Read an Excerpt
Narrative Policy Analysis
Theory and Practice
By Emery Roe Duke University Press
Copyright © 1994 Duke University Press
All rights reserved.
ISBN: 978-0-8223-8189-1
CHAPTER 1
Deconstructing Budgets, Reconstructing Budgeting: Contemporary Literary Theory and Public Policy in Action
Text and reading. What, you might well ask, do these have to do with making policy and its analysis more useful? The answer: Text and reading are core to contemporary literary theory's focus on the narrative, and this theory and focus prove immensely helpful in addressing the major policy issues of our day. To see how, we consider one of those issues that confronts readers in the United States and elsewhere, namely, the worldwide disarray in national budgeting systems.
This chapter is a first step toward understanding chapter 2's discussion of what a policy narrative is. I show that even a rudimentary understanding of contemporary theory can improve our understanding and making of public policy.
The Problem
It is not just the so-called Third World countries that cannot cope budgetarily, but the United States as well. We here in the States have had an increasingly hard time keeping to the federal budget. "How we Americans used to deride the 'banana republics' of the world for their 'repetitive budgeting' under which the budget was reallocated many times during the year, until it became hardly recognizable.... Yet resolutions that continue last year's funding for agencies, for want of ability to agree on this year's, are becoming a way of life in the United States," as Aaron Wildavsky put the problem a decade ago. Since then, Wildavsky has argued that continuously remaking the national budget "has now become standard practice in relatively rich nations."
Descriptions of national budgeting systems occur again and again in the work of Aaron Wildavsky, particularly The Politics of the Budgetary Process, Budgeting: A Comparative Theory of Budgetary Processes, and Naomi Caiden and Wildavsky's Planning and Budgeting in Poor Countries. These documents are regarded as the most influential post–World War II books on U.S. budgeting and comparative budgeting generally and provided some of the most recent information we have had on national budgeting systems. I have distilled seven major features of current national budgets from this material. After discussing these features, I draw out some of their important implications. The seven features underscore why current budgetary practice has a great deal in common with contemporary literary practice, in particular, deconstruction.
A qualification before proceeding further. What follows is not a comprehensive review of the literature on government budgeting. What is reviewed scarcely touches on the project implementation and program evaluation literatures. The sole aim of the next section is to distill features of budgeting that reproduce concerns in a field we normally in no way associate with budgeting, namely, literary theory.
Key Features of National Budgets
Budgets are, first of all, texts. Usually budgetary figures are "embodied in a document that may be called the budget, but the budget is much more than that. It is the outcome of a process," comments one expert. Yet for all the accent on budgetary process and politics in the political science literature, the budget as written text is virtually always the starting point in that literature. Almost unavoidably, the first chapter of The Politics of the Budgetary Process begins, "In the most literal sense, a budget is a document, containing words and figures, which proposes expenditures on certain items." So central is the printed budget to our understanding of what government is all about that the inability to publish a national budget is easily one of the best measures we have of a government whose very existence is under threat. Politically plagued Angola, for example, apparently did not publish a budget for years.
National budgeting extends, of course, beyond the written word. Budgeting is a way to set priorities, a mechanism for expenditure control, a means of staff coordination, and more. Still, the printed budget and its documentation remain core even to these other efforts. The national budget is, in reality, often not just one published text but many. The Government of Kenya's budget, for instance, has been a set of publications covering annually its development estimates, recurrent estimates, supplementary estimates, ministerial budget speech, revenue estimates, and survey of current and future economic conditions. The U.S. federal budget has, in contrast, summarized many of these same topics in two documents, the budget and its appendix. In addition, each of the conventional stages of the budget process—compilation, approval, execution, and audit—requires its own documentation. The more or less dispersed nature of the printed budget within and over its various stages and functions has profound implications for how closely budgets, money, and power are related, as we shall see.
Budget texts are increasingly fictional National budgets are notorious for trying, by way of figures and statistics, to simplify, quantify, and commodity into commensurable units a reality that revolts against such reductionism. The fictional character of national budgets also derives from areas other than this inherent problem of using numbers on the page to refer to things out there. For Caiden and Wildavsky,
the failure of budgets to have predictive value, that is, to calculate expected national income accurately, to relate expenditures to it over the year, to allocate these resources to various purposes, and to have them spent as authorized is a noteworthy phenomenon in many poor countries. To speak of a budget as "a great lie" ... is sometimes no exaggeration.
The description still holds for many Third World countries and now presents a fair picture of parts of the U.S. federal budget as well. The more written texts there are to the national budget, the more the sense of fiction amplifies this sense not of lies or deliberate distortion as much as drastic simplification and inherent nonreferentiality. For instance, substantial differences have existed between what was printed in the five-year National Development Plan of the Government of Kenya (GoK) and what the GoK eventually budgeted in its three-year forward budget, between that printed forward budget and what was eventually budgeted annually in the published estimates, between what the ministries formally requested to have budgeted in the annual estimates and what was finally allocated them by way of official treasury warrants, between that allocation and the documented funds made available to be spent, and between what was available and what audit reports subsequently show was in fact spent.
Budget texts have no author. Casual observers would likely say a national budget is produced by anonymous decision makers on behalf of "government." More fundamentally, though, a national budget has no author. Budgetary priorities are often implicit and emerge from the push and pull of decision making and bargaining. "The budget that emerges from this process is nobody's ordering of preferences," as one long-time observer put it. "It does not conform in all its particulars to any [Congress] Member's comparative preferences, let alone those of a congressional majority." The same point is made by Wildavsky: There "is no one person, the president and congressional leaders included, who is charged with the task of dealing with the 'budget as a whole' and who is able to enforce his preferences." "After a number of entitlements and tax expenditures have been decided upon at different times, usually without full awareness of the others, implicit priorities are produced ipso facto [in the budget], untouched, as it were, by human hands." The budgets we are talking about, in other words, have many narrators, but few, if any, authors.
The texts are open to (mis)reading only. There is probably no better or more universal example of the inherent conflict between readerly and writerly versions of a text, to use Roland Barthes' distinction, than national budgets. According to Barthes, readerly texts are those that call for a reader who is "serious," that is, one who is "intransitive ... [and] left with no more than the poor freedom either to accept or reject the text"; writerly texts, on the other hand, require a reader who is "no longer a consumer, but a producer of the text." When it comes to national budgets, a government's chief executive or treasury would prefer their world to be full of readerly (take-it-or-leave-it) budgets. From the treasury's perspective, its budgets should be unquestioned and unchanged by the audience for which they are intended, namely, civil servants whose task it is to treat budgets seriously by spending only what is budgeted on only those activities budgeted. Such budgets are, in short, "authorized" texts.
In practice, many governments produce writerly (always revisable) versions of a national budget. A host of government departments and decision makers rewrite the budget once it is published, best evidenced by all those "unauthorized" operations found across rich and poor nations alike, for example, widespread misappropriations, underexpenditures, overexpenditures, uncontrollable expenditures, and failed implementation (more about budget-induced implementation failure in chapter 2). Much of this revision is carried out unofficially and in the form of reinterpretation, much is officially sanctioned and in the form of amendment. Both forms go by the name repetitive budgeting. Under this budgetary practice, the "budget is not made once and for all when estimates are submitted and approved. Rather, as the process of budgeting is repeated, it is made and remade over the course of the year.... The entire budget is treated as if each item were supplemental, subject to renegotiation at the last minute.... Its most extreme manifestation is as cash-flow budgeting where changes may be made from day to day or even from one hour to the next."
Governments are understandably reluctant to admit how widespread repetitive budgeting is, but one candid admission comes from the Government of Kenya in its Report and Recommendations of the Working Party on Government Expenditures:
Finally, but perhaps most damaging to the effectiveness of Government, many policies agreed by Cabinet have been unnecessarily delayed or distorted during implementation. An examination of the policy chapter in the current Development Plan reveals that little work has yet been done to implement about half of the Cabinet decisions recorded there. Delays in implementation often cost Government millions.... A careful review of the Reports of the Controller and Auditor-General reveals a systematic tendency in many ministries of ignoring financial regulations and instructions from the Treasury. Indeed, the proliferation of Treasury circulars on many aspects of financial management and responsibility testifies to the breakdown in management in the Public Service.
Note the reference to the "proliferation" of budget documents. Each document in this proliferation is itself a reminder of the aforementioned gap between what starts on paper and what ends in practice. The more documents there are, the greater the gap seems, and the greater the sense of breakdown in public management under the writerly revision of national budgets.
Budgets are by definition intertextual One year's budget infiltrates and is infiltrated by earlier budgets. Vincent Leitch, in his early book on poststructuralist criticism, put the matter generally:
The predecessor-texts themselves operate intertextually, meaning that no first, pure, or original text ever can or did rule over or delimit the historical oscillations at play in texts. Thus all texts appear doubled: they are uncontrollably permeated with previous texts.
The subject of intertextuality is taken up in chapters 6 and 7. Here note only that national budgets are interwoven with many documents in the same manner as Leitch describes. They interpenetrate each other not only because a given annual budget is itself a set of interconnected documents frequently referring to and depending upon the others for that year. Equally important, budgets are publications produced incrementally in a series over time, where in many cases this year's proposed figures are contrasted on the same page with last year's figures for the same item. A national budget is really one long, multiyear compendium, where each new annual budget gives its readers the opportunity to reinterpret it as well as preceding annual budgets in light of each other.
The terms within the text define each other. Budgets live and breathe semiotic contraposition. Each and every item budgeted defines and is defined by all other items through the budget total. More money for x means less money for y, if the two items are to equal the budget total, z. In the past, this contraposition of x being defined by what it is not, namely, y, has been reinforced by a wider contraposition between budget guardians (those conservers of the budget such as the treasury) and budget advocates (those spenders of the budget whose expenditures the treasury was meant to control):
[A]gencies are advocates of their own expenditures, not guardians of the nation's purse.... And the job of the finance ministry is to see that they don't get it.... Each role implies its opposite, guardianship expects advocacy to provide a choice among items to cut, and advocacy needs guardianship to supply at least tacit limits within which to maneuver.... [These] spending-saving roles are one of the constants of budgeting.
Less constant today, however, under repetitive budgeting. As Wildavsky and his colleagues have made clear, worldwide inflation, recession, and pervasive economic uncertainty have over time fueled the dispersal and erosion of the underlying trust, expectations, and rules of the game that once made budgets predictable, readerly texts.
As a result, a profound gap exists today between what we thought the budget text was and what it now is. Many still expect a national budget to have figures that are more or less accurate, priorities that are more or less deliberately chosen, and an implementation that is more or less as stated. Yet, the reality is that national budgets have been and are now being rewritten in a writerly fashion all over the place. Repetitive budgeting is not merely contrary to the expectation of many, it is at every point orthogonal to that expectation. Published figures, priorities, and plans have no stable interpretation and mean nothing apart from their readers' rewriting, the only constant of this revision being its repetitive nature.
Policy Implications of Repetitive Budgeting
Where do the seven features get us? If the budgeting literature is correct in identifying these features, then repetitive budgeting is a deconstructive practice that treats budgets primarily as texts whose reading is always open-ended, if not undecidable. The political and institutional implications are considerable.
Budgets, Power, and Income. Because repetitive budgeting is an unstable process, the conventional link between money and power is drastically altered. True, the older syllogism still holds: money is power, budgets are money, so budgets represent power. It is realized in a vastly different way than before, however. The power we see in repetitive budgeting is that of different budget (re)writers at different times and in different ways remaking the budget, in whole and in part. Budgetary power is dispersed and fragmented, and the ability to make or influence decisions—that is, to interpret action and act upon such interpretations—is spread through the budget process. Yes, material interests matter, but they matter all over the place and in many competing ways.
Moreover, this interpretative power is dispersed precisely in terms familiar to deconstructionists. Contrary to received wisdom, there is no Grand Author articulating national budgets. Instead, budget texts are multiple, intertextual, and constantly revised and altered by their readers. Just as the reader has a major say in what the literary text means, so too for readers of the budget text in repetitive budgeting. In repetitive budgeting, this power of readers to "reinterpret" is quite literally, as the saying has it, "good as money."
(Continues...)
Excerpted from Narrative Policy Analysis by Emery Roe. Copyright © 1994 Duke University Press. Excerpted by permission of Duke University Press.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
