Excerpted from - University of California Press · 2015-11-27 · Barzelay_Text 11/7/00 12:50 PM...

Excerpted from

©2001 by the Regents of the University of California. All rights reserved.

May not be copied or reused without express written permission of the publisher.

click here toBUY THIS BOOK

http://www.ucpress.edu/books/pages/8969.html

CHAPTER SIX

CONCLUSION

New Public Management is a field of discussion largely aboutpolicy interventions within executive government. The charac-teristic instruments of such policy interventions are institutionalrules and organizational routines affecting expenditure planningand financial management, civil service and labor relations, pro-curement, organization and methods, and audit and evaluation.These instruments exercise pervasive influence over many kindsof decisions made within government. While they do not deter-mine the scope or programmatic content of governmental activ-ity, these government-wide institutional rules and organizationalroutines affect how government agencies are managed, operated,and overseen: they structure that part of the governmentalprocess usefully described as public management.1 In recentyears, political executives, central agency leaders, and legislatorsin numerous settings have demonstrated a sustained interest inpolicies affecting public management, the best-known cases ofwhich are the United Kingdom, New Zealand, and Australia.

156

Barzelay_Text 11/7/00 12:50 PM Page 156

NPM AND ITS PROBLEMS

New Public Management (NPM) is concerned with the system-atic analysis and management of public management policy. Thispolicy domain relates to all government-wide, centrally managedinstitutional rules and routines affecting the public managementprocess. For this reason, the domain encompasses multiple or-ganizations within government, including central agencies re-sponsible for budgeting, accounting, civil service and labor rela-tions, efficiency and quality, auditing, and evaluation. Systematicanalysis involves clear argumentation about the relationship be-tween context, goals, policy instruments, and choices. Systematicmanagement is a process of decision making that is both in-formed by analysis and well adapted to the political and organi-zational forces that shape decisions and their downstream ef-fects. NPM as a field of discussion is thus rooted in that ofsystematic management and policy analysis.

NPM’s Twin Elements

If NPM is a field of discussion about public management policy,it is important to be clear about its elements. This book proposestwo main elements, which policy-makers need to consider ifpublic management policy is to be placed on a sound footing.The first element focuses on the political and organizationalprocesses through which policy change takes place. Theseprocesses are influenced by a host of conditions, both institu-tional (such as the overall structure of the governmental systemand the specific organization of central administrative responsi-bilities) and noninstitutional (such as policy spillover and inter-ference effects). Policy dynamics can be analyzed in terms of

Conclusion / 157


specific mechanisms and patterns through which policy-makingprocesses operate. The key analytic issues linked to this elementof NPM discussion include estimating the feasibility of policychange and crafting lines of action to satisfy the situation-specific requirements of policy entrepreneurship.

The focus of the second element is the substantive analysis ofpublic management policy. This analysis concerns the advan-tages and disadvantages of various combinations of government-wide institutional rules and routines within specified contexts.Analysis is best regarded as a process of argumentation, for tworeasons. First, policy conclusions—even retrospective, evaluativeones—are supported by beliefs about government that are plau-sible rather than definitively true. Second, analysis takes place ina dialectical context where disagreement arises because of the va-riety of beliefs, expertise, and interests that are relevant to thechoice of management controls in government.

These broad categories—process and substance—give somestructure to the abstract conception of NPM proposed here.Thinking of NPM in these terms helps to focus inquiry on eachof two key issues of public policy analysis—feasibility and desir-ability—that are relevant for policy-makers. Focusing on theseanalytic issues also provides a more definitive context for discuss-ing methods for conducting research and argumentation on pub-lic management policy. This conception of NPM is superior forlearning from experience than was NPM’s initial formulation.

NPM’s Origins and Problems

To benefit fully from previous work on NPM, scholars and policy-makers need to be familiar with how this field of discussion

158 / Conclusion


has evolved. In sketching the evolution of the NPM discussion,it is useful to distinguish scholarship from both professionalcommentary and actual policy-making activity. Discussion tak-ing place within governments—for example, those taking placewithin the New Zealand Treasury in the 1980s and presented inits postelection briefing, Government Management—will be re-ferred to as NPM1. Professional commentary—exemplified byOsborne and Gaebler’s (1992) Reinventing Government and illus-trated by publications of the OECD Public Management Ser-vice—is denoted by NPM2. Finally, academic scholarship—thecategory to which nearly all works cited in the present volumebelong—is referred to as NPM3.

At the risk of oversimplification, I present the following ac-count of NPM’s evolution:

1. The concept of New Public Management originated inNPM3 (Hood and Jackson 1991; Hood 1991).

2. NPM was initially characterized as an internationaltrend. The essence of the trend was distilled from an array ofspecific ideas about management and government drawn fromNPM1 and NPM2 (Hood 1991). An influential accountidentified two paradigms of ideas: public choice and manageri-alism (Aucoin 1990).

3. The main empirical referents of the trend were theUnited Kingdom, Australia, and New Zealand in the 1980s.

4. The case of New Zealand (NPM1) acquired specialsignificance in both NPM2 and NPM3 for two main reasons.First, policy change took place across a wide range of distinctareas—expenditure planning, financial management, organiza-tion, civil service, and labor relations—within a single three-

Conclusion / 159


year parliamentary mandate (Boston et al. 1991). The NewZealand case thereby demonstrated even more clearly than theUnited Kingdom under Margaret Thatcher that public man-agement had become a policy domain. Second, the New Zea-land Treasury’s deliberations and policy arguments were framedin terms of economic theories of organization and government.This style of argumentation was highly unconventional in pub-lic management policy-making. The conjuncture of rapid com-prehensive change in public management policies and an un-conventional pattern of argumentation made the New Zealandcase (NPM1) especially noteworthy. Numerous scholars com-mented (NPM3) that economic theories of organization andgovernment (New Institutional Economics) constituted the in-tellectual foundations of New Public Management.2

5. In professional and academic discussion, countries wherepublic management policy change has been less than compre-hensive were labeled as “laggards” (Aucoin 1995).

6. The notion that the NPM is a widely applicable blue-print for the organizational design of the public sector is com-monplace in professional discussion (NPM2).

7. Some scholars in continental Europe argue that NPM is an Anglo-American model whose relevance outside its corecases is highly questionable.3

While points 4 through 7 are all worrisome, the last twopoints evince the most severe present limitations of the NPMfield of discussion. A blueprint approach to policy design ishighly questionable: the functioning of a given system of formalarrangements, such as management controls, depends on thecontext in which it operates. On the other hand, the equation of

160 / Conclusion


NPM with an Anglo-American approach to public managementpolicy is hardly a recipe for policy analysis and learning on an in-ternational scale. While these two extreme views about NPM areequally unsatisfactory, they grow out of the history of this field ofdiscourse. In particular, these limitations can be attributed to theinitial conception of NPM as a trend, centered on three similarcases.

AN EMERGING POLICY APPROACH TO PUBLIC MANAGEMENT

If the NPM field of discourse is to be more useful for practition-ers, a decisive turn away from its initial contours is required.Adopting a public policy approach to this subject constitutessuch a turn. The policy approach directs attention toward ex-plaining change in public management policy on a comparativebasis; it would also place a high value on rigorously argued eval-uative discussions of policy choices.

The policy approach can build on the accumulated strengthsof the NPM3 literature. As shown in chapter 2, scholars have ex-plored policy-making episodes and sequences in a number ofcases. This literature provides a starting point for explainingsimilarities and differences in public management policy changeacross cases. To accelerate research progress, it is advantageousto apply well-honed explanatory frameworks to such case evi-dence, as shown in chapter 3. Processual models of agenda set-ting and alternative generation explain, for instance, how ideasfrom economics and management contributed to policy changein the benchmark cases of NPM. Processual models also explainthe effects of interventions in the policy-making process by ex-

Conclusion / 161


ecutive authorities, such as prime ministers and top officers ofcentral agencies.

A policy approach can build on the evaluative literature dis-cussed in chapters 4 and 5 as well. These works provide com-mentary on policy choices, such as output budgets, accrual ac-counting systems, performance agreements with chiefexecutives, and division of executive government into myriadsingle-purpose organizational units. This recent developmentmoves beyond the initial conception of NPM as a trend by en-gaging in argumentation about the desirability of specific publicmanagement policies (institutional rules and organizationalroutines). Notably, this literature is yielding controversy overpublic management policies. For instance, Schick (1996) ex-presses strong reservations about specific aspects of New Zea-land’s institutional rules and organizational routines in the pub-lic management domain. In contrast, Aucoin (1995) offers fewdoubts about the desirability of public management policies inthe United Kingdom, Australia, and New Zealand. The basisfor such disagreements can be traced to the analytic frameworksemployed. One major difference lies in the specific strands ofmanagerial thought that the authors rely upon to evaluate pub-lic management practices.

The Policy Approach Exemplified

Schick’s study of New Zealand’s reforms serves as an exemplar ofpolicy-oriented, academically rigorous analysis within the NewPublic Management. In analyzing policy choices, the study pen-etrates through the fog of NPM themes (such as an outputorientation) and buzzwords (such as performance contracting).

162 / Conclusion


The object of evaluation is the system of management con-trols—broadly defined—operating throughout the departmentsof government. These controls are described in terms of theirroutine—as opposed to hypothetical—operation. Moreover, thedescription of the system of controls illuminates how organiza-tional factors—such as the cultures of central agencies—influence the evolving operationalization of public managementpolicies at the level of routines and transactions.

The standpoint from which Schick evaluated the observedsystem of controls is another facet of the study deserving seriousconsideration by practitioners and scholars alike. By standpoint,I mean general ideas that serve as a plausible basis for evaluatingor designing public management policies in particular settings.4

The ideas are drawn from bodies of thought and knowledge re-lated to both government and management. The thesis that gov-ernment is potentially an effective instrument of collective prob-lem solving, which plays a role in Schick’s critique of the patternof attention allocation at high levels of government and admin-istration, is embedded within the public philosophy of govern-ance sometimes referred to as Progressive Public Administration(PPA). A contemporary statement of this public philosophy iscontained in Moore’s Creating Public Value: Strategic Managementin Government (1995).

Schick draws ideas from two major schools of thought onmanagement: strategic management and management account-ing and control. These schools of thought are neither scientifictheories in the usual sense nor simply professional wisdom. Theyare well-developed traditions of argumentation about how com-plex organizations should be managed. Each school of thought isexpressed in a substantial professional literature, much of it

Conclusion / 163


written by academics. Indeed, it could be argued that each schoolof thought is a discipline within the field of management.

Incorporating Managerial Schools of Thought

Strategic management is centered on the executive function as awhole. This function includes the formation of strategies for or-ganizations at the business and corporate levels. In the privatesector context, a strategy is often conceived as a plan for achiev-ing sustainable competitive advantage. In government, a strategyis sometimes conceived as a pattern of decisions geared to creat-ing public value. Some strands of strategic management thoughtargue that strategies should be consciously formulated and vis-ibly endorsed by top decision makers. This approach also acceptsthe presumption that the effect of any given policy choice de-pends on other choices; in other words, policies are potentiallycomplements. The executive function, to be well performed, re-quires that a wide array of choices—for example, overall objec-tives, measures of merit for business processes, managementcontrol systems, and improvement plans—should be aligned.

Management accounting and control encompasses a largeproportion of the executive function—all but the making of fun-damental strategic decisions, on the one hand, and productionmanagement, in the narrowest sense of the term, on the other(Anthony 1965). Core interests of management accounting andcontrol include the systematic generation of nonfinancial ac-counting information, to be used internally for purposes of at-tention directing, decision making, and score keeping (Simon1954). A closely related interest is in the design and operation of

164 / Conclusion


management control systems, especially in the context of com-plex, decentralized organizations. Such control systems includebudgeting, cost accounting, reporting, and performance ap-praisal processes. Both scientific and practical developmentshave produced significant change in the field of management ac-counting and control over recent decades (Emmanuel and Otley1996; Kaplan and Cooper 1998).

Learning from Experience

From Schick’s theoretical standpoint, the first wave of reforms inNew Zealand earned a clean—if qualified—bill of health.Changes in public management policies provided a basis for im-proved performance planning, including requirements that min-isters and chief executives of government departments togetherformulate annual performance objectives for chief executives.The revamped budgetary process also provided several tools tosupport improved performance planning. These tools includeddescribing organizational activity in terms of “outputs” andadopting accrual accounting methods so that budgetary chargeswould reflect the rate of consumption of fixed assets. With thesepractices, budgets authorized the consumption of a given valueof resources to produce a certain type of output within aspecified time frame, and in this sense, budgets constituted per-formance plans.

Although these measures provided a platform for performanceplanning on a large scale, Schick identified two main limitations tothe technique. First, performance plans encoded in budgets weretypically formulated without the benefit of cost accounting infor-

Conclusion / 165


mation, which concerns the value of resources consumed in per-forming activities and/or producing outputs. Output budgetingand cost analysis, in other words, are not equivalent practices.Schick argued that output budgets are somewhat arbitrary per-formance plans, since they are not necessarily informed by sys-tematic measurement and analysis of the cost of government’sbusiness processes. This argument was rooted in the discipline ofmanagement accounting and control. Second, by focusing on out-puts, ministers and top officials paid insufficient attention to out-comes—that is, planning for policy accomplishments. This con-cern was primarily rooted in the PPA thesis that government ispotentially an instrument of collective betterment, reinforced bythe strategic management thesis that all organizational activity isideally geared to achieving strategic, corporate goals. Schick wenton to argue that the unbalanced attention to outputs was beingpartially redressed by the launching of a government-wide strate-gic planning process in the early 1990s.

TAKING NPM FORWARD

Schick’s study of New Zealand demonstrates that rigorous argu-ments about public management policies can be made on thebasis of what economists often refer to as the traditional man-agement literature. In this way, Schick casts doubt on the claimthat the New Institutional Economics (NIE) should be regardedas the intellectual foundation of New Public Management (seepoint 4, above). However, the question remains whether rigor-ous argumentation about public management policy can takeplace on a large scale without the benefit of a “paradigm” such asthe NIE.

166 / Conclusion


From Paradigmatic Coherence to Interdisciplinary Dialogue

The crucial property of the economic approach is that it pro-vides a basis for a coherent discussion of the full range of man-agement controls in government, including organizational struc-ture. The unifying force of the economic approach isattributable to the paradigmatic coherence of economics, gener-ally, and to the specific practice of modeling organizations as col-lections of individuals whose choices are determined by the in-centives they face. This conceptual framework ignores fields ofdiscourse that grew up around specific administrative functions,such as budgeting, personnel management, and evaluation. Afterall, management controls generate incentives irrespective of thespecific administrative function to which they primarily relate.This framework helped formulate the policy question of how tooptimize the whole system of incentives and controls operatingin government—a different question from how to improvebudgeting, personnel management, and evaluation as isolatedelements. The same framework provided a basis for answeringthis policy question as well.

The vitality of NPM as a field of policy research depends cru-cially on broadening its intellectual foundations beyond eco-nomic theories of organization while safeguarding the advan-tages of the economics approach. As a practical matter, thischallenge falls to specialists in public management based in thedistinct fields of public administration, management, and ac-counting. However, they face the inherent problems of conduct-ing an interdisciplinary policy dialogue. Such difficulties havebeen redressed to some extent in other policy fields where policyknowledge is drawn from several different academic disciplines

Conclusion / 167


and professional communities; they should not be insuperable inthe public management policy field, as I suggested in chapters 4and 5. If the challenge of interdisciplinary dialogue is not met,the economics approach to public management policy is likely topredominate.

The method used in chapters 4 and 5 to discuss arguments onNPM is a potentially valuable tool. In some ways, the method iselementary. One requirement is to be clear about the subject, na-ture, and scope of claims (see chapter 4). Do the claims relate topublic management policy or to the exercise of executive leader-ship in government? Are the claims theoretical, or are they par-ticular evaluations? Are theoretical claims meant to apply to alltypes of governmental systems, or just some? Another guidelineis borrowed from economics (as well as political theory): for pur-poses of effective communication, outline the structure of thearguments in a general way before specifying the details. A thirdguideline is to debug arguments that would otherwise founderon unacknowledged paradigm conflicts; this guideline appliesforcefully to ambitious arguments that draw together multiplefields of discourse, as demonstrated in chapters 4 and 5. By fol-lowing these guidelines, it may be possible for the NPM field ofdiscourse to (NPM2, NPM3) draw on a suitably wide range ofideas about government and management while sustaining thesort of coherent discussion of public management policy thatwas achieved in practice in New Zealand (NPM1).

Beyond the Anglo-American Model

This method for discussion has the additional advantage of ad-dressing the complaint, mentioned earlier, that NPM is the

168 / Conclusion


Anglo-American model of public management. If NPM is de-fined as a field of discussion, its essence lies in a commitment toserious argumentation, not to particular substantive foundationsor conclusions. The broad structure of a serious argument mightbe expressed as follows:

(1) Pi � A (Ei, Fi )(2) Ei � A (Si, Ti )(3) Ti � A (•), wherePi means policy proposals for a given country iEi means the evaluation of current policy in country iFi means the feasibility of policy change in country iSi means a survey of current policy in country iTi means the theoretical basis for assessing policy in country i(•) means the bodies of thought on which Ti draws

The complaint that NPM is an Anglo-American model mightbe interpreted as an assumption that (•) in expression (3) must bespecified in a particular way—for example, in terms of a particu-lar public philosophy of governance. However, an abstract con-ception of NPM means that a policy analyst simply needs to beclear about the basis for Ti.

The unit of argument represented by expression (3) resemblesthe broad approach taken in chapter 5 if (•) is specified as fol-lows:

(3�) Ti � A (PPG, MAN, KG), wherePPG means a public philosophy of governance MAN means schools of thought about managementKG means empirical knowledge about governmentalprocesses, including policy implementation

Conclusion / 169


All three terms can be specified or “customized” to fit the ana-lyst’s view about the context in which public management poli-cies are to be evaluated and devised. The term knowledge of gov-ernment (KG ), for instance, allows account to be taken ofspecificities in the structure of the governmental system underanalysis. Differences in governmental systems are pronounced,even within the so-called Anglo-American context (as betweenthe Westminster-type parliamentary and the U.S. separation-of-powers systems). The term PPG allows consideration of nationaltraditions of thought about government. For example, publicphilosophies of governance may differ substantially between in-stitutional states, sovereign states, and corporate-bargainingstates (March and Olsen 1989). Differences in the specificationof MAN, on the other hand, may be relatively modest, althoughit is well known that the specific content of management thoughtvaries among national settings (Guillén 1994). Therefore, theobjection that NPM is an Anglo-American model can be laid torest provided that NPM is conceptualized abstractly as a field ofdiscussion about policy interventions within government andprovided that high standards of argumentation are routinelypracticed.

From Isolated Case Studies to Comparative Research

A critical factor for enriching policy debates about NPM is todevelop and assimilate high-grade knowledge about the policy-making dynamics that drive the public management policy inno-vation. This knowledge can be developed through the compara-tive study of public management policy change. The immediatetask of such studies is to explain similarities and differences

170 / Conclusion


among cases. A logical way forward is for individual studies tocompare one or more cases with the NPM Benchmark Case—the composite of the United Kingdom, Australia, and New Zea-land cases developed in chapter 3. The specific results of re-search along these lines are limited historical generalizations,arrived at through comparative research methods. The field ofNPM will then possess an understanding of the causal effects ofsuch factors as policy images (such as systemic organizationalinefficiency), policy spillover and interference effects, policysubsystems, and policy entrepreneurship. Such an understandingcan enlighten analysis of the feasibility of policy change in par-ticular settings.

The major obstacle to policy research along these lines is per-haps the relative lack of interest in the subject of public manage-ment policy displayed by political scientists (for notable excep-tions, see Campbell and Halligan 1992; Schwartz 1994a, 1994b;Zifcak 1994; Hood 1996; and Kettl 1998). For their part, publicadministrationists have not dedicated themselves to explainingpolicy change in as systematic a manner as is required of a prac-ticing political scientist in the field of comparative politics/pub-lic policy. I hope that both political scientists and public admin-istrationists will be drawn to this line of research.

In sum, the fundamental limitation of scholarship on the NewPublic Management is that it has yet to become a vital area ofpolicy research. This book has sought to demonstrate that mak-ing a decisive turn toward a policy approach is desirable and fea-sible, at least in the area of public management policy. To adoptthe policy approach is a way for scholars to contribute to the policy-making process while also strengthening the intellectualtradition that Aaron Wildavsky did so much to shape.

Conclusion / 171

Barzelay_Text 11/7/00 12:50 PM 1

Excerpted from - University of California Press · 2015-11-27 · Barzelay_Text 11/7/00 12:50 PM...

Documents

Transcript of Excerpted from - University of California Press · 2015-11-27 · Barzelay_Text 11/7/00 12:50 PM...