Environment

30 views

A Review of:�Politics, Policy, and Organizations: Frontiers in the Study of Bureaucracy, edited by George A. Krause and Kenneth J. Meier�

This is a valuable book for several reasons�perhaps most importantly because of the exemplars it provides showing the use and power of state-of-the-art formal models in public management. It is less valuable than it might be, however, because its
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Share
Tags
Transcript
  BOOK REVIEW POLITICS, POLICY, AND ORGANIZATIONS: FRONTIERS IN THE STUDY OF BUREAUCRACY, EDITED BYGEORGE A. KRAUSE AND KENNETH J. MEIER FRED THOMPSON WILLAMETTE UNIVERSITY  Politics, Policy, and Organizations : Frontiers in the Study of Bureaucracy . George A.Krause and Kenneth J. Meier (editors), Ann Arbor: University of Michigan Press,2003, 352 pages.The group on the scientific study of bureaucracy (S 2 OB) workshop at the fifthbiennial conference of the Public Management Research Association in 1999 pro-duced this volume. It includes a preface, an overview, srcinal articles, a conclusionby the editors, consolidated references, contributor information, author index,and subject index. The srcinal articles are organized into three sections: theory,methodological technology, and empirical studies. The contributors are DanielP. Carpenter, Krause; Thomas H. Hammond; David B. Spence; John Brehm, ScottGates, and Brad Gomez; Andrew B. Whitford; Steven J. Balla and John R. Wright;Lael R. Kaiser; Kevin Corder; Kevin B. Smith; and Michael J, Licari. The authorindex expresses the flavor of this volume nicely. Aside from the contributors, themost frequently cited authorities are Terry Moe, Herbert Simon, Barry Weingast,James Q. Wilson, and B. Dan Wood.This is a valuable book for several reasons—perhaps most importantly becauseof the exemplars it provides showing the use and power of state-of-the-art formalmodels in public management. It is less valuable than it might be, however,because its contributors are American political scientists and the models tendto reflect their idiosyncratic concerns. As an academic field, public management is lar-gely concerned with a set of doctrinal issues: What is the role of public managers?What should be the design of a programmatic organization? How should governmentoperations be managed? What public management policies (e.g., financial manage-ment, human resource management, procurement, etc.) should be chosen? 1 For themost part, these issues are given short shrift in this volume.Carpenter’s article, ‘‘Why Do Bureaucrats Delay? Lessons from a Stochastic Opti-mal Stopping Model of Agency Timing, with Applications to the FDA,’’ beautifully International Public Management Journal  International Public Management Journal, 8(3), pages 439–445 Copyright # 2005 Taylor & FrancisAll rights of reproduction in any form reserved. ISSN: 1096-7494  illustrates both the value and the foibles of this volume. Carpenter constructs a soph-isticated learning model to explain bureaucratic delay. Analysts have tended to treatthis phenomenon as a queuing problem. One might call this approach the LieutenantKije model of bureaucratic delay, in which a number of harried officials must pro-cess an application, for a license or permit say, in a particular order, which mayor may not be well understood by the participants. Under this approach, eachparticipant processes the application on either a first-in first-out or a last-in first-out basis, and passes applications that raise unfamiliar or difficult issues to anotherofficial, perhaps a supervisor, for resolution. The consequences of such a process area very high ratio of cycle time to processing time, high mean delays, and considerablevariance. Carpenter’s model produces the last two outcomes, but it presumes a farmore sensible process.Carpenter predicts time to approval of new drugs by the United States Food andDrug Administration (FDA). He treats this decision as a real option 2 in which theFDA knows the benefits of approval (expressed in terms of the drug’s efficacy,the size of the afflicted population, and the availability of substitutes, plus the polit-ical influence of the potential users and of the drug company submitting the appli-cation), but not its costs (expressed in terms the drug’s safety and, thereby, thethreat error poses to FDA’s reputation).Uncertainty can be expressed as a probability distribution. Complete ignorance,for example, implies that every possible outcome is equally likely. Analysts pricingfinancial options usually model future asset values using normal, triangular, orbinomial distributions, depending upon the fit to historical data. In this case, how-ever, costs are truncated at zero. Consequently, Carpenter assumes that safety is astandard Brownian motion or a continuous-time Wiener process, a statistical distri-bution that exhibits intermittent exponential growth and decay, with an absorbingbarrier at zero, which is appropriate given a concern with adverse drug reactions.Carpenter further assumes that time and repeated drug trials will produce better esti-mates of cost, i.e., they will reduce the FDA’s uncertainty about a drug’s safety,which is to say the FDA will learn. Finally, the value Carpenter assigns to the FDA’saversion to adverse drug reactions drives the model’s optimal stopping point.Implicitly, Carpenter says that the FDA will (should) approve a new drug whenthe payoff to doing so is equal to or greater than the expected costs of approval plus the value to the FDA of further review.Reading this model is worth the price of the book. It is a revelation. Moreover,contrary to Carpenter’s claim (31), this is a typical problem of administrative choice.Many, if not most, important choices involve real options. Only projects that arenow-or-never decisions, or projects that can be undone without cost, resolve to asimple question of (risk-adjusted) expected value. All others require a further test:Will the project have greater value in the future than it has now? The answer to thisquestion will usually depend upon whether time will resolve some uncertainty aboutthe benefits or costs of the project. In other words, many organizational processescan and probably should be modeled as learning problems (employ real optionanalysis). 440 International Public Management Journal Vol. 8, No. 3, 2005  On the other hand, I am not convinced that the addition of public choicevariables—the political influence of potential users or drug companies or even thereputation of the FDA—contributes much of value. As a first cut, I would takethe FDA’s charge seriously: insuring the safety and efficacy of pharmaceuticals.Both can be measured in terms of quality-adjusted life years (QALY), a measure thatcombines health status and years of health.Using QALYs to measure safety and efficacy would have the analytical advantageof applying the same metric for benefits and costs. (Presumably the latter would haveto be weighted to reflect FDA’s emphasis on safety.) Having done this analysis, Imight then be prepared to ask the secondary question of how much political influ-ence adds to our understanding of the timing of FDA approvals or, perhaps, whatdoes the timing of FDA approvals tell us about political influence. These questionshave some intellectual bite. Nevertheless, putting them first is a case of the tailwagging the dog.Moreover, many of the chapters in this book are even more idiosyncratic in theirperspective than Carpenter’s. The theme that runs through the book is congressionalcontrol of executive branch agencies. 3 This is a uniquely American obsession. Of course, the volume has other themes, including several of interest to students of pub-lic management. In their conclusion, ‘‘An Agenda for the Scientific Study of Bureaucracy,’’ Meier and Krause suggest that we would progress faster if we focusedour attention on five areas: ‘‘getting inside the black box of bureaucracy, takingtheory more seriously, exploiting the variance in bureaucracies, making an explicitreturn to the structural aspects of bureaucracy, and adopting methods that corre-spond to the data generating processes of bureaucracy’’ (293). They further callfor a ‘‘new organizations approach’’ to the study of administrative organizations,one that is rigorous, logical, analytical, and statistical, but applied to timeless issuesof administrative organization. Sure! But, what’s new? How is this agenda differentfrom the one that academics publishing in journals such as Administrative ScienceQuarterly , Academy of Management Review , and Academy of Management Journal  have been pursuing all along?For example, ‘‘Donut Shops, Speed Traps, and Paperwork: Supervision and theAllocation of Time to Bureaucratic Tasks’’ by Brehm, Gates, and Gomez, andWhitford’s ‘‘Adapting Agencies: Competition, Imitation, and Punishment in theDesign of Bureaucratic Performance’’ would both look smart in AdministrativeScience Quarterly . Meier and Krause emphasize their methodological contributions.Brehm, Gates, and Gomez model the allocation of a fixed total amount of laboreffort as a Dirichlet process, a conjugate form of the multinomial distribution thatestimates the probability density of a particular set of discrete frequencies p 1 , . . . ,p n ,where the frequencies sum to 1. To say the very least, this is an intelligent, albeitrisky, solution to a difficult compositional data problem. Whitford uses computersimulation to model organizations as complex adaptive systems. Computer simula-tion is not a new approach to organizational analysis, but arguably, we finally havethe tools we need to use it well—both of these articles do so. Lacking the kinds of game-theoretic models needed to pursue fully the rational-choice agenda with BOOK REVIEW  441  respect to organizational design (Gibbons 2003), perhaps the time has come toreconsider alternatives. Among the most attractive of these are the organizationalprocess models elaborated by scholars in the 1960s and 1970s: Herbert Simon, JohnPatrick Crecine, John Padgett, and others. They deserve a second look and, for thatlook, computer simulation is the tool of choice (Bender 1990, 2003; Green andThompson 2001; Jones 2002, 2003).However, interesting as these articles are for their cutting-edge methodologicaltools, they are, perhaps, even more so for what they say about the design of programmatic organizations. Brehm, Gates, and Gomez show how people,organizations, and tasks interact to create and sustain the kinds of purposeful effortsand cooperative relationships needed for organizations to outperform markets (157-8). They conclude that functional organizations characterized by a high degree of task specialization should select employees with the right mix of functional prefer-ences, and design contracts and remuneration schemes that sufficiently reward themfor performing intrinsically unsatisfying tasks. In contrast, organizations designedaround interdisciplinary teams must recruit employees who have heterogeneous pre-ferences. Mutual observability and supervision are also important, but the lattershould be focused on the employees who are most amenable to supervision.Whitford’s findings are directly relevant to performance measurement,benchmarking, and collaboration. He models a set of organizations in which goalsetting, search and innovation, and linking performance and search are the keysto increasing performance, and then subjects them to three levers of redirection:comparison, punishment, and imitation. He concludes that competition andcomparison will better the average performance of these organizations, but thisachievement has two costs (182). One is short term instability (even the best perfor-mers will experience oscillations). The other is that some organizations may lag everfurther behind performance leaders.The articles in the last section of the volume rely on standard research methodsand generally arrive at unexceptionable conclusions. In ‘‘Consensual RuleMaking and the Time it takes to Make Rules,’’ Balla and Wright find that bothagency and rule characteristics influence the time elapsed between publication of aproposed rule and its promulgation. Deadlines, fewer rules at the last juncture of therule-making process, and agency independence all result in swifter promulgation of final rules, but not consensual rule making.Keiser explains ‘‘Why it Matters Whether State Bureaucrats as Opposed to Fed-eral Bureaucrats Administer Federal Programs.’’ The answer is that local and stateofficials are not exclusively concerned about compliance with national regulations,and they may also take the costs to the local unit and the benefits to its residents intoaccount. Actually, aside from the effect of a moralistic local culture, the estimatedcoefficients reported in this article seem surprisingly small. It may be there just isn’tvery much variance that can be explained by pursuit of local interests, because,perhaps, officials in all states rationally pursue approximately the same fiscal andlocal benefit objectives. It is hard to tell. I found Keiser’s descriptions of data used,derivation of specifications tested, and analysis performed somewhat cryptic, evenopaque. 442 International Public Management Journal Vol. 8, No. 3, 2005  According to Korder, in ‘‘Structural Choice and Political Control of Bureaucracy:Updating Federal Credit Programs,’’ ceilings and funding sources affect the growthof these programs, congressional commitments are subject to revision, and the choiceof policy instrument—direct loans, loan guarantees, or insurance—matters. Loanguarantees, for example, appear to be especially responsive to political control byelected officials. Moreover, Congress frequently reorganizes and updates these pro-grams to amend eligibility, terms, and instruments and, thereby, who gets what,when and how.In the antepenultimate article, ‘‘Administrative Structure and Social DemocraticResults: The Case of Education,’’ Smith demonstrates that the governance structureof schools (public or private), although not the degree of discretionary authorityvested in teachers, affects the objectives they pursue. Private school teachers aretwice as likely to stress academic excellence over basic literacy skills as public schoolteachers and a whopping 22 times more likely to rank moral = religious indoctrinationahead of basic literacy. Smith concludes that a likely consequence of school privati-zation, especially schemes featuring portable vouchers, would be increased servicedifferentiation, with most institutions oriented to specific market segments and seg-regated along cultural, ethnic, religious, or social class lines. Finally, Licari describesan interesting case of ‘‘Bureaucratic Discretion and Regulatory Success withoutEnforcement,’’ that suggests that regulators may sometimes be given more creditfor programmatic compliance than they earn.Judged by its provenance and aims, the only real weakness of this volume is onethat Meier and Krauss frankly acknowledge: ‘‘Bureaucracy scholars overstudy fed-eral regulatory agencies. . . . Federal regulatory agencies, however, have characteris-tics distinctly different from those of other bureaucracies. . . . This overfocus onregulation calls into question how general the findings in the literature are.’’ (301)They suggest two solutions: greater attention to public agencies at the state and locallevel, especially schools, and additional cross-sectional, cross-national research. Toboth suggestions, particularly the second, I can only say, amen. I would also notethat, in my opinion, the articles in the empirical section of the volume tend to usetoo many words and not enough equations to explain = derive the models theyestimate.Finally, reading these papers, especially Carpenter’s and Krause’s, 4 reminds meof a minor but longstanding source of irritation. I wish that the authors of articlesmaking rigorous formal arguments would provide tables listing all the notationalsymbols they use in the order of their appearance and that editors would imposecommon notational usage and definitions throughout the volumes they edit. Nowthat I no longer edit this journal, I think journal editors should do likewise. NOTES 1. Michael Barzelay (2001) formulated this statement of the public management agenda. Itdistinguishes public management from the academic fields of public administration, publicpolicy, political science, and public economics and finance, which it intersects, based in parton its greater overlap with the academic fields of strategic management, business process BOOK REVIEW  443
Advertisement
Related Documents
View more
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks