more speedily when the key performance indicators go off-trend. Consequently, middle management in modern corporations has reduced, with substantial delayering leading to wider, flatter hierarchies" (Margetts and Dunleavy 2013: 7). Translating these insights into the public realm, the authors anticipate "a move towards an 'intelligent centre/devolved delivery' (IC + DD) design in the allocation of functions across tiers of government," especially for those governments facing austerity pressures (Margetts and Dunleavy 2013: 8).

A strict application of Margetts and Dunleavy's model, which is focused on service delivery, is a poor fit for an analysis of digital era PAS from a ministerial perspective. However, if we extend the idea of a data empowered centre beyond the service delivery realm and into the policy process, and if we assume that existing processes of policy consultation with "frontline workers" and "clients" of public policy are seen as an integral part of the existing public policy model, the idea of an intelligent centre should immediately strike the reader as potentially hostile to expectations of representativeness in the policy deliberation process (Martin 2011). Though opportunities to feed into the policy process may grow all-round as the result of digitization and enhanced information flow, it may not be equally shared, or beneficial. For instance, if those at the very "centre" (top) of the policy-making process have unique capabilities to collect, aggregate, and quickly analyze large volumes of data compared to those policy actors further from the centre, the latter may have their policy voice delegitimized or at least drowned-out due to its inability to compete with sophisticated volumes of policy input wielded by a select few.

From the perspective of ministers, policymaking already requires significant coordination not only within one's department but across to other departments, ministers' offices, cabinet and its committees, the prime minister and central agencies, and outward--to stakeholders, citizens, and pertinent international players (Craft 2016; Wilson 2016). Furthermore, notwithstanding what Margetts and Dunleavy have said about downward pressure on middle management in service delivery sectors, it appears to be the case that the middle managers in a PAS system (i.e. bureaucrats) enjoy a kind of institutional stickiness or dependence that would make them difficult to displace. Again, it is important to underscore the political-administrative boundaries that ministers must navigate and need to attend to. Control and coordination take special meaning for ministers who must look to their advisers to help avoid policy, programmatic, and political policy failures (Kemp 1986; Craft 2017; McConnell 2010).

Of course the first minister is a minister too, and the control and coordination dynamics of cabinet government have received much attention within Canada to date. The "center," including public service central agencies, has always served a coordination role. But a strong case for the growing centralization of power and concomitant displacement of others, including parliament, in the legitimate exercise of democratic authority exists (Savoie 1999; 2015). Given this risk of centralization, it will be important to determine who exactly forms the core of our redacted understanding of an intelligent centre. Who will be extended membership in the digital era, which advisers will be influential? The seeds of disruption from the digital governance era are already playing out at the center of government. For example, via the digitization of Prime Minister's notes and returns, policy and operational advice from PCO to the Prime Minister along with his responses. According to PCO, "The project resulted in significant changes to analysts' work practices" (Privy Council Office 2016: 7). Another project piloting e-cabinet within a cabinet committee "is intended to improve efficiency in the preparation and distribution of Cabinet meeting material; simplify access to Cabinet meeting material, including when Ministers are travelling across Canada; improve the consistency and reliability of communication with departments and within PCO, by providing a common environment to share information and collaborate in the preparation of meeting material; and improve security of information" (Ibid).

Discussion and research agenda

We have taken the position that some of what "digital" brings to the governance table is new while some is just a technologically updated manifestation of enduring governance dilemmas, "dilemmas 2.0." These involve questions as to the optimal distribution and relative influence of various advisory supplies, the capacity of individual units and entire systems to support effective policymaking, and the ability of PASs to evolve in keeping with the norms and exigencies of governance. The above makes plain that the channels, substance, and configurations of actors are all susceptible to disruption. This article is full of questions because so much remains unknown about DEG, particularly when we look through the prism of ministers. Case studies, surveys, ethnographic studies, and experiments offer critical avenues to test the impacts of digitization and PAS. Longitudinal approaches will help to track patterns and dynamics in advisory systems and the effectiveness of specific interventions. More specific recommendations, aligned with our three conceptual touchstones, would be:

Study demand dynamics explicitly

We need to update, and apply a Canadian lens, to ministerial role type classification schemes. Basic descriptive research is needed to identify how ministerial type may vary in a PAS context. How do ministers engage with advice, where do they go for supply, what types of supply do they need, and when? How well served are they? Not to mention the significant research questions which remain about the effects and implications of DEG on ministers within PAS. Some of these questions have been explored in other Westminster systems, such as whether and how political executives and ministers cope, how ministers learn, and how they can be best supported to undertake their responsibilities (Dahlstrom, Peters, and Pierre 2011; Tiernan and Weller 2010; Davison 2015). Canadian studies however lag well behind. A series of focus groups, workshops, surveys, interviews, or other research may help to better understand what ministers want in terms of policy advice, how ministers engage with advisory systems, and how digital governance impacts the nature of being a minister.

Experimenting with new interfaces

We think it valuable to test some of our theories of information processing dynamics with controlled experiments involving high-knowledge political actors and various digitally-enabled user-interfaces. Such experiments would seek to measure the capacity of legislators (or experimental proxies) to adapt to new framing techniques and perhaps new quantities and qualities of information. Though our hunches on information processing may bear fruit and show a general reluctance to new interfaces, we would anticipate political preferences for different interfaces to exist on a continuum thus signaling what new forms of ministerial communication may be least disruptive.

The creation of a digital competencies map

There would be much value in having improved survey accounts of politicians' self-perceptions of digital literacy and their perceptions of the digital literacy of those in the surrounding advisory system. We would then encourage, perhaps through an extension of the survey to other actors or through review of job descriptions and credentials amongst surrounding actors, the creation of a performance-based competency map that reveals actual patterns of literacy within the advisory system. Mapping these constellations on top of one another will help to identify some of the inherent biases of politicians when trying to establish the authoritativeness of digitally-enabled advice. The map could identify characteristics of proximity and representativeness for individuals who are seen to be authoritative interpreters of digital advisory inputs, outputs, and analytics. Are such authorities generally proximate to politicians, or are digitally-enabled advice givers more distant from government and politicians? Are advice givers also the idea generators? How does one build credibility as a digital interpreter? Are certain groups (e.g. bureaucrats) less able to develop credibility in this area than others? Do characteristics such as age, gender, ethnicity, level of non-technical education, or geographic status (urban or rural) automatically afford someone more credibility as a digital interpreter than others? Critically, are those who are considered authoritative sifters of content also consider representative of the citizenry and/or authoritative identifiers of representativeness? Given the tendency to associate digital literacy with younger age groups (Delli Carpini 2000), we would anticipate the competencies map generating some interesting demographic themes. Such findings might help to counterbalance accusations of intergenerational imbalance within the Canadian public policy systems (Kershaw and Anderson 2016).

An advisors' code of digital conduct

As a final category of inquiry, we would encourage researchers to consider the more normative implications of digital policy advice. Questions to consider would include: how will issues of digitization interact with the transparency and secrecy imperatives of advising political decision makers; what implications does digital era governance raise for advisory work in particular domains or trans-boundary or multi-subsystemic advisory work; how has digitization affected the compression or speed of policy advisory activities; how do notions of alignment or congruence between advisory units change when advice is digitized (either the processes of production or consumption, or the content of advice itself); will the attention considerations noted above lead to predominance or predispositions for proactive or reactive types of advice? As a potential ethnographic study, it may be valuable for a researcher to inhabit a political office for the purposes of identifying and understanding an emerging ethics of deliberation, particularly as it relates to governance in the digital era.

Notes

(1) We do not adopt the specific definition of "digital era governance" set out by Dunleavy et al. (2007) but rather follow the lead of the editors of this issue in understanding governance in the digital era as a set of activities, phenomenon, and societal trends.

(2) Unbeknownst to the authors at the time of writing, reviewers noted that the hashtaghashbrown pun appeared in the television sitcom Parks and Recreation.

References

Bakvis, H. 1997. "Advising the executive: Think tanks, consultants, political staff and kitchen cabinets." In The Hollow Crown: Countervailing Trends in Core Executives, edited by P. Weller, H. Bakvis, & R. A. W. Rhodes. Basingstoke: MacMillan, pp. 84-125.

Bates, Benjamin ]. 1990. "Information as an economic good: A re-evaluation of theoretical approaches. Mediation, information, and communication." Information and Behavior 3: 379-94.

Bawden, David, and Lyn Robinson. 2009. "The dark side of information: Overload, anxiety and other paradoxes and pathologies." Journal of Information Science 35 (2): 180-91.

Beland, Daniel, and Micheal Howlett. 2016. "How solutions chase problems: Instrument constituencies in the policy process." Governance 29 (3): 393-109.

Botti, Simona, and Sheena S. Iyengar. 2006. "The dark side of choice: When choice impairs social welfare." Journal of Public Policy & Marketing 25 (1): 24-38.

Chadwick, Andrew, and Christopher May. 2003. "Interaction between states and citizens in the age of the Internet: "e-Government" in the United States, Britain, and the European Union." Governance 16 (2): 271-300.

Clarke, Amanda, and Jonathan Craft. 2017. "The vestiges and vanguards of policy design in a digital context." Canadian Journal of Public Administration.

Craft, Jonathan, and John Halligan. 2017. "Assessing 30 years of Westminster policy advisory system experience." Policy Sciences 50 (1): 47-62.

Craft, Jonathan, and Matt Wilder. 2017. "Catching a second wave: Context and compatibility in advisory system dynamics." Policy Studies Journal 45 (1): 215-39.

Craft, Jonathan. 2015. "Conceptualizing partisan advisers as policy workers." Policy Sciences journal 48 (2): 135-58.

--. 2016. Backrooms and Beyond: Partisan Advisers and the Politics of Policy Work in Canada. Toronto: University of Toronto Press.

--. 2017. "Partisan advisers and political policy failure avoidance." Public Administration 95 (2): 327-41.

Craft, Jonathan, and Michael Howlett. 2012. "Policy formulation, governance shifts and policy influence: Location and content in policy advisory systems." Journal of Public Policy 32 (2): 79-98.

--. 2013. "The dual dynamics of policy advisory systems: The impact of externalization and politicization on policy advice." Policy and Society 32 (3): 187-97.

Dahlstrom, Carl, and B. Guy Peters, and John Pierre, eds. 2011. Steering from the Centre: Strengthening Political Control in Western Democracies. Toronto: University of Toronto Press.

Davison, Nehal. 2015. Supporting Politicians to Lead in Government. London: Institute for Government.

Davoudi, Simin. 2006. "Evidence-based planning: Rhetoric and reality." disP--The Planning Review 42 (165): 14-24.

Delli Carpini, Michael X. 2000. "Gen.com: Youth, civic engagement, and the new information environment." Political Communication 17 (4): 341-49.

Di Francesco, Michael. 2012. "Grand designs? The 'managerial' role of ministers within westminster-based public management policy." Australian Journal of Public Administration 71 (3): 257-68.

Dunleavy, Patrick, Hellen Margetts, Simon Bastow, and Jane Tinkler. 2006. "New public management is dead--Long live digital-era governance." Journal of Public Administration Research and Theory 16 (3): 467-94.

Fishenden, Jerry, and Mark Thompson. 2013. "Digital government, open architecture, and innovation: Why public sector IT will never be the same again." Journal of Public Administration Research and Theory 23 (4): 977-1004.

Fisher, Matthew, Mariel K. Goddu, and Frank C. Keil. 2015. Searching for Explanations: How the Internet Inflates Estimates of Internal Knowledge, Journal of Experimental Psychology, 144 (3): 674-687.

Forester, John. 1993. The Argumentative Turn in Policy Analysis and Planning. Durham, NC: Duke University Press.

Greenhalgh, Trisha, and Jill Russell. 2009. "Evidence-based policymaking: A critique." Perspectives in Biology and Medicine 52 (2): 304-18.

Gregory, Robert, and Zsusanna Lonti. 2008. "Chasing shadows? Performance measurement of policy advice in New Zealand government departments." Public Administration 86 (3): 837-56.

Halligan John. 1995. "Policy advice and the public sector." In Governance in a Changing Environment, edited by B. Guy Peters and D. T. Savoie. Montreal: McGill-Queen's University Press, pp. 138-72.

Heikka, Taneli. 2015. "The rise of the mediating citizen: Time, space, and citizenship in the crowdsourcing of Finnish legislation." Policy & Internet 7 (3): 268-91.

Hochtl, Johan, Peter Parycek, Ralph Schollhammer. 2015. "Big data in the policy cycle: Policy decision making in the digital era." Journal of Organizational Computing and Electronic Commerce 26 (1-2): 147-69.

Howlett, Michael, and Evert Lindquist. 2004. "Policy analysis and governance: Analytical and policy styles in Canada." Journal of Comparative Policy Analysis 6 (3): 225-49.

Howlett, Michael, Adam Wellstead, and Jonathan Craft, eds. 2017. Policy Work in Canada: Professional Practices and Analytical Capacities. Toronto: University of Toronto Press.

Hustedt, Thurid, and Sylvia Veit. 2017. "Policy advisory systems: Change dynamics and sources of variation." Policy Sciences 50 (1): 41-46.

Iyengar, Sheena S., and Mark R. Lepper. 2000. "When choice is demotivating: Can one desire too much of a good thing?" Journal of Personality and Social Psychology 79 (6): 995-1006.

Janssen, Marijn, and Natalie Helbig. 2016. "Innovating and changing the policy-cycle: Policy-makers be prepared!" Government Information Quarterly, doi: 10.1016/ j.giq.2015.11.009.

Janssen, Marijn, Yannis Charalabidis, and Anneke Zuiderwijk. 2012. "Benefits, adoption barriers and myths of open data and open government." Information Systems Management 29 (4) 258-68.

Jones, Bryan D. 2001. Politics and the Architecture of Choice: Bounded Rationality and Governance. Chicago: University of Chicago Press.

Just, Natasha, and Micheal Latzer. 2016. "Governance by algorithms: Reality construction by algorithmic selection on the Internet." Media, Culture & Society 39 (2): 238-58.

Kemp, David. 1986. "The recent evolution of central political control mechanisms in parliamentary systems." International Political Science Review 7 (1): 56-66.

Kershaw, Paul, and Lynell Anderson. 2016. "Measuring the age distribution in Canadian social spending." Canadian Public Administration 59 (4): 556-79.

Kingdon, John. 1995. Agendas, Alternatives, and Public Policies, 2nd ed. New York: Longman.

Linders, Dennis. 2012. "From e-government to we-government: Defining a typology for citizen coproduction in the age of social media." Government Information Quarterly 29 (4): 446-54.

Lindquist, Evert, and Anne Tiernan. 2011. "The Australian Public Service and policy advising: Meeting the challenges of 21st century governance." Australian Journal of Public Administration 70 (4): 437-50.

Lindquist, Evert. 2015. "Visualization meets policy making: Visual traditions, policy complexity, strategic investments." In Governance in the Information Era: Theory and Practice of Policy Informatics, edited by Erik W. Johnston. New York: Routledge, pp. 62-82.

Majone, Giancomo. 1989. Evidence, Argument, and Persuasion in the Policy Process. New Heaven: Yale University Press.

Margetts, Helen, and Patrick Dunleavy. 2013. "The second wave of digital-era governance: A quasi-paradigm for government on the Web." Philosophical Transactions of the Royal Society A 371: 20120382. doi: 10.1098/rsta.2012.0382.

Martin, Graham P. 2011. "Public deliberation in action: Emotion, inclusion and exclusion in participatory decision making." Critical Social Policy 32 (2): 163-83.

McConnell, Allen. 2010. "Policy success, policy failure and grey areas in-between." Journal of Public Policy 30 (3): 345-62.

Miller, Kristina C. 2009. "The limitations of heuristics for political elites." Political Psychology 30 (6): 863-94.

Parsons, Wayne. 2004. "Not just steering but weaving: Relevant knowledge and the craft of building policy capacity and coherence." Australian journal of Public Administration 63 (1): 43-57.

Peters, B. Guy, and Anthony Barker. 1993. Advising West European Governments: Inquiries, Expertise and Public Policy. Edinburgh: Edinburgh University Press.

Pierre, John. 1998. "Public consultation and citizen engagement: Dilemmas of policy advice." In Taking Stock: Assessing Public Sector Reforms, edited by B. G. Peters and D. J. Savoie. Ottawa: Canadian Centre for Management Development, pp. 137-63.

Plowden, William, ed. 1987. Advising the Rulers. Oxford: Basil Blackwell.

Pollitt, Christopher, ed. 2013. Context in Public Policy and Management. Cheltenham: Edward Elgar.

Prince, Michael. 2007. "Soft craft, hard choices, altered context: Reflections on 25 years of policy advice in Canada." In Policy Analysis in Canada: The State of the Art, edited by L. Dobuzinskis, M. Howlett, and D. Laycock. Toronto: University of Toronto Press, pp. 95-106.

Privy Council Office [Canada]. 2016. "Blueprint 2020 annual report." Available at https:// www.canada.ca/en/privy-council/services/reports/blueprint-2020/annual-report-2016. html.

Raphael, Chad, Christine Bachen, Kathleen M. Lynn, Jessica Baldwin-Philippi, and Kristen A. McKee. 2010. "Games for civic learning: A conceptual framework and agenda for research and design." Games and Culture 5 (2): 199-235.

Rhodes, R. A. W., John Wanna, and Patrick Weller. eds. 2010. Comparing Westminster. Oxford, UK: Oxford University Press.

Rhodes, R. A. W. 2011. Everyday Life in British Politics. Oxford: Oxford University Press.

Savoie, Donald. 1994. Thatcher, Reagan, Mulroney: In Search of a New Bureaucracy. Toronto, ON: University of Toronto Press.

--. 1999. Governing from the Centre: The Concentration of Power in Canadian Politics. University of Toronto Press.

--. 2003. Breaking the Bargain: Public Servants, Ministers, and Parliament. Toronto: IPAC/ University of Toronto Press.

--. 2015. "The Canadian Public Service: In search for a new equilibrium." In The International Handbook of Public Administration and Governance, edited by A. Massey and K. Johnson. London: Edward Elgar, pp. 182-98.

Schwartz, Barry. 2004. The Paradox of Choice: Why More Is Less. New York: Ecco.

Scott, Claudia, and Karen Baehler. 2010. Adding Value to Policy Analysis and Advice. Sydney, Australia: University of New South Wales Press.

Shapiro, Carl, and Hal R. Varian. 1999. Information Rules. Boston: Harvard Business School Press.

Simon, Herbert A. 1956. "Rational choice and the structure of the environment." Psychological Review 63 (2): 129.

Thornton, Daniel, and Lucy Campbell. 2017. Improving the Management of Digital Government. London: Institute for Government.

Tiernan, Anne. 2015. "Craft and capacity in the public service." Australian Journal of Public Administration 74 (1): 53-62.

--2011. "Advising Australian federal governments: Assessing the evolving capacity and role of the Australian public service." Australian journal of Public Administration 70 (4): 335-346.

Tiernan, Anne, and Patrick Weller. 2010. Learning to be a Minister: Heroic Expectations, Practical Realities. Melbourne: Melbourne University Press.

Tversky, Amos, and Daniel Kahneman. 1974. "Judgment under uncertainty: Heuristics and biases." Science 185 (4157): 1124-31.

Vesely, Arnost. 2016. "Policy advice as policy work: A conceptual framework for multi-level analysis." Policy Science 50 (1): 139-54.

Voss, Jan P., and Arno Simons. 2014. "Instrument constituencies and the supply side of policy innovation: The social life of emissions trading." Environmental Politics 23 (5): 735-54.

Walgrave, Stefaan, and Yves Dejaeghere. 2016. "Surviving information overload. How elite politicians select information." Governance 30 (2): 229-44.

Weaver, Kent, and Paul B Stares. 2001. "Overview." In Guidance for Governance: Comparing Alternative Sources of Public Policy Advice, edited by R. K. Weaver & P. Stares. Tokyo: Japan Center for International Exchange, pp. 71-88.

Weible, Christopher. 2008. "Expert-based information and policy subsystems: A review and synthesis." Policy Studies journal 36 (4): 615-35.

Weller, Patrick. 2015. "Policy professionals in context: Advisors and ministers." In Policy analysis in Australia, edited by B. Head & K. Crowley. Bristol: Policy Press, pp. 23-36.

West, D. M. 2005. Digital Government: Technology and Public Sector Performance. Princeton, NJ: Princeton University Press.

Wilson, R. P. 2016. "Trust but verify: Ministerial Policy advisors and public servants in the Government of Canada." Canadian Public Administration 59 (3): 337-356.

Workman, S., Brian D. Jones, and Asheley E. Jochim. 2009. "Information processing and policy 1044 dynamics." The Policy Studies journal 37 (1): 75-92.

Zavattaro, Staci M., and Arthur J. Sementelli. 2014. "A critical examination of social media adoption in government: Introducing omnipresence." Government Information Quarterly 31 (2): 257-64.

Zussman, David. 2015. "Public policy analysis in Canada: A forty-year overview." In A Subtle Balance: Expertise, Evidence, and Democracy in Public Policy and Governance 1970-2010, edited by E. Parsons. Montreal/Kingston: McGill/Queen's University Press, pp. 11-36.

Dylan Marando is a PhD Candidate, Department of Political Science, University of Toronto. Jonathan Craft is Assistant Professor, Department of Political Science, School of Public Policy and Governance, University of Toronto.