abstract. A large body of literature in administrative law discusses presidential control of executive agencies through centralized review of regulations in the Office of Information and Regulatory Affairs (OIRA), part of the White House’s Office of Management and Budget (OMB). Largely overlooked in this literature is how the President’s budget acts as a source of agency policy control—in particular, how the White House exercises control through OMB’s authority to prepare the budget, oversee agencies’ execution of the budget, and create and implement management initiatives through the budget process. This Article identifies seven levers associated with OMB’s work on budget preparation, budget execution, and management and shows how these levers can control agency policymaking. These levers have some salutary aspects, especially in their valuable coordination work throughout the administrative state, but they also raise a series of accountability concerns related to opacity, the extensive discretion afforded to civil servants and lower-level political appointees, and the potential for substantive policy (and political) choices to be obscured by technocratic-sounding work. The Article concludes with a reform agenda, mapping out ways that the President, OMB, Congress, and civil society should respond to these accountability problems. Future analyses of OIRA’s authority should incorporate discussion of the complementary power of OMB to use the budget as a source of agency policy control.

author. Associate Professor, Georgetown University Law Center. For helpful conversations, comments, and critiques at different stages of this project, I thank Judy Appelbaum, Bob Bullock, Bill Buzbee, Dan Chenok, Martha Coven, Dan Ernst, Nora Gordon, Robert Gordon, Seth Harris, Lisa Heinzerling, John Hudak, Melissa Junge, Sheara Krvaric, Rich Loeb, Tom Lue, Sunil Mansukhani, Katherine McFate, Adam Neufeld, Jennifer Nou, Kathy Peroff, Aaron Saiger, Phil Schrag, Mark Seidenfeld, Miriam Seifter, Jason Snyder, David Super, Michael Vandenbergh, David Vladeck, and Tim Westmoreland. I am also grateful to current and former OMB and agency officials and staff members at civil society organizations who provided me with necessary background and/or reviewed drafts, as well as to several anonymous reviewers for the Yale Law Journal, all of whom provided helpful insight and feedback. I also benefitted from exchanges during faculty workshops at the Georgetown University Law Center and the University of Colorado Law School. For helpful research assistance, I thank Amanda McGinn, Georgetown Law ’16; Michelle Willauer, Georgetown Law ’16; Felycia Itza, Georgetown Law ’17; Shanna Holako, Georgetown Law ’17; and Morgan Stoddard and her team at the Georgetown Law Library. I also thank Monica Martinez and Angie Villarreal for ongoing administrative support. This Article represents my own views, and any errors remain entirely my own.

Introduction

One of the secrets only the initiated know is that those who labor here [at the Office of Management and Budget] for long do so because the numbers are the keys to the doors of everything. Spending for the arts, the sciences, foreign policy and defense, health and welfare, education, agriculture, the environment, everything—and revenues from every source—all are reflected, recorded, and battled over—in numbers. And the sums of the numbers produce fiscal and monetary policy. If it matters—there are numbers that define it. And if you are responsible for advising the president about numbers, you are—de facto—in the stream of every policy decision made by the federal government.

— Paul O’Neill, Former Deputy Director of OMB.1

Scholarship on administrative law is replete with analysis of presidential control of executive agencies through centralized review of regulations in the Office of Information and Regulatory Affairs (OIRA), part of the White House’s Office of Management and Budget (OMB). While the literature is sharply divided as to whether OIRA’s control is salutary or dangerous,2 the literature largely shares an underlying framework within which the subject matter is discussed: it tends to focus on regulations as the primary policy lever through which OMB affects agencies’ policy choices.3

This portrayal of OMB as an institution for asserting presidential control over the administrative state is incomplete. Reviewing regulations is not the only policy lever OMB has to control executive agencies’ policy choices. In fact, it may not even be the main one. The budget itself—the core reason for OMB’s existence4—is a key tool for controlling agencies.5 Yet the mechanisms of control through the executive budget process remain little discussed and insufficiently understood.

This Article seeks to expand the view of centralized control of the administrative state by describing, categorizing, and analyzing the operations surrounding the President’s budget. It maps out the legal documents that govern this work—some statutes, but primarily documents produced by OMB and the White House more generally—as well as the OMB offices and personnel behind this work. These sources help to explain the mechanisms and processes by which OMB uses the budget to get “in the stream of every policy decision made by the federal government.”6

The Article advances three kinds of arguments: descriptive, normative, and prescriptive. The core descriptive claim is that understanding OMB’s budget operations is fundamental to understanding centralized control of agencies’ decision making because OMB’s work on the budget has important policymaking effects. This insight provides a new perspective on the federal budget process. Much writing on the budget process focuses solely on legislative procedures and general fiscal policy, attending very little to the executive’s role.7 When the administrative law literature discusses the budget, it tends to do so through the lens of institutional battles between Congress and the President rather than by examining the budget as a method through which the White House can control agencies’ policymaking.8 When the literature does discuss the intra-executive role of the budget, it tends to focus on blunt tools and discrete moments in time: the President’s ability to propose the funding levels and associated policy choices that Congress acts on,9 to “recommend budget cuts for agencies that fail to follow administration preferences (and budget increases for those that comply),”10 and ultimately to veto appropriations legislation not to his liking.11

This Article expands this view of the intra-executive budget process, arguing instead that OMB’s budget work serves as a regularized and pervasive form of agency control. For each component of the budget process in the executive branch—the preparation of the President’s budget, the execution of the budget that Congress eventually passes and the President signs, and the implementation of presidential management initiatives that are embedded in the budget—this Article identifies and names levers that function as a form of policy control. In preparing the budget, OMB uses the form-and-content lever to tell agencies what to put in their budget requests to OMB in the first instance, the approval lever to require that the substance of agency budget requests passes muster with OMB, and the confidentiality lever to direct agencies to remain silent about any policy preference that may differ from what the President’s budget ultimately presents to Congress.12 In executing the budget, OMB uses the specification lever to define how agencies may spend their appropriated money and the monitoring lever to ensure that agencies’ ongoing work is acceptable.13 And in overseeing management initiatives, OMB uses the Presidential Management Agenda lever to develop agency-specific versions of those initiatives, and the budget-nexus lever to ensure that the initiatives are realized throughout the budget process.14 Collectively, these levers reach widely and deeply into agency policy choices.

In identifying and examining these levers, this Article focuses not on the appropriations process but instead on the periods leading up to the annual submission of the President’s budget to Congress and following the passage of the budget.15 This is not to say that the congressional appropriations process is irrelevant.16 Rather, OMB’s power is rooted more in the system of executive authority that has developed around the budget cycle than in the ultimate appropriation.

As a central part of describing and analyzing this power, the Article surfaces the role of the Resource Management Offices (RMOs), a critically important but understudied part of OMB. The five RMOs collectively contain more than four times as many staff members as OIRA.17 Working directly with budget and policy officials in each agency, the RMO staff play a large role in overseeing—indeed, at times in directing—the work of agencies throughout the administrative state because they have primary responsibility for pulling the aforementioned levers associated with budget preparation, budget execution, and management initiatives.18 Yet despite the broad scope of their authority, a recent search of Westlaw’s database of law reviews and journals identified only seven references to the RMOs,19 in contrast to over a thousand articles discussing OIRA during that same time period.20 Given the omnipresence of the RMOs in agency oversight and direction, the inattention in the literature to the RMOs is remarkable.21 OIRA is important, but that does not mean that the rest of OMB is not worthy of study.22

Three key points about centralized executive control emerge from this study of the RMOs. First, the RMOs provide a direct line into agencies. Each agency has identifiable RMO staff responsible for its work and a regular mode of communication with that staff.23 The RMOs therefore can serve as a conduit for policy and political direction from the President, the White House policy councils and other White House political advisors, and the OMB Director. If there is a message to be conveyed to agencies, the RMOs are a good way to convey it. The RMOs therefore work to ensure conformity with the President’s policy program and political interests.24 In this sense, the RMOs’ work through the budget process reflects presidential, or at least White House, control of the administrative state.

Second, the RMOs are not simply a conduit of information from the top down. They also serve as a source of deep and valuable knowledge of agency programs and practices, and busy senior political officials can accept their judgment calls as final.25 Thus, the RMOs’ work also reflects the power of the RMO staff members to play a large role in determining what presidential control of the administrative state will look like.

Third, the RMOs reach many decisions about agency action on their own, since much agency oversight does not require elevation.26 This does not mean that RMO staff members advance their own political preferences; the staff prides itself on being apolitical and working as hard for one administration as it does for the next.27 Given the extremely high caliber of the RMO staff,28 decisions the RMOs make with and for agencies may be “better,” at least against some metrics, than decisions the agencies would reach on their own. At the same time, it is clear that a subset of RMO decisions have policy import;29 that institutional and interpersonal dynamics mean that agencies will not always elevate these decisions outside the RMOs;30 and that RMO staff may not always be aware that they are making policy-inflected decisions, rather than neutral and technocratic ones.31

In this sense, whether the RMOs’ work is a form of presidential control is less clear. At times, the RMOs’ work may instead reflect OMB control, or RMO-intuited versions of presidential control as applied to particular situations, with case-specific value judgments obscured. Accordingly, this Article is titled “The President’s Budget as a Source of Agency Policy Control,” without further identifying the actor with the ultimate control.

My portrayal of the RMOs’ work necessarily paints with a broad brush. I offer a sketch of how OMB’s policy levers generally secure agency compliance, even if, as in any human institution, the dynamics will not hold true in every instance.32 While this account leaves much open for future work, it provides an analytic framework for understanding the policy control OMB can exercise through the budget process.

Parts I and II elaborate on this descriptive argument. Part I places the RMOs in the context of the larger OMB and explains how the RMOs’ work is integrally related to agency policymaking. This Part also compares the scope of the RMOs’ authority with OIRA’s, showing that, in some ways, the RMOs extend more deeply and broadly throughout the administrative state than OIRA does. Part II then details how OMB’s budget process puts the office “in the stream of every policy decision made by the federal government,”33 by describing how each lever operates in practice.

Part III turns to the Article’s normative argument. This Part evaluates OMB’s levers and, more generally, the role of the RMOs in the budget process, concluding that they have both benefits and drawbacks. One benefit is that the RMOs’ work, unlike OIRA’s, is undoubtedly legal. Indeed, the questions about the legality of OIRA’s work that have dogged that office may explain why much more attention has been paid to OIRA than the RMOs.34 There is also little doubt that the RMOs play an important role in coordinating the sprawling administrative state. In doing so, the RMOs further core administrative law values of efficiency, effectiveness, and to some extent, accountability.

On the negative side, however, three aspects of the RMOs’ work collectively weaken their accountability. First, the RMOs’ work is far too opaque. The lack of transparency surrounding the RMOs’ interactions with agencies and third parties makes it difficult for the public and for Congress to monitor their actions. Second, the structure of the RMOs’ work empowers OMB’s civil servants relative to politically appointed agency officials and obscures ultimate responsibility for agency decisions. Third, because the RMOs’ work seems dry and technical from the outside—the kind of work associated with the bean-counter, green-eyeshade stereotype of budget bureaucrats—its substantive nature and potential for partisan politicization are ignored.

Part IV sets forth my prescriptive argument, although my suggestions are meant to start a conversation rather than to present a perfect package of solutions. I first consider how actors inside the executive branch, namely the President and OMB itself, should respond to the RMOs’ weak accountability. I argue that Presidents should issue executive orders governing the RMOs’ work, thereby claiming ownership of it, just as they issue executive orders governing OIRA’s work, thereby setting forth their regulatory philosophies. I propose a variety of transparency requirements that could be embedded in such an executive order and assess their pros and cons. Additionally, I suggest ways that OMB could make its own work more transparent and participatory.

I then turn to actors outside the executive branch, namely Congress and civil society organizations. I consider how Congress could attempt to increase the RMOs’ accountability through additional oversight. I also consider ways that civil society organizations could increase their monitoring of the RMOs’ work and expand their efforts to influence its work.

The Article concludes with a cautionary note for OIRA’s critics, who have sometimes suggested that OIRA’s role in regulatory review ought to be eliminated. Because OIRA’s work could be accomplished through the RMOs, which are less transparent and accountable, reform—not elimination—is the better option. More generally, future analysis of OIRA’s interactions with agencies should include consideration of the RMOs’ complementary power.

I. the role of budget oversight in omb

As recent shutdowns dramatically illustrate,35 the federal budget is indispensable to the government’s work. The budget also serves as a statement of national priorities. OMB plays a critical role in developing this statement and overseeing its implementation through three related activities: preparing the budget, executing the budget, and working on management initiatives tied to the budget.

This Part explains the importance of OMB’s budget work, laying the groundwork for the more detailed analysis in Part II of the budgetary levers OMB can use to influence agency policymaking. Section I.A maps out OMB’s basic structure, showing that management and budget are integrally related to policy choices. Section I.B introduces the RMOs as central to OMB’s control of agencies’ policy choices through the budget process. This Section explains in broad strokes the work of these offices and their policymaking effect. Section I.C situates OMB’s work in the context of executive branch oversight. It compares these OMB offices to OIRA because OIRA’s power is much better understood in the literature. Drawing on this descriptive work, I argue that the scope of the RMOs’ work is in some ways even greater than OIRA’s.

A. The Office of Management and Budget—and Policy

OMB dates back to the Budget and Accounting Act of 1921, which created OMB’s predecessor, the Bureau of the Budget.36 The Act was intended to rationalize the uncoordinated process in which individual federal agencies presented their budget requests seriatim to Congress with no big-picture, national view.37 The Act located the Bureau, colloquially known as the BOB, in the Treasury Department but created a Director and Assistant Director who reported directly to the President.38 Originally these positions were simply the President’s own confidential appointees; only later would they come to require Senate confirmation.39

In 1939, during President Franklin D. Roosevelt’s Administration, the BOB’s role of providing staff assistance to the President was formalized when the BOB was moved out of the Treasury Department and into the newly created Executive Office of the President (EOP).40 Although the BOB focused on the national budget at a macro-level and on reducing government waste on a micro-level,41 its work soon expanded to providing broader policy advice to the President on all sorts of matters.42 In 1970, as part of a reorganization plan put forth by President Nixon, the BOB’s name was changed to the Office of Management and Budget.43

Today’s OMB is an office of around 435 full-time employees,44 making it the largest unit in the EOP.45 More than ninety percent of OMB’s employees are career civil servants,46 further distinguishing it from most other EOP offices, which tend to be staffed more heavily by political appointees.47 Of the approximately forty political appointees in OMB, very few are Senate-confirmed: only the Director, Deputy Director, Deputy Director for Management, Administrator of OIRA, and heads of two other offices.48

OMB is often said to be divided into an “M” side (for management) and a “B” side (for budget).49 In principle, the “M” side consists of several offices created by statute that oversee matters such as federal financial management, procurement, e-government, and information technology.50 OIRA is one of these offices.51 The “M” side also includes a non-statutory office overseeing performance and personnel management in the agencies.52 In principle, the “B” side consists of five RMOs, organized by agency and program area, which oversee budget development and execution for the agencies under their purview. It also includes a separate Budget Review Division, which coordinates the President’s budget as a whole and analyzes budget policy and trends at an aggregate level.53 The organizational chart below maps out this world, where the “M” units are the Statutory Offices (on the right), along with the Performance and Personnel Management unit (one of the OMB-wide support offices on the left), and the “B” units are the RMOs (at the bottom) along with the Budget Review Division (one of the OMB-Wide Support Offices on the left).

Figure 1.

omb organization54

In reality, however, the story is more complicated. OIRA’s work on regulatory and information policy cannot fairly be described as management-related. Rather, these are policy functions. The current White House seems to acknowledge this by presenting OIRA’s work in a distinct tab on OMB’s website, separate from the two tabs on management and budget.55 In addition, another important aspect of OMB’s job is to centralize agencies’ views on legislation and their interactions with Congress; the current website again provides a separate tab for this policy-laden task.56

For their part, the RMOs deal with far more than the agency budgets they oversee. As the rest of this Article demonstrates, their authority over budget preparation, budget execution, and related management initiatives gives them a wide purchase over agency policy decisions.57 Dividing OMB into a management side and a budget side thus obscures the role of policy in the office.58

In some ways, the process of obscuring the role of policy in OMB began with President Nixon’s 1970 reorganization plan. In renaming the BOB the Office of Management and Budget, President Nixon also created what became the Domestic Policy Council, explaining that “the Domestic Council will be primarily concerned with what we do; the Office of Management and Budget will be primarily concerned with how we do it and how well we do it.”59 This plan to keep policy out of OMB was impossible from the start.60 But while it is generally understood that the job of the White House policy councils is to coordinate the President’s policy,61 it is less widely discussed that this, too, is the task of the budget side of OMB. To the extent that it is understood at an abstract level,62 the mechanisms by which OMB’s budget side does this work remain underexplored. Accordingly, the rest of this Article turns to demonstrating how OMB’s budget side does this work.

B. The Resource Management Offices

As Figure 1 indicates, the RMOs are central to OMB’s operation. In 1994, the RMOs were introduced in their current form as part of an internal OMB reorganization.63 The RMOs grew out of longstanding budget-focused “program divisions,” which even in their narrower focus were “the ‘heart and soul’ of the institution” dating back at least to the World War II era.64 Together, the RMOs oversee the entire administrative state—cabinet departments, other executive agencies, and independent agencies—in five groups organized by subject matter: Natural Resource Programs; Education, Income Maintenance, and Labor Programs; Health Programs; General Government Programs; and National Security Programs.65

Almost half of OMB’s 435 employees work in the RMOs.66 At the helm of each RMO is a political appointee called a Program Associate Director or PAD.67 But unlike the heads of OIRA, the Office of Federal Financial Management, and the Office of Federal Procurement Policy, the PADs are not Senate-confirmed.68 The RMOs are further organized into distinct divisions, each run by a career member of the Senior Executive Service, called a Deputy Associate Director, or DAD.69 Each division is then split into branches run by a career official called a branch chief.70 The remainder of the staff members within each branch are program examiners, with primary oversight responsibility over part of a large agency, several smaller agencies, or some combination thereof.71 In keeping with the high expectations for RMO staff in general, program examiners tend to be highly credentialed.72 They are also often (although not always) relatively junior in their careers.73

The RMOs have broad authority over the agencies they oversee. As the Director and Deputy Director of OMB explained in describing the 1994 transformation of the program divisions into the RMOs, the RMOs would “better integrate our budget analysis, management review and policy development roles,” assessing how well agency programs work and making program and policy plans for the future.74 This open-ended portfolio covers almost anything agencies could conceivably want to do.

The core of the RMOs’ work tracks three distinct parts of the budget process: budget preparation, during which the RMOs work with the agencies under their authority to guide the development of their budget proposals; budget execution, during which the RMOs ensure that agencies implement the budget in accordance with legislative requirements and the President’s priorities; and management implementation, which requires the RMOs to ensure that agencies implement various management requirements as the new budget is prepared and the previous budget is executed.75 These three aspects of the budget process structure the relationship between the RMOs and agencies and give the RMOs a great deal of authority over agency action.

A recent publication providing advice for new political appointees underscores the importance of the RMOs: “There is one certainty in Washington: You will be dealing with the Office of Management and Budget throughout your tenure as an agency head. Nearly every major issue you will face will pass through OMB.”76 While “[y]ou will have to work with OMB in a variety of areas,” including regulatory review, “the budget process is the main arena of engagement,” and “[y]our lead OMB policy official for most budget and program policy matters will be the program associate director (PAD) with jurisdiction over your agency.”77 And as one former PAD explains, “You sit at the pure epicenter of policy. You’re in a position to make a difference. And eventually, everything will come across your desk.”78 Given the importance of the RMOs to agency decision making, their work requires more attention.

C. Comparing Power Centers for Review of Budgets and Regulations

Before turning to the levers the RMOs use to control agency policymaking through the budget process, it is worth underscoring the influence of those offices as compared to OIRA, which provides a more familiar frame of reference.

As is well known in the academic literature and to Washington insiders, OIRA’s role in regulatory review gives it significant authority over agency policymaking.79 Every President since Reagan has required executive agencies to submit significant regulatory actions to OIRA for approval and to conform those regulations to various cost-benefit principles as justified in a Regulatory Impact Statement.80 Because OIRA ultimately determines whether a regulatory action is significant, in practice OIRA at least initially investigates a large portion of regulatory actions.81 OIRA’s review can result in a regulation being significantly delayed, never being published at all, or being published in a dramatically different form.82 For this reason, OIRA is routinely referred to as “the most important government office you’ve never heard of.”83

The RMOs, even less well known, are equally deserving of this superlative. Indeed, in some ways, the RMOs are more influential than OIRA: (1) they penetrate deeper into agency practice; and (2) they have broader purview over the executive establishment. This influence is a function of both their institutional attributes and the nature of their work.

The RMOs are able to push deeper into agency practice for two reasons. First, they have more staff with which to do so. OIRA is an office of around forty-five people,84 divided, like the RMOs, into different branches that each oversees a subset of agencies.85 The RMOs collectively have more than four times as many staff members, and three of the RMOs are each larger than OIRA itself.86 With a greater number of staff members assigned to each agency, the RMOs have more capacity to engage with agency work.87

Second, the RMOs are more deeply involved because the core of their work is proactive, rather than reactive. OIRA largely responds to what agencies bring to it.88 Although President George W. Bush’s OIRA Administrator introduced the practice of prompting agencies to consider promulgating a particular regulation, OIRA rarely uses this tool.89 In contrast, the RMOs are proactive by, for example, telling agencies the kinds of policy choices they expect to see in agencies’ budget submissions, in keeping with OMB Directors’ budget instructions,90 and detailing how agencies may spend the money allocated to them.91 By instigating agency action rather than merely responding to it, the RMOs have the capacity to affect a greater variety of agency work.

Relatedly, the RMOs extend oversight more broadly throughout the executive establishment than OIRA does. Most importantly, while independent agencies need not submit their regulations to OIRA for review,92 they must participate in the annual budget cycle under RMO oversight.93 The absence of authority over independent agencies through regulatory review poses an important constraint on OIRA, so important that some commentators have called “being or becoming an ‘independent’ agency” a potential “tactic” for agencies to use in order to avoid OIRA oversight.94 Conversely, independent agencies are subject to RMO oversight95 (although there are some variations in how independent agency budgets are constructed and submitted96). The RMOs thus provide a powerful tool for presidential control over independent agencies that OIRA does not offer.97

The RMOs may also hold particular sway over a subset of traditional executive agencies over which OIRA has less control: those agencies that do more of their work through spending programs than through regulation. While regulatory programs “employ regulatory action to achieve program and agency goals,”98 spending programs use federal money to achieve their goals. Pervasive throughout the administrative state, spending programs include competitive grant programs, block or formula grant programs, capital assets and service acquisition programs, credit programs, direct federal programs, and research and development programs.99 At least some agencies that primarily operate spending programs tend to regulate less frequently100 and under less expansive statutory authority,101 giving OIRA fewer opportunities to engage with their policymaking. When agencies do regulate under spending programs, OIRA tends to review their analysis less stringently, separating its analysis of so-called “transfer regulations” from traditional regulations,102 reserving its deepest review for the latter.103 For agencies primarily operating spending programs, then, the RMOs are a comparatively greater source of centralized control than OIRA is.

But even as to those agencies for which OIRA’s oversight is strongest—traditional executive branch regulatory agencies such as the EPA and the FDA—the RMOs play a powerful complementary role. Presidents have long used budget cuts as a deregulatory strategy to limit the capacity of these agencies to act,104 thereby empowering the RMOs that work closely with these agencies throughout the budget process to make decisions about how to prioritize their resources. Presidents also use existing budget amounts to re-allocate priorities within agencies, thereby changing policy directions within agencies.105 In addition, as Part II shows in more depth, the entire budget process empowers the RMOs to shape agency policy choices regardless of the eventual appropriations decisions by Congress. Because budgets are critical to the work of traditional executive branch regulatory agencies, the RMOs play a significant role in shaping their policy choices, just as OIRA does.

This comparison between OIRA and the RMOs is not intended to downplay the importance of OIRA’s control or to suggest that OIRA is unworthy of the vast amount of attention it receives. Rather, the goal is to illustrate the important and underappreciated role that the RMOs play in the administrative state. Their role needs to be better understood. The next Part begins this task.

II. omb’s control of agency policymaking through the budget process

This Part argues that OMB’s budget and management authority provides the opportunity for significant control over agency policymaking. OMB’s role in budget preparation (Section II.A) and budget execution (Section II.B) affects how agencies prioritize, justify, and make decisions about the policies under their purview. So, too, does OMB’s related power to develop agency-specific versions of management initiatives (Section II.C). Collectively, these aspects of the budget process provide OMB with seven levers to control agency action. Rooted variously in statutes, OMB circulars,106 memoranda, or simply practice, these levers influence and achieve particular outcomes in agency policy choices.

The fact that these levers provide OMB the opportunity to control agency policymaking through the budget process does not mean that OMB always uses this opportunity to the full extent of its authority or that when it does, its actions are taken only at its own behest. As to the first caveat, the use of these levers varies by personnel in agencies and OMB.107 The interaction between agencies and OMB is sometimes combative (if OMB supersedes what agencies wish to do), but at other times collaborative (if agencies and OMB work together to reach consensus about the best way forward) or even collusive (if agencies ask OMB to give it a particular direction that it would have a hard time implementing if the instruction did not seem to come from the top).108 As to the second caveat, the levers available to OMB are embedded in a broader set of interactions between OMB and other parts of the EOP as well as between other parts of the EOP and agencies. Accordingly, not everything OMB conveys to agencies is a product of OMB’s own decision, and agencies may at times hear directly from other White House offices rather than OMB itself.109 This Part acknowledges these nuances while providing a general map of how the levers operate to strengthen OMB’s control. It leaves the task of refining and building on this initial sketch to future work.

A. The Mechanisms of Control Through Budget Preparation

The Congressional Budget Act of 1974 requires the President to submit a detailed budget proposal for the following fiscal year110 to Congress annually “on or after the first Monday in January but not later than the first Monday in February.”111 OMB does the bulk of this work on behalf of the President.112 In anticipation of the statutory deadline, agencies submit their budget requests to OMB in early fall.113 OMB then spends the next few months considering these requests, asking agencies to justify them, and often ultimately modifying them as OMB consolidates a budget proposal for the whole federal government.114

OMB has three levers that affect agency policymaking during the budget-preparation process, regardless of Congress’s subsequent action on the budget: (1) a form-and-content lever, under which OMB sets ex ante requirements for the budget and policy proposals that agencies must submit for OMB’s review; (2) an approval lever, under which OMB must consent to those budget and policy requests ex post; and (3) a confidentiality lever, under which OMB restricts what agencies may disclose about this process.

1. The Form-and-Content Lever

The first lever that OMB can use to control agency policymaking through budget preparation is the ability to tell agencies what they should put in their budget requests in the first place (the content) and how they should convey this information (the form).115

OMB operationalizes its form-and-content lever through two sets of documents. The first is OMB Circular A-11, titled The Preparation, Submission, and Execution of the Budget.116 This 900-page document is issued each summer to federal agencies to guide their budget requests,117 although the circular does not change dramatically from year to year. Large parts of it are technical and do not play a substantial role in controlling agency policymaking.118 Two parts of Circular A-11, however, do have a major effect on agency policymaking. One part requires agencies to keep the substance of the budget process confidential. As I discuss below, this confidentiality lever means that “[i]t is not uncommon for someone to find himself publicly saying the opposite of what he thinks because he lost a battle with OMB.”119 A second part is the requirement to embed the administration’s various management initiatives in agency budget requests.120 As I explain later, many management initiatives are actually substantive policy choices without being denoted as such. The requirement to tie these initiatives to the budget gives OMB an enormous lever to shape how agencies dedicate their resources.121

The other set of documents through which OMB uses the form-and-content lever are memoranda issued by the OMB Director to provide more specific guidance to agencies on what their budget submissions should include.122 These memoranda can play a significant role in shaping agency policymaking.

One way in which these memoranda guide agency action involves the budgeting method selected to develop the President’s budget. In the 1970s, for example, President Carter’s OMB Director required that agencies use Zero-Based Budgeting to prepare their budget requests—that is, to prepare each year’s request as if it were starting at zero.123 More recent budget memoranda have instead required Incremental Budgeting,124 a budget method that assumes that last year’s budget is the starting point for incremental adjustments up or down.125 It is not difficult to see how these distinct approaches affect agencies differently. Budget scholars have connected Incremental Budgeting to more gradual and modest change within agencies, while Zero-Based Budgeting encourages more radical rethinking about agency priorities.126

Directors’ budget memoranda can also instruct agencies to justify their programs in light of particular presidential priorities. The memoranda indicate that agencies are more likely to be successful in their budget requests to the extent the agencies can shape their priorities to match those of the President. This guidance therefore tells agencies where to direct their internal efforts. Not surprisingly, these initiatives vary significantly according to the preferences of the current President. President George W. Bush’s OMB Director focused on Bush’s priorities after September 11, including homeland defense and national security,127 while President Obama’s OMB Director has focused on priorities that Obama established in the wake of the financial crash and the Great Recession, including domestic matters such as health, education, energy reform, and fiscal discipline.128

Directors’ memoranda, sometimes co-signed with other officials in the White House, can also address a narrower set of agencies to instruct them to emphasize new presidential priorities in their budget submissions. One recent example instructed the Departments of State, Defense, Justice, Agriculture, Commerce, and others on preparing budget submissions for particular programs designed to counter biological threats.129 Another instructed a similar but smaller set of agencies on preparing budget submissions for programs designed to combat antibiotic resistant bacteria.130 Each memorandum included approximately five single-spaced pages of factual findings and instructions on exactly what kinds of substantive budget proposals to offer.131

Budget memoranda can further instruct agencies about the administration’s priorities in the management of their internal affairs. For example, recent memoranda have directed agencies to catalogue and take specific steps toward expanding their employee health and wellness programs and to reform their hiring practices.132 These instructions involve inward-focused rather than outward-focused policymaking but can nonetheless be significant with respect to opportunity costs (as money and time focused on these initiatives means less money and time focused elsewhere), the number of people affected by the policies (as by some estimates, 2.85 million civilians work in federal agencies133), and potential ripple effects (as internal federal efforts have sometimes been harbingers of broader social change134).

In sum, OMB’s form-and-content lever helps shape where and how agencies focus their efforts before any money is even requested.

2. The Approval Lever

The form-and-content lever derives its strength from the fact that OMB must ultimately approve the agencies’ budget requests. The approval lever is thus an ex post complement to the ex ante instructions OMB issues in Circular A-11 and the Directors’ budget memoranda. In other words, under the form-and-content lever, OMB tells agencies what to include in their budget requests before agencies draft them, while under the approval lever, OMB tells agencies how those initial drafts must be modified before they can be transmitted to Congress. The approval lever functions both at a broad level, securing overall agency compliance with the President’s general policy preferences,135 and at a narrow level, governing budget and policy choices in discrete line items.136

The approval lever is central to OMB’s power over agencies. This Section first describes the way the structure of the RMOs affects the way the approval lever is operationalized and therefore the relationship between OMB and agencies; second, it explains how budget numbers work with budget language to set forth substantive policy; and finally, it discusses variations in the way the approval lever operates for several subsets of agencies.

a. The RMOs’ Pyramid Structure

The pyramid structure of the RMOs affects how the approval lever operates in practice. The Program Associate Director (PAD), a political appointee, oversees one or more divisions, each run by a Deputy Associate Director (DAD) who, in turn, oversees several branches.137 Each branch is run by a branch chief who, in turn, oversees a group of program examiners.138 This structure gives a lot of authority to the program examiners, who provide the first review of the agency’s budget submission, hold hearings on or otherwise request additional information about the agency’s submission, and reach preliminary conclusions about what should be funded and at what amount.139 Although a pyramid, it is fairly flat as hierarchies go, with program examiners typically providing key briefings to senior career and political staff. Program examiners’ knowledge of the programs in question and assessments of the policy options are thus important to the decision-making process.140 Program examiners can also influence agencies’ budget and policy proposals even before the budget justification is submitted, as agencies may shape their proposals in anticipation of their program examiner’s expectations.141

This is not to say that program examiners routinely impose their own policy preferences on agencies through the budget process; professional norms and examiners’ on-the-job training work against this possibility.142 The pyramid structure with its tiers of review and collaboration among all levels of the hierarchy also militates against the imposition of personal policy priorities.143 At the same time, program examiners must use their discretion in interpreting how to implement what they perceive to be the President’s program,144 and their role as gatekeepers makes them influential.145

To be sure, the OMB Director makes the ultimate decision about what to recommend to the President about each agency’s budget request.146 The White House policy councils and other White House offices may also get involved in specific budget decisions related to high-profile policy issues.147 However, the number of issues open for discussion and debate shrinks as the agency’s budget request moves up the chain of command from the RMOs to the Director.148 Even fewer issues reach the President for decision.149

After “passback”—the formal process by which OMB, typically through the RMOs, informs the agency about the budget and associated policy choices that OMB has approved for that agency150—agency officials may appeal to the Director or even the President for more or differently allocated money.151 Under some administrations and in some economic circumstances, appeals tend to be “serious matters and often involve millions or billions of dollars and major policy choices,”152 while at other times appeals are more routine.153 Disagreements about passback are commonly negotiated at the staff level, although more senior OMB officials may get involved with particular agencies, on particular issues or at different stages.154

The pyramid structure encourages agency officials to choose their budget battles carefully on the way to obtaining OMB approval. Not everything can be elevated, and even things that are elevated may not be resolved in the agency’s favor, leaving the agency under the day-to-day oversight of the OMB staff whose views may have prevailed.

b. The Relationship Between Budget Numbers, Budget Language, and Substantive Policy

The approval lever affects substantive policy choices because OMB’s approval is not simply about an overall funding amount for each agency or even each program or function within an agency. Instead, OMB’s approval is linked to policy decisions about executive branch priorities.

These policy decisions are first reflected in passback, “which includes the policy decisions and corresponding appropriations language.”155 When passback directs spending cuts, OMB can give greater or lesser degrees of policy flexibility to agencies in allocating the cuts depending on how much the administration favors their work.156 Where passback is furthering presidential priorities, passback language setting forth the administration’s policy preferences can be quite detailed.157

The policy decisions are further reflected, after any post-passback appeals, in the proposed appropriations language in the President’s budget that is ultimately submitted to Congress. This language typically builds on the previous year’s appropriations language but can include additional proposed substantive limitations on or uses of funds that reflect the administration’s policy goals.158

The approval lever thus extends not just to monetary amounts but also to the specific policies tied to the amounts because budget language supports the underlying budget numbers at a fine-grained level. These are the policy choices that the administration—including agency officials themselves—will advocate for during the subsequent congressional budget process.159

c. Variations in the Approval Lever

There are some formal variations in how the approval lever functions for different agencies. For example, program examiners work alongside officials in the Department of Defense to develop that agency’s budget proposals much more collaboratively and much earlier in the process than with any other agency.160 This variation in the budget process does not obviously weaken OMB’s approval lever. After all, even a cooperative process depends on ultimate OMB approval. Moreover, the early integration of program examiners presents another avenue to ensure that White House priorities are baked into the budget request, while information that program examiners have gleaned during the internal deliberations can influence what OMB subsequently approves.161

Another variation in the operation of the approval lever exists for the subset of agencies that have “budget bypass authority,” either because they can submit their budget request to Congress at the same time they submit it to the President162 or because the President must present their original proposal to Congress unchanged.163 The ability to have a direct line to Congress without first obtaining White House approval somewhat weakens the force of the approval lever,164 but not entirely. The President can still submit his own proposals for these budget bypass agencies.165 OMB even instructs such agencies that “OMB may provide you additional materials supporting the President’s Budget request that you will forward to the Congress with the agency testimony” and directs agency witnesses to be able to “explain . . . the request in the President’s Budget” along with their own.166 Moreover, having the White House’s support in a budget request can be valuable.167

A third variation in the approval lever may exist for agencies that run programs rooted in “mandatory spending” authority, rather than discretionary spending that goes through the annual appropriations process.168 In principle, the approval lever might be weaker for agencies running programs that are not subject to the annual appropriations process. OMB, however, still retains significant oversight of these agencies and programs. The discretionary part of the agency’s budget is still subject to OMB approval, so priority setting (for example, in the allocation of staff among activities and divisions) continues to be part of the annual appropriations process.169 Moreover, the President’s budget may propose changes to mandatory spending, proposals that OMB has by definition approved.170 At least in the current and previous administrations, OMB has required agencies to submit, as part of their annual budget requests, a description of any effort to take discretionary action that would increase mandatory spending, and has strictly limited its approval of these efforts.171

Only a small subset of agencies are not affected by OMB’s approval lever: those that are largely self-funded and obtain their budgets from non-governmental sources rather than the annual appropriations process.172 But these agencies are exceptions rather than the norm.173 The approval lever is generally applicable to and influential in both independent and executive agencies.

3. The Confidentiality Lever

A third lever OMB uses to control agency policymaking through the budget preparation process is the confidentiality lever: the requirement that agency officials silence their own differing preferences and, if those preferences become known, distance themselves from them.174 The confidentiality lever seeks first to promote open, vigorous internal debate and then ultimately to ensure that the administration speaks publicly with one voice.175 As a result, the confidentiality lever limits agencies’ ability to state publicly their own views of alternative budget and policy priorities.

The confidentiality lever overlaps with a more general set of clearance requirements OMB uses for all agency communication with Congress.176 But the confidentiality lever is broader because it applies to agency disclosures to anyone outside the executive branch,177 including the media, interest groups, academics, and others. The confidentiality lever also overlaps with the President’s constitutional appointment and removal powers because administrators’ relationships with the President and interests in keeping their jobs also limit disclosure of policy preferences at odds with the President’s program.178 But again, the confidentiality lever is broader because it applies to independent agency officials179 and to civil servants,180 who are not subject to presidential removal.

When agency officials testify before Congress after the President’s budget has been submitted, Circular A-11 instructs that “[w]itnesses will give frank and complete answers to all questions,” “avoid volunteering personal opinions that reflect positions inconsistent with the President’s program or appropriation request,” and “will not provide the agency’s request to OMB or plans for the use of appropriations that exceed the President’s request.”181 Agency officials may speak only in support of the President’s budget, even if they strenuously argued for different sums of money or different policy priorities up until the time the President’s budget became final.182 RMO staff listen to agency officials testifying before Congress about the budget,183 and efforts to circumvent the confidentiality requirement may have negative consequences for both officials and their agencies.184

The confidentiality requirement applies to written material submitted to Congress as well as to testimony; the circular requires that agencies submit all “budget-related materials to OMB for clearance prior to transmittal to congressional committees, individual Members of the Congress or their staff, or the media.”185 This rule limits even informal communication between congressional staff and agency staff.186 Further, although the circular requires agencies to post on their websites all of the material underlying their part of the budget request to Congress,187 it prevents agencies from posting any material that they originally submitted to OMB.188

There are exceptions to the general rule of confidentiality. For example, the Inspector General Reform Act of 2008 provides that the President must include in his budget “any comments” from an Inspector General who concludes that the budget proposal “would substantially inhibit the Inspector General from performing the duties of the office.”189 More generally, the Whistleblower Protection Act limits adverse employment actions against civil servants who disclose information that they “reasonably believe[] evidences . . . a gross waste of funds, an abuse of authority, or a substantial and specific danger to public health or safety.”190 In addition, by practice although not by statute, members of the House and Senate Armed Services committees have typically asked the top military officials in the Department of Defense for their “unfunded priority lists”—that is, a list of desired items that did not make it into the President’s final budget.191

But these are exceptions that prove the rule,192 and even these exceptions can be limited in scope.193 In general, agency officials are reluctant to try to circumvent the confidentiality lever, even in back channels, for fear of adverse consequences from OMB.194

B. The Mechanisms of Control Through Budget Execution

OMB’s role in the budget process does not end when Congress passes and the President signs the annual appropriations bills (or, in more recent years, the continuing resolutions to provide funding for a limited period of time after the fiscal year until the appropriations bills are agreed upon).195 OMB is intimately involved in budget execution—the way federal agencies carry out their work under the budgetary authority they have been granted.196 OMB affects budget execution through two different levers: the formal specification lever, through which it “apportions” and otherwise defines how agencies spend the funds Congress has appropriated, and the informal monitoring lever, through which it oversees agencies’ implementation of their programs.

1. The Specification Lever

While OMB “is much too small to oversee all transactions,” nevertheless “[o]n any particular matter it may intervene to influence the use of federal dollars.”197 The specification lever provides four main tools for OMB influence: it must (1) apportion agency spending; (2) approve requests to transfer or reprogram funds; (3) approve requests to defer or rescind funds; and (4) oversee decisions regarding a government shutdown in the event of a failure to reach a budget agreement.

a. Apportionment

Before agencies can spend the funds that Congress has appropriated, OMB must apportion them by specifying how much may be expended, when it may be expended, and even to some extent how it may be expended.198 The Anti-Deficiency Act199 requires agencies to spread out appropriated funds so that they do not spend them too quickly and come back asking for more.200 Under this authority, OMB limits how much agencies can spend either by time period, project, or both.201 It also reviews apportionments at least four times a year,202 with the potential to reapportion funds under certain circumstances.203 The apportionment power gives OMB a regular opportunity to control how agencies conduct their operations. The RMOs take the lead in this responsibility.204

“Although apportionment is largely a technical procedure,” Allen Schick explains, “it is the last point at which OMB formally controls agency spending.”205 Therefore, “OMB sometimes uses apportionment to impose conditions on agency spending or to demand changes in agency practices.”206 For example, the RMO may include a footnote placing further limitations on a particular apportionment amount,207 such as requiring an agency to spend its funds on particular activities.208 The RMO may also require that an agency take some action before it receives its apportionment.209

The power of the footnote should not be understated: apportionment footnotes are subject to the requirements of the Anti-Deficiency Act,210 and agency officials disregard them at their peril. Violations of the Act are punishable by adverse employment actions (including suspension without pay or removal from office)211 and criminal penalties (including a fine of up to $5,000 and up to two years imprisonment).212 In addition to apportionment itself, then, apportionment footnotes offer OMB an opportunity to specify how agencies spend their money and thus the actions agencies take.

The effect of OMB’s apportionment power increases when the government operates under continuing resolutions. As a reporter recently wrote, the number one “hidden cost[] of continuing resolutions” from an agency’s perspective is that “OMB gets in your face.”213 OMB directs agencies to “operate at a minimal level until after your regular [fiscal year] appropriation is enacted” and oversees their choices to implement that direction214 to prevent a situation where a subsequently enacted regular appropriation provides less funding than the agency had expected.215 While the OMB Director provides a formula that automatically apportions amounts provided under the continuing resolution,216 the RMOs may further limit this amount,217 deploy footnotes to specify additional restrictions on its use,218 and grant requests for sums beyond the automatic apportionment only in “extraordinary circumstances.”219 The uncertainty about what will happen at the end of a continuing resolution thus amplifies the RMOs’ attention to agency spending.

b. Transfers and Reprogramming

Another tool of the specification lever allows OMB to exert influence when agencies seek to change an aspect of Congress’s appropriation. For example, agencies may seek to transfer funds from one account to another or to reprogram funds from one purpose to another within the same account.220 OMB must approve the request before the agency can discuss transferring or reprogramming funds with the relevant congressional committees.221 This process, too, gives the RMOs a way to influence where the agency directs its funds.222

c. Deferral and Rescission

The specification lever is also at work in the less frequent instances when the President proposes to defer or rescind the use of appropriated funds.223 The Congressional Budget and Impoundment Control Act of 1974,224 which Congress passed after a showdown with President Nixon over the President’s efforts to impound funds for policy reasons, governs this process.225 The Act forbids unfettered policy impoundments and provides for more limited deferrals and rescissions, which must be grounded in more than simply the President’s distaste for the program in question and which Congress must approve.226

Given this history, deferral and rescission are most often seen as political battles between Congress and the White House.227 However, deferrals and rescissions are also a way for the White House to control agencies, as an individual agency must provide OMB with the material it requests in support of the President’s formal “special message” to Congress requesting deferral or rescission.228 Because of the need for presidential involvement, policy-inflected decisions around this tool are less within the province of the RMOs and more clearly within White House itself,229 although the impetus for a particular proposal may come from within OMB in the first instance.230

d. Government Shutdowns

Finally, the specification lever is at work in the lead-up to a government shutdown, and if efforts to reach a continuing resolution fail, during the shutdown itself.231 While a shutdown was historically an extraordinary event, the possibility and reality of shutdowns have loomed large in recent years,232 so this tool is more than merely hypothetical.

OMB requires agencies to develop shutdown plans and specifies what these plans must contain,233 with different OMB Directors providing different kinds of instructions.234 At bottom, the plans must comply with the requirements of the Anti-Deficiency Act and associated interpretations,235 which puts OMB in the position of making decisions about which agency employees and activities are “essential” and should continue to operate during a shutdown.236 Because of the discretion and judgment involved in making these decisions,237 approval of agencies’ shutdown plans can have substantive policy effect.238 Decisions about the shutdown plans are also rooted in the political environment, which includes sensitivity to ongoing congressional-White House budget negotiations that could avert a shutdown.239

OMB talks with agencies as the plans are being prepared, reviews the plans and any updates to them, and may require changes240 in advance of making the plans public.241 OMB “holds meeting[s] or teleconference[s] with agency senior officials” to discuss “shutdown plans” both before and during a shutdown.242 OMB also maintains regular contact with agencies during a shutdown,243 in large part because OMB is the point of contact for agencies on any aspect of a shutdown, including legal questions about the Anti-Deficiency Act.244

Even in the absence of appropriations, then, OMB uses its responsibility to execute the budget to control agency action through the specification lever. OMB’s senior political appointees make decisions regarding shutdown plans, because of their high-profile nature; however, the RMO staff plays a supporting role in communicating with the agencies and evaluating plans.245

2. The Monitoring Lever

The RMOs also become intimately involved with agencies’ policy choices using the monitoring lever, through which the RMOs oversee agencies’ implementation of their programs.246 This lever is among the most ambiguous because it is informal; it is not governed by any particular legal source but exists in light of the RMOs’ formal duties.

The monitoring lever can manifest itself in frequent communication between agency policymaking officials and RMO program examiners.247 Agency documents reflecting policy choices, such as grant criteria and other allocative decisions, may be significantly revised by the RMOs and sent back to the agency to incorporate changes.248 Even when the program examiner does not actively change documents, the program examiner may ask questions that require agency policy officials to justify or modify their initial decisions.249

Agency policy officials may also reach decisions in anticipation of the RMOs’ requests or collaboratively, as part of a regular phone call or meeting.250 As one analysis suggests, “[E]fficiency is gained by understanding and possibly preemptively incorporating OMB preferences and expectations into outcomes.”251

Much is left to the discretion of the individual program examiner in this relationship.252 Of course, the civil servant branch chiefs, DADs, and the political PADs play something of a unifying role. But not every issue will be elevated, and agency officials may be reluctant to go over the heads of the program examiners with whom they work on a regular basis. Similarly, while DADs tend to have long tenures, PADs tend to change multiple times in an administration.253 Different PADs may have different priorities; agencies subject to significant PAD oversight under one PAD might receive less attention when a new PAD takes over. Different agencies may also receive more or less stringent RMO review depending on the status of the agency’s head within the White House.254 In addition, different RMOs have different longstanding relationships with the agencies under their purview that may transcend administrations.255

To be sure, the program examiner’s job is to effectuate the President’s policy priorities, but it can be difficult to translate big-picture presidential views into reality in each policy decision before an agency. As such, program examiners necessarily use their own judgment.256 Similarly, while PADs are political appointees with views that are supposed to reflect those of the President, the PADs do not get White House clearance for every decision they make, so again, these officials must use their judgment. Again, this is not to suggest that the RMO staff, whether civil servants or political appointees, regularly implement their own policy priorities257 but rather to note that independent judgment calls may have substantive effect.258

As in the budget preparation process,259 other White House offices may also get involved in different decisions related to budget execution. For example, the relevant staff member on the Domestic Policy Council may weigh in on particular matters of social policy as grant criteria are being developed, communicating either directly with the agency or through the RMOs.260 But the monitoring lever gives the RMOs the opportunity for the most regular interaction with the agencies on policy decisions.261

C. The Mechanisms of Control Through Management Initiatives

OMB has often been said to neglect management in favor of its work on the budget.262 Over the last 25 years, however, management has become a more integral part of OMB’s work, “provid[ing] a way for the White House to influence the implementation of its policy agenda.”263 This expanded attention to management is a result of several factors. First, Congress has increasingly delegated particular management tasks to OMB.264 Second, there are an increasing number of statutory offices devoted to particular management issues, from financial management265 to e-government.266 Third, internal executive branch efforts have contributed, including the 1994 reorganization of OMB to integrate management more thoroughly into the office’s budget work267 as well as more recent Presidents’ development and implementation of their own management agendas.268 This Section focuses on these executive branch efforts as providing particularly strong examples of the RMOs’ ability to influence agencies’ policy choices through management initiatives.

Two such levers exist: the Presidential Management Agenda lever, which sets forth presidential initiatives ostensibly designed to improve the administration of government but that often have a substantive policy overlay, and the budget-nexus lever, which connects these management initiatives to the budget process. These levers are related: the budget-nexus lever provides the procedural hook for the more substantive Presidential Management Agenda lever.

1. The Presidential Management Agenda Lever

Management initiatives are not simply neutral, technocratic procedures. As political scientist Andrew Rudalevidge put it when describing the way President Nixon’s political advisors originally viewed the “M” in the new OMB, management was not to be “boring public administration theory” but rather “‘management in the get-the-Secretary-to-do-what-the-President-needs-and-wants-him-do-do-whether-he-likes-it-or-not sense.’”269To that end, management initiatives often either explicitly contemplate substantive policy choices or implicitly lead to them. The Presidential Management Agenda (PMA) exemplifies this dynamic.

Consider, for example, President George W. Bush’s Faith-Based and Community Initiative, which the Administration presented as a management initiative to break down bureaucratic barriers limiting religiously-affiliated organizations from engaging with government.270 But the initiative was not simply bureaucratic; it attempted to weaken the wall between church and state.271

Other initiatives sound more technocratic but end up driving agencies’ substantive choices. President Bush’s Program Assessment Rating Tool’s (PART) stated goal was to integrate budget and performance evaluation to allow for continuation and expansion of well-functioning programs and reform or removal of poorly performing programs.272 One aspect of the PART process asked program examiners to evaluate agencies’ program purpose and design, including the soundness of choices in the underlying legislation.273 But statutes often contain multiple purposes, and the PART process sometimes led agencies to emphasize one over another in an effort to secure program examiner approval.274

Likewise, some of President Obama’s management initiatives sound technocratic but have substantive effects. For example, the Evidence and Evaluation Agenda seeks to move the federal government towards evidence-based policymaking.275 This agenda directs investment of federal dollars into programs that have proven effective on the ground and, once the money is awarded, requires ongoing evaluation of the program’s implementation and outcomes.276 The initiative has affected agency policy choices in a number of ways. It has pushed agencies to adopt some policy goals and approaches to service delivery over others.277 It has prioritized competitive grants over formula grants,278 and transformed decades-old accountability systems.279 One study of the Evidence and Evaluation Agenda goes so far as to call it a “vast attempt to change the foundation of American social policy.”280

These management initiatives are also intricately intertwined with political decisions. For example, some have charged that President Obama’s evidence-based initiative relies more heavily on evidence that supports the administration’s preferred policy decisions.281 Others suggested that President Bush’s PART program was politically motivated to cut the budgets of disfavored programs282 and that the faith-based initiative was a political maneuver without any meaningful policy analysis or apparatus.283 In earlier administrations, President Nixon’s management reforms were described as intending at once to “improve governmental management,”284 “redistribute power in the intergovernmental system,”285 and ensure “political direction” over agencies’ activities.286 And a Reagan Administration official reportedly joked after leaving office that the name for President Reagan’s management initiative was I-D-E-O-L-O-G-Y (riffing on the acronym-heavy titles of the previous Presidents’ management initiatives).287

Management reforms are thus not always simply neutral technocratic reforms; they reflect the substantive policy interests of different administrations and are tied to political contexts as well. Implementing these reforms through the budget process is another lever by which OMB can control agency policymaking, with the RMOs playing an important role.288

2. The Budget-Nexus Lever

Management initiatives also serve as a form of policymaking control because they are directly tied to the RMOs’ work on the budget, and the budget has the levers for policymaking control described in Sections I.A and I.B above.

PMA is tied to the budget in part because OMB’s budget instructions direct agencies to embed the initiatives set forth in the PMA in their budget requests. For example, OMB Circular A-11 requires that agencies present budget requests that “reflect [their] efforts and planned action to strengthen management” in keeping with a series of stated “Administration’s commitment[s].”289 More specifically, the latest version of Circular A-11 explains that “[a]gency [budget] requests are more likely to be fully funded if proposed funding increases and policy changes are grounded” in the Obama Administration’s Evidence and Evaluation Agenda.290 Directors’ budget memoranda similarly provide specific instructions for agencies to take particular actions to implement management initiatives.291

The PMA is also tied to the budget because of requirements set forth in the PMA itself. For example, President George W. Bush’s PMA required each agency to incorporate its substantive policies in its budget requests.292 One of these initiatives, PART, was itself designed to transform agencies’ budget requests by connecting them to program evaluation.293 President Obama continued and perhaps even expanded this trend by setting forth his management agendas directly in his budget.294

In practice, the link between the budget process and management initiatives means that program examiners can play a large role in influencing agencies’ policy choices through executive management initiatives.295

III. assessing omb’s control of agency policymaking through the budget process

This Part turns to a normative assessment of OMB’s control of agency policymaking through the budget process. Section III.A makes the case that there are some salutary aspects of this power, including its firm legal basis (a strength in comparison with OIRA’s regulatory review) and its promotion of coordination across the expansive administrative state. Section III.B nonetheless argues that three aspects of the RMOs’ work raise accountability concerns: the RMOs’ lack of transparency; the delegation of significant policy responsibility to the RMOs’ civil servants and non-Senate-confirmed officials; and the potential for the RMOs’ technocratic-sounding work to obscure policy and political decisions.

A. Salutary Aspects

1. Legality

Unlike the OIRA regulatory review process, the legality of which continues to be debated,296 there is little doubt that the RMOs’ work on budgets and management is legal.

The RMOs’ work is defensible under both major understandings of executive power. Proponents of the unitary executive, focusing on the “pre-ratification historical context” as an aid to understanding the original public meaning of the Constitution, might point to Alexander Hamilton’s listing with “no distinction” the “command of foreign negotiations, preparation of a budget, spending appropriations, direction of the army and navy, direction of a war, ‘and other matters of a like nature’” as core to the executive power of Article II.297

The RMOs’ work would likely pass muster with pluralists as well. Focusing on the extent to which Congress has invested the President with authority to “control the policy discretion of other administrators,”298 a pluralist might point to the delegation to the President to “prepare budgets of the United States Government,”299 “prescribe the contents . . . in the budget,”300 and “change agency appropriation requests.”301 The President has statutory authority to delegate these tasks to the Director of OMB, as a Senate-confirmed official, under the Presidential Subdelegation Act of 1950.302 OMB has specific statutory authority both to work under the President’s “direction” to administer the office303 and to promulgate and oversee management policies for the executive branch.304 Even the authority to exert control over independent agencies’ budgets is calibrated by statute.305

One might nevertheless question whether the full extent of policymaking control that OMB can exert through the budget process is within the scope of this delegation. Critics might note that no statute determines OMB’s role in budget execution. The Anti-Deficiency Act requires OMB to apportion agency spending but does not indicate that OMB may place additional conditions on apportionment.306 Nor is the regular communication between RMO program examiners and agency officials on their policy decisions and implementation set forth anywhere.

The better argument, however, is that, the RMOs’ work on budget, management, and policy stands on firm legal footing. As a textual matter, the expansive scope of authority otherwise given to the President and to OMB on both budget preparation and management contemplates the RMOs’ role in budget execution. As a functional matter, it is hard to imagine a sensible system that could permit OMB’s annual assessments of agency activity in order to devise a budget and strategic plan while not also ensuring appropriate implementation. As a historical matter, the Brownlow Committee on Administrative Management—whose 1937 recommendations resulted in a reorganization of the executive branch and OMB’s predecessor office307—told Congress: “[t]he execution, as well as the preparation, of the budget should be supervised by the Bureau of the Budget and should be closely correlated with fiscal programs and plans.”308 That Congress explicitly rejected many of the Brownlow Committee’s recommendations309 but not this one further supports the idea that Congress did not mean to leave the President and OMB without this authority.

2. Coordination

OMB’s control of agency policymaking through the budget process can also be praised for its coordinating effects.310 The administrative state is a sprawling behemoth, employing millions of people in hundreds of agencies.311 It creates and implements policies that affect every aspect of American daily life (and the world beyond the United States as well). The RMOs’ work usefully coordinates this endeavor. At its best, this coordination provides value not only to the President (whose duty to take care that the laws are faithfully executed is supported by this work) and to Congress (which benefits from this work to such an extent that it continually increases OMB’s statutory duties312), but also to “We the People,” in whose name the government operates in the first place.313

Coordination is apparent in all aspects of the budget process. As to budget preparation, for example, the form-and-content and approval levers give OMB the opportunity to ensure that agencies are not working at odds with each other, especially when the form-and-content lever directs a group of agencies to share approaches to particular problems. The confidentiality lever supports coordination by requiring agencies to trade their own goals for the broader whole.

As to budget execution, the specification lever can support coordination if the RMOs use tools like apportionment to ensure consistency across agencies. But it is really the monitoring lever, in conjunction with the pyramid structure, that has the greatest potential as a coordination tool. The RMOs’ monitoring permits regular assessment of where agencies might be drifting away from their commitments, while the pyramid structure can ensure that the right people are talking to each other about related matters.

As to management, presidential management initiatives can promote coordination, whether through streamlining government-wide interactions with faith-based organizations or requiring agencies to embed evidence-based policymaking throughout their programs, while the budget-nexus lever helps ensure the implementation of shared initiatives as part of the budget process.

To be sure, the RMOs’ coordination is not perfect. The broad discretion of individual program examiners leaves the potential for agencies to receive disparate rather than unifying instructions, whether as to budget preparation, budget execution, or management initiatives.314 Because the individual RMOs are structured to reflect agency-by-agency oversight, they may suffer from the same “stovepiping” problem as the rest of the executive branch, effectively limiting “cross-agency perspectives and knowledge.”315 The RMOs may also fail to coordinate with other OMB offices, including the statutory offices that are responsible for distinct management issues throughout the executive branch. This lack of coordination could contribute to the dissemination of conflicting instructions to agencies.316

That said, the pyramid structure of the RMOs works to counter these concerns by letting the officials at each higher level know what is going on below, allowing the officials to share information across program areas.317 The goal of increasing OMB’s coordination has long been a focus of OMB’s internal reform efforts.318 And because the RMOs’ coordination serves the President’s political interests—after all, Presidents can be judged by how well the administrative state functions and how successfully they implement their policies—there are strong forces supporting ongoing fortification of the RMOs’ coordinating efforts.

From the perspective of administrative law values, however, coordination itself is not the final goal; coordination is useful to the extent it furthers other goals supporting the legitimacy of the administrative state, such as efficiency, effectiveness, and accountability.319 Much of the RMOs’ coordinating work supports the first two of these values and to some extent the last one as well.

For example, some of the RMOs’ coordinating work promotes efficiency,320 sometimes within OMB itself. The structure of the RMOs’ budget preparation review—with clear deadlines and hard decisions made behind closed doors—“move[s] decisions along quickly” and permits “budget cuts that enable the government to hit the bottom line,” promoting both “micro- and macro-efficiency.”321 The speed with which the RMOs can work to respond to policymakers’ questions, and the lack of ceremony involved in getting all levels of the hierarchy together to make a decision, also further the efficiency of the operation.322

Other aspects of the RMOs’ work promote efficiency in agency action. The approval lever and form-and-content lever both signal to agencies what work the White House and OMB will support as Congress begins the appropriations process, reducing the need for agencies to engage in unproductive wish lists or to try to read the “political tea leaves.”323 The pyramid structure promotes efficiency by letting only the most contentious issues rise to the top. The specification and monitoring levers encourage matters to be resolved before they become problems, so an upfront investment of time can save time on the back end.324 Management initiatives focus on efficient conduct in agency action overall.

The RMOs’ coordinating work may also support the effectiveness of agency action. The whole point of this work is to ensure that agencies have the resources they need to do their jobs well and that they are managed and monitored appropriately. The high caliber of the OMB staff—more than half have master’s degrees, more than ten percent have doctorates, and they are often seen as among “the best and brightest” in government325—helps further this goal. The depth of program expertise, especially in the long-time RMO employees that tend to hold the branch chief and DAD positions, provides institutional memory and subject-matter expertise,326 both of which aid sensible decision making.

Still other aspects of the RMOs’ coordinating work support some form of accountability.327 The President’s budget is a public statement that represents a coordinated synthesis of past activities and forward-looking goals, allowing the President to take ownership of the activities of the administrative state. The accompanying PMA furthers this goal. Agencies’ budget justification materials submitted to Congress and posted on their websites publicly state what the agencies’ goals are and justifies both why they are the right goals and the amounts requested to support those goals.328 The RMOs are critical in both putting these documents together and making the programs reflected in the documents work, and to that extent can be said to support accountability.

However, the RMOs’ coordinating work supports accountability only at a high level of generality related to the published products that result. As the rest of this Part argues, the RMOs’ work raises significant accountability concerns in terms of its process. Overall, these concerns undercut the general accountability that might be associated with the budget, the PMA, and agencies’ congressional justifications.

B. Troublesome Aspects

There are three troublesome aspects of the RMOs’ work, all related to the issue of accountability, by which I mean “the ability of one actor to demand an explanation or justification of another actor for its actions and to reward or punish that second actor on the basis of its performance or its explanation.”329 First, there is a lack of transparency in the way the RMOs’ work is conducted and the substance of what they discuss. Second, the RMOs’ work can elevate OMB’s civil servants and lower-level political appointees over Senate-confirmed agency officials, and the ultimate lines of responsibility are ill-defined. Third, the RMOs’ seemingly technocratic work on the budget can obscure value-driven or partisan decision making.

All three of these issues make it difficult for Congress and the American public to hold agencies, OMB, and the White House more generally accountable. It is hard “to demand an explanation or justification”330 when it is not clear what to ask about, whom to ask, or when or why an explanation or justification would be needed, or when it is impossible to ask the right person or to get a straight answer. It is similarly hard “to reward or punish . . . on the basis of . . . performance or . . . explanation”331 when it is not clear whose performance is fundamentally at issue. The rest of this Section elaborates on the accountability problems associated with each of these troublesome aspects of the RMOs’ work.

1. The Lack of Transparency

At first blush, it may seem strange to suggest that the RMOs’ work suffers from a lack of transparency. After all, OMB posts much material online, from directors’ memoranda to OMB circulars; agencies post documents explaining and justifying their congressional budget requests; and the budget itself is a voluminous public document that details the policy choices embedded therein.

But as suggested by the earlier discussion of the confidentiality lever,332 the RMOs’ work does exhibit a lack of transparency.333 While the details of the budget process are known in broad brush strokes, there are nevertheless many parts that remain hidden. We do not know, for example, when, which kind, and how many meetings between the RMOs and the agencies occur over the course of the budget preparation season and throughout budget execution; what interest groups or other administration officials meet with the RMOs, what the meetings are about, and who is present during such meetings; what kinds of agency policy work interest the RMOs, and what kinds do not; how often apportionment footnotes are used and the kinds of demands that are made therein; and how all of the above might vary by administration, by OMB Director, by PAD, by program examiner, or by agency.

Much of the substance of these interactions remains hidden, too. For example, only the agency and its RMO see the contents of the agency’s original budget request to OMB, which includes the amount requested, the agency’s proposed allocation among different programs, its assessment of its own capacities, and its own priorities. PADs offer no public statement of their different priorities when they step into their roles. Additionally, there is no public documentation or acknowledgment when agency policies change in response to the RMOs’ encouragement or requirement.

OMB offers two rationales for the confidentiality it requires of agencies. First, it contends that “[p]olicy consistency” is necessary within the executive branch, particularly when speaking or giving documents to Congress and the media.334 Second, it suggests that the “institutional interests . . . implicated by [the] disclosure” of confidential budget documents militate in favor of confidentiality.335 Such institutional interests include protecting “the deliberative process of the government” by permitting government officials “to express their opinions freely . . . without fear of publicity [that might] . . . inhibit frank discussion of policy matters and likely impair the quality of decisions.”336 To that end, the Freedom of Information Act (FOIA) exempts from disclosure documents that are deliberative and predecisional, like agency-OMB budget discussions.337

However, while each of these rationales has some validity, neither can actually justify the extent of opacity in the budget process, as the next two Subsections argue. In addition, neither rationale appropriately distinguishes the substance of predecisional deliberation from information about the procedural aspects of deliberation and the final post-deliberation decisions, as the third Subsection below explains.

a. Protecting Policy Consistency

First, “policy consistency” does not require a pretense that an agency and OMB never diverged over the appropriate agency budget and policy request. Currently, OMB directs agency witnesses testifying before Congress, if asked about their interest in appropriations beyond the scope of the President’s request, to explain that such interest is “not appropriate,” since “witnesses are responsible for one or a few programs, whereas the President is responsible for all the needs of the Federal Government.”338 This explanation could also disclose the backstory of the agency’s budget request, without sacrificing policy consistency. Witnesses could disclose prior views while avowing conversion to the President’s proposals. Such disclosure need not undercut OMB’s goal of policy consistency because agency officials could explain why they came to believe the final decision was the right one.

In some cases, though, such masking of disagreements about the implications of different budgetary choices might place agency officials in the position of speaking untruths to Congress. Officials do not always come to believe that the final decision was the right one.339 On the one hand, OMB’s reminder to agency witnesses testifying before Congress that the President has responsibility for the whole government while agency witnesses have responsibility for a limited number of programs is surely right and surely justifies the President’s ultimate decision-making authority. But agency officials could say that they understand the President’s request for their agency in light of the entire federal government’s needs—acknowledging that the President and Congress alike face hard choices in the budgetary process—without having to claim that the President’s request for their agency will accomplish what they believe that it will not.340 OMB’s current confidentiality requirements discourage this kind of honesty.

To be sure, such honesty might present several dangers. First, agency officials might simply push for more funding, rather than carefully discussing the nuances of policy tradeoffs. Second, despite agency officials’ best efforts, Congress and the media might focus on the fact of intra-executive squabbles instead of the substance of the discussion. Rather than a thoughtful discussion of policy tradeoffs, we might be left with politicized soundbites for ideological spin. It is hard enough for the President to move anything through Congress as it is, without permitting the exploitation of a history of internal executive disagreement.

This concern is valid enough. But, at the same time, Congress has mandated by statute that agencies do a particular job. In all likelihood, it wants that job done. So, presumably, do the people for whose benefit the job is undertaken. The confidentiality requirements rooted in the goal of policy consistency make it hard for agency officials to say anything meaningful about the extent to which the agency is up for the task, the tradeoff among different priorities, or the real resource constraints that agencies face.341 These requirements therefore make it hard for Congress, much less the public, to know and understand what is actually happening.

Permitting such a disclosure, either where agency officials are asked directly or where they think in particular instances that disclosure is valuable, would accordingly help promote accountability—one of the primary values a well-designed transparency regime can serve.342 It could help give Congress information that it needs in order to set overall funding levels for individual agencies and allocate across competing priorities within each agency, thereby improving Congress’s ability to hold agencies accountable. It could also help give the public information that would make concrete the policy choices that are reflected in the budget, as such choices are often best understood within the context of the alternatives that were rejected.343

It is true that disclosure of the final decisions permits some level of post-hoc accountability.344 But the budget is so many thousands of pages and requires so much analysis and translation that the cost of finding, much less understanding, alternatives that were rejected along the way is extremely high. Individual agency and sub-agency budget justifications to Congress can themselves reach hundreds or thousands of pages. A clear articulation of points of difference, and why they were resolved as they were, can help citizens and civil society organizations better identify specific decisions for which to hold officials accountable.345 This is especially so because voters often need concrete disputes to crystallize their own views about policy choices.346

Nor is it enough to say that interest groups will provide sufficient information about the implications of different budgetary decisions to Congress and the public throughout the appropriations process. Interest groups will not have all of the inside information that agencies do; they are unlikely to capture the full scope of issues that may be important; they may be biased; and agencies may be forced to deny their accuracy.347 Similarly, back channels between agency officials and Hill staffers may get some information across, but inconsistently, given the formal limits the confidentiality requirements place on these back channels.348 Moreover, back channels do little for the public’s ability to hold anyone in either branch accountable.

A second reason that greater disclosure of the intra-executive budget process may promote accountability, regardless of OMB’s interest in policy consistency, is that such disclosure could deter self-dealing or one-sided dealing, which is no less a danger in the budget context than in the regulatory context.349 For example, recent political science scholarship has shown a correlation between the President’s political interests and the distribution of federal funds. One study found that swing states receive more grants and a greater dollar amount in grants than non-swing states;350 this result was more pronounced in the two years leading up to a presidential election than in the two years after one.351 As the author of this study concludes, “[P]residents engage in pork barrel politics.”352 Another group of political scientists found that districts receive more federal funding when they are represented in Congress by members of the President’s own party.353 These authors explain, “For an artful president intent upon redirecting federal outlays to a preferred constituency, ‘the opportunity for mischief is substantial.’”354

These studies do not directly tie OMB to changing expenditure levels, but researchers have found evidence suggestive of such a link. For example, in a related study, agency officials whose job involved decisions about procurement awards, licenses or loans, or grants—identified as “distributors”355—reported more levels of “policy influence” from OMB than did nondistributors.356 A related set of interviews provided some anecdotal evidence that “after the peer review process produces a ranking of grant applications by quality scores, OMB can change the order in which proposals are funded in many programs, thereby influencing the timing, location, and likelihood of funding.”357 In these ways, the conductor of these interviews suggests, OMB “facilitates presidential control of federal agencies, thereby enhancing the ability of the White House to affect micro-level public policy decision making.”358

While these studies do not disaggregate which office in OMB is at issue, the RMOs are a likely candidate, given the direct ongoing interactions about the budget and related policy matters between the RMOs and the agencies they oversee.359 Moreover, the extent of interest group lobbying of, or even informational meetings with, the RMOs is unknown.360 Recent research on OIRA shows that lobbying that office can affect regulatory policy.361 The lack of transparency around who is meeting with RMO officials, when, and about what limits our understanding of the factors and actors that influence budget and related policy decisions.362

b. Protecting the Deliberative Process

OMB’s second rationale for requiring secrecy in the intra-executive budget context—protecting the integrity of the government’s decision-making process—similarly does not justify the full extent of secrecy employed though it, too, is rooted in valid and important concerns.

Preliminarily, the mere existence of a FOIA exemption for documents that reveal the government’s deliberative process—Exemption 5363—is no reason to require withholding of those documents. As Attorney General Holder explained in a memorandum sent to agency heads early in the Obama Administration, “an agency should not withhold information simply because it may do so legally.”364 OMB has thus disclosed material protected by Exemption 5 because “disclosure would not create a harm protected by that exemption.”365 Similarly, the executive order that governs regulatory review requires agencies to disclose both their original regulatory proposals and the substantive changes OIRA requested and to explain to the public “in a complete, clear, and simple manner” the differences between the two366 even though OMB during the Reagan Administration had successfully used Exemption 5 to block disclosure of much more information about the agency-OIRA regulatory process.367

The real question, therefore, is whether it would help or hinder deliberation to disclose predecisional budget documents that reveal the development of OMB’s and agencies’ thinking,368 in keeping with the purpose of Exemption 5.369 The deliberative costs associated with too much transparency often include entrenching positions rather than letting parties develop more nuanced ideas through conversation;370 silencing good ideas for fear of being publicly rejected or pilloried;371 and driving deliberation underground, further out of sight, through mechanisms developed to avoid whatever transparency regime is imposed on unwilling participants.372 These interests may be heightened in politically polarized times, when interest groups monitor their own party for orthodoxy and the other party for everything, and when Congress and the executive are at war with each other.

These are serious concerns that we should not dismiss lightly. However, there are costs to the current system of opacity with respect to accountability. As Lisa Heinzerling has noted in the context of OIRA’s lack of transparency, opacity in government limits “people from understanding the way their government operates, how they can intervene and at what points, what the government is up to, who is making important decisions, [and] why the government has made those decisions.”373

Moreover, the problems associated with too much transparency may already be present under the current non-transparent system. We just do not know about it. For example, agency officials and the RMOs may already find themselves in entrenched positions simply by virtue of the structure of their relationship.374 Agency officials may already silence good, creative ideas if they constrict their proposals based on what they think their program examiner wants to see.375 And the pyramid structure of OMB’s operation may drive deliberation underground. That is, program examiners may reach agreements with agency officials on certain matters on the condition that agency officials take certain actions; subsequently, those agreed-upon actions may not become an issue as the agency’s budget request moves up the chain of command within OMB.

These costs suggest that some recalibration of the current regime is worthwhile. In another context, the D.C. Circuit has rejected the contention that “the congressional goal of centralized budget formulation cannot be achieved without secrecy,” reasoning that the requirement that “the President submit a single, unified executive branch budget proposal to Congress for consideration” does not “require that the President’s proposals be the only budgetary information available to the public.”376 While formally discussing the public-meeting requirement for multi-member agencies, the D.C. Circuit used language about good public policy that could apply more generally: “disclosure of budget deliberations would . . . inform the public ‘what facts and policy considerations the agency found important in reaching its decision, and what alternatives it considered and rejected,’ and thereby . . . permit ‘wider and more informed public debate of the agency’s policies.’”377

Indeed, in this same litigation, OMB attempted to justify secret deliberations about the budget without distinguishing between multi-member and executive agencies. Its arguments relied on the general “importance of the budget process” and the fact that internal “budget discussions lead to presidential recommendations reflecting the President’s ‘best judgment of how the nation’s fiscal resources should be allocated to meet its future economic and social needs’” with respect to “‘vital policies and billions of dollars . . . at issue every year.’”378 The D.C. Circuit roundly rejected this argument and demanded disclosure: “The public can reasonably be expected to have an interest in matters of such importance.”379

The question then is how to design a more nuanced approach to transparency that would accommodate the public interest while still taking seriously the need to protect the deliberative process.

c. Disclosing Procedural Aspects of the Budget Process and Post-Deliberative Decisions

At the very least, it should be clear that OMB’s reliance on the interests implicated by the deliberative process applies only to the content of documents.380 It does not apply to the confidentiality that exists around the RMOs’ interactions with agencies and with outside interest groups or to other procedural aspects of the intra-executive budget process that remain hidden. OMB’s interest in protecting policy consistency by forbidding the disclosure of predecisional budget material also does not extend to these procedural aspects. More disclosure of this procedural information could be a valuable source of accountability because it would permit better monitoring of who is participating in the process and when.

In addition, OMB’s confidentiality rationales do not extend to the various post-deliberative decisions that are nonetheless not routinely disclosed, such as its budget execution decisions. For example, while apportionment requests and decisions are formal documents made on a standardized government form and transmitted via a standardized web-based portal,381 there is no public collection of apportionment decisions, much less apportionment footnotes, which makes it difficult for the public to track OMB’s directions to agencies. Similarly, there is no centralized compendium of OMB’s approved requests to Congress to transfer or reprogram sums, making it difficult for the public to track how agencies’ spending diverges from Congress’s original appropriations decisions or the extent of Congress’s approval and disapproval of these requests.382 Here, too, more disclosure would improve accountability.383

2. The Role of Civil Servants and Political Officials

A second concern about OMB’s control of agency policymaking through the budget process involves the players engaged in effectuating that control. Contrary to the usual understanding of power in the administrative state, where higher-level political officials have authority over both lower-level political officials and the civil service,384 civil servants and lower-level political appointees in OMB can supersede the policy goals of Senate-confirmed agency officials. This reversal of expectations impedes accountability and is exacerbated by the lack of transparency discussed above.385

One conventional concern about White House control over agency policymaking is that high-level political advisors close to the President may direct agency officials, whether political appointees or civil servants, to take actions that are illegitimate. For example, these advisors may direct officials to take actions going beyond the agency’s legal authority, or beyond the facts, in the service of a pre-ordained position driven by private political interests.386 More generally, these advisors may direct officials to act on the basis of “pure partisanship or raw politics” instead of some notion of the public interest.387 Another view is that the White House achieves some of these same goals by nominating ideologically partisan political appointees to head the agencies—appointees who will loyally align their policies with the President’s goals without being swayed by “civil-service-led resistance to their preferred policies.”388

Relatedly, the value of civil servants in agencies is thought to be their ability to “resist and redirect agency leaders intent on shortchanging procedures, ignoring or downplaying congressional directives or scientific findings, or championing unvarnished partisan causes.”389 “[U]nlike the political leadership beholden to a particular presidential agenda,” this view holds, “the civil servants are . . . generally understood to be animated by professional norms and legal commitments to fair administration and enforcement of the laws.”390

The structure and work of the RMOs complicate this view. On the one hand, it is civil servants, not political appointees, who take a front-line position in directing agency action.391 To be sure, these civil servants are also bound by professional norms,392 with loyalty to the institution of the presidency rather than to any political party.393 And most of the time, RMO staff work with agency staff, rather than directly with Senate-confirmed agency officials. But RMO staff hardly serve a checking function over political or politicized activity; to the contrary, their very job is to ensure that agency policy is consistent with presidential priorities.394 Their portfolio is vast and full of discretion, and they play an unusual role for civil servants in high-level policy decisions,395 especially with respect to their power over activities taking place elsewhere in the government.396 There is accordingly a danger, more than theoretical, that their role will be co-opted in the service of partisan action.397 Moreover, while senior RMO civil servants like branch chiefs and DADs often have deep policy expertise in the areas they oversee, that is not always the case for the more junior program examiners who tend to have front-line interactions with agency staff.398

On the other hand, while there are political appointees at the top of the RMOs, they are not Senate-confirmed.399 The PADs work at the “middle level of management,”400 between Senate-confirmed officials and civil servants. The PADs are even less transparent and more powerful than the so-called “czars” that received so much attention, much of it negative, in the early Obama Administration.401 They are less transparent because their appointment is routine, so their existence and portfolio is not scrutinized. Their work remains shrouded in secrecy because of the confidentiality lever. They are more powerful because they have clear responsibilities and duties that operate under statutory- and OMB-driven deadlines through all of the budget preparation, budget execution, and management levers discussed above. They do not fall under either category of political official presented in the conventional view402—they are neither senior White House political advisors nor Senate-confirmed agency officials.

One potential answer to this conundrum is that RMO officials are merely an extension of the President.403 It would thus not matter that middle-management political appointees and civil servants were playing a major role in controlling agency policy choices. This answer seems unlikely, however, both as a descriptive matter and as a legal matter.

As a descriptive matter, the hundreds of program examiners and their civil servant supervisors do not sit in the White House and have no regular contact with the President;404 their instructions to agency officials most often come from channeling the President’s expected views, sometimes as conveyed by the PADs.405 Given the politicization of agency officials documented by others,406 there is no particular reason to think that these civil servants better represent the President’s actual views than these agency officials.

The descriptive case for the five PADs as an extension of the President fares only slightly better. While the PADs are politically appointed and sit near the White House in the Eisenhower Executive Office Building,407 their access to the President is sporadic,408 and they do not have the same intimate advisory connection enjoyed by other executive officials sometimes viewed as speaking for the President. To be sure, PADs do have more regular conversations with White House advisors than agency officials do,409 but that does not ensure that PADs are actually speaking for the President.410 Given the information flow from the RMOs up to White House advisors and the President,411 it is possible that decisions made by White House advisors are actually just what the PADs (or even the RMO civil servant staff) advised them to decide. And some subset of decisions made by PADs does not get further elevated.412

As a legal matter, the fact that OMB is considered an “agency” under FOIA means that it is considered “substantially independent” rather than meant “solely to advise and assist” in a manner akin to “the President’s immediate personal staff.”413 Further, while the boundaries of the presidential communications privilege remain unsettled,414 one might see its narrow scope as related to the question of who speaks for the President. The D.C. Circuit has held that the privilege “should be construed as narrowly as is consistent with ensuring that the confidentiality of the President’s decision-making process is adequately protected,” and so it should be available only to “those members of an immediate White House adviser’s staff who have broad and significant responsibility for investigating and formulating the advice to be given the President . . . .”415 The few cases to have considered the question have concluded that OMB as a whole is too far removed from the President’s inner circle for its officials to receive the privilege.416 Even if the privilege were to apply to OMB in some instances, it seems unlikely that the RMO staff, whether civil servants or the PADs, are high level enough to qualify as “immediate White House adviser[s]” and their staff.417

The implausibility of regarding PADs and program examiners as extensions of the President has troubling implications for accountability, especially in light of the transparency problems discussed above. One way to secure accountability is through “the complex structure of the administrative hierarchies that constitute our basic mechanism for governing ourselves.”418 The structure of the RMOs and the RMOs’ interactions with agencies pose challenges for accountability at each level of the hierarchy.

At the agency level, one “[k]ey element[] of accountability” is “the requirement[] that administrators appear annually before Congress in order to justify their budget requests and respond to periodic demands from congressional oversight committees to explain and justify their decision making in public testimony.”419 But if all agency officials can offer is what OMB has told or permitted them to say, any reward or punishment the agency receives will not be fully grounded in reality. To hold agency officials accountable requires understanding what they wanted to do and what OMB told them to do.420

Accountability is also compromised within OMB itself, in large part because the public and Congress have no way of knowing what the PADs and RMO staff are doing behind the scenes. Like OIRA review, the RMO process “offers the tantalizing possibility of influence without fingerprints.”421 It is no answer to say that accountability is satisfied because the RMOs report through the Senate-confirmed OMB Director to the President, given the limited number of RMO actions that reach the Director, much less the President.

At the top of the hierarchy, the President can use the RMO process to avoid accountability. As Richard Neustadt observed almost sixty years ago, when “[t]he voice that speaks is not the President’s . . . [but] the Budget Bureau’s[,] . . . when need be, the Budget serves as whipping-boy.”422 Instead of claiming the RMOs’ decisions, the President can distance himself from the RMOs, “blaming ‘a nameless OMB bureaucrat five levels down from the top.’”423

3. The Policy and Political Implications of Technocratic Decisions

These critiques would matter much less if RMOs simply applied neutral expertise, if the budget were simply a dry document about numbers, and if program examiners were simply bean counters. But, of course, none of that is the case. “[B]udgeting is a political decision influenced by the political content of programs themselves and the political predispositions of key actors in the budgeting process.”424 Even if every decision is not a political one, neither is it a technical application of objective principles. It involves complex, value-laden decisions about how to “confront tradeoffs and project them into an uncertain future.”425 This reality underscores a third problem with the RMOs’ work: its complexity allows a technocratic appearance to obscure underlying substantive choices, thereby reducing accountability.426

Some substantive choices, such as tradeoffs among competing interests, might be appropriate for a budget because it is a public statement of national priorities. At the same time, it is not clear that those who are making the decisions are the right people to make those decisions, especially given the broad scope of authority held by the RMOs’ civil servants.427 Alternatively, even if making these decisions is a valid part of their job, it might not always be clear, even to them, that the decisions they are making are actually policy choices.428 Both of these problems reflect what Wendy Wagner in another context has called “the unintentional science charade”—the false belief that science alone can answer all of the questions related to scientific policy decisions, accompanied by the unintentional substitute of technocrats’ own unarticulated value choices at those junctures where science cannot provide an answer.429

Sometimes, however, substantive choices made in the budget are more problematic, such as those based on “pure partisanship”430 or political pressure beyond the public interest, at least when made by the RMOs and couched in the language of technocracy.431 Such decisions would be examples of what Wagner calls the “intentional” or even “premeditated science charade”—when “bureaucrats consciously disguise policy choices as science” or first make a decision and only then selectively introduce scientific evidence to justify it.432

To illustrate how substantive policy choices underlie technocratic-sounding budgetary decisions, consider, for example, a decision to cut back on NIH training grants, making them available only in fields with a shortage of researchers.433 This approach may seem like a neutral way to decide how to allocate a limited sum of money. But the program examiners asked NIH to justify its proposal by explaining the different scientific accomplishments it expected from training people in different fields.434 Making decisions on the basis of answers to these questions is necessarily a value-laden choice about the relative merits of different scientific outcomes. It could also lead to a more partisan decision about what kinds of outcomes would be acceptable.

The same set of circumstances can exist in the budget execution context. For example, grant competition priorities can appear neutral while in fact privileging certain sets of applicants, whether those whose work is favored on substantive policy grounds435 or those who are politically important.436 And examples of both value-laden policy decisions and the possibility of partisan decision making through the RMOs’ implementation of management initiatives abound, as I explained earlier.437

To be clear, value-laden decisions are perfectly appropriate in preparing and executing a budget, as well as in designing and overseeing management initiatives. Indeed, value judgments are inseparable from those activities. The problem arises when the language of technocracy obscures value choices, whether intentionally or unintentionally. Such technocratic cover hides the fact that people are making choices, conceals who is making them, and opens the door to partisan decision making. Under any of these scenarios, accountability suffers.

IV. responding to omb’s control of agency policymaking through the budget process

Any response to OMB’s control of agency policymaking through the budget process must be nuanced, mitigating the system’s problematic lack of accountability while protecting its valuable coordinating work. Section IV.A offers two variations on this effort from inside the executive branch, mapping out ways that the President and OMB itself could reform OMB’s work. Section IV.B sets forth potential responses from outside the executive branch, suggesting ways for Congress and civil society organizations to better engage with the RMOs’ work. Finally, Section IV.C broadens the lens to OIRA reform, explaining that the RMOs’ authority should be considered in any discussion of reforming OIRA, since the work of those offices is complementary and could involve spillover.

A. Inside the Executive Branch

To increase accountability of the RMOs, the President could issue an executive order that both sets forth how he or she intends to use the RMOs to work with agencies on setting and implementing policy and establishes various transparency requirements. At a smaller scale, OMB could also usefully take steps to increase its own transparency and engagement with the public.

1. The President

An executive order governing the RMOs’ work and making it more transparent would enhance accountability in two ways. First, the mere fact of its existence would provide an opportunity for Presidents to claim the RMOs’ work as their own. Second, the executive order’s substantive transparency requirements would provide opportunities for the public to better monitor the RMOs’ work.

As to presidential claiming, an explanatory executive order would enhance accountability by requiring the President to take ownership of the RMOs’ actions. New Presidents could put their own stamp on the process, furthering accountability. The executive order would parallel Presidents’ other executive orders detailing how they intend to use OIRA for regulatory review,438 a process that itself developed out of originally uncodified practices on the budget side of OMB.439

For all of the criticisms of its substance and implementation,440 the executive order governing OIRA’s regulatory review at least provides a sequence and scope of activities that the public can expect. The absence of such a document on the RMO side means that a set of offices more than four times as large as OIRA, with oversight over more of the federal executive establishment, operates with more opacity. Such a document would also set clear guidelines for the RMOs and for agencies beyond what Circular A-11 already requires, to the extent that it would set forth a high-level vision of budgetary policymaking and of the relationship between the RMOs and agencies. In a similar way, while various OMB circulars govern the details of OIRA’s work,441 the executive orders nonetheless provide an overall presidential vision of that office’s scope.

The substantive details of the executive order would also seek to increase the RMOs’ transparency.442 Several options for how to do so exist. The rest of this Section considers three possibilities, in order of their likely level of controversy: (a) transparency of procedural aspects of the budget process; (b) transparency of final budget execution decisions; and (c) transparency of predecisional budget preparation information.

a. Transparency of Procedural Aspects of the Budget Process

The least controversial option would require transparency as to procedural aspects of the RMOs’ work alone. The executive order could, for example, explain what the interactions between the RMOs and the agencies should involve over the course of the year, with a rough timeline of when the different steps will take place.443 It could clarify which kinds of policy decisions the RMOs will get involved with and which kinds they will not.444 It could clarify appropriate rationales for decisions.445 And it could require both logging and disclosure of meetings the RMOs have with entities outside the executive branch and preparation of summaries of their agendas.446

Requiring disclosure of the RMOs’ processes would do some work to increase accountability.447 Much of what we know about OIRA comes from the disclosure requirements about its procedures.448 Disclosing who meets with OIRA and when provides important facts about who influences the office and the scope of its power.449 Even where OIRA’s process disclosure is incomplete or misleading, the expectation of disclosure gives the public a metric against which to measure compliance, permitting the public to track discrepancies between OIRA’s promises and its reality.450 An executive order requiring similar process-based disclosure for the RMOs would therefore add value.

b. Transparency of Final Budget Execution Decisions

A second option would require increased transparency about budget execution decisions under formal mechanisms such as apportionment and requests to Congress about transfers or reprogramming. These are final decisions, so nothing predecisional would be released. Apportionment decisions are legal requirements and are subject to the Anti-Deficiency Act. Requests to transfer or reprogram have legal effect on agency action once Congress approves them. These requirements thus govern agency spending just as appropriations acts do, and their regular disclosure would serve similar values as publication of appropriations acts themselves. Disclosure would be especially valuable if the information is presented to the public in a way that permits targeted review by affected interests: in searchable formats, organized by agency and by subject-matter.451

This option might be more controversial than mere disclosure of the RMOs’ general process, in large part because it would reveal some substantive decisions made by the RMOs, potentially unsettling any claim that the RMOs do not make policy.452 On the other hand, the majority of these decisions are likely to be fairly straightforward, so disclosure might on balance support the RMOs’ contention that the bulk of their work is not policy-oriented. Either way, this information would be useful information to the public. Moreover, in the context of a presidential executive order asserting ownership of the RMOs’ work, this disclosure would frame budget execution decisions as the President’s decisions rather than the RMOs’ decisions, just as the budget itself is. In so doing, disclosure might support presidential claiming, one of the goals of the executive order in general.

c. Transparency of Predecisional Budget Preparation Information

A third, even more controversial, option would increase transparency about the substance of interactions between the RMOs and agencies as to predecisional budget and policy deliberations.

A mandatory version of this transparency option would require disclosure of RMO-agency communications about budget and policy. For example, after the President’s budget is submitted to Congress and appropriations decisions are made, RMOs and agencies would have to make available, perhaps by posting online, the agencies’ original requests, any related documents from the RMOs, and a summary of what changes OMB made during the approval process. This mandatory disclosure would parallel what the OIRA executive order currently requires as to regulations,453 even though two of OIRA’s early administrators originally dismissed calls to make that information routinely public because of the possible harm to the deliberative process.454

A permissive version of this transparency option would simply allow disclosure of this substantive information. Agency heads would be permitted to disclose the original requests, related documents, and summary of changes to discuss the evolution of their thinking, their understanding of the implications of various funding levels and policy alternatives, and the policy directions they received during the execution process. The executive order could provide some guidance setting forth circumstances when disclosure would be permitted, such as the standard Congress has given to Inspectors General. Under that standard, Inspectors General must provide independent comments to Congress when they perceive that the President’s Budget request “would substantially inhibit the Inspector General from performing the duties of the office.”455 Or the executive order could give the agency head discretion in deciding when to disclose, providing broader opportunities to discuss policy differences.

Requiring the disclosure of the substance of RMO-agency interactions is likely to be more controversial, and ironically might not even accomplish its goal of improving transparency and accountability. For one thing, the equivalent mandatory disclosure in the OIRA process has little compliance.456 Moreover, full compliance might result in a data dump that would not be useful for the public.457 Mandatory disclosure might also push conversations between the agencies and the RMOs underground, resulting in fewer documents available for disclosure.458 Or it might lead agencies to use their initial budget requests to posture for their clientele rather than to make hard decisions themselves. This behavior might in turn give more power to the RMOs to construct workable realistic budgets. It might also lead Congress to exploit differences between agency goals and administration preferences for political gain, rather than for meaningful accountability.

Permissive disclosure might get around some of these concerns, especially because agency officials and the RMOs would not know ex ante what internal conclusions will be reached. Thus, there might be less opportunity to game the system and less danger of a data dump. At the same time, permissive disclosure could pose its own problems. Agency heads loyal to the President may not wish to disclose any policy differences at all,459 limiting the value of this intervention in terms of accountability.460 Alternately, agency heads may wish instead to use the potential for disclosure as a threat to gain leverage in the budget process, leading to worse decision making inside OMB and the White House in hopes of avoiding a public intra-executive dispute.

Because these potential issues point in opposite directions, a permissive disclosure regime might not be clearly problematic. Indeed, the same pressures that might dissuade agency heads from disclosing policy differences might also keep them from threatening disclosure as leverage; the removal power (for executive branch officials)461 and the power of relationships (for all appointees, whether in independent or executive agencies)462 might further limit the leverage problem. Moreover, the structural forces that might dissuade disclosure could make the disclosure that happens that much more useful to the public—less boy-who-cries-wolf and more watchdog. Even loyal agency officials sometimes find themselves wanting to share information that they believe would be useful to Congress and the public.463 A permissive regime could allow them to do so in a way that would appropriately calibrate the interests of transparency with the interests of the deliberative process.

At this preliminary stage of sketching what an executive order governing the RMOs’ work might look like, an acknowledgment that different administrations would likely take different positions on these various transparency options is more useful than a delineation of the ideal transparency regime. The choice among various options would itself give the public valuable information about the administration’s priorities.464

Overall, the lesson from the expansion of OIRA’s transparency obligations alongside its transparency problems is not that promoting transparency in the RMOs’ work is doomed to failure but rather that transparency is a goal worth pursuing, with the devil in the details.465 An executive order would provide a good opportunity to work these details out in a manner that would promote accountability.

2. OMB

Another set of reform possibilities lies within OMB’s own control. OMB itself can improve its transparency and increase its accountability, even in the absence of presidential claiming.466

First, OMB could provide more and better information online. Despite valid charges that OIRA’s dashboard is incomplete,467 the dashboard is nonetheless valuable for capturing at least some important information about what OIRA is reviewing from which agencies and how long draft regulations have been under review.468 This information both informs the public and allows for better public critique and engagement.469 In a similar capacity, OMB could present in visually helpful ways where the budget process is—for example, the steps being taken to execute last year’s budget, prepare this year’s budget, and plan for next year’s budget in addition to the status of congressional action. Instead, the website is largely a compendium of documents,470 making OMB’s work appear static and leaving out the important interactions OMB has with agencies throughout the year. OMB would not have to disclose everything about the substance of the RMOs’ work with agencies to make such a dashboard valuable; information about process and scope alone would be a big improvement.471

Second, OMB could solicit input from the public on its major policy choices, targeting underrepresented voices.472 The challenges of both engaging the public and gleaning information that is likely to be useful for government decision making are well documented.473 But OMB could do more than it currently does, particularly with respect to policies that are government-wide and not likely to be the subject of large-scale notice-and-comment rulemaking anywhere.

For example, in the new Evidence and Evaluation Agenda, which is having a substantial effect on domestic social policy,474 OMB has not publicly engaged with critiques of evidence-based policymaking that characterize the approach as unreasonable and misleading.475 It is possible that OMB is aware of the critiques and has simply discounted them,476 but the public has no way of knowing that OMB has done so and no mechanism to engage with or suggest improvements to the policy. Public comment on individual agencies’ proposed regulations that incorporate OMB’s evidence-based policymaking requirements is no substitute, as there is little chance that OMB will shift the direction of the government-wide initiative based on an objection raised in a particular rulemaking. Soliciting early feedback on large-scale government-wide policy choices could both improve the quality of OMB’s decision making and enhance OMB’s accountability.

B. Outside the Executive Branch

Structural reforms from within the executive branch are not the only potential responses to the accountability concerns related to the scope and extent of OMB’s ability to control policy through the budget process. Both Congress and civil society have an important role to play as well.477

1. Congress

Congress took a more active role in attempting to oversee the work of OMB in the 1970s478 and of OIRA in the 1980s479 as the scope of those offices grew and as Congress grew concerned about reports of their politicization. That is not to suggest that the RMOs are currently particularly politicized, either in the abstract or in reference to any particular administration. But the scope of the RMOs’ policymaking effect is large, and the potential for politicization is present. As such, Congress should increase its monitoring and oversight across administrations and do so in a public forum that permits citizen review. It could do so by attempting to get more information about the extent of the RMOs’ work either from agencies or directly from the PADs themselves.

If Congress were to focus on agencies, it might request agencies to provide information about policy alternatives or designs that were considered and rejected in addition to the congressional justifications submitted to Congress along with the President’s overall budget. Congress might also ask this question of officials testifying during appropriations hearings. Given the confidentiality lever, however, these requests are not likely to produce much information—unless the executive branch has committed to the disclosure of predeliberative budget information discussed above480—as agency officials’ responses would be cleared by OMB.

Congress might instead turn to the PADs, seeking to learn through their testimony how the RMOs influenced agency policy goals. The PADs are not currently among the OMB officials who testify before Congress.481 Typically, as is the case with White House staff members in general, congressional testimony is reserved for, or at least standard for, those officials who are confirmed by the Senate,482 which the PADs are not. Occasionally, though, Congress has created an OMB position that is subject only to presidential appointment, and yet the official in that position is still expected to testify, as is the case for the Chief Information Officer, who runs the Office of E-Government.483 Accordingly, Congress could choose to require PADs to testify before Congress when asked.484 This requirement would speak to two of the three accountability concerns discussed above: it would make the scope of the RMOs’ work more transparent by bringing to light the work that these offices do, and it would help make more perceptible the values-based decisions underlying seemingly technocratic budget work.

Alternatively, Congress could also require Senate confirmation for the PADs, which would additionally address the third accountability concern: it would limit the elevation of a low-level political appointee over Senate-confirmed officials in agencies. Beyond securing the PADs’ testimony, the extra requirement of Senate confirmation would provide an opportunity to probe the PADs’ different policy commitments and goals. In so doing, it would increase both the transparency and the accountability of the RMOs’ work.485

Nevertheless, this proposal for Senate confirmation is not a perfect solution. For one thing, it runs counter to trend. The Presidential Appointment Efficiency and Streamlining Act of 2011 removed the requirement of Senate confirmation for 163 positions in the executive branch.486 This Act was in keeping with a growing body of scholarship critiquing the appointment process as cumbersome and full of delays487 and as leading to too much politicization of the executive branch.488

But the positions recently removed from advice-and-consent were generally those with “little or no policy role” or “lower-level or administrative positions.”489 Assistant secretaries for public affairs, directors and deputy directors of single-issue bureaus, and members of various boards and advisory councils are among those no longer subject to Senate confirmation.490 The PADs exercise significantly more policy authority than these positions do. To be sure, also eliminated from Senate confirmation were a number of assistant secretaries in low-level management,491 which arguably bear some parallel to the PAD positions. But other management and budget positions still require Senate confirmation,492 as do most policy positions with which the PADs interface.493 The recent reduction in Senate-confirmed positions does not, then, suggest that it is implausible to imagine Senate confirmation for the PADs. Similarly, while Congress might not wish to highlight the importance of the PADs through the confirmation process, in an effort to downplay the importance of the President’s budget as compared to its own appropriations authority, Congress might nonetheless value the enhanced oversight that comes with the ability to question directly.

Of course, this enhanced oversight is exactly why presidentialists, not to mention the President, would likely resist this proposal. And that reaction would have some substantive merit. The delays associated with the appointment process,494 one of Congress’s concerns in passing the 2011 Act,495 could hamper OMB’s ability to meet the deadlines associated with putting the President’s budget together, especially when a new President takes office. It might result in submission of less careful, less vetted proposals to Congress. These delays could result in vacancies at the head of the RMOs and, therefore, decreased accountability,496 since there would be no one answering to the President at the helm of the RMOs other than the OMB Director and Deputy Director. Further, these two leaders could be so overloaded with PAD-level decisions that higher-level work would suffer or they could simply ignore the PAD-level decisions and delegate them to the civil servant DADs and branch chiefs; neither result would provide any accountability benefit.

Similarly, the cumbersome nature of the confirmation process could deter high-quality applicants from pursuing the PAD position as opposed to other policymaking positions in the EOP that require no confirmation, such as the policy councils. The process of confirmation could itself enhance the politicization of the positions rather than cabin it,497 leading PADs to make decisions that are rooted more in politics than in the merits.

At the same time, these problems are not sure to arise with any consistency, and when they do, there might be workarounds. For example, on the issue of delay, the Senate has often confirmed OMB Directors efficiently,498 and it might extend the same courtesy to the PADs in light of Congress’s own interest in the budget process. Presidents could also request that candidates for the PAD positions commit to staying in place for a significant length of time if confirmed, thus reducing frequent turnover and associated delays.499 Review of agency budget requests among the civil servants in the RMOs would be less affected by delays in the Senate, and some short-term slippage in accountability (due to branch chiefs and DADs making more of the decisions) might be tempered by the increased systemic accountability a confirmation requirement would bring.

Although the increased burden associated with Senate confirmation might deter qualified candidates, it might also expand the pool, given the prestige associated with Senate confirmation. Nor would Senate confirmation necessarily increase politicization; the PADs are already political appointees who serve at the pleasure of the President. Requiring their confirmation would arguably make the fact of their work more public without increasing their politicization.

Thus, despite some justifiable skepticism, there are good reasons to think that Senate confirmation for the PADs would be valuable. At the very least, the possibility is worth further discussion, especially in light of precedent for turning OMB’s high-level policy positions into Senate-confirmed ones. The Director and Deputy Director of OMB were not originally subject to Senate confirmation, but Congress turned them into Senate-confirmed positions as the scope of their policymaking authority grew and as the President started to use OMB more politically.500 Congress similarly turned the Administrator of OIRA into a Senate-confirmed position out of concern that the position’s vast authority required more congressional oversight.501 The analogy is not perfect because these positions all began as congressionally created positions, unlike the PADs, but the history reflects congressional interest in expanding oversight of OMB. Indeed, two other Senate-confirmed positions in OMB oversee offices that are much smaller than the RMOs and have a narrower purview.502 Against this backdrop, making the PAD positions Senate-confirmed in an effort to enhance transparency and accountability could be a natural evolution.

2. Civil Society Organizations

Civil society organizations could also take steps both to monitor and influence OMB in an effort to improve the system’s accountability.

As to monitoring, civil society organizations could expand their oversight of what is already public about OMB’s actions through the budget process. For example, it is typically a major news story when the President releases the budget, but the OMB directors’ release of budget or other memoranda is not often a story, at least not outside the Beltway. It should be.

Civil society organizations should call for more transparency in the RMOs’ process overall, including on the budget execution side. Civil society groups might even play a role in bringing about the kind of pro-transparency executive order proposed above.503 Because a President once in office might find it difficult to resist the status quo of the non-transparent RMO process, open-government groups might work to secure a campaign promise to commit to such a reform.

Moreover, if the RMOs are making policy, it is important to ensure the RMOs are hearing from a broad base of interests.504 OMB budget review is an insider’s game.505 There is a small group of D.C. lobbyists with specialties in OMB, who serve a client base that is likely dominated by well-heeled corporate clients.506 While less is known about the RMOs’ meetings with outsiders because there is no requirement to document them, there is no reason to believe that the imbalance is any less present with the RMOs than it is with OIRA.507

Civil society organizations could help redress this likely imbalance. Publicizing the importance of what the RMOs do and how to engage with them would be one small step towards encouraging greater participation. More broadly, civil society organizations that already meet with agency officials and congressional staff members on policy matters could ensure they have the RMOs in their sights as well. Former OMB officials could take on pro bono projects with civil society organizations seeking to influence the RMOs or run training sessions on how to incorporate the RMOs into a federal lobbying strategy. The goal would be not to politicize the RMOs but rather to ensure the RMOs are hearing from a broad base of affected interests. This strategy would ultimately increase accountability to the public as a whole, rather than to the segment of the public that knows how to gain access.

C. A Cautionary Note

Critics of OIRA, concerned that it has become too powerful, have sometimes suggested returning final rulemaking authority to agencies.508 Understanding the broader scope of OMB’s work through the RMOs should give these critics pause in suggesting the elimination of OIRA’s review of regulations as a cure for its ills.

Much of the effect OIRA currently has on agencies’ regulations could be implemented through the RMOs’ work on budget preparation, budget execution, and management.509 The approval lever and form-and-content lever could direct which regulations agencies should and should not prioritize, while the pyramid structure of the RMOs could let political officials maintain plausible deniability. The monitoring lever could ensure that agencies take the steps OMB directs. The Presidential Management Agenda lever could demand that particular regulations receive more attention than others. And the confidentiality lever could keep much of this secret. Affecting agencies’ regulations through these other means instead of through OIRA would simply drive OMB’s policy control even further underground.

To be sure, a President hoping to rescind the practice of OIRA’s regulatory review would be unlikely to shift the work to the RMOs. At the same time, a President might wish to capitalize on public praise for returning authority to the agencies while nonetheless gaining from the RMOs’ less public ability to effect control. A subsequent President might then wish to keep the whole process of influence secret. Alternatively, the RMOs themselves might fill the gap of their own accord.

Discussions about reforming OIRA should thus incorporate analysis of the RMOs’ authority to avoid the “‘whack-a-mole’ effect,” where a restriction on agency practice simply leads to experimentation to get around the restriction.510 Attention to the RMOs’ work more generally is critical for understanding OMB’s capacity to control the administrative state.511

Conclusion

This Article began with the observation from Paul O’Neill, former deputy director of OMB, that policy debates are “reflected, recorded, and battled over” in budget numbers and that “the numbers are the keys to the doors of everything.”512 By identifying and elucidating the levers OMB has at its disposal to control agency policymaking through the President’s budget process, I have sought to show that this observation is correct. It is through the budget that OMB finds itself “in the stream of every policy decision made by the federal government.”513 While OIRA’s control of agency policymaking through regulatory review is important, it is only one mechanism through which OMB may exercise policymaking authority over federal agencies.

Beyond this descriptive analysis, the Article sketched the various ways in which OMB’s budget work is simultaneously salutary and concerning. This work appears to go no further than is legally authorized, and it plays a valuable role in coordinating the expansive administrative state. But at the same time, by operating non-transparently, by giving so much discretion to lower-level political appointees and civil servants, and by making it possible for values-based decision making to be obscured by technocratic-sounding analysis, OMB’s Resource Management Offices present troubling challenges to accountability.

The Article therefore offered a series of potential reforms that would improve accountability while still maintaining OMB’s beneficial coordinating role: an executive order governing the RMOs’ process and requiring more transparency; increased OMB efforts to make its budget work more transparent and to engage the public on its government-wide policy decisions; greater congressional monitoring and oversight of the RMOs’ budget work; and expanded attention and engagement from civil society organizations.

In the end, however, the Article is not intended to provide the last word, but rather to open a conversation, on the President’s budget as a source of agency policy control.