Going by the book: Academic guides for public servants

Kathy MacDermott

George Argyrous (ed) Evidence for Policy and Decision-Making: A Practical Guide, Sydney, UNSW Press, 2009 (304 pp). ISBN 9-780868409-030 (paperback) RRP $49.95.

Claudia Scott and Karen Baehler Adding Value to Policy Analysis and Advice, Sydney, UNSW Press, 2010 (288 pp). ISBN 9-780868408-590 (paperback) RRP $49.95.

Governments are responsible for policy; public servants assist and advise on it. In carrying out these tasks, public servants are bound by section 10 of the Public Service Act 1999 to be ‘responsive to the Government in providing frank, honest, comprehensive, accurate and timely advice’. Meeting all of these requirements at once has never been straightforward. Most recently, there has been concern about the proper balance between responsiveness and comprehensiveness. ‘Responsiveness’ has a range of meanings, but is generally used to mean sensitivity to the intent and direction of government policy (MacDermott 2008a, p. 4ff.). ‘Comprehensiveness’ is generally taken to mean telling government all the facts and considerations relevant to a particular policy decision. The shorthand for these two values is ‘telling the government what it wants to know’ and ‘telling the government what it needs to know’.

The failure of the public service to strike a balance between these two values was cited by academic and media commentators with increasing frequency over the course of the Howard Government, beginning with the ‘Children Overboard’ affair (known to the Senate as ‘A Certain Maritime Incident’), which involved the Departments of Defence, Immigration and the Prime Minister and Cabinet. This was followed by the cases of the detention of Cornelia Rau and the deportation of Vivian Solon, involving the then Department of Immigration and Multicultural and Indigenous Affairs; the payments made by the Australian Wheat Board to the regime of Saddam Hussein to obtain contracts for the sale of Australian wheat to Iraq, involving the Department of Foreign Affairs and Trade; the detention of Dr Mohamed Haneef, involving the Australian Federal Police; and the involvement of a senior public servant in the Employment and Workplace Relations portfolio as the face of the Howard Government’s WorkChoices media campaign. By the time of the 2007 election campaign the then Opposition was able to argue with some force that ‘the ability and expectation the [sic] public servants give frank and fearless advice has been threatened by the implicit understanding that government only wants to receive advice they want to hear’ (Wong 2007, p. 5). There was talk of the politicisation of public service policy advising (see, for example, Mulgan 2007; McGuinness 2007; Manne 2006) and of the replacement of evidence-based policy by policy-based evidence (Banks 2009, p. 16).

Following the 2007 election, the Rudd Government tackled re-balancing of public service advising by changing a number of the drivers with which the previous government had exerted pressure on the public service. As Special Minister of State, Senator Faulkner pushed through reforms promised by the incoming Labor government. By Faulkner’s own tally (2010), these included: new merit selection processes for public service agency heads and the removal of performance pay for departmental secretaries and senior statutory office holders; the establishment of the Ethics Advisory Service for public servants; changes to the Standards of Ministerial Ethics; the introduction of a Code of Conduct for ministerial staff; and changes to Freedom of Information provisions and government advertising guidelines whose effect was to increase government accountability and transparency. These reforms were important. However, they were not comprehensive; and after Senator Faulkner’s transfer to the Defence portfolio, some of them were rolled back (MacDermott 2008b, p. 98; Mulgan 2010), while the push to reform public service advising has been moved away from government and into the public service itself.

In 2009, Rudd passed the torch of public sector reform to the secretary of his department—the Department of the Prime Minister and Cabinet—who chaired an Advisory Group that managed a process that in March of this year issued a Blueprint for the Reform of Australian Government Administration. One of the questions posed by the Blueprint is whether the balance between responsiveness and comprehensiveness in public service policy advising can be improved by ignoring politicians and focusing on internal changes to public service machinery and skill sets. Recommendation 3 of the Blueprint, which deals with strengthening strategic policy, calls for a service-wide committee (a ‘Strategic Policy Network’) and agency-specific strategic policy committees (‘project teams’), both of which are intended to broaden policy perspectives. In addition, the Blueprint foreshadows making available to policy advisers ‘a policy tool kit … that provides guidance on strategic thinking, problem definition, stakeholder engagement, data analysis, project management, implementation, regulation, design and evaluation’ (Advisory Group on Reform of Australian Government Administration 2010, p. 42).

Governments are responsible for policy; public servants assist and advise on it.

How helpful these initiatives are likely to prove can be assessed by looking at two new books on policy advising just published by UNSW Press. Evidence for Policy and Decision-Making, edited by George Argyrous, and Adding Value to Policy Analysis and Advice, by Claudia Scott and Karen Baehler, provide just the type of guidance the Blueprint calls for—on strategic thinking, problem definition, stakeholder engagement, data analysis, project management, implementation, regulation, design and evaluation. Because of the ground they cover between them, the books raise also the broader question of whether policy advising can be fixed with a tool kit.


Evidence for Policy and Decision-Making aims to improve policy by improving its evidentiary basis. The book is effectively skills-based rather than theoretical and, with some exceptions, is pitched to public service practitioners rather than students, consultants or other academics. The skills on offer include evaluation principles and practice, cost-benefit analysis, the use of sample surveys and secondary data, the use and management of consultants, economic modelling and forecasting, scenario analysis and a very useful introduction to basic statistical methods. The collection is a sort of manual for those public servants whose policy advising responsibilities take them beyond their own specialist subject matter skills. It does not deal substantially with policy advising itself but does assist in dealing with policy’s underpinnings, and in this respect could prove very useful. Its contribution is particularly important because of the damage done to the research capacity of the public service by the preference of recent governments for market research over other types of evidence, and by the associated tendency for public service budget cuts to disproportionately affect agency research staff (Briggs 2005). Several of the practitioners’ guides in the Argyrous book should be well suited to the policy toolkit proposed by the Blueprint to strengthen public servants’ strategic policy capability.


Adding Value to Policy Analysis and Advice addresses the more theoretical components of a policy advising toolkit: strategic thinking, problem definition and stakeholder engagement. To these it applies the ‘two principal variant streams in the scholarly discussion of public policy’:

One is the ‘rational comprehensive’ approach, which gives rise to policy cycle models and linear, multi-step models of decision-making and problem solving. The other major stream, which focuses on actors and institutions rather than processes, has come to be called the ‘network participatory’ category of approaches, reflecting an emphasis on participatory democracy and horizontal networks of influence rather than the vertical exercise of authority (p. 26).

The ‘rational comprehensive’ approach is grounded in scholarly efforts to bring order to the generally untidy business of public service policy advising. These efforts give rise to models of the policy process which have an ex post facto and idealised air to them that does not sit comfortably with the cut and thrust of day-to-day policy advising. There are maps and flow charts setting out analytical sequences for developing and analysing policy from problem to solution. These maps and flowcharts go by names unheard of by most policy advisers, such as Bardach’s eightfold path of policy analysis (‘assemble evidence / define problem / select criteria / construct alternatives / project outcomes / confront tradeoffs / decide / tell story’) and the Althaus, Bridgman and Davis cycle model (‘identifying issues / policy analysis / policy instruments / consultation / coordination / decision / implementation / evaluation / identifying issues …’). There are basic logic chains, fishbone diagrams and a policy hexagon—all of them intended to assist policy advisers systematise their thinking and ensure that all relevant factors are taken into account in an orderly fashion before deciding on the best policy options.

In 2009, Prime Minister Rudd passed the torch of public sector reform.

As the quotation above explains, the ‘network participatory’ approach focuses on actors and institutions rather than intellectual processes. The authors adopt what they call a ‘systems perspective’, by which they mean ‘connecting up actors and institutions on the inside and outside of government’. Politicians and public servants are grouped together as insiders, while outsiders are taken to include ‘everything and everyone that can affect or be affected by policy’ (p. 6). Not surprisingly, the ‘network participatory’ approach to policy advising features comprehensive consultation.

The ‘rational comprehensive’ and ‘network participatory’ approaches are not alternatives: each step of Bardach’s eightfold path of policy analysis could, for example, become the subject of an exhaustive public consultation process under the ‘network participatory’ approach. But it is unlikely that many public servants would be able pursue such a thorough policy development process unless the policy in question was so very long term that the government of the day was not expected to have to reach any actual policy decision during its current term of office—or unless the real point of consulting ‘everyone that can affect or be affected by policy’ is just to draw the public’s attention to the government’s democratic credentials. If government is taking a policy issue seriously, then any theoretical models of policy development will have to be applied by public servants inside a system that resists linear, multi-step models of problem solving and assumes a more complex, dynamic and sometimes even Machiavellian understanding of ‘network participatory’ approaches.

Adding Value to Policy Analysis and Advice recommends applying a systems perspective to policy analysis and advising, but it does not really take much interest in the Westminster system within which Australian public servants work. In the Westminster system, politicians and public servants should not be automatically and invariably grouped together as insiders: one group is elected and the other is not, and the whole point of the institutional arrangements making up this system is to keep these two groups distinct. The elected group makes decisions while the non-elected group provides advice beforehand and carries them out afterwards. Each group ‘has distinctive rights, responsibilities and bases of legitimacy’ (Rhodes & Wanna 2007, p. 408).

These distinctive rights, responsibilities and bases of legitimacy have, in their turn, shaped the governance, conventions, and procedures that serve the Westminster system. Those governance arrangements, conventions, and procedures are not really, as the Scott and Baehler book regards them, ‘insider games’. They have rules that have grown out of a considered framework for the conduct of democratic government in a system involving the balanced interaction of elected politicians and appointed officials. Some of these rules may be too influential and some may not be influential enough, but between them they shape policy advising. They condition what evidence can be gathered, those from whom it can be gathered, what attention can and will be paid to that evidence, what options government has an interest in hearing about, whether any policy losers can be tolerated and who they are, how any policy options are formally communicated to government, how such policy options are considered by government, how they are communicated to the public and the parliament, and what time and resources are available to public servants to make all this happen.

Can policy advising
be fixed with a
tool kit?

Take the very simple case of public servants gathering data to serve as evidence for developing a policy. Gathering data can be costly, and is generally not initiated without consulting the portfolio minister. Alternatively, the minister himself or herself may initiate a data gathering exercise when, for example, a survey is wanted to demonstrate the value of a particular policy that a new government took to a recent election. In either case the minister would be asked to approve the expenditure. If the minister wants to be able to point to the independence of the data collection exercise, he or she is likely to direct public servants to commission an academic—often one from a pre-set departmental research panel who is known to sympathise with the government’s position. If the minister wants to ask leading questions that are not consistent with the usual academic standards for survey instruments, but are consistent with the government’s reasons for wanting the data collected, he or she may direct public servants to commission a private sector organisation that will conduct a survey that includes any question for which it is paid. If the minister wants to look balanced, he or she may also decide to engage an eminent person who supports a much more radical version of the policy, and ask that person to undertake a study that draws on existing or overseas data and advocates the more extreme view and, thereby, make the government’s position look moderate.

Once the evidence is gathered it could become apparent, for example, that the policy has been found to benefit men in the mining sector and adversely affect women in the retail and hospitality sectors. The department may provide an oral briefing to this effect, and the minister may then express an interest in hearing formally from the department about mining sector impacts only. Alternatively, the minister may be given a broad briefing but may instruct the department to publish only the mining sector impacts. In this case the publication would have to be a government publication and not a departmental one, as this kind of selectivity is not consistent with public service standards. Throughout, both the department’s actions and the minister’s instructions will take into account their common understanding that questions about the full extent and findings of the research may arise as part of a freedom of information request or in a Senate Estimates hearing, and what form and kind of advice is and is not exempt from such questioning.

These are the obvious features of the system that affect all policy advising; but there are others that are less obvious but widely acknowledged by public servants. They include inter- and intra-agency silos that inhibit ‘big picture’ thinking and stop up the flow of information necessary to underpin such thinking (Public Service Commissioner 2009, p. 121). They also include risk aversion. There is ‘a culture of under-rewarding success and over-penalising failure’ that tends to inhibit an expansive approach to policy development and that permeates not only public service accountability requirements but also agency performance management schemes (Australian Public Service Commission 2009a, pp. 17–18; 2009b, p. 3). Policy innovation can be expensive and very disruptive to the public, and for both these reasons is highly likely to attract the attention of such guardians of public accountability as audit offices, ombudsmen and parliamentary committees (not to mention the media). Where innovation fails, everyone is likely to know about it. Agencies can lose funding; senior managers can lose face; and the responsible officer is likely to lose both departmental standing and performance pay. Arguably, this contributes to a culture in which the game is not felt to be worth the candle.

Policy innovation
can be expensive andvery disruptive
to the public.

These are not simple one-off issues that can be resolved by prudent public servants. They are procedural realities that shape the day-to-day process of providing policy advice. Even pre-set formats affect what can be said. Cabinet submissions require, or in the past have required, a one-line summary of the impact of the proposal on the government’s favoured community or professional groups (and not on ‘everything and everyone that can affect or be affected by policy’). Some ministers do not want any brief that is longer than one page, regardless of the complexity of the matter under consideration. Many agencies’ briefs automatically incorporate a grid that invites the minister or a ministerial office adviser to rate the material provided on a scale of 0 to 4 for its quality, timeliness and relevance to the government’s policy directions, and provide the contact numbers of the public servant who prepared it. These features of the briefing format are likely to focus the minds of public servants on what the minister wants to hear as opposed to what the minister needs to hear.

If some features of the system resist ‘linear, multi-step models of problem solving’, others resist ‘network participatory’ activities like consultation. These include the speed at which reactive policy options have to be prepared (Public Service Commissioner 2009, p. 142) and the secrecy that surrounds the budget and other critical policy development processes. They also include the requirement to be responsive to the government’s political preferences. Following the 1996 election, the public service rarely consulted the unions about industrial relations change because the Howard Government wished to demonstrate the irrelevance of such ‘third parties’ to a system that would be based on individual bargaining (MacDermott 2008a, Ch 5). Scott and Baehler argue that ‘policy must consult the outside game … to retain legitimacy’ (p. 60) but in this case consulting the outside game would, in the government’s view, have undermined the legitimacy of the policy. In the Westminster system, consultation—however theoretically desirable—is not a form of voting. Voting has already taken place.


Improving the skill sets of individual public servants will not change the systems they have to negotiate as they exercise their skills. These systems have evolved to serve the operational realities of the government/administrative interface. They can be changed at the margins by the efforts of project teams and the study of toolkits, but they will only change substantively if there is change at political level. Government not only has to signal a desire for change—as the Rudd Government certainly did (Rudd 2008; Rudd 2009)—but it also has to underwrite that change. There would have to be money for additional staff so that there would be sufficient time for proactive rather than reactive policy work. There would have to be money for research, both ‘in house’ and commissioned. There would have to be time for training in research skills and, beneath that, a genuine interest by government in having relevant data collected and used transparently. There would also have to be a genuine interest in consultation (beyond consultation about details of implementation), and government would have to be willing to act, even where policies are unpopular, or have significant losers.

Systems will only change substantively if there is change at political level.

If the political will were there, the governance arrangements, conventions, and procedures that serve the Westminster system would adapt to increase the transparency, comprehensiveness and directness of strategic policy advising. If this were to occur, the theoretical models employed in the scholarly discussion of public policy and described by Scott and Baehler would be closer to those actually employed by public servants in their daily work. Since it is unlikely to occur, the process of improving strategic policy advising will have to make do with what can be gleaned from the teams, networks and toolkits proposed in the Blueprint for the Reform of Australian Government Administration. Both of the books considered here could make a contribution to make to these, particularly the Argyrous collection.

Does this mean that academics can only contribute to strategic policy advising at the margins? No: none of the preceding discussion applies to the contribution that academics have to make to strategic policy development though their subject matter expertise. It simply means that while public servants with an interest in such expertise would benefit from immersion in research institutions, academics with an interest in the policy advising process would benefit from immersion in the public service. Former prime minister, Kevin Rudd, recently argued that:

… for too long in Australia, thick walls have existed between places of research and learning, and places of policy making and implementation. Those thick walls do not enhance either the quality of public administration or the quality of academia (2009).

The Blueprint for the Reform of Australian Government Administration proposes several responses to this concern. In addition to establishing the usual project team, it calls for the settling of more formal policy networks with academic institutions, including the Australia and New Zealand School of Government, and think tanks such as the Lowy and Grattan Institutes. It is to be hoped that these stronger links include the cross-placement of public servants and academics.


Advisory Group on Reform of Australian Government Administration 2010, Ahead of the Game: Blueprint for the Reform of Australian Government Administration, Canberra [Online], Available: http://www.dpmc.gov.au/publications/aga_reform/aga_reform_blueprint/docs/APS_reform.pdf [2010, Jun 23].

Australian Public Service Commission 2009a, Delivering Performance and Accountability, Canberra [Online], Available: http://www.apsc.gov.au/publications09/performanceandaccountability.pdf [2010, Jun 23].

Australian Public Service Commission 2009b, The Senior Executive Service Census Survey: The Next Five Years, Australian Public Service Commission, Canberra.

Banks, G. 2009, Evidence-Based Policy Making: What Is It? How Do We Get It? ANU Public Lecture Series, presented by ANZSOG, 4 February, Productivity Commission, Canberra [Online], Available: http://www.pc.gov.au/__data/assets/pdf_file/0003/85836/20090204-evidence-based-policy.pdf [2010, Jul 8].

Briggs, L. 2005, A Passion for Policy? ANU Public Lecture Series, presented by ANZSOG, Canberra, 29 June [Online], Available: http://www.apsc.gov.au/media/briggs290605.htm [2010, Jul 7].

Senator John Faulkner, n.d. [Online], Available: http://senatorjohnfaulkner.com.au/file.php?file=/ministry_archives.html [2010, Jul 7].

MacDermott, K. 2008a, Whatever Happened to ‘Frank and Fearless’?: The Impact of New Public Management on the Australian Public Service, Canberra [Online], Available: http://epress.anu.edu.au/anzsog/frank_fearless/pdf/whole_book.pdf [2010, Jul 7].

MacDermott, K. 2008b, Marketing Government, Canberra [Online], Available: http://www.democraticaudit.anu.edu.au/papers/focussed_audits/200810macdermottmarktng.pdf [2010, Jul 7].

Manne, R. 2006, ‘The Nation Reviewed’, The Monthly, no. 12, May.

McGuinness, P. 2007, ‘Editorial: A Politicised Public Service?’, Quadrant, no. 93, April [Online]. Available: http://www.henrythornton.com/article.asp?article_id=4647 [2010, Jul 7].

Mulgan, R. 2007, ‘Truth in government and the politicization of public service advice’, Public Administration, vol. 85, no. 3, pp. 569–586.

Mulgan, R. 2010, ‘Selling credibility: How Rudd botched his advertising pitch’, The Public Sector Informant, July, p. 4.

Public Service Commissioner 2009, 200809 State of the Service Report, Australian Public Service Commission, Canberra, p. 121 [Online], Available: http://www.apsc.gov.au/stateoftheservice/index.html [2010, Jun 23].

Rhodes, R. & Wanna, J. 2007, ‘The limits to public value, or rescuing responsible governments from the platonic guardians’, Australian Journal of Public Administration, vol. 66, no.4, pp. 406–421.

Rudd, K. 2008, Address to Heads of Agencies and members of Senior Executive Service, Canberra, [Online], Available: http://www.apsc.gov.au/media/rudd300408.htm [2010, Jun 23].

Rudd, K. 2009, ‘John Paterson Oration’, Australia New Zealand School of Government Annual Conference, Canberra [Online], Available: http://news.anu.edu.au/?p=1638 [2010, Jun 23].

Wong, P. 2007, Labor’s approach to the Australian Public Service, Speech to the Institute of Public Administration Australia, p. 7.

Dr Kathy MacDermott has taught in universities in Australia and the United States and worked in the senior executive service of the Australian Public Service in the areas of industrial relations policy and public sector governance. Her most recent publications include Whatever Happened to ‘Frank and Fearless’? (2008, ANU EPress) for the Australia and New Zealand School of Government and Marketing Government for the Democratic Audit of Australia.