LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... ·...

24
chapter 22 ............................................................................................. LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC AGENCIES ............................................................................................. george a. krause Delegation is a foundational characteristic of American representative democracy. Citizens delegate responsibility to legislatures, which, in turn, choose whether to delegate policymaking responsibility to the executive branch. Moreover, determining how institutions allocate policymaking authority is a puzzle of governance that extends well beyond the founding of the American republic. For example, to solve the constitution crisis in seventeenth-century England, the executive (Crown) was afforded the responsibility to offer spending proposals, while the legislature (Parlia- ment) possessed authority to both authorize and appropriate funds in response to these proposals (North and Weingast 1989, 818). In fact, the foundational roots of delegation decisions can be traced even further back to ancient biblical times (Bendor, Glazer, and Hammond 2001, 235; Bendor and Meirowitz 2004, 293). However, delegation of authority in the American political system affords both benefits and challenges. Separating power among different government institutions can diffuse conflict and thus induce stability requisite for effective, albeit imperfect, governance. Nonetheless, this diffusion of policymaking authority comes at the expense of an efficient system whereby policy change is challenging, electoral accountability

Transcript of LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... ·...

Page 1: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

c h a p t e r 2 2

.............................................................................................

LEGISLATIVEDELEGATION OFAUTHORITY TOBUREAUCRATIC

AGENCIES.............................................................................................

george a. krause

Delegation is a foundational characteristic of American representative democracy.

Citizens delegate responsibility to legislatures, which, in turn, choose whether to

delegate policymaking responsibility to the executive branch. Moreover, determining

how institutions allocate policymaking authority is a puzzle of governance that

extends well beyond the founding of the American republic. For example, to solve

the constitution crisis in seventeenth-century England, the executive (Crown) was

afforded the responsibility to offer spending proposals, while the legislature (Parlia-

ment) possessed authority to both authorize and appropriate funds in response to

these proposals (North and Weingast 1989, 818). In fact, the foundational roots of

delegation decisions can be traced even further back to ancient biblical times

(Bendor, Glazer, and Hammond 2001, 235; Bendor and Meirowitz 2004, 293).

However, delegation of authority in the American political system affords both

benefits and challenges. Separating power among different government institutions

can diffuse conflict and thus induce stability requisite for effective, albeit imperfect,

governance. Nonetheless, this diffusion of policymaking authority comes at the expense

of an efficient system whereby policy change is challenging, electoral accountability

Page 2: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

is blurred, and coordination among government branches often becomes exceed-

ingly difficult (Wilson 1973).

Scholars of U.S. political institutions have made both great theoretical and empir-

ical strides examining the causes of legislative delegation to the executive branch.

This literature reflects a strong affinity to a Madisonian conception of the separation

of powers insofar as the legislative branch is a separate institution with a unique

power base that distinguishes it from the executive. Although James Madison

(Madison, Hamilton, and Jay 1987/1788) in “Federalist 51” makes it abundantly

clear that this distribution of power is necessary to thwart abuse of authority, he

nonetheless states that the legislative branch is predominant by constitutional design

(see also Wilson 1973/1885; see Rosenbloom in this volume for different perspectives

on this claim). As a matter of fact, the preeminent role of Congress in the adminis-

trative state has been viewed as desirable on a normative level (Rosenbloom 2000).

It is thus hardly surprising that modern theories of legislative delegation view the

legislature as a principal (i.e., the supplier of policymaking authority) and the

executive as an agent (i.e., the recipient of policymaking authority). In this view,

the delegation problem simply entails the legislature making strategic choices

regarding when it should delegate authority to the executive versus when it should

not. Focused on this meta-question, scholarly research on legislative delegation has

developed both a sophisticated and a well-grounded view of how delegation works

from the legislature’s perspective in the form of (1) the tension between committee

and floor preferences (e.g., Epstein and O’Halloran 1999; Ferejohn and Shipan

1990), (2) the importance of legislative capacity (Huber and Shipan 2002) and

bureaucratic capacity (Huber and McCarty 2004), (3) the principal’s strategic

decision calculus under general conditions (Bendor and Meirowitz 2004), and

(4) the legislature’s ability to mitigate credible commitment problems through

delegation (Falaschetti and Miller 2001).

Generally, however, the key contributions made to the scientific study of legislative

delegation to the bureaucracy emanate from scholars whose primary interests are

intended to provide a better understanding of legislative institutions (see Moe 1990).

I argue that while this legislative-centric perspective has netted valuable insights into

delegation problems, it has also had important repercussions in limiting the devel-

opment of our social scientific understanding of this topic. In particular, the execu-

tive’s problemwhen it comes to delegation has typically been given short shrift in this

social scientific perspective, thus precluding a better understanding of delegation

problems through that lens (but see, e.g., Bertelli and Feldmann 2007; Krause 2003;

Lavertu 2009; Miller 2000; Moe 1990). Thus, although the marriage between the

Framers’ intent and our current treatment of legislative delegation is compatible in

normative terms, it misses the practical complications that arise for an executive

whose power and structure are purposely defined as ambiguous by the Framers

(James 2005). For this reason, greater attention and effort are needed to theorize

(both analytically and verbally) the role of the executive in the analysis of

decisions related to legislative delegation if scholars pursuing the scientific study of

bureaucracy are going to make meaningful future strides on this research program.

522 george a. krause

Page 3: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

In making this argument, I discuss and critique what are deemed three “genera-

tions” of legislative delegation literature in the modern era. “Modern” is defined as

research published over the last thirty years. My use of the term “generation” is not

meant to convey temporal order but rather shifts in the substantive focus of research

in the modern era. I start with the “first generation” of research on the core rationale

underlying the delegation choice confronting legislatures vis-a-vis executive discre-

tion. Approached as a principal–agent problem with implications solely for legisla-

tive choice, this first generation focused on legislatures’ balancing of the benefits and

costs associated with the delegation choice and identified various mechanisms for

maintaining accountability. However, while transaction costs were implicit in these

studies (see, for example, Dodd and Schott 1979; Fiorina 1986; Ogul 1976; Weingast

1984), they were not made explicit.

Next, I discuss and critique some of the major theoretical and empirical develop-

ments of a “second generation” of research in the legislative delegation literature. Still

within a principal–agent framework, this research placed the delegation choice

within a separation-of-powers framework that explicitly incorporated transaction

costs. Legislatures desire policy control, but they must and do temper this desire with

the need for greater policy expertise and institutional capacity derived from the

executive. Second-generation research also questioned the assumption that informa-

tion asymmetries always favor bureaucratic agents over their principals and intro-

duced the idea that bureaucracies also affect the delegation decision. They have also

shown that agencies are not totally reliant on elected principals for discretion.

I conclude with four recommendations regarding how the literature on delegation

studies can most fruitfully develop from a social scientific study of bureaucracy per-

spective. Serving collectively as a “third-generation” research agenda, each recommen-

dation seeks to advance Terry Moe’s (1990) call two decades ago for greater theoretical

incorporation of executive branch actors in delegation studies. My prescriptions are

distinct yet complement Moe’s by emphasizing the importance of more fully understa-

nding executive agents’ incentives and policy actions within the context of delegation.

THE “F IRST-GENERATION” STUDY

OF LEGISLATIVE DELEGATION

TO THE BUREAUCRACY................................................................................................................

As noted earlier, the modern delegation literature in political science views legisla-

tures as the “principal” and the executive branch as the “agent” (see Wood in this

volume for an extensive discussion and full-throated advocacy of the logic of

principal–agent theory). The determination as to whether the agent is the president

or bureaucracy depends upon the context by which delegation is being studied. As a

result, I will use both actors as agents throughout this chapter depending upon

legislative delegation of authority 523

Page 4: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

context. I argue that, from a normative perspective, the Framers of the U.S. Consti-

tution thought it desirable for legislators to act as “principals” by possessing the

authority to choose whether or not to delegate authority to executive branch policy

actors (“Federalist 51”). Because legislatures possess diffuse authority relative to

popularly elected executives, this view of delegation seeks to ensure against tyranny

by a singular (elected) executive. Furthermore, legislators are elected representatives

who are comparatively closer to the will of the people than are executive branch

agents, given legislators’ reelection motives. With this said, however, legislators

possess limited means to handle policymaking authority because the executive

branch possesses greater expertise and capacity in absolute terms. As a result, the

legislative branch has an incentive to delegate responsibility to the executive branch

under conditions deemed favorable to the former branch (see Workman, Jones, and

Jochim in this volume for an information processing-based normative theory of

bureaucracy that builds on this insight).

More specifically, the benefits of legislative delegation are: (1) increased policy

expertise resulting in better policy outcomes; (2) greater time and resources that can

be spent on other more preferred legislative activities, since oversight and monitoring

are costly with little electoral return; (3) mitigation of policy responsibility, and hence

blame, for unpopular policies or outcomes—especially when the opposition party

controls the executive branch; and (4) mitigation of dynamic problems relating to

policy instability and a lack of credible policy commitment. Meanwhile, the costs of

legislative delegation, in principal–agent terms, entail a loss of policy control and/or

effectiveness generated from adverse selection problems (i.e., delegating to the “wrong”

agent) and moral hazard problems relating to bureaucratic drift (i.e., agent shirking).

In this section, I review the intellectual development of the first generation of

modern delegation literature. Examined using the lens of how legislatures theoreti-

cally weigh the benefits and costs of delegation are four primary foci of that research

genre: oversight mechanisms as latent controls, legislative “stacking of the deck”

through procedural controls, budgetary controls on policy discretion, and structural

controls on agency discretion. While Dan Wood touches on each of these areas,

I restrict my attention to how they relate to delegation problems and also discuss

some of these limitations in the extant research program.

Oversight Mechanisms of Policy Control

First-generation research on modern delegation was predicated on explaining two

puzzles. First, can legislators control policymaking and implementation when they

apparently spend little effort at such tasks? Further, can legislators design institu-

tional mechanisms (via rules, procedures, and structures) to ensure that those agents

possessing delegated authority act in a manner consistent with legislative prefer-

ences? In tandem, both puzzles are at the core of the modern study of legislative

delegation. The former puzzle pertains to the optimal minimization of monitoring

costs incurred frommoral hazard (i.e., agency loss resulting from shirking), while the

524 george a. krause

Page 5: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

latter puzzle pertains to maximizing incentive compatibility designed to ensure that

executive branch agents act in a manner consistent with legislative intent.

Noting the demise of legislative oversight activity during the postwar era, many

congressional scholars claimed that legislative oversight is a low payoff activity for

legislators seeking electoral rewards and policy influence (Aberbach 1990; Ogul 1976;

Scher 1963). This, however, did not necessarily mean that legislative abdication of

policymaking had occurred. In a series of published works, MathewMcCubbins (and

his various collaborators) explored the incentives legislatures had for delegating

policymaking authority to the executive branch, without abdicating control over it.

In the realm of legislative oversight, members of Congress created “fire alarms”

that would enable organized interests to notify them of when executive branch actors

were deviating from legislative intent (McCubbins and Schwartz 1984). The use of fire

alarms allows legislators to save their scarce time and resources for electoral and

policy activities that accrue higher payoffs to them. The logic of fire alarm oversight

means that legislators can minimize investment in monitoring costs (oversight

activities) while simultaneously maximizing incentive compatibility with organized

interests so that the latter can provide surrogate labor in monitoring executive action.

In the event that organized interests have similar preferences to those of the agency,

yet diverge from those of the legislature, then fire alarm oversight may be ineffectual

since the principal’s (legislature) and de facto monitor’s (organized interests) policy

preferences are not aligned with one another.

David Epstein and Sharyn O’Halloran (1995) have extended this logic by claiming

that, in equilibrium, no fire alarm is touched off. Moreover, they claim that legislators

can effectively control multiple competing interest groups by making decisions based

on the range of moderate groups that straddle along the lines of support/no support

for agency actions. Therefore, a legislature may find it easier to ensure compliance

when interest groups are biased against agency policy than when they share identical

preferences. However, this logic assumes, first, that there are multiple groups so that

organized interest power is sufficiently diffuse. When power is not diffuse, then the

legislature may experience agency losses from having an “opposing” organized inter-

est that does not share its own policy aims. Second, the above logic assumes that

organized interests possess equal power. If some groups, however, are advantaged in

the struggle for obtaining policy benefits, one cannot presume that the legislature will

split differences down the middle among moderate factions if the more powerful

groups oppose legislative preferences. More generally, a focus on the concentration/

diffuseness of interest group power and their relative preference location vis-a-vis the

legislature and agency is critical for further extending the logical implications of fire

alarm oversight beyond what we know at present.

Procedural and Rule-Based Control

The specific issue of procedural and rule-based constraints on the exercise of

bureaucratic discretion is critical for understanding the precise mechanisms

legislative delegation of authority 525

Page 6: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

intended to obtain agency compliance. As Wood points out in his chapter,

McCubbins, Roger Noll, and Barry Weingast (subsequently McNollGast) (1987,

1989) claimed in a seminal pair of articles that legislators can minimize monitor-

ing costs while maximizing incentive compatibility by constraining future execu-

tive action via the enactment of rules and procedures (see also Moe 1989).

Procedures and rules, in particular, can be used not only to mitigate information

advantages enjoyed by bureaucratic institutions but also to embolden important

legislative constituencies (i.e., organized interests). Moreover, by legislators allowing

those constituencies to participate in agency activities, organized interests can ensure

that the agency is responsive to the legislators’ preferences (McNollGast 1987, 244).

In this manner, legislatures rely on the electoral connection between them-

selves and organized interests who are beneficiaries of policymaking as a means

of incentive compatibility designed to reduce Congress’s efforts at labor-inten-

sive oversight activities. “Deck-stacking” thus means that the legislature is

essentially delegating its monitoring function to loyal organized interests that

possess shared incentives. Because the Administrative Procedure Act of 1946

standardized baseline administrative procedures across all federal agencies, Con-

gress can more easily constrain executive policymaking authority by requiring

all promulgated rules to pass enhanced legislative scrutiny and uniform guide-

lines for procedural compliance (McNollGast 1987, 255–8; see the chapters in

this volume by Rosenbloom, Mashaw, and Kerwin and his associates for an

extensive discussion of this dynamic, divergent perspectives on it, and its

consequences for American bureaucracy). Such procedural constraints are tight-

ened when executive agencies are granted more policy responsibilities and also

when policy conflict among legislators rises (McCubbins 1985).

The key insight drawn from this particular line of inquiry is that much of the

political control that takes place is latent (i.e., ex ante) insofar as agencies choose

policies sufficiently closer to legislative preferences in order to avoid sanctions (e.g.,

Calvert, McCubbins, and Weingast 1989). Yet the empirical evidence testing legis-

lative control of the bureaucracy via rules and procedures has, at best, provided

mixed evidence in favor of the deck-stacking hypothesis (e.g., Balla 1998; Hamilton

and Schroeder 1994; Potoski 1999; Spence 1997). Perhaps one reason for the paucity

of clear and consistent evidence supporting the efficacy of procedural constraints as

conceptualized by McNollGast is that it fails to consider legislators’ desires to trade

off between statutory (ex ante) and oversight (ex post) controls (Bawn 1997). That

is, legislators not on the relevant standing committee prefer statutory control to

diminish the power of “cozy iron triangles,” while those members serving on the

relevant standing committee prefer oversight control whenever they can accrue

benefits from constituents for their agency casework activities (Bawn 1997, 119).

Still another reason for the chasm between theory and evidence in the realm of

procedural rule-type controls is that they can be circumvented by executive agents

for reasons discussed later in this chapter (see section on “third-generation”

research).

526 george a. krause

Page 7: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

Budgetary Mechanisms of Policy Control

In the realm of budgetary politics, D. Roderick Kiewiet and Matthew D. McCubbins

(1991) argue that Congress’s desire to delegate construction of the U.S. federal

government budget to the executive branch is not abdication (also see Mullins and

Mikesell in this volume for a discussion of how political rationality and technical

rationality intersect in this process). According to the logic of their bilateral bargain-

ing veto model, presidents are constrained in offering budget proposals that exceed

what Congress wishes to appropriate. This is because the president’s veto authority

regarding appropriation bills can only be effective when Congress prefers a higher

level of spending. Therefore, the president’s veto authority is rendered ineffective

when executive budget proposals exceed congressional appropriations. Alternatively

stated, Congress gladly cedes discretion to executive budgetary preferences so long as

the executive request does not exceed legislative appropriations. Whenever this

condition is not met (i.e., the president wishes to spend more than Congress), the

legislature takes the reins of budget policymaking away from the executive branch.

Subsequent research by Nolan McCarty (1997) has shown that a president’s veto

reputation is not immutably fixed through time but instead wanes across the lifecycle

of the administration. In a related vein, Brandice Canes-Wrone (2006) shows that

popular support of the president can afford chief executives a tangible advantage in

extracting the resources that they desire from the budget process. In sum, then, the

legislature enjoys the upper hand in presidential–congressional budgetary relations,

albeit it can be conditionally weakened by popular presidents during the early

portion of their tenure in office.

While this work tells us a great deal about legislative delegation in terms of

presidential influence over budgetary outcomes, it does not speak to how budgetary

outcomes are translated into bureaucratic action. Furthermore, even if the legislature

does possess strong discretionary authority over resources between these two

branches, this policymaking tool may be of limited influence due to an inherent

tradeoff between desirable policy outcomes that provide additional (slack) agency

resources and less preferable outcomes that come at a lower cost (Ting 2001).

Moreover, the boundedly rational nature of information processing yields a sluggish

agency response to congressional budgetary signals (Carpenter 1996; also see Work-

man and his associates, as well as Bendor and Hammond, in this volume on the

impact of bounded rationality generally). Further, agency policy outputs endoge-

nously affect resource growth. Thus, legislators are merely reacting to bureaucratic

autonomy displayed by administrative agencies (Krause 1999). Therefore, even if

legislators enjoy a robust position in determining budgetary outcomes as presumed

in legislative delegation models, how this is translated into bureaucratic policymak-

ing remains rather ambiguous as it relates to policy control.

Recent work by Jason MacDonald (2009) has offered a novel extension of the

legislative delegation literature in the realm of budgetary control. MacDonald claims

that past studies understate legislative control of the bureaucracy since they fail to

consider “limitation riders,” which forbid agencies from spending money on specific

legislative delegation of authority 527

Page 8: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

tasks, as a tool of political control. As a result, the importance of bureaucratic

expertise as a mechanism for enhancing bureaucratic discretion is overstated, while

preference configurations as a mechanism of legislative incentives understate politi-

cal control in existing theories and tests of legislative discretion.

MacDonald’s logic centered on limitation riders provides a more stringent (and

cleaner) test of legislative control over the bureaucracy than those focused on proce-

dural constraints because these can be more easily circumvented by organizations

relative to exacting resource constraints (Pfeffer and Salancik 1978). Nonetheless,

since limitation riders are predicated on inaction, political control through this

mechanism presumes that agencies monotonically desire more tasks (e.g., enforce-

ment). This is not always the case since regulatory or administrative retrenchment may

occur (e.g., Durant 1992; Nathan 1983; Waterman 1989). In such instances, what may

appear as legislative control via limitation riders will be observationally equivalent to

the agency’s preferred outcome. Also, while these limitation riders may signal political

control through bureaucratic inaction, they may be observationally equivalent to intra-

branch conflict between presidents (via political executives) and agencies (via agency

executives and other careerists). Therefore, while this test adds an important advance

to our understanding of legislative choice in delegation studies, it remains far from

being a definitive test of legislative control over the bureaucracy.

Structural Mechanisms of Policy Control

Finally, first-generation studies also emphasized that delegation choices were affected

by venue options: the choice between entrusting policymaking authority to an

executive branch agency or an independent commission (Bernstein 1955; Fiorina

1982, 1986). I hasten to add that research has gone beyond this choice in institutional

venue. Legislators also can determine whether to divide policy tasks among two or

more agencies or let a single agency possess sole jurisdictional authority over it (see

Ting 2002). However, I shall focus primarily on research studying the executive

branch/independent commission choice.

Independent commissions enjoy greater insulation from executive politics due to

having multi-member leadership, staggered appointment terms that do not coincide

with a president’s tenure in office, and, in some instances, partisan balancing require-

ments (Lewis 2003). Executive agencies are muchmore subject to presidential influence

since leadership at the agency comprises a single agency head, presidents can “hire and

fire” agency heads at will, and these agencies are subject to legislative and budgetary

oversight restrictions by the Office of Management and Budget (see Durant and Resh,

as well as Rosenbloom, in this volume for extensive discussions and critiques of efforts

to “presidentialize” the federal bureaucracy). Indeed, counterintuitively, Anthony Ber-

telli and Sven Feldmann (2007) even argue that presidents have an incentive to appoint

political executives whose preferences are less closely aligned to their own in order to

offset pressures from organized interests. It comes as no surprise, then, that legislatures

seeking policy control typically prefer independent commissions to executive agencies

528 george a. krause

Page 9: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

as the institutional structure that houses policy administration activities (McNollGast

1987, 1989; Moe 1989). Moreover, from an interest group perspective (that legislators

certainly pay attention to), Rui de Figueiredo (2002) makes the positive assertion that

interest groups’ preference for insulating policy structures will occur when they are

politically weak and, hence, do not expect to have sufficient influence over the gover-

nance of the administrative agency in question.

Empirical evidence on the executive/commission choice shows that this logic is

borne out. If left to their own devices, legislatures prefer to “hardwire” agencies by

creating institutional structures that are purposely insulated from executive influ-

ence—especially if their policy preferences diverge from one another (e.g., Lewis

2003; Volden 2002a; Wood and Bohte 2004). Furthermore, insulated institutional

structures have the added benefit of being better able to absorb policy conflict and

volatility arising from both politicians and organized interests (Lavertu 2009).

Stephan Lavertu (2009) extends this delegation logic by claiming that presidents

(or governors) will often prefer an insulated design so as to considerably reduce

administrative uncertainty induced from conflict among organized interests.

Besides setting ex ante constraints via institutional design choices, legislative

delegation decisions are also affected by the available venue options that the legisla-

ture has at its disposal at any point in time. Craig Volden (2002a) has formally shown

that legislatures are prone to delegating authority to executive agencies under unified

government when preferences between the executive and legislative branches are

aligned with one another; independent commissions are more apt than executive

agencies to receive delegated authority under divided government when these elec-

toral institutions possess divergent preferences (see also Volden 2002b). Volden’s

distinction between type of delegatory institution is a vital contribution, since policy

delegation of legislative authority is not merely an “all or nothing” proposition and

because some minimal level of legislative delegation is necessary in practically all

policy matters. The delegation decision entails the legislature’s rational pursuit of

maximizing incentive compatibility by selecting the “correct” policymaking venue—

executive agency or independent commission—in accordance with the alignment of

legislative and executive branch preferences.

THE “SECOND-GENERATION” STUDY

OF LEGISLATIVE DELEGATION

TO THE BUREAUCRACY................................................................................................................

As noted, much of the first-generation modern delegation literature focused on the

legislature’s choice as driven by substantive policy concerns. As such, it implicitly

considered only the political costs to the legislature of delegation, ignoring the

legislative delegation of authority 529

Page 10: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

political costs and policy benefits of the presence or absence of bureaucratic exper-

tise. Moreover, it portrayed bureaucratic capacity as fixed, as bureaucrats always

enjoying information advantages over elected principals, and as bureaucratic capa-

city being exogenously determined by political principals rather than as potentially

dynamic and endogenously driven by the agencies themselves (at least on the

margins). Second-generation researchers realized that whether delegation was

grounded in bilateral bargaining behavior whereby the legislature is in a “favored”

position or organized interests perform de facto monitoring activities on behalf of

the legislature, the true costs of delegation were being underestimated. In this section,

I focus on two key areas of theoretical development pursued in second-generation

research related to the scientific study of bureaucracy: refinements of our under-

standing of (1) how delegation decisions are informed as well by considerations of

bureaucratic expertise and (2) how bureaucratic capacity is neither always a “given”

nor exogenously determined. Both illustrate how a broadening of the conceptualiza-

tion of the transaction costs of delegation have advanced theory building on the

delegation of authority to American bureaucracy.

Getting Beyond Policy Alignment

The second generation of modern delegation literature has typically centered on

legislators balancing the tension between working with bureaucrats (whose prefer-

ences are often not closely aligned with their own) and delegating their policymaking

authority to agencies in order to reduce policy uncertainty and enjoy the benefits of

policy expertise. This tension’s basic logic is best summarized by Epstein and

O’Halloran: “But it is incorrect to conclude that Congress wants only to restrict

agency discretion; legislators want to strike a balance between granting agencies too

much leeway and constraining them so tightly that there is no room to incorporate

bureaucratic expertise into policy outcomes” (1999, 27). Epstein and O’Halloran’s

basic argument regarding the range of delegation afforded by the legislature to the

executive is simple and powerful. When the principal’s (legislature) and the agent’s

(executive) policy preferences are sufficiently distinct under a separating equilibri-

um, the legislature will not delegate policymaking authority to the executive branch.

Conversely, when the principal’s and the agent’s policy preferences are sufficiently

similar under a pooling equilibrium, the legislature will delegate policymaking

authority to the executive branch (67–8).

Perhaps the most important testable implications derived from Epstein and

O’Halloran’s (1999) particular logic are twofold. First, divergence between legislative

(median floor) and executive (presidential) preferences results in a lower level of

discretion granted to public agencies (Proposition 2, 75). Second, greater policy

uncertainty (often but not always pertaining to task complexity and at other times

to blame avoidance) will also increase the odds that the legislature chooses to

delegate policymaking authority to the bureaucracy (Proposition 1, 75). Scholars

such as John Huber and Charles Shipan (2002) also persuasively claim that legislators

may possess a rational incentive to delegate even when executive branch preferences

530 george a. krause

Page 11: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

sharply diverge from their own. This is because low-capacity legislatures incur large

costs by not exploiting bureaucratic expertise in policy administration, costs that

outweigh any accrued benefits from maintaining policy control. This is true, they

argue, even when policy conflict between the executive and legislative branches is

robust and political uncertainty surrounding enacting coalitions is sufficiently high

due to electoral turnover.

In addition, recent theoretical research has begun to emphasize the institutional

capacity of the agent—viz., the bureaucracy. For instance, Huber and McCarty (2004)

offer an analytical model of delegation that explains what occurs when bureaucratic

capacity (defined as bureaucratic competence involved in policy implementation) is

low. Their theory predicts that low bureaucratic capacity not only yields poorer policy

performance resulting from lower technical expertise but also has the negative second-

order effect of making bureaucratic compliance to legislative intent less likely. Contrary

to the standard predictions arising from delegation theories, Huber and McCarty

predict that governments with low levels of bureaucratic capacity will choose to

delegate less authority to bureaucrats possessing similar preferences. This theoretical

result is of vital importance on a normative level. It suggests, first, that legislators do

care about successful implementation of the legislation they enact and, second, that

they recognize (as must researchers) that the vicious cycle of low bureaucratic capacity

cannot be addressed by the standard principal-based remedies of placing a premium

on political loyalty and oversight mechanisms. Rather, the solution to this problem

must entail much more costly structural reforms that enhance the governance capacity

of both elected and non-elected government institutions.

In turn, restrictive procedures accompanying delegation hinder the capacity of

public agencies to weigh all relevant policy information and thus make an unfettered

policy decision. Loose procedures, of course, adversely affect politicians’ capacity to

direct policymaking. The crux of Kathleen Bawn’s (1995) argument in this regard

offers yet another testable hypothesis: the legislature’s granting of discretion to a

public agency—in the form of relaxed administrative procedures—increases as

uncertainty over policy outcomes exceeds uncertainty arising from agency action.

While not directly related to the legislature but certainly having implications for

members, this tradeoff logic is further developed by Lavertu (2009) from the very

different perspective of the president. He implicitly offers a fourth testable hypothesis

by arguing that presidents will prefer to trade off policy control in favor of reducing

policy outcome uncertainty by requesting less policy discretion. They do so because

they do not wish to deal with the “costs” associated with interest group conflict and

the resulting policy outcome uncertainty that arises from it.

Rethinking Information Asymmetries

The theme emphasized in research surveyed to this point has assumed that the

institutional capacity of both political and bureaucratic institutions is immutably

fixed. Simply, this research theme assumes a generic legislature and a generic agency.

legislative delegation of authority 531

Page 12: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

Each is assumed to have a fixed level of institutional capacity. Moreover, by defini-

tion, the agency possesses greater capacity (in the form of technical expertise) than

the legislature, making delegation a potentially desirable choice for legislators. In

recent years, however, scholars pursuing second-generation research have been

examining how varying (as opposed to static) institutional capacity also affects the

delegation decision, as well as where incentives for building agency capacity (at least

at the margins) come from. The bottom line, however, is that assumptions about the

advantages of delegating authority to agencies because they hold greater fixed levels

of expertise (i.e., they are favored by information asymmetry) do not always hold and

thus also become part of delegation decisions.

For example, as noted, previous research assumed bureaucratic expertise was

exogenously determined vis-a-vis the legislature, with technical expertise and skills

attributed to bureaucrats primarily through the channels of organizational solidarity,

professional networks, and educational background (e.g., Carpenter 2001a; Mosher

1982; Rourke 1984; Wilson 1989). More recent scholarly attention has focused on

whether bureaucrats are more likely to alter their costly marginal investment in

acquiring additional information under a variety of circumstances related to their

political environment. Put another way, administrative agencies as organizational

bodies possess a reservoir of knowledge and expertise due to such things as organi-

zational continuity and standard operating procedures. However, agencies’ incen-

tives for devoting the marginal additional effort of being informed about a given

policy matter in ways that preserve information asymmetries over elected officials are

neither “givens” nor always endogenously determined by their political environment.

Matthew Stephenson (2007), for instance, argues that an “uninformed” agency

will, in fact, only increase its technical expertise when its leaders prefer to adopt new

regulations because of their political environment. Conversely, this means that

agencies are willing to make costly research investments in expertise (i.e., request

funding from Congress) when they prefer a new regulation to the status quo policy

and when they have a better sense of learning the true effects of that particular new

regulation. Sean Gailmard and John Patty (2007, 2008) also endogenize bureaucratic

expertise in a sophisticated dynamic legislative delegation model. They show that

civil service job protections and lower outside wage offers, coupled with granting

bureaucrats discretion over policy, will produce a regime of politicized competence.

Gailmard and Patty also provide an innovative theoretical contribution to the study

of delegation by focusing on agents’ implicit incentives attributable to career con-

cerns (for more general applications to public bureaucracy, see Alesina and Tabellini

2007; Dewatripont, Jewitt, and Tirole 1999). This differs markedly from the standard

approach, which posits that explicit incentives bound directly to ex ante and ex post

contractual-based mechanisms are dispositive.

Although these conceptions of endogenous bureaucratic expertise hold consider-

able promise by shifting focus from bargaining among politicians to those responsi-

ble for utilizing delegated policymaking authority, this nascent strand of inquiry

must tackle a few critical remaining issues before its promise can be fully realized.

First, the human capital stock residing within agencies is effectively ignored, since

532 george a. krause

Page 13: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

these theories are limited to examining only marginal investments in expertise that

reflect human capital flows. This focus on relatively small changes in bureaucratic

expertise can provide a highly misleading sense of bureaucratic capacity because

most of this commodity has already been exogenously established prior to the arrival

of individual bureaucrats in a given agency. Second, any incremental change in

human capital investment flows may be offset by agencies having to function

under a prescribed set of norms for administering policy consistent with the agency’s

mission or the broader professional culture (Carpenter 2001a; Rourke 1984; Wilson

1989). Third, and relatedly, this human capital stock (which occurs exogenous to the

legislative–agency bargaining relationship) varies considerably across agencies based

upon the structural demand for bureaucratic expertise and the institutional context

in which the exercise of bureaucratic expertise is insulated from political influence

(Lewis 2003; Moe 1989; Wood and Bohte 2004). Fourth, endogenous human capital

investment theories ignore variations in organizational stability and, in turn, their

effect on the marginal costliness of investment in expertise. That is to say, low

personnel turnover and routinized procedures lower the costly investment of addi-

tional information needed to augment bureaucratic expertise. Finally, by internaliz-

ing the costs of marginal information search to the agency/bureaucrats in question,

these models fail to consider explicitly the well-established informational role played

by organized industry and professional groups that are capable of lowering the

marginal costs associated with bureaucratic decision making (see Carpenter 2002,

2004; also see the chapters by Riccucci and Keiser in this volume for broader notions

of the role of representation, recapitalization of assets, and legal requirements).

Nonetheless, the study of endogenous bureaucratic expertise has considerable

promise for understanding legislative delegation since it creates a more sophisticated

portrait of incentive compatibility problems facing both principal and agent. It also

has helped refine delegation models and mindsets that assume fixed and uniform

information advantages for the bureaucracy. And by introducing the idea of a

proactive rather than merely reactive bureaucracy, it has helped propel a third

generation of research that gives bureaucracy a more agentic role in delegation

studies.

TOWARD A “THIRD GENERATION” OF RESEARCH?RECONSIDERING THE NATURE OF LEGISLATIVE

DELEGATION TO THE BUREAUCRACY................................................................................................................

Even with its limitations, the modern delegation literature has provided both a

constructive and an informative research program for understanding the allocation

of power among governmental institutions. From the core insights of McNollGast

legislative delegation of authority 533

Page 14: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

(1987, 1989) through the most elegant (and general) treatment of stylized delegation

models by Jonathan Bendor and Adam Meirowitz (2004), the modern delegation

literature has shed important light on the incentives, mechanisms, and circumstances

by which legislative control of, or delegation to, bureaucratic agencies occurs.

Specifically, legislators possess the conflicting incentives of: (1) having policy out-

comes come as close to their most preferred outcome (ideal point) as possible,

(2) seeing successful policy outcomes that are closely dependent on levels of bureau-

cratic expertise and discretion, and (3) doing all this while minimizing the costs

associated with both sides of these conflicting incentives.

Legislators use various mechanisms when delegating authority. These mechanisms

range from administrative rules and procedures to budgetary resources to institu-

tional structures. The general hypotheses that can be drawn from this extant litera-

ture are that legislatures should be more willing to delegate policymaking authority

to the bureaucracy when: (1) their (principal’s) monitoring costs rise, (2) the

incentive compatibility for the bureaucracy (agent) increases, (3) the policy conflict

between the legislature and agency declines, (4) policy uncertainty rises, (5) the

agent’s expertise and/or effort rises, and (6) an increasing desire exists to “lock in”

policy to solve credible commitment problems.

However, for advances in scientific theory building on legislative delegation choices

to occur, a more explicit and sophisticated treatment of the role of the executive

branch in bureaucratic policymaking has to advance beyond what has been done to

date. Because most of the advances in legislative delegation research emanate from

students of legislative politics in political science, it is hardly surprising that consider-

ably more effort has been expended in modeling the role of the legislature than of the

executive branch (Moe 1990). The extent of modeling of the executive branch in

separation-of-powers models of delegation is often relegated to providing a unique

ideal point for agency heads and presidents (for notable exceptions, see Bendor and

Meirowitz 2004; Gailmard and Patty 2007; Ting 2002). Such a focus, however, comes

at a considerable expense to theory. Specifically, the modern delegation literature is

primarily focused on understanding the supply of bureaucratic discretion, with little

explicit concern for either the demand for or actual exercise of bureaucratic discretion.

The importance of focusing on the agent’s problem in matters of delegated

authority is most articulately captured by Moe in the following quote:

But alas, not only is control destined to be imperfect in political practice, it is

destined to be doubly imperfect: there is slippage as groups try to control

politicians, and there is slippage as politicians try to control their bureaucratic

subordinates. Bureaucrats will therefore have a measure of autonomy, perhaps a

substantial measure—and they can use the coercive power of public authority to

pursue their own interests at the expense of their creators. (1990, 234)

Put simply, a fundamental criticism of the extant legislative delegation literature is

that it makes implied statements about bureaucratic policymaking. It does so by

telling us little about executive action and, hence, about what bureaucrats actually do

with the policy discretion that they are granted by the legislature and the resulting

534 george a. krause

Page 15: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

policy outcomes (for a similar point, see Durant and Resh in this volume). In

contrast, and as many of the chapters in this handbook attest, public administration

and public management scholars have focused largely on what bureaucrats do with

their discretion and much less on the conditions whereby their discretion is expand-

ed or restricted. In light of all this, I offer four broad recommendations for reorient-

ing the research attention of delegation scholars toward topics critical for our

understanding to advance appreciably on how delegation decisions get translated

into actual bureaucratic policy actions.

Recommendation # 1: Strengthen the Presumed

“Weak” View of Executive Authority

Existing theories of legislative delegation assume that grants of executive discretion

occur only within the legislative process via statutory means. In reality, presidents

can circumvent a legislature’s decision to withhold policymaking authority by

employing executive orders to grant themselves greater policymaking discretion

(Howell 2003; see Durant 2008, 2009, as well as Durant and Resh in this volume,

for constraints on these actions during implementation). In such instances, legisla-

tures are almost always powerless since they have little chance of passing legislation

that overturns unilateral executive actions (Moe and Howell 1999). Future research

on legislative delegation that explicitly accounts for the possibility of unilateral action

as a “side option” at the president’s disposal may unearth previously unmodeled

limits of legislative policy control. For instance, when legislatures do not have veto

override majorities in both chambers and the executive cares enough about the

policy in question, it is quite possible that a strategic president or governor will

circumvent the legislature’s delegation decision by issuing an executive order de-

signed to grant themselves greater policy control.

A second manner in which executive branch actors can be fortified in delegation

models is to account for how they endogenously create greater autonomy from the

legislative branch. Specifically, bureaucratic agencies at various organizational levels

can cultivate autonomy from political institutions through their own actions. These

include through selective recruitment of individuals who share the agency’s core

mission and through building a reputation for policy competence and credibility

(Carpenter 2001a, 2001b). Not only is bureaucratic autonomy distinct from the

bureaucratic discretion that is bestowed upon agencies by politicians, but it lies

outside a formal contractual relationship that is the embodiment of the canonical

legislative delegation problem (Carpenter 2001a, 16–18). The difficulty of remedying

this problem using the existing principal–agent-based framework rooted in explicit

incentives is that bureaucratic autonomy lies outside the contractual relationship

(Carpenter 2001a; Krause 1999; also see Frederickson and Stazyk in this volume for a

broader and biting critique of principal–agent theory more broadly).

Daniel Carpenter and I (2009) recently proposed a possible alternative theoretical

framework for studying such problems. An audience-based framework, it allows for

legislative delegation of authority 535

Page 16: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

bureaucratic agents to be motivated outside the explicit, formal contract with

political institutions. These “external” sources of influence include implicit contracts

such as career concerns (e.g., Alesina and Tabellini 2007; Dewatripont, Jewitt, and

Tirole 1999; Holmstrom 1982), reputation building among professional peer associa-

tions and groups (e.g., Balla 2001; Carpenter 2001a; Frumkin and Galaskiewicz 2004;

Krause and Douglas 2006), the vitality of local institutions directly affected by

bureaucratic policy choices (e.g., Scholz and Wang 2006), and organizational identi-

fication with both the agency’s goals and the expectations that arise from sources

beyond the parameters set by formal authority and contractual relationships (e.g.,

Akerlof and Kranton 2005; Hannan 2005; Simon 1991).

Certainly, public administration researchers have recognized for at least six decades

that these factors have both negative and positive effects on accountability (e.g., Finer

1941; Friedrich 1940; Mosher 1982; Khademian in this volume). So, too, have sociolog-

ical institutionalists placed legitimacy at the heart of organizational behavior (Hall

and Taylor 1996). Yet they have been lacking in legislative delegation research. Failing

to account for such “exogenous” shifts induced by bureaucratic agents’ political

legitimacy may erroneously equate high levels of policy discretion granted by the

legislature—reflecting latent political control—with an absence of bureaucratic au-

tonomy (Carpenter 2001a, 357). For legislative delegation research, such a focus

requires examining the endogenous symbiotic interaction of organized interests and

professional associations with bureaucratic agencies. This is a relationship distinctly

at odds with the current predominant perspective on theories of delegation, which

treats legislatures and interest groups as natural allies (e.g., the logic of fire alarm

oversight). Nor is this a trivial concern. If one views policy discretion granted by

legislators to bureaucrats as representing a minor portion of agency policymaking

authority (Carpenter 2001a, 17), then it naturally follows that the current state of

legislative delegation research still has considerable room for development so as to

enhance our understanding regarding the actual exercise of policymaking authority.

Recommendation # 2: Provide a Richer Portrait of

Hierarchical Relations within the Executive Branch between

Presidents and Administrative Agencies

As noted, most legislative delegation theories and empirical tests generally focus

on the nature of conflict and uncertainty regarding legislative–executive relations

(e.g., Epstein and O’Halloran 1999). Subsequent theoretical advances have distin-

guished between agency and presidential preferences (e.g., Volden 2002a). Delegation

research has also focused on hierarchical distinctions between floor and committee

preferences (e.g., Epstein and O’Halloran 1999; Ferejohn and Shipan 1990). Such a

focus on hierarchy within legislatures is valuable since it is essential for understand-

ing the focal point of coordination within a given institution (Cox and McCubbins

1993; Weingast and Marshall 1988).

536 george a. krause

Page 17: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

However, the legislative delegation research program has spent little effort on

examining these vertical or hierarchical coordination dilemmas within the executive

branch among presidents, political executives, and career executives. Because these

dilemmas are difficult to resolve (Miller 1992, ch. 5) and any particular solution has

unintended consequences (Hammond and Thomas 1989), addressing this matter is

a daunting challenge for students of legislative delegation research. Successful reso-

lution often requires trust and cooperation between superiors and subordinates

(Brehm and Gates 1997; Miller 1992), factors that are eschewed in existing legislative

delegation models. These models are typically centered on the dual pillars of formal

authority and coercive means—whether explicit or implicit—to obtain compliance.

Nonetheless, these hierarchal dilemmas are of even greater consequence in execu-

tive branch than legislative politics. Because actors at different levels of the executive

branch possess varying incentives and preferences (Krause 2009; Miller 2000),

vertical coordination problems make cohesive executive action much more difficult

than envisioned in current theories and empirical tests of legislative delegation. In a

recent paper analyzing the staffing of different rungs within an administrative

hierarchy, Stuart Jordan (2009) advances a clever argument. He argues that even if

bureaucrats at these different organizational levels (e.g., federal executives versus

street-level bureaucrats) possess identical preferences, they will exhibit internal

disagreement because their subunit environments are sufficiently different from

one another. This, he argues, is by organizational design by Congress as a means of

avoiding “extremist” administrative behavior that either favors or rejects client

requests in every instance.

Clearly, bureaucrats’ preferences do systematically vary across these organizational

subunits within public agencies (e.g., Aberbach and Rockman 2000; Seidman and

Gilmour 1986). Indeed, internal disagreements—and, hence, hierarchical coordina-

tion problems—are even more severe in practice than characterized by Jordan’s

(2009) conservative theoretical model. To understand properly how policy discretion

is actually used by bureaucrats, one must focus on such thorny intra-branch relation-

ships if the legislature’s delegation choice is going to be explicitly linked to actual

policy outcomes involving bureaucratic action.

Thus, while the view of a unified executive adopted in many delegation theories may

be consistent with constitutional design features (e.g., Epstein and O’Halloran 1999), it

does not closely mirror the practice of governance within the executive branch (West

2006). In reality, political executives (appointees) often serve as brokers between pre-

sidents seeking to exert their will over policy administration and the permanent career

bureaucracy (Aberbach and Rockman 2000). That is, formal authority has limits in

obtaining compliance in hierarchical relationships (Miller 1992, 120; also see especially

Durant and Resh, Bendor and Hammond, and Rosenbloom in this volume for diverse

perspectives on these dynamics). In U.S. executive branch politics, presidents must rely

on bargaining strategies to ameliorate the information asymmetries that they encounter

in relation to their staff and agency-level subordinates (Rudalevige 2005). Because the

severity of this information asymmetry problem increases as one moves from the

president to street-level bureaucrats responsible for enforcing laws and administering

legislative delegation of authority 537

Page 18: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

public policies (see Maynard-Moody and Portillo in this volume), the slippage that

occurs in the chain of command within the executive branch is more pervasive than

what transpires within the legislative branch among chambers, parties, and committees.

Further exacerbating executive branch coordination problems is the fact that pre-

sidents and political executives possess much shorter policy time horizons than do

both the career executives and street-level bureaucrats who comprise the permanent

bureaucracy. Presidential attempts at controlling bureaucratic policymaking are ham-

pered by their short time in office (Waterman 1989, 189). Likewise, the “revolving

door” for political executives (Heclo 1977) adversely affects both organizational conti-

nuity and expertise. This problem is further exacerbated by presidents who strategically

manage political executive positions through cyclical creations and reductions in the

proportion of appointees in order to establish policy and administrative control over

agencies (Lewis 2008; also seeWest 2006 for a contrarian perspective that challenges the

notion that presidents do, or can, think strategically in this fashion). This tension

between personnel with differing time horizons means that when legislatures choose to

delegate policymaking authority to the executive branch, it is quite uncertain how that

authority will be exercised.

The implications of intra-branch coordinationwithin the executive branch are thus

far-reaching for understanding the consequences of legislative delegation. This is

because delegation choices often have unintended policy consequences since hierar-

chical conflict and instability make the exercise of policy discretion allocated by the

legislature rather difficult within agencies. Understanding these intra-branch coordi-

nation dilemmas is necessary for properly modeling the linkage between policy

discretion supplied by the legislature and subsequent policy outcomes involving

bureaucratic action.

Recommendation # 3: Better Characterize the “Demand Side”

of Executive Discretion

As I have argued elsewhere (Krause 2003), classic treatises on public organizations view

discretion as a variable commodity that can be either spurned or embraced by admin-

istrators depending on self-interest (Rourke 1984, 41–2; Thompson 1967, 118; Wilson

1989, 251). Unfortunately, the modern delegation literature neither shares nor models

this understanding of policy discretion. Canonical treatments of legislative delegation

presume that the executive branch monotonically obtains greater utility from increases

in policy discretion (but, again, see Bendor and Meirowitz 2004 and Ting 2002 for

important exceptions). Yet contrary to the prevailing view of modern delegation studies,

bureaucratic agencies are, in fact, proactive in shaping the contents of legislation

(Rieselbach 1995, 212-14; see also Wilson 1989, 251). And in shaping legislation, several

reasons exist as to why bureaucratic agencies may seek less, not more, policy discretion.

Bureaucratic agencies may not only be risk-averse with respect to policymaking

uncertainty. They also may find themselves in a discretionary context where the

marginal net benefits associated with discretion are negative (Krause 2003, 46).

538 george a. krause

Page 19: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

Under these conditions, agencies may seek less discretion whenever the following

conditions exist: high levels of political fragmentation (e.g., divided government),

high task complexity, low organizational (agency) stability, and high issue salience.

In addition, the emergence and corresponding growth of the executive branch

arising from the modern institutional presidency (Burke 1992; Dickinson 1997) have

meant that many bureaucratic agencies are sufficiently occupied with present policy

and administrative tasks such that they are not in a position to want or seek

additional responsibilities. As noted, it is even possible that agencies may be seeking

less discretion as a way to economize scarce resources. At the same time, executive

branch actors may seek less discretion as a means to avoid conflict with and, hence,

blame from organized interests (Lavertu 2009). In addition, higher levels of legisla-

tive capacity relative to executive capacity may also result in the executive branch

actively seeking less discretion since they are less sufficiently equipped to handle

policymaking responsibility without restrictions emanating from the legislative

branch.

But if bureaucrats do not uniformly seek greater discretion (see below), what

implications does this have for understanding legislative control over the bureaucra-

cy? When the legislature chooses not to supply policy discretion to an agency, by

definition the appearance of political control is merely illusory if public agencies are

not demanding more discretion. Put differently, if bureaucrats want less policy

discretion (control) and legislators oblige by granting them less discretion, then

bureaucrats are the party getting their desired outcome. Conversely, if bureaucrats

want less policy discretion and legislators do not oblige by granting them additional

discretion, then the legislature is imposing its preferences for policy discretion over

the agency.

As a consequence, the need for future research to focus on demand for executive

discretion is critical. To do otherwise is to only know of legislative intentions without

directly observing those actors most proximate to policy outcomes—bureaucrats. A

greater focus on the executive branch may even eventually yield a unified theory of

policymaking in which legislative delegation is a key component that intersects with

executive action. At present, the delegation literature provides a very useful lens into

bureaucratic policymaking regarding the legislature’s incentives for allocating formal

policymaking authority to the executive branch. Nonetheless, this literature has yet

to constitute a holistic view of policymaking and implementation within the modern

administrative state.

Recommendation #4: Discern Whether or Not, and Why,

Delegation Makes a Policy Difference

Greater research attention to the first three recommendations might also provide an

ancillary benefit of insight into whether the commonly accepted “first-principle”

assumptions motivating the legislative delegation decision are, in fact, empirically

legislative delegation of authority 539

Page 20: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

valid. For example, does legislative delegation necessarily result in reducing oppor-

tunistic policy manipulation as commonly presumed in extant research? A recent

empirical study of official general fund revenue forecasts in the American states

conducted by David Lewis, James Douglas and myself (Krause, Lewis, and Douglas

2010) reveals that legislative delegation to either the executive branch (i.e., governors)

or an independent commission (i.e., consensus groups) will only insulate policy-

making from opportunistic policy manipulation when the political system itself fails

to check either legislative excesses that are attributable to unified party government

or executive myopia resulting from gubernatorial term limit restrictions.

If not, however, a more direct approach to assessing these assumptions is vital for

theory building to proceed profitably. Basically, these are the “so-what” questions

related to delegation. That is, does legislative delegation to a public agency truly lead

to superior policy performance? If not, why not? Further, does legislative delegation

to an insulated agency (e.g., independent commission) provide superior perfor-

mance vis-a-vis a less insulated agency under the direct aegis of the executive branch?

If not, why not? Does a lack of legislative delegation necessarily result in a loss of

bureaucratic influence over policymaking? If not, why not? Regardless of the ques-

tions asked or results discerned, however, viewing legislative-executive policymaking

relations as a two-way street is a necessary condition for advancing the scientific

study of legislative delegation in America.

REFERENCES

Aberbach, J. D. 1990. Keeping a Watchful Eye: The Politics of Congressional Oversight. Washing-

ton, DC: Brookings Institution Press.

——and Rockman, B. A. 2000. TheWeb of Executive Politics: Three Decades of the U.S. Federal

Executive. Washington, DC: Brookings Institution Press.

Akerlof, G. A., and Kranton, R. E. 2005. Identity and the Economics of Organizations.

Journal of Economics Perspectives, 19/1: 9–32.

Alesina, A., and Tabellini, G. 2007. Bureaucrats or Politicians, Part I: A Single Policy Task.

American Economic Review, 97/1: 169–79.

Balla, S. J. 1998. Administrative Procedures and Political Control of the Bureaucracy.

American Political Science Review, 92/3: 663–73.

——2001. Interstate Professional Associations and the Diffusion of Policy Innovations.

American Politics Research, 29/3: 221–45.

Bawn, K. 1995. Political Control Versus Expertise: Congressional Choices About Administrative

Procedures. American Political Science Review, 89 (March): 62–73.

——1997. Choosing Strategies to Control the Bureaucracy: Statutory Constraints,

Oversight, and the Committee System. Journal of Law, Economics, and Organization,

13/1: 101–26.

Bendor, J., Glazer, A., and Hammond, T. 2001. Theories of Delegation. Annual Review of

Political Science, 4/1: 235–69.

Bendor, J., andMeirowitz, A. 2004. Spatial Models of Delegation. American Political Science

Review, 98/2: 293–310.

540 george a. krause

Page 21: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

Bernstein,M. H. 1955. Regulating Business by Independent Commission. Princeton, NJ: Prince-

ton University Press.

Bertelli, A. M., and Feldmann, S. 2007. Strategic Appointments. Journal of Public Adminis-

tration Research and Theory, 17/1: 19–38.

Brehm, J., and Gates, S. 1997. Working, Shirking, and Sabotage: Bureaucratic Response to a

Democratic Public. Ann Arbor: University of Michigan Press.

Burke, J. P. 1992. The Institutional Presidency. Baltimore, MD: Johns Hopkins University

Press.

Calvert, R.,McCubbins, M. D., andWeingast, B. R. 1989. A Theory of Political Control of

Agency Discretion. American Journal of Political Science, 33/3: 588–610.

Canes-Wrone, B. 2006. Who Leads Whom? Presidents, Policy, and the Public. Chicago, IL:

University of Chicago Press.

Carpenter, D. P. 1996. Adaptive Signal Processing, Hierarchy, and Budgetary Control in

Federal Regulation. American Political Science Review, 90/2: 283–302.

——2001a. The Forging of Bureaucratic Autonomy: Reputations, Networks, and Policy Innova-

tion in Executive Agencies, 1862–1928. Princeton, NJ: Princeton University Press.

——2001b. The Political Foundations of Bureaucratic Autonomy. Studies in American Political

Development, 15/1: 113–22.

——2002. Groups, the Media, Agency Waiting Costs, and FDA Drug Approval. American

Journal of Political Science, 46/3: 490–505.

——2004. Protection without Capture: Product Approval by a Politically Responsive,

Learning Regulator. American Political Science Review, 98/4: 613–31.

——and Krause, G. A. 2009. Bureaucratic Reputation and Democratic Governance in the

United States: An Audience-Based Framework. Presented at the Annual Meeting of the

Midwest Political Science Association, Chicago, IL, April 2–5.

Cox, G. W., and McCubbins, M. D. 1993. Legislative Leviathan: Party Government in the

House. Berkeley: University of California Press.

de Figueiredo, R. J. P. 2002. Electoral Competition, Political Uncertainty, and Policy Insula-

tion. American Political Science Review, 96/2: 321–33.

Dewatripont, M., Jewitt, I., and Tirole, J. 1999. The Economics of Career Concerns, Part

II: Applications to Missions and Accountability of Government Agencies. Review of Eco-

nomic Studies, 66/1: 199–217.

Dickinson, M. J. 1997. Bitter Harvest: FDR, Presidential Power, and the Growth of the

Presidential Branch. Cambridge, MA: Harvard University Press.

Dodd, L. C., and Schott, R. 1979. Congress and the Administrative State. New York:

John Wiley.

Durant, R. F. 1992. The Administrative Presidency Revisited: Public Lands, the BLM, and the

Reagan Revolution. Albany: State University of New York Press.

——2008. Sharpening a Knife Cleverly: Organizational Change, Policy Paradox, and the

“Weaponizing” of Administrative Reform. Public Administration Review, 68/2: 282–94.

——2009. Back to the Future? Toward Revitalizing the Study of the Administrative Presidency.

Presidential Studies Quarterly, 39/1: 89–110.

Epstein, D. L., and O’Halloran, S. 1995. A Theory of Strategic Oversight: Congress,

Lobbyists, and the Bureaucracy. Journal of Law, Economics, and Organization, 11/2: 227–55.

————1999.Delegating Powers: ATransaction Cost Politics Approach to Policy Making Under

Separate Powers. New York: Cambridge University Press.

Falaschetti, D., andMiller, G. J. 2001. Constraining Leviathan: Moral Hazard and Credible

Commitment in Constitutional Design. Journal of Theoretical Politics, 13/4: 389–411.

legislative delegation of authority 541

Page 22: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

Ferejohn, J. A., and Shipan, C. 1990. Congressional Influence on the Bureaucracy. Journal of

Law, Economics, and Organization, 6 (January): 1–20.

Finer, H. 1941. Administrative Responsibility in Democratic Government. Public Administra-

tion Review, 1/4: 335–50.

Fiorina, M. P. 1982. Legislative Choice of Regulatory Reforms: Legal Process or Administra-

tive Process? Public Choice, 39/1: 33–66.

——1986. Legislator Uncertainty, Legislative Control, and the Delegation of Legislative Power.

Public Choice, 2/1: 33–51.

Friedrich, C. J. 1940. Public Policy and the Nature of Administrative Responsibility. Public

Policy, 1: 3–24.

Frumkin, P., and Galaskiewicz, J. P. 2004. Institutional Isomorphism and Public Sector

Organizations. Journal of Public Administration Research and Theory, 14/3: 283–307.

Gailmard, S., and Patty, J. W. 2007. Slackers and Zealots: Civil Service, Policy Discretion,

and Bureaucratic Expertise. American Journal of Political Science, 51/4: 873–89.

——2008. Strategic Preference Conflict and Endogenous Information in Bureaucracies.

Presented at the Annual Meeting of the American Political Science Association, Boston,

MA, August 28–31.

Hall, P. A., and Taylor, R. C. R. 1996. Political Science and the Three New Institutionalisms.

Political Studies, 44/5: 936–57.

Hamilton, J. T., and Schroeder, C. H. 1994. Strategic Regulators and the Choice of

Rulemaking Procedures: The Selection of Formal versus Informal Rules in Regulating

Hazardous Waste. Law and Contemporary Problems, 57 (Winter/Spring): 111–60.

Hammond, T. H., and Thomas, P. A. 1989. The Impossibility of a Neutral Hierarchy. Journal

of Law, Economics, and Organization, 5/1: 155–84.

Hannan, M. T. 2005. Ecologies of Organizations: Diversity and Identity. Journal of Economic

Perspectives, 19/1: 51–70.

Heclo, H. A. 1977. A Government of Strangers: Executive Politics in Washington. Washington,

DC: Brookings Institution Press.

Holmstrom, B. 1982. Moral Hazard in Teams. Bell Journal of Economics, 13/2: 324–40.

Howell, W. G. 2003. Power Without Persuasion: The Politics of Direct Presidential Action.

Princeton, NJ: Princeton University Press.

Huber, J. D., and McCarty, N. 2004. Bureaucratic Capacity, Delegation, and Political

Reform. American Political Science Review, 98/3: 481–94.

Huber, J. D., and Shipan, C. R. 2002. Deliberate Discretion? The Institutional Foundations of

Bureaucratic Autonomy. New York: Cambridge University Press.

James, S. C. 2005. The Evolution of the Presidency: Between the Fear and the Promise.

In Institutions of American Democracy: The Executive Branch, ed. J. D. Aberbach and

M. A. Peterson. New York: Oxford University Press.

Jordan, S. V. 2009. Internal Disagreement. Typescript, University of Rochester, NY.

Kiewiet, D. R., and McCubbins, M. D. 1991. The Logic of Delegation: Congressional Parties

and the Appropriations Process. Chicago, IL: University of Chicago Press.

Krause, G. A. 1999. ATwo-Way Street: The Institutional Dynamics of the Modern Administra-

tive State. Pittsburgh, PA: University of Pittsburgh Press.

——2003. Agency Risk-Propensities for the Demand for Bureaucratic Discretion. In Politics,

Policy, and Organizations: Frontiers in the Scientific Study of Bureaucracy, ed. G. A. Krause

and K. J. Meier. Ann Arbor: University of Michigan Press.

——2009. Organizational Complexity and Coordination Dilemmas in U.S. Executive Politics.

Presidential Studies Quarterly, 39/1: 74–88.

542 george a. krause

Page 23: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

——and Douglas, J. W. 2006. Does Agency Competition Improve the Quality of Policy

Analysis? Evidence from OMB and CBO Current Year Fiscal Projections. Journal of Policy

Analysis and Management, 25/1: 53–74.

Krause, G. A., Lewis, D. E., and Douglas, J. W. 2010. Does Delegation Insulate Policymaking

from Politics? Evidence from Official Revenue Forecasts in the American States, 1987–2004.

Typescript, University of Pittsburgh.

Lavertu, S. 2009. Domain-Specific Conflict and Presidential Demand for Politically Exposed

Agencies. Typescript, University of Wisconsin-Madison.

Lewis, D. E. 2003. Presidents and the Politics of Agency Design: Political Insulation in the United

States Government Bureaucracy, 1946–1997. Palo Alto, CA: Stanford University Press.

——2008. The Politics of Presidential Appointments: Political Control and Bureaucratic Perfor-

mance. Princeton, NJ: Princeton University Press.

MacDonald, J. A. 2009. Limitation Riders and Congressional Influence over Bureaucratic

Policy Decisions. Typescript, West Virginia University.

Madison, J., Hamilton, A., and Jay, J. 1987. The Federalist Papers. London: Penguin Books.

First published 1788.

McCarty, N. 1997. Presidential Reputation and the Veto. Economics and Politics, 9/1: 1–26.

McCubbins, M. D. 1985. The Legislative Design of Regulatory Structure. American Journal of

Political Science, 29/4: 721–48.

——Noll, R., and Weingast, B. R. (McNollGast). 1987. Administrative Procedures as

Instruments of Political Control. Journal of Law, Economics, and Organization, 3/2: 243–77.

——————1989. Structure and Process, Politics and Policy: Administrative Arrange-

ments and the Political Control of Agencies. University of Virginia Law Review,

74 (March): 431–82.

McCubbins, M. D., and Schwartz, T. 1984. Congressional Oversight Overlooked: Police

Patrols versus Fire Alarm Oversight. American Journal of Political Science, 28/1: 165–79.

Miller, G. J. 1992. Managerial Dilemmas: The Political Economy of Hierarchy. New York:

Cambridge University Press.

——2000. Above Politics: Credible Commitment and Efficiency in the Design of Public

Agencies. Journal of Public Administration Research and Theory, 10/2: 289–328.

Moe, T. M. 1989. The Politics of Bureaucratic Structure. In Can the Government Govern?, ed.

J. E. Chubb and P. E. Peterson. Washington, DC: Brookings Institution Press.

——1990. Political Institutions: The Neglected Side of the Story. Journal of Law, Economics,

and Organization, 6 (January): 231–53.

——and Howell, W. G. 1999. The Presidential Power of Unilateral Action. Journal of Law,

Economics, and Organization, 15/1: 132–79.

Mosher, F. 1982. Democracy and the Public Service. 2nd edn. New York: Oxford University

Press.

Nathan, R. P. 1983. The Administrative Presidency. 2nd edn. New York: John Wiley & Sons.

North, D. C., and Weingast, B. R. 1989. Constitutions and Commitment: The Evolution of

Institutional Governing in Public Choice in Seventeenth-Century England. Journal of

Economic History, 49/4: 803–32.

Ogul, M. 1976. Congress Oversees the Bureaucracy. Pittsburgh, PA: University of Pittsburgh

Press.

Pfeffer, J., and Salancik, G. R. 1978. The External Control of Organizations: A Resource

Dependence Perspective. New York: Harper and Row.

Potoski, M. 1999. Managing Uncertainty through Bureaucratic Design: Administrative

Procedures and State Air Pollution Control Agencies. Journal of Public Administration

Research and Theory, 9/4: 623–39.

legislative delegation of authority 543

Page 24: LEGISLATIVE DELEGATION OF AUTHORITY TO BUREAUCRATIC …gkrause/Krause.Legislative Delegation... · DELEGATION OF AUTHORITY TO BUREAUCRATIC ... Citizens delegate responsibility to

Rieselbach, L. 1995. Congressional Politics: The Evolving Legislative System. 2nd edn. Boulder,

CO: Westview Press.

Rosenbloom, D. H. 2000. Building a Legislative-Centered Public Administration: Congress and

the Administrative State, 1946–1999. Tuscaloosa: University of Alabama Press.

Rourke, F. E. 1984. Bureaucracy, Politics, and Public Policy. 3rd edn. Boston, MA: Little, Brown.

Rudalevige, A. C. 2005. The Structure of Leadership: Presidents, Hierarchies, and Informa-

tion Flows. Presidential Studies Quarterly, 35/2: 333–60.

Scher, S. 1963. Conditions for Legislative Control. Journal of Politics, 25/3: 526–51.

Scholz, J. T., and Wang, C.-L. 2006. Cooptation or Transformation? Local Policy Networks

and Federal Regulatory Enforcement. American Journal of Political Science, 50/1: 81–97.

Seidman, H., and Gilmour, S. 1986. Politics, Position, and Power: From the Positive to the

Regulatory State. 4th edn. New York: Oxford University Press.

Simon, H. A. 1991. Organizations and Markets. Journal of Economic Perspectives, 5/2: 25–44.

Spence, D. B. 1997. Agency Policymaking and Political Control: Modeling Away the Delega-

tion Problem. Journal of Public Administration Research and Theory, 7/2: 199–219.

Stephenson, M. 2007. Bureaucratic Decision Costs and Endogenous Agency Expertise.

Journal of Law, Economics, and Organization, 23/2: 469–98.

Thompson, J. D. 1967. Organizations in Action: Social Science Basis of Administrative Theory.

New York: McGraw-Hill.

Ting, M. M. 2001. The “Power of the Purse” and Its Implications for Bureaucratic Policy-

making. Public Choice, 106/3–4: 243–74.

——2002. A Theory of Jurisdictional Assignments in Bureaucracies. American Journal of

Political Science, 46/2: 364–78.

Volden, C. 2002a. A Formal Model of the Politics of Delegation in a Separation of Powers

System. American Journal of Political Science, 46/1: 111–33.

——2002b. Delegating Power to Bureaucracies: Evidence from the States. Journal of Law,

Economics, and Organization, 18/1: 187–220.

Waterman, R. W. 1989. Presidential Influence and the Administrative State. Knoxville: Univer-

sity of Tennessee Press.

Weingast, B. R. 1984. The Congressional-Bureaucratic System: A Principal–Agent Perspec-

tive (with Applications to the SEC). Public Choice, 44/1: 147–91.

——and Marshall, W. J. 1988. The Industrial Organization of Congress; or, Why

Legislatures, like Firms, are not Organized as Markets. Journal of Political Economy,

96 (February): 132–63.

West, W. F. 2006. Presidential Leadership and Administrative Coordination: Examining the

Theory of a Unified Executive. Presidential Studies Quarterly, 36/3: 433–56.

Wilson, J. Q. 1989. Bureaucracy: What Government Agencies Do and Why They Do It.

New York: Free Press.

Wilson, W. 1973. Congressional Government: A Study in American Politics. Gloucester, MA:

Peter Smith. First published 1885.

Wood, B. D., and Bohte, J. 2004. Political Transaction Costs and the Politics of Administra-

tive Design. Journal of Politics, 66/1: 176–202.

544 george a. krause