Agency Design as an Ongoing Tool of Bureaucratic Influenceagency s structure and procedures in an...

383Agency Design

LEGISLATIVE STUDIES QUARTERLY XXIX 3 August 2004 383

CHRISTOPHER REENOCKFlorida State University

SARAH POGGIONEFlorida International University

Agency Design as an Ongoing Toolof Bureaucratic Influence

Theoretical work assumes that legislators use ex ante design to gain bureau-cratic influence not only at an agencyrsquos appointment stage but also as an ongoingtactic Yet no empirical work has investigated whether or not legislators prefer to usedesign to exert influence after an agencyrsquos appointment stage Using a mail survey ofmore than 2500 legislators we model legislatorsrsquo preferences for ex ante design as afunction of both institutional factors and individual legislatorsrsquo characteristics Ourresults suggest that the feasibility of agency design as an ongoing tactic of bureau-cratic influence is more limited than theoretical work indicates and that bothinstitutional- and individual-level factors explain legislatorsrsquo preferences

Introduction

Recent studies of legislative control of the bureaucracy stress thevalue of ex ante agency design as a tool for legislators seeking bureau-cratic responsiveness Ex ante tactics allow legislators to alter anagencyrsquos structure and procedures in an attempt to shape bureaucraticdiscretion and ultimately decrease the likelihood of undesirable policyoutcomes Theoretical treatments have produced various expectationsregarding the conditions that influence legislatorsrsquo choices either toenhance or to restrict bureaucratic discretion through agency design(Bawn 1995 Epstein and OrsquoHalloran 1994 1999 McCubbins Nolland Weingast 1987 1989) Empirical work on this matter suggests thatinstitutional- (Huber Shipan and Pfahler 2001 Potoski 1999) andindividual-level factors (Balla 2000 Bawn 1997) influence the level ofdiscretion that legislators seek to obtain through agency design

Yet both the theoretical and empirical studies simply assume thatlegislators prefer ex ante design as a tactic of bureaucratic influenceThere is reason to believe however that given the relatively rareopportunities to alter agency structure and procedures (Arnold 1987

Reenock and Poggionep65 62204 152 PM383

384 Christopher Reenock and Sarah Poggione

Spence 1997) legislators may not prefer ex ante design as an ongoingtactic to achieve influence (Robinson 1989) If this is the case then theutility of ex ante design may be overstated in theoretical treatments(Robinson 1989)

In this paper we seek to answer two questions to what extent dostate legislators prefer to use ex ante design as an ongoing tactic of bureau-cratic influence and what factors condition these preferences Focusingon environmental policy we use a mail survey of more than 2500 legisla-tors in 24 states to collect data on state legislatorsrsquo preferences towardvarious ex ante tactics of bureaucratic control their policy preferencesand other individual-level characteristics Using the survey data we createan index that measures state legislatorsrsquo preferences for using ex antetactics to influence state environmental agencies We then combine thesedata with state-level institutional data to test whether or not individual andinstitutional differences explain legislatorsrsquo preferences for ex ante tactics

Exploring the Assumptions of Ex Ante Design Tactics

The literature on bureaucratic responsiveness suggests that exante design reduces the costs that legislators incur when they delegateauthority to bureaucratic agents (Bawn 1995 1997 Calvert McCubbinsand Weingast 1989 Epstein and OrsquoHalloran 1994 1995 1996 1999Huber and Shipan 2000 McCubbins Noll and Weingast 1987 1989)Such ex ante limitations on bureaucratic behavior involve the construc-tion of bureaucratic procedures organizational mandates and a varietyof other passive attempts at control (Calvert McCubbins and Weingast1989) Proponents argue that ex ante design is a cheaper more effi-cient alternative to other tactics that require costly monitoring and expost punishment With ex ante design legislators are said to be able todecrease the informational asymmetries and agent shirking inherent indelegation they do so by structuring an agencyrsquos rules and proceduresin such a way as to reduce ldquothe ability of an agency or other executiveactors to enact outcomes different from the policies preferred by thosewho originally delegated powerrdquo (Epstein and OrsquoHalloran 1994 699)

Empirical research suggests that the legislative application of exante tactics to alter agentsrsquo discretionary environments is generallyinfluenced by both institutional- and individual-level characteristics Takentogether these factors condition the level of discretion that legislaturesseek to obtain through their design of an agencyrsquos structures andprocedures State-level research capitalizes on the institutional variationacross the states This research suggests that the saliency and complexityof the policy area (Potoski 1998) the institutional characteristics of the


385Agency Design

legislature [such as legislative professionalism (Huber Shipan andPfahler 2001 Potoski 1998] and the nature of partisan agreementbetween the legislature and the executive may condition the discretionarylevels that legislators permit when they employ ex ante design (HuberShipan and Pfahler 2001)1 In contrast national-level studies suggestthat characteristics specific to individual legislators also influence theirdecisions to enhance or restrict bureaucratic discretion Membershipon oversight committees (Bawn 1997) personal ideology ideologicalposition within committees and in Congress more generally and electoralsecurity (Balla 2000) have all been linked to representativesrsquo opinionsas to the optimal levels of discretion to establish through ex ante design2

Both the theoretical and empirical studies assume that legislatorsroutinely apply ex ante tactics In fact compared to the more expen-sive and resource-intensive ex post tactics of monitoring and oversightex ante design is portrayed as the optimal tactic (McCubbins Noll andWeingast 1987) Calvert McCubbins and Weingast (1989) suggestthat ex ante methods used during the appointment stage of an agencywhich includes the ldquostructuring of the agency itself the denominationof its powers and jurisdiction the specification of administrative proce-dures to be followed and the type of personnel with which the agencyis to be staffedrdquo (Calvert McCubbins and Weingast 1989 604) arethe most efficient influence tactics that legislators can pursue Proponentssuggest that even after the appointment stage design alterations affordlegislators the same benefits as ex post techniques The theoreticalworks on ex ante design further assume that legislators are aware ofdesign as a tool of influence and prefer to use it as an ongoing controldevice after an agencyrsquos appointment stage3 Yet the empirical worksprovide no evidence of this preference On the contrary there is reasonto suspect that the use of ex ante design may be less attractive thanother methods as an ongoing tactic of bureaucratic control

There are at least two reasons why the assumptions made in thisprevious research may be problematic First individual legislators maynot have the motivation to engage in ex ante design Personal factorsmay influence legislators to be more or less active in attempting toinfluence bureaucratic policymaking through ex ante design A varietyof individual-level characteristics including membersrsquo policy prefer-ences in relation to those of the agencyrsquos can affect whether or notlegislators deem ex ante design a worthy expenditure of their time andresources Even within a single chamber not all legislators are equallylikely to use ex ante design If legislators with particular policy prefer-ences are more likely to use design then this tactic may afford smallernumbers of legislators with shared interests or preferences the opportunity



to influence policy We know from other research that varying motivationallevels encourage some legislators to be more active in the origin and markupof legislation thereby giving them a greater impact on the eventualproduct (see for example Hall 1996 and Hall and Wayman 1990)

If membersrsquo motivations and therefore participation in ex antedesign tactics do vary systematically then ex ante design and thebureaucratic responsiveness it engenders may reflect particular policypreferences and not the preferences of the legislature as a whole oreven of the majority party Work that ignores legislatorsrsquo motivations to useex ante tactics fails to account for the fact that some legislators play moreactive roles in agency design than others because of a desire to accomplishpolicy goals As a result such work may mistakenly conclude that the useof ante design reflects the collective preferences of the chamber whenin fact it reflects the preferences of a few highly motivated activemembers pursuing their own or their constituentsrsquo policy preferences

Second state legislators have vastly different resources to drawon for the costly activities of bureaucratic monitoring and oversight Insome states legislators have resources rivaling those of the USCongress in others resources are quite limited Given that legislativeresources like professional staff and legislative review powers maycondition legislatorsrsquo decisions to use ex ante tactics we believe thatlegislatorsrsquo preferences toward ex ante design should vary across statesaccordingly Lacking the resources to pursue more costly monitoringand ex post punishment legislators may opt for the more attractivealternative of ex ante design to gain influence over the bureaucracy

Both these reasons are consistent with the criticism by some schol-ars that design although useful in the creation of an agency is lesspractical as an ongoing tactic of influence over the bureaucracy(Robinson 1989) Furthermore unlike reauthorization and budgetarydecisions that typically require annual or biennial debate and actionopportunities to alter agency structure and procedures are relativelyrare events (Arnold 1987 Spence 1997) If legislators do not prefer exante design as a routine tactic of bureaucratic influence then propo-nents may have overstated its utility (Robinson 1989)

In this paper we test the assumption that legislators prefer exante design as a tactic of bureaucratic influence First we describe thepreferences of legislators for ex ante mechanisms to address the pos-sibility that legislators do not prefer ex ante design as an ongoing tacticof bureaucratic influence Second we use an individual-level approachto examine how the institutional characteristics of the legislature andthe individual characteristics of legislators influence membersrsquo attitudestoward ex ante tactics of bureaucratic influence


387Agency Design

Explaining Preferences for Ex Ante Tactics

Extrapolating from prior work we expect that legislatorsrsquo prefer-ences toward ex ante design will reflect its relative costs and benefits(Bawn 1997)4 We apply a cost-benefit framework to individual statelegislatorsrsquo preferences ensuring that the costs and benefits ofbureaucratic delegation vary not only across states and institutions butalso across members themselves Generally these factors include thecosts of an agency that deviates from legislatorsrsquo preferences theopportunity costs and direct costs of engaging in ex ante design thecosts associated with the technical expertise trade-off the costs ofpolitical drift and the benefits associated with using ex ante design inthe absence of ex post options

Costs of an Agencyrsquos Activities

The costs of delegated authority to an agency differ among legis-lators As a legislatorrsquos ideal policy location moves away from theagencyrsquos ideal point that legislator fearing political costs will be inclinedto restrict the discretion with which that agency can make policydecisions (Epstein and OrsquoHalloran 1994 1999) Ex ante design is oneof the tactics that legislators may use to accomplish these restrictionsLegislators with conservative preferences on environmental policy willlikely view the regulatory agency of a more liberal gubernatorialadministration as politically unappealing and as a result will be motivatedto alter that agencyrsquos design Therefore we expect that members whoseideal policy preferences are further away from the ideal location of theagency will be more likely to exhibit favorable preferences toward exante design

Hypothesis 1 Legislators whose policy preferences are furtherremoved from the ideal location of the agency will be more likelyto prefer ex ante design as a bureaucratic influence tactic

Costs of Ex Ante Design

Ex ante tactics also have costs associated with them For instanceoverhauling the procedures that an agency employs to modify or initiatepolicy proposals presents a substantial cost to legislators attempting toinfluence an agency Engaging in such tactics may require a legislatorto possess a higher level of expertise about the policy area as well as theagencyrsquos status quo structure and procedures (Spence 1997) Experienced



legislators with policy-specific knowledge and expertise may be able toovercome the costs in time and resources associated with researchingand writing detailed legislation that alters an agencyrsquos structure andprocedures Senior legislators should have greater experience andexpertise in the policymaking process (Hibbing 1991 1993 see alsoBratton and Haynie 1999) and as a result they should be better able toovercome the costs of engaging in ex ante design Therefore we expectthat more-senior legislators will prefer ex ante tactics to influencebureaucratic policymaking

Hypothesis 2 Legislators with greater tenure in office will be morelikely to prefer ex ante design as a bureaucratic influence tactic

Opportunity Costs

Ex ante design has been portrayed as a cheap and efficient alter-native to ex post oversight tactics (Calvert McCubbins and Weingast1989 McCubbins Noll and Weingast 1987 1989) Legislators whohave limited resources particularly staff will value ex ante design asan attractive alternative to resource-intensive monitoring as well aspotential hearings and investigations (Elling 1984) The availability ofstaff lowers the opportunity costs of engaging in ex post tactics therebymaking ex ante techniques less attractive Thus we expect that legis-lators who have access to staff will be less likely to prefer ex antetactics of bureaucratic control

Hypothesis 3 Legislators who have access to staff resources willbe less likely to prefer ex ante design as a bureaucratic influence tactic

The Technical Expertise Trade-off

With ex ante design legislators can alter regulatory agentsrsquodiscretionary environments Restrictions on bureaucratic discretion how-ever may hinder policy production and implementation under highlyrestrictive conditions bureaucratic agents cannot capitalize on theirexpertise to deliver informed policy and the resulting inefficiency pro-duces distributive losses (Bawn 1995)5 Legislators face the dilemmaof balancing the transaction costs of high discretion with the distributivelosses incurred by an overly restrictive design (Bawn 1995 1997Ringquist Worsham and Eisner 2002) We therefore expect that whenagencies deal with technically complex issues and the possibility fordistributive losses is great legislators will hold less favorable prefer-ences toward ex ante design tactics6


389Agency Design

Hypothesis 4 As the complexity of the policy area increaseslegislators will be less likely to prefer ex ante design as a bureau-cratic influence tactic

The Potential for Political Drift

Another cost that legislators face is associated with the politicaldrift of agency policy or the unfavorable policy changes that may beimplemented by new political coalitions that come to power (Horn andShepsle 1989 Shepsle 1992) In a politically competitive environmentlegislators will be less certain about the stability of bureaucratic outputsand their continued ability to monitor and influence bureaucraticagencies They may therefore seek to insulate themselves from suchpolitical drift If legislators believe that an agencyrsquos policies are likelyto deviate because the political environment is unstable then they willbe more likely to prefer ex ante design as a mechanism for restrictingthe agencyrsquos discretion

Hypothesis 5 As the potential for political drift increases legislatorswill be more likely to prefer ex ante tactics as a bureaucraticinfluence tactic

The Existence of Ex Post Options

Having the opportunity to correct the actions of an agency thatdeviates from a legislatorrsquos ideal policy location lowers the marginalbenefits of that legislator engaging in ex ante tactics (Bawn 1997) Theopportunity to use ex post options such as legislative vetoes or legislativereviews of agency rules (Balla 2000) will condition a legislatorrsquosdecision to redesign an agency because for legislators ldquoensuring thatthe agency behaves in the desired way is less urgent when there willbe opportunities to correct for lsquomistakesrsquo after the factrdquo (Bawn 1997108) Hence we expect that legislators will hold less favorable prefer-ences toward ex ante design tactics when they have the ability to reviewand correct agency decisions

Hypothesis 6 If legislators have the opportunity to engage in expost review of agency policy then they will be less likely to preferex ante design as a bureaucratic influence tactic



Research Design

To test our expectations we collected data on state legislatorsrsquoindividual characteristics policy preferences and their attitudes towardvarious tactics of bureaucratic control7 The survey conducted in thesummer and fall of 2000 was mailed to more than 2500 state legislatorsmdashall members of the lower houses of 24 state legislatures8 We randomlyselected the 24 state legislatures from several strata organized accordingto such contingencies as party control of the chamber and legislativeprofessionalism thus ensuring variation in the resulting sample of statelegislators More than 500 legislators responded to the survey an overallresponse rate of about 21 Although we took numerous steps toencourage survey response 9 this response rate is somewhat low evenfor a survey of political elites We thus employed a Heckman selectionmodel (discussed in our ldquoStatistical Modelrdquo section) to correct forpossible selection bias resulting from the low response rate

Using survey data we created an index that assesses legislatorsrsquopreferences toward ex ante tactics We then modeled the effects oflegislative context including legislative capacity and ex post veto poweras well as the effects of individual-level factors including membersrsquopolicy preferences and membersrsquo preferences for ex ante tactics ofbureaucratic control

Dependent Variable

Our dependent variable measures the preferences of an individuallegislator toward ex ante control tactics We asked state legislatorsfive questions to determine their preferences for ex ante tactics forinfluencing environmental agencies such as altering the organizationof the agency changing the method of selecting the head of the agencyaltering the regulatory scope of the agency changing the agencyrsquosdecision-making procedures and requiring the agency to report to thelegislature10 For each tactic respondents who indicated they werevery likely to use the tactic scored two points those who said theywere somewhat likely scored one point and those who were not verylikely scored zero points To measure membersrsquo preferences towardex ante tactics we constructed a simple additive scale weighting eachof the five items equally11

Table 1 lists the five ordinal items that compose the index as wellas their means standard deviations and corrected scale-item correla-tions All five items are positively correlated with each other and allthe corrected scale-item correlations (the correlation of the item with


391Agency Design

TABLE 1Items and Index of Legislatorsrsquo Preferences

toward Ex Ante Tactics

Standard CorrectedVariable Mean Range Deviation Correlationa

Alter the organization of the agency 57 0ndash2 69 53

Change method of selecting agency head 29 0ndash2 62 39

Alter the agencyrsquos scope of authority 89 0ndash2 69 49

Require the agency to report to the legislature 160 0ndash2 63 31

Alter the agencyrsquos decision-making procedures 77 0ndash2 72 47

Index of preferences for ex ante controlb 410 0ndash10 222

Note All items are ordinal variablesaCorrelation between the item and the corrected index of preferences for ex ante control(the index computed without the item)bα coefficient of reliability is 68

the scale computed without the item) are also positive indicating thatthe items reflect an underlying dimension The large and significantcoefficient of reliability for the scale (α = 68) also supports this conclusionThe index ranges from zero indicating that the member does not at allprefer to engage in ex ante tactics to 10 indicating that the member greatlyprefers ex ante tactics as a means for influencing the bureaucracy

Independent Variables

Environmental Policy Preferences Relative to the Agency Weassess the political costs of having an active agency by measuring thelegislatorsrsquo environmental policy preferences relative to the position ofthe agency First we measured legislatorsrsquo environmental policy pref-erences using an additive index of environmental policy conservatismcomposed of six ordinal items indicating legislatorsrsquo support or oppositionto principles of environmental regulation12 The resulting index rangesfrom 0 the most liberal position on environmental policy to 12 indicatingthe most conservative environmental policy preferences

Next we determined the position of the environmental agencywith a dichotomous variable indicating the presence of a Democraticgovernor in the state Agencies operating under Democratic governors



should be more active in regulating environmental conditions Weassessed the legislatorsrsquo preferences relative to the agency by includingthe interaction of membersrsquo environmental policy preferences with thepresence of a Democratic governor13 We expect that conservativelegislators will be motivated to engage in ex ante tactics when a Demo-cratic governor holds office since it is likely that the governor will pursuegoals divergent from these legislatorsrsquo in agency policy

Legislative Seniority To assess the costs associated withengaging in ex ante design we included a measure of legislative senioritydefined as the number of continuous years of service in the legislatureWe collected this measure from the official websites of the variousstate legislatures and the Project Vote Smart database of state legislators

Legislative Staff To assess the opportunity costs of engaging inex ante design we measured the legislative capacity or the resourcesavailable to legislators by noting whether or not individual membershave staff working in their capitol offices We collected this dichoto-mous measure of staff from the survey of state legislators14

Entropy We measure the technical expertise tradeoff that alegislator faces when engaging in ex ante design by measuring theenvironmental policy complexity within the state Specifically wemeasure the entropy of the statersquos air emissions15 If a statersquos air emis-sions derive from a single source then this measure is zero the measureincreases as both the total number of categories increases and thedistribution across different categories becomes more evenly distrib-uted16 This measure is essentially a diversity index of the state airemissions sources within each state Higher state entropy scoresrepresent extremely complex policy environments suggesting thatlegislators are not likely to have the technical knowledge to understandall the ramifications of their decisions Increased complexity should berelated to a decreased preference on the part of legislators toward exante design We calculated the measure using data from the EPArsquosToxic Release Inventory CD-ROM (EPA 1999)

Potential for Political Drift Political Uncertainty We measurethe potential for the bureaucracy to drift by assessing the uncertaintythat arises from political competition We use the Ranney index ofparty competition in each state between 1994 and 1998 (Bibby andHolbrook 1999) The Ranney index incorporates the proportion of partysuccesses in executive and legislative elections and the duration of


393Agency Design

party control of these bodies (see Bibby and Holbrook 1999 for a com-plete description) The folded Ranney index or the Ranney competi-tion index indicates the degree of competition in the state political insti-tutions regardless of party This measure ranges from 5 to 1 where 5represents no competition and 1 perfect competition We use the dataas collected and reported by Bibby and Holbrook (1999)

Legislative Review Powers We measure the availability of statu-torily defined ex post powers by including a three-category ordinalvariable that measures the powers of the legislature to review adminis-trative rules This variable ranges from the legislature having no agencyreview powers (0) to advisory powers only (1) to formal review powersof agency rules (2) We collected these data from the Book of theStates (Council of State Governments 2000ndash01)

Statistical Model

Because we rely on survey data especially for the measure ofthe dependent variable we think it important to consider how the choiceof respondents to participate in the survey potentially influenced thedata and the results of the analysis If certain types of respondentswere less likely to participate in the survey then survey nonresponsecan introduce selection bias into even a random sample and ultimatelythreaten the inferential value of survey analysis (see Groves Cialdiniand Couper 1992) In order to test and correct for possible selectionbias we used a Heckman selection model The Heckman selectionmodel estimates a system of equations while using full informationmaximum likelihood In this case the system includes a selection equa-tion predicting the probability that a member of the sample respondedto the survey and is included in the second equation which predictsmembersrsquo preferences for ex ante measures of bureaucratic controlThis procedure corrects for the selectivity bias and produces consistentasymptotically efficient parameter estimates (Greene 1993 708ndash14)

In order to use a Heckman selection model one must first specifythe selection equation which predicts whether or not a member of thesample will respond to the survey Literature on survey nonresponsesuggests that sociodemographic characteristics like age gender andsocioeconomic status are important predictors of the decision to par-ticipate in a survey (see for example DeMaio 1980 and Smith 1983)In addition to these general characteristics the professionalism of thelegislature and membersrsquo legislative seniority may also influence thelikelihood that members will participate in the survey We expect thatmembers who serve in more professional legislaturesmdashinstitutions that



generally have higher workloads and meet for longer sessionsmdashmaybe less likely to complete and return the survey given the numerousdemands on their time17 More-senior members may also be less likelyto participate because they may be more involved in the decision makingof legislature In addition party may affect the decisions of membersto participate in the survey Consequently we model the decision toparticipate in the survey as a function of age gender socioeconomicstatus the professionalism of the legislature legislative seniority andparty affiliation18 The overall model estimates robust standard errorsthat are corrected for clustering by state to account for the fact thatobservations may not be independent within states

Results

We begin with a brief discussion of the legislatorsrsquo preferencestoward ex ante design First we find that legislators on average do notexhibit extraordinarily favorable attitudes toward ex ante design tacticsFigure 1 shows the distribution of membersrsquo scores on the index Thehistogram suggests that there is a good deal of variation along thisdimension but most of the legislatorsrsquo scores fall in the lower half ofthe scale indicating that a substantial number of state legislators reportedthey were not very likely to use ex ante tactics as an ongoing methodof influence This pattern suggests that previous theoretical work may haveoverstated the uniformity and strength of legislatorsrsquo preferences for exante tactics as an ongoing tool of influence Far from appearing to embraceex ante design as the optimal control tactic after the appointment stagestate legislators seem to be generally reluctant to engage in it

In addition Table 2 presents both the mean and standard devia-tion of index scores for legislators by state Average scores differ widelyacross the states from a low of 300 in Colorado indicating that mem-bers of the Colorado legislature have less favorable views of ex antedesign to a high of 543 in Idaho suggesting that legislators in Idahohave much more favorable views of ex ante controls Table 2 alsodemonstrates the marked differences in preferences for ex ante designwithin states The relatively large standard deviations suggest that thecosts and benefits associated with agency design vary across legislatorswithin the same institution The variation both across and within statessuggests that institutional and individual factors likely explain membersrsquopreferences toward ex ante controls

To explore this possibility we turn now to our multivariate analysisTable 3 lists the results of our analysis Note that age and educationhave significant effects on the likelihood of responding to the survey


395Agency Design

Older and more highly educated respondents are more likely to partici-pate in the survey and more senior members and those from moreprofessional legislatures are less likely In addition party also has aneffect on the likelihood of members participating in the survey Non-partisan legislators and those affiliated with third parties are less likelyto participate in the survey

The estimate of ρ reflects the correlation of the error terms ofthe selection model and the substantive regression equation If ρ equalszero then ordinary least squares (OLS) estimation is appropriate ifhowever ρ does not equal zero then OLS will produce biased param-eter estimates In this case the estimated ρ is 11 indicating that theerror terms are related although not strongly In addition the Wald testof independent equations (the χ2 statistic presented in the final row ofthe first column) is not large enough to reject the null hypothesis that ρequals zero and that the two equations are independent Although theoverall selection model parameters suggest that it was not necessarywe employed a Heckman selection model for this analysis because ofthe significance of the individual coefficients in the selection model andthe correlation between some of these significant factors and key inde-pendent variables in the substantive equation

FIGURE 1Distribution of Preferences for Ex Ante Control

over State Environmental Agency

21

34

58

92

70

87

4137

179 8

0

10

20

30

40

50

60

70

80

90

100

0 1 2 3 4 5 6 7 8 9 10

Preferences for Ex Ante Control N = 474

Num

ber

of R

espo

nden

ts



TABLE 2Legislator Preferences toward Ex Ante Tactics

by State

State Meana Standard Deviationb Valid N

Arkansas 412 252 17

California 336 196 11

Colorado 300 166 9

Connecticut 377 180 31

Delaware 440 207 5

Idaho 543 191 14

Illinois 388 181 8

Louisiana 457 98 7

Maine 433 204 45

Michigan 437 248 19

Missouri 336 222 22

Montana 525 256 24

North Dakota 423 184 31

Nebraska 435 166 17

New Jersey 382 271 11

Ohio 457 271 14

Pennsylvania 361 236 46

South Carolina 328 285 18

South Dakota 305 165 22

Tennessee 400 173 17

Texas 475 238 12

Vermont 461 258 36

West Virginia 500 232 17

Wyoming 352 169 21

All states 410 222 474

aMean score on index of ex ante control for all legislators within a particular statebStandard deviation of scores on index of ex ante control for all legislators in a state


397Agency Design

TABLE 3State Legislatorsrsquo Preferences toward Ex Ante Tactics

Variable

Ex Ante Preferences Indexa

b se(b)

Entropy 98 (76)Costs of an active agency

Environmental policy conservatism 08 (05)Democratic governor ndash163 (137)Env policy conservatism times Dem governor 15 (20)

Legislative staff ndash48 (28)Legislative seniority 01 (01)Ranney index of party competition ndash64 (154)Review power ndash52 (23)Constant 492

Uncensored observations = 353χ2 = 7069

Selection Modelb se(b)

Age 02 (004)Education 27 (09)Gender (female) ndash01 (08)Party

Democrat ndash03 (09)Other party ndash594 (15)

State legislative professionalism ndash55 (102)Years served in legislature ndash01 (01)Constant ndash168

Total observations = 1852ρ11χ2 (ρ = 0)06

Note Robust standard errors in parentheses corrected for clustering on states Resultsgenerated using Heckman selection modelaFive-item additive index of membersrsquo willingness to engage in ex ante control of stateenvironmental agencyp lt 10 p lt 05 p lt 01 two-tailed tests



The results of the substantive equation predicting the membersrsquopreferences for ex ante design suggest that both individual-level legis-lator characteristics and institutional context have a significant influ-ence on membersrsquo preferences toward ex ante control Legislatorsrsquoenvironmental policy opinions and whether or not they have staff in thecapitol are factors that affect membersrsquo preferences for agency designWhether or not the legislature has the power to review agency decisionsa characteristic that differs only across legislatures also significantlyconditions membersrsquo preferences toward agency design to influencestate environmental agencies

We now turn to our specific hypotheses on individual legislatorcharacteristics Members who hold more conservative opinions onenvironmental policy have more favorable preferences for ex antetactics to influence state environmental agencies than do their liberalcounterparts The interaction of membersrsquo preferences and the pres-ence of a Democratic governor however is not statistically significantThe preferences of legislators with conservative environmental policypreferences are not significantly more inclined toward ex ante designwhen the legislators face a Democratic governor than when they facea Republican governor This finding suggests that perceived policy dif-ferences between a legislator and an environmental agency do notinfluence legislator preferences It does appear however that conser-vative legislators are likely to incur greater costs for active environ-mental agencies and therefore to favor ex ante tactics to control theseagencies more than their liberal counterparts For example an averagelegislator with the most liberal environmental policy preferences has apredicted score on the ex ante index of only 346 indicating that he orshe is unlikely to employ these tactics19 An average legislator with themost conservative environmental preferences has a predicted score of442 indicating that he or she is more likely to consider agency designas a strategy for influencing bureaucratic decisions This disparity sug-gests that legislators may very well consider the use of ex ante tacticsas a means of restricting rather than enhancing agency discretionTherefore regardless of who is in the governorrsquos mansion a legislatorwith conservative views in environmental policy prefers to use ex antedesign to gain influence over the agency

A legislatorrsquos tenure in office does not affect the costs that areassigned to ex ante tactics Although the coefficient is in the predictedpositive direction legislative seniority does not appear to significantlyreduce the costs of ex ante design Therefore enhanced skill andexperience in the policy area does not seem to offset the costs ofengaging in ex ante design20


399Agency Design

With regard to legislative context the negative coefficient forlegislative staff is statistically significant suggesting that access to staffmakes ex ante design substantially less attractive to legislators Thisfinding is particularly interesting when we consider expectations aboutthe importance of legislative resources in conditioning legislatorsrsquo useof ex ante design (McCubbins Noll and Weingast 1987 1989) Indeedrecent research suggests that enhanced resources provide legislatorswith the tools they require in the face of partisan conflict to producelengthy and detailed legislative grants of discretionary authority to stateagencies (Huber Shipan and Pfahler 2001) Counter to these findingsour results suggest that legislative resources increase the opportunitycosts of engaging in ex ante design by increasing legislatorsrsquo ability tomonitor agencies directly and employ ongoing ex post tactics (Aberbach1990) The impact of staff would be particularly felt among legislatorsattempting to engage in those design tactics that are higher in relative cost

The effect of entropy is not negative as we expected or statisti-cally significant It appears that the complexity of state environmentalpolicy at least as it varies across the states in a single policy area doesnot decrease legislatorsrsquo preferences toward ex ante design Thegreater possibility of distributive losses associated with an overlyrestrictive design posed by ex ante manipulation of agencies does notappear to inhibit membersrsquo preferences toward ex ante controls in theenvironmental policy arena This null finding may however reflect thefact that variation in policy complexity may be greater across differentpolicy areas than across states within the same area

Also contrary to our expectations the coefficient for the Ranneyindex of party competition is not statistically significant signifying thatthe increased potential for political drift that accompanies greater partycompetition does not appear to motivate legislators to pursue ex antedesign as a tactic to influence bureaucratic affairs This finding supple-ments previous research which showed that the potential for politicaldrift influences how much discretion legislators allow when they alteran agencyrsquos design (Epstein and OrsquoHalloran 1999 Huber Shipan andPfahler 2001) We take these results to mean that although uncer-tainty about the political future affects the application of design it doesnot influence legislatorsrsquo willingness to use it

Finally whether or not the legislature has the ability to review andcorrect agency decisions after the fact significantly influenceslegislatorsrsquo preferences toward ex ante measures As expectedmembers in states with legislative review powers are significantly lesslikely to pursue agency design than are members without these reviewpowers For example an average legislator who has formal review



powers has an index score of 349 indicating a less favorable view ofex ante tactics In comparison an average member with only advisorypowers has a score of 400 on the index and an average member withno review powers has a score of 452 Members with weaker or no expost powers at their disposal are more likely to turn to ex ante means toguarantee their interests are represented in agency policy

Conclusion

Our analysis has implications for studies of legislative control ofthe bureaucracy First our analysis demonstrates that the feasibility ofagency design as a tactic of ongoing legislative influence over thebureaucracy may be more limited than national-level theoretical workwould suggest This finding is consistent with the previously untestedexpectations of scholars who argue that the availability of opportunitiesto alter design is limited and that theoretical work must consider thisissue (Arnold 1987 Robinson 1989 Spence 1997) Our findings gener-ally support such critiques and suggest that while ex ante design istheoretically appealing legislators do not seem to prefer ex ante designas an ongoing means of reining in the bureaucracy We do not believethat legislators simply abdicate program responsibility however (Kiewietand McCubbins 1991) it may be that legislators prefer to make use ofother perhaps more accessible methods of influence rather than exante design For example legislators may simply wait for the opportu-nity to punish or reward agencies through the relevant budgetaryauthorizations or some other oversight instrument

Second our individual-level analysis allowed us to test for theimpact of both statewide contextual and legislator-specific factors onlegislatorsrsquo preferences for ex ante tactics We found that characteristicsthat vary across states and individual legislators are important in under-standing legislatorsrsquo preferences over ex ante design Preferencestoward the use of design are a function of a legislatorrsquos environmentalpolicy preferences Legislators with conservative environmental pref-erences appear more likely to prefer ex ante design to influence envi-ronmental agencies than their more liberal colleagues

In addition access to legislative resources (such as staff) and thepower to review agency decisions reduce the marginal costs of engagingin ex post tactics making ex ante tactics less attractive When legislatorshave the resources they may be more likely to assign staff to monitoringand oversight than to rewriting complex design legislation (Aberbach1990) We do not wish to argue however that staff does not behave inthe manner suggested by Huber Shipan and Pfahler (2001) Staff


401Agency Design

operates differently in these two contexts In our study staff indepen-dently lowers the costs of engaging in other control mechanisms suchas ex post tactics thereby diminishing preferences toward ex ante designas an ongoing tool of influence In Huber Shipan and Pfahlerrsquos studystaff conditionally reduces the costs that a legislator bears for writingthe extensive language required to restrict agency discretion21 There-fore state legislators with greater legislative staff are less likely toprefer ex ante design as an ongoing tactic of influence but when giventhe opportunity (for example a bill initiating a new program in an agencycomes under review) and if motivated by political conflict they areable to write more detailed legislation Simply put staff operates in amore complex manner than previous theory suggested

Christopher Reenock is Assistant Professor of Political ScienceFlorida State University 570 Bellamy Building Tallahassee Florida32306-2230 Sarah Poggione is Assistant Professor of PoliticalScience Florida International University DM480 Miami Florida33199

NOTES

1 See also Epstein and OrsquoHalloran 1999 and Ringquist Worsham and Eisner2002 for recent evidence at the national level

2 Many of these assumptions may be valid for the highly professional environ-ment of the US Congress but they are unverified assertions for most state legislaturesIndividual-level surveys of state legislatorsrsquo perceptions of effective control instru-ments have found that legislatorsrsquo attitudes toward various ex post tactics vary thesestudies do not however include ex ante design tactics (Elling 1984 Pearson and Wigginton1986)

3 Most of the work that refers to the appointment stage applies this term toboth the initial creation of an agency and the addition or transfer of a new program to apre-existing agency

4 Bawnrsquos model derives predictions for the ideal level of statutory control thata legislator would like to see established in the legislation affecting an agencyrsquos mission(Bawn 1997 107) We extend her general framework to a legislatorrsquos preference towardex ante methods

5 Terry Moe describes the extreme case of such overrestriction ldquoIn the interestof public protection agencies are knowingly burdened with cumbersome complicatedtechnically inappropriate structures that undermine their capacity to perform theirjobs wellrdquo (Moe 1990 228)

6 Here complexity refers to ldquothe degree to which specialized technical knowl-edge is necessary to craft effective policy solutions or understand the policy areardquo(Ringuist Worsham and Eisner 2002 6ndash7 see also Gormley 1986)



7 The use of either macro-level (Balla 1998 Epstein and OrsquoHalloran 1999Huber Shipan and Pfahler 2001 Potoski 1998 1999 Spence 1999) or individual-levelresearch (Balla 2000 Bawn 1997) at the national level does not provide sufficientvariation in important explanatory variables such as legislative resources and policysalience These factors vary considerably across the states permitting us to test howvariation in these factors influences legislative decisions about ex ante design Thusstates are a more appropriate setting for tests of ex ante design tactics

8 The set of 2526 legislators includes the populations of legislators from thelower houses of the following state legislatures Arkansas California ColoradoConnecticut Delaware Idaho Illinois Louisiana Maine Michigan Missouri MontanaNevada New Jersey North Dakota Ohio Pennsylvania South Carolina South DakotaTennessee Texas Vermont West Virginia and Wyoming

9 First we gave legislators advance notice of the survey Approximately oneweek later we mailed the survey questionnaire along with a cover letter indicating thatwe would make a report of the results available to the legislators upon completion ofthe project We mailed a reminder postcard and second letter and copy of the surveyquestionnaire to nonrespondents approximately two and four weeks later respectively

10 Members were given the following instructions ldquoWe are interested in thetechniques that you use to ensure that state government agencies do what you wantthem to do For each of the three state government agencies listed below please circle a(1) if you are not likely a (2) if you are somewhat likely or a (3) if you are very likelyto use the following techniques to keep that agency in checkrdquo

11 We recognize that the costs associated with using ex ante tactics differ acrosstypes of individual control technology Our survey included ex ante tactics with higherrelative costs such as reorganizing an agency and lower relative costs such as requiringa report from an agency Because of this variation in cost we created an additive scaleto measure a legislatorrsquos underlying propensity toward these tactics generally

12 We asked legislators if they favored neither favored nor opposed or opposedsix principles regarding environmental policy in their states Legislators scored twopoints per principle when they opposed funding for pollution prevention programssupported requiring cost-benefit analysis of proposed environmental regulations beforeimplementation supported pollution credit programs that allow industries to purchaseadditional pollution beyond their permitted levels supported ldquoself-auditrdquo legislationthat creates incentives for industries to audit themselves and clean up pollution sup-ported requiring the state to compensate citizens when environmental regulations limituses of privately owned land and opposed state environmental regulations being stricterthan federal law Legislators scored one point when they neither favored nor opposeda principle and zero otherwise All six items were positively correlated with oneanother and all their corrected scale-item correlations were positive These resultsalong with the high and significant coefficient of reliability (α = 66) lead us to concludethat the items do reflect membersrsquo underlying attitudes about environmental policy

13 We also included an indirect measure of membersrsquo policy preferences regardingthe environment their constituentsrsquo environmental policy preferences We use the per-cent of the state house legislative district employed in the manufacturing sector as a proxyfor constituency interests regarding this policy area This variable reflects 1990 figures forstate house districts that appear in Lilley DeFranco and Diefenderfer 1994 Data for 1999through 2000 are not available but other demographic characteristics with comparable


403Agency Design

data for both time periods appear to be relatively stable over time For example thecorrelation of 1990 and 1997 figures for district average income and the percent of thepopulation with college education is approximately 97 for all 2526 state house districtsin the study Using the indirect measure of member policy preferences and its interactionwith the presence of a Democratic governor yields nearly identical results to our originalfindings so we report the result using the direct measure of membersrsquo preferences

14 We asked members ldquoAbout how many full-time part-time or volunteerstaff do you have working in your district or capitol officesrdquo We then created adichotomous variable indicating whether or not members had staff in their capitoloffices We chose to use the availability of staff in the capitol office to assess eachmemberrsquos individual resources rather than a measure indicating the staffing for eachlegislature In some legislatures members can choose how to apportion their staffbetween their district and capitol offices We also used an updated professionalizationindex which measures the aggregate resources of the state legislature (Squire 1992) Themeasures produced similar results

15 The complexity of a statersquos environmental policy area is arguably morecomprehensive than only those complexities resulting from air pollution regulationNevertheless the technical challenges that state legislators face with air pollution con-trol require the highest level of technical knowledge Compare air pollution control tomining regulation for example (Gormley 1986 Ringquist Worsham and Eisner 2002)Using this measure of policy complexity allows us to assess the high end of thecomplexity scale for each state

16 The formula used to calculate Potoskirsquos entropy measure is

1ln( )

n

iE p pi i

== minus sum where p represents the probability of the ith Standard Industrial

Classification (SIC) source category Each statersquos total air emissions are broken down intotwo-digit SIC codes These two-digit codes represent the ith category in the equation

17 In addition to typically higher workloads and longer sessions professionalstate legislatures are also characterized by higher salaries greater financial benefits andmore staff and support facilities (Squire 1988a 1988b)

18 The measure of socioeconomic status is a dichotomous measure of educationtaken from the survey the variable is coded as one if the member has at least a collegedegree and as zero otherwise The measure of state legislative professionalism devel-oped by Squire (1992) reflects three main components member salary session lengthand staff support States are compared to the US Congress on all three attributes thestate legislaturersquos score on each attribute was converted into a percentage of Congressrsquosscore For each state the three percentages were then averaged We updated this profes-sionalism index using data on Congress and the 50 state legislatures from 1993 to 1997Squirersquos original measure using 1986ndash88 data and our updated measure computedusing 1993ndash97 data are highly correlated (96)

19 An average legislator is defined as one with mean values on all remaininginterval-level independent variables median values on all ordinal-level variables andmodal values on dichotomous variables In this study an average legislator has anentropy score of 68 a score of 68 points on the environmental policy index and ascore of 86 on the Ranney index of party competition This average legislator also hasapproximately eight years in the state legislature access to legislative staff at thecapitol a Republican governor and the power to review agency decisions



20 For an alternative measure of expertise we used a dummy variable to indicateif the legislator served on a relevant oversight policy committee (Bawn 1997) Thisvariable was not significant either

21 Unlike our measure which varies across individual legislators Huber Shipanand Pfahlerrsquos (2001) measure varies only across chambers As a result the difference inthese findings may be a function of both the level and form of analysis

REFERENCES

Aberbach Joel 1990 Keeping a Watchful Eye The Politics of Congressional Over-sight Washington DC Brookings Institution

Anderson James E 1975 Public Policy Making New York PraegerArnold R Douglas 1987 ldquoPolitical Control of Administrative Officialsrdquo Journal of

Law Economics and Organization 3219ndash86Balla Steven J 1998 ldquoAdministrative Procedures and Political Control of the Bureau-

cracyrdquo American Political Science Review 92663ndash73Balla Steven J 2000 ldquoLegislative Organization and Congressional Review of Agency

Regulationsrdquo Journal of Law Economics and Organization 16424ndash48Bawn Kathleen 1995 ldquoPolitical Control versus Expertise Congressional Choice about

Administration Proceduresrdquo American Political Science Review 8962ndash73Bawn Kathleen 1997 ldquoChoosing Strategies to Control the Bureaucracy Statutory

Constraints Oversight and the Committee Systemrdquo Journal of Law Economicsand Organization 13101ndash26

Bibby John F and Thomas M Holbrook 1999 ldquoParties and Electionsrdquo In Politics inthe American States 4th ed ed Virginia Gray Herbert Jacob and Kenneth NVines Boston Little Brown

Bratton Kathleen A and Kerry L Haynie 1999 ldquoAgenda Setting and LegislativeSuccess in State Legislatures The Effects of Gender and Racerdquo Journal ofPolitics 61658ndash79

Calvert Randall L Mathew D McCubbins and Barry R Weingast 1989 ldquoA Theoryof Political Control and Agency Discretionrdquo American Journal of Political Science33588ndash611

Council of State Governments 2000ndash01 The Book of States Lexington KY Council ofState Governments

DeMaio WE 1980 ldquoRefusals Who Where and Whyrdquo Public Opinion Quarterly 44223ndash33

Dodd Lawrence C and Richard L Schott 1979 Congress and the AdministrativeState New York John Wiley

Elling Richard C 1984 ldquoState Legislative Influence in the Administrative ProcessConsequences and Constraintsrdquo Public Administration Quarterly 7457ndash81

Epstein David and Sharyn OrsquoHalloran 1994 ldquoAdministrative Procedures Informationand Agency Discretionrdquo American Journal of Political Science 38697ndash722

Epstein David and Sharyn OrsquoHalloran 1995 ldquoA Theory of Strategic OversightCongress Lobbyists and the Bureaucracyrdquo Journal of Law Economics andOrganization 11227ndash55

Epstein David and Sharyn OrsquoHalloran 1996 ldquoDivided Government and the Design ofAdministrative Procedures A Formal Model and Empirical Testrdquo Journal ofPolitics 58373ndash97


405Agency Design

Epstein David and Sharyn OrsquoHalloran 1999 Delegating Powers A Transaction CostPolitics Approach to Policy Making under Separate Powers Cambridge UKCambridge University Press

Gerber Brian 2000 ldquoStructuring Policy Access State-level Interest Groups Institu-tional Venues and Air Quality Regulationrdquo Presented at the annual meeting ofthe American Political Science Association Washington DC

Gormley William T Jr 1986 ldquoRegulatory Issue Networks in a Federal SystemrdquoPolity 18595ndash620

Greene William H 1993 Econometric Analysis 2d ed Englewood Cliffs Prentice-HallGroves Robert M Robert B Cialdini and Mick P Couper 1992 ldquoUnderstanding the

Decision to Participate in a Surveyrdquo Public Opinion Quarterly 56475ndash95Hall Richard L 1996 Participation in Congress New Haven Yale University PressHall Richard L and Frank W Wayman 1990 ldquoBuying Time Moneyed Interests and

the Mobilization of Bias in Congressional Committeesrdquo American PoliticalScience Review 84797ndash820

Hamilton James T and Christopher H Schroeder 1994 ldquoStrategic Regulators and theChoice of Rulemaking Procedures The Selection of Formal versus Informal Rulesin Regulating Hazardous Wasterdquo Law and Contemporary Problems 57111ndash60

Hibbing John R 1991 Congressional Careers Contours of Life in the US House ofRepresentatives Chapel Hill University of North Carolina Press

Hibbing John R 1993 ldquoCareerism in Congressrdquo In Congress Reconsidered edLawrence C Dodd and Bruce I Oppenheimer Washington DC CQ Press

Horn Murray J and Kenneth A Shepsle 1989 ldquoCommentary on lsquoAdministrativeArrangements and the Political Control of Agenciesrsquo Administrative Processand Organizational Form as Legislative Responses to Agency Costsrdquo VirginiaLaw Review 75499ndash508

Huber John D and Charles R Shipan 2000 ldquoThe Costs of Control LegislatorsAgencies and Transaction Costsrdquo Legislative Studies Quarterly 2525ndash52

Huber John D Charles R Shipan and Madelaine Pfahler 2001 ldquoLegislatures andStatutory Control of Bureaucracyrdquo American Journal of Political Science 45330ndash45

Kaufman Herbert 1981 The Administrative Behavior of Federal Bureau ChiefsWashington DC Brookings Institution

Kiewiet D Roderick and Mathew D McCubbins 1991 The Logic of DelegationChicago IL University of Chicago

Lilley William III Laurence J DeFranco and William M Diefenderfer III 1994 TheAlmanac of State Legislatures State Data Atlas Washington CQ Press

Lupia Arthur and Mathew D McCubbins 1994 ldquoDesigning Bureaucratic Account-abilityrdquo Law and Contemporary Problems 5791ndash126

Macey Jonathan R 1992 ldquoOrganizational Design and Political Control of Administra-tive Agenciesrdquo Journal of Law Economics and Organization 893ndash111

McCubbins Mathew D Roger G Noll and Barry R Weingast 1987 ldquoAdministra-tive Procedures as Instruments of Political Controlrdquo Journal of Law Economicsand Organization 3243ndash77

McCubbins Mathew D Roger G Noll and Barry R Weingast 1989 ldquoStructure andProcess Politics and Policy Administrative Arrangements and the PoliticalControl of Agenciesrdquo Virginia Law Review 75431ndash82



Moe Terry 1989 ldquoThe Politics of Bureaucratic Structurerdquo In Can the GovernmentGovern ed John Chubb and Paul Peterson Washington DC BrookingsInstitution

Moe Terry 1990 ldquoPolitical Institutions The Neglected Side of the Storyrdquo Journal ofLaw Economics and Organization 6(special issue)213ndash53

North Douglas 1990 Institutions Institutional Change and Economic PerformanceCambridge UK Cambridge University Press

Pearson William M and Van A Wigginton 1986 ldquoEffectiveness of AdministrativeControls Some Perceptions of State Legislatorsrdquo Public Administration Review46328ndash31

Potoski Matthew 1998 ldquoManaging Uncertainty in Environmental Policy Air Pollu-tion Regulation in the American Statesrdquo PhD diss Indiana University

Potoski Matthew 1999 ldquoManaging Uncertainty through Bureaucratic DesignAdministrative Procedures and State Air Pollution Control Agenciesrdquo Journalof Public Administration Research and Theory 9623ndash39

Ringquist Evan J 1993 Environmental Protection at the State Level Armonk NYME Sharpe

Ringquist Evan J Jeff Worsham and Marc Allen Eisner 2002 ldquoSalience Complexityand the Legislative Direction of Regulatory Bureaucraciesrdquo Unpublishedmanuscript

Robinson Glen O 1989 ldquoCommentary on lsquoAdministrative Arrangements and thePolitical Control of Agenciesrsquo Political Uses of Structure and Processrdquo VirginiaLaw Review 75483ndash98

Rourke Francis E 1976 Bureaucracy Politics and Public Policy Boston Little BrownShepsle Kenneth A 1992 ldquoBureaucratic Drift Coalitional Drift and Time Consis-

tency A Comment on Maceyrdquo Journal of Law Economics and Organization 8111ndash25

Smith TW 1983 ldquoThe Hidden 25 Percent An Analysis of Nonresponse on the 1980General Social Surveyrdquo Public Opinion Quarterly 47386ndash404

Spence David B 1997 ldquoAgency Policy Making and Political Control Modeling awaythe Delegation Problemrdquo Journal of Public Administration Research and Theory7199ndash219

Spence David B 1999 ldquoAgency Discretion and the Dynamics of Procedural ReformrdquoPublic Administration Review 59425ndash42

Squire Peverill 1988a ldquoCareer Opportunities and Membership Stability in Legisla-turesrdquo Legislative Studies Quarterly 1365ndash82

Squire Peverill 1988b ldquoMember Career Opportunities and the Internal Organizationof Legislaturesrdquo Journal of Politics 50726ndash44

Squire Peverill 1992 ldquoLegislative Professionalization and Membership Diversity inState Legislaturesrdquo Legislative Studies Quarterly 1769ndash79

US Environmental Protection Agency (EPA) 1999 Office of Pollution Preventionand Toxics Toxics Release Inventory Wahington DC Environmental Protec-tion Agency

Weingast Barry R and Mark J Moran 1983 ldquoBureaucratic Discretion or Congres-sional Control Regulatory Policymaking by the Federal Trade CommissionrdquoJournal of Political Economy 91765ndash800



Spence 1997) legislators may not prefer ex ante design as an ongoingtactic to achieve influence (Robinson 1989) If this is the case then theutility of ex ante design may be overstated in theoretical treatments(Robinson 1989)

In this paper we seek to answer two questions to what extent dostate legislators prefer to use ex ante design as an ongoing tactic of bureau-cratic influence and what factors condition these preferences Focusingon environmental policy we use a mail survey of more than 2500 legisla-tors in 24 states to collect data on state legislatorsrsquo preferences towardvarious ex ante tactics of bureaucratic control their policy preferencesand other individual-level characteristics Using the survey data we createan index that measures state legislatorsrsquo preferences for using ex antetactics to influence state environmental agencies We then combine thesedata with state-level institutional data to test whether or not individual andinstitutional differences explain legislatorsrsquo preferences for ex ante tactics

Exploring the Assumptions of Ex Ante Design Tactics

The literature on bureaucratic responsiveness suggests that exante design reduces the costs that legislators incur when they delegateauthority to bureaucratic agents (Bawn 1995 1997 Calvert McCubbinsand Weingast 1989 Epstein and OrsquoHalloran 1994 1995 1996 1999Huber and Shipan 2000 McCubbins Noll and Weingast 1987 1989)Such ex ante limitations on bureaucratic behavior involve the construc-tion of bureaucratic procedures organizational mandates and a varietyof other passive attempts at control (Calvert McCubbins and Weingast1989) Proponents argue that ex ante design is a cheaper more effi-cient alternative to other tactics that require costly monitoring and expost punishment With ex ante design legislators are said to be able todecrease the informational asymmetries and agent shirking inherent indelegation they do so by structuring an agencyrsquos rules and proceduresin such a way as to reduce ldquothe ability of an agency or other executiveactors to enact outcomes different from the policies preferred by thosewho originally delegated powerrdquo (Epstein and OrsquoHalloran 1994 699)

Empirical research suggests that the legislative application of exante tactics to alter agentsrsquo discretionary environments is generallyinfluenced by both institutional- and individual-level characteristics Takentogether these factors condition the level of discretion that legislaturesseek to obtain through their design of an agencyrsquos structures andprocedures State-level research capitalizes on the institutional variationacross the states This research suggests that the saliency and complexityof the policy area (Potoski 1998) the institutional characteristics of the


385Agency Design












387Agency Design












Opportunity Costs






389Agency Design










Research Design



Dependent Variable




391Agency Design























393Agency Design



Statistical Model






Results





395Agency Design





21

34

58

92

70

87

4137

179 8

0

10

20

30

40

50

60

70

80

90

100

0 1 2 3 4 5 6 7 8 9 10


Num

ber

of R

espo

nden

ts




by State


Arkansas 412 252 17


Colorado 300 166 9


Delaware 440 207 5

Idaho 543 191 14

Illinois 388 181 8

Louisiana 457 98 7

Maine 433 204 45

Michigan 437 248 19

Missouri 336 222 22

Montana 525 256 24


Nebraska 435 166 17


Ohio 457 271 14





Texas 475 238 12

Vermont 461 258 36


Wyoming 352 169 21




397Agency Design


Variable


b se(b)

















399Agency Design








Conclusion





401Agency Design



NOTES

















403Agency Design





1ln( )

n

iE p pi i










REFERENCES



















405Agency Design










































385Agency Design












387Agency Design












Opportunity Costs






389Agency Design










Research Design



Dependent Variable




391Agency Design























393Agency Design



Statistical Model






Results





395Agency Design





21

34

58

92

70

87

4137

179 8

0

10

20

30

40

50

60

70

80

90

100

0 1 2 3 4 5 6 7 8 9 10


Num

ber

of R

espo

nden

ts




by State


Arkansas 412 252 17


Colorado 300 166 9


Delaware 440 207 5

Idaho 543 191 14

Illinois 388 181 8

Louisiana 457 98 7

Maine 433 204 45

Michigan 437 248 19

Missouri 336 222 22

Montana 525 256 24


Nebraska 435 166 17


Ohio 457 271 14





Texas 475 238 12

Vermont 461 258 36


Wyoming 352 169 21




397Agency Design


Variable


b se(b)

















399Agency Design








Conclusion





401Agency Design



NOTES

















403Agency Design





1ln( )

n

iE p pi i










REFERENCES



















405Agency Design

















































387Agency Design












Opportunity Costs






389Agency Design










Research Design



Dependent Variable




391Agency Design























393Agency Design



Statistical Model






Results





395Agency Design





21

34

58

92

70

87

4137

179 8

0

10

20

30

40

50

60

70

80

90

100

0 1 2 3 4 5 6 7 8 9 10


Num

ber

of R

espo

nden

ts




by State


Arkansas 412 252 17


Colorado 300 166 9


Delaware 440 207 5

Idaho 543 191 14

Illinois 388 181 8

Louisiana 457 98 7

Maine 433 204 45

Michigan 437 248 19

Missouri 336 222 22

Montana 525 256 24


Nebraska 435 166 17


Ohio 457 271 14





Texas 475 238 12

Vermont 461 258 36


Wyoming 352 169 21




397Agency Design


Variable


b se(b)

















399Agency Design








Conclusion





401Agency Design



NOTES

















403Agency Design





1ln( )

n

iE p pi i










REFERENCES



















405Agency Design













































Opportunity Costs






389Agency Design










Research Design



Dependent Variable




391Agency Design























393Agency Design



Statistical Model






Results





395Agency Design





21

34

58

92

70

87

4137

179 8

0

10

20

30

40

50

60

70

80

90

100

0 1 2 3 4 5 6 7 8 9 10


Num

ber

of R

espo

nden

ts




by State


Arkansas 412 252 17


Colorado 300 166 9


Delaware 440 207 5

Idaho 543 191 14

Illinois 388 181 8

Louisiana 457 98 7

Maine 433 204 45

Michigan 437 248 19

Missouri 336 222 22

Montana 525 256 24


Nebraska 435 166 17


Ohio 457 271 14





Texas 475 238 12

Vermont 461 258 36


Wyoming 352 169 21




397Agency Design


Variable


b se(b)

















399Agency Design








Conclusion





401Agency Design



NOTES

















403Agency Design





1ln( )

n

iE p pi i










REFERENCES



















405Agency Design










































389Agency Design










Research Design



Dependent Variable




391Agency Design























393Agency Design



Statistical Model






Results





395Agency Design





21

34

58

92

70

87

4137

179 8

0

10

20

30

40

50

60

70

80

90

100

0 1 2 3 4 5 6 7 8 9 10


Num

ber

of R

espo

nden

ts




by State


Arkansas 412 252 17


Colorado 300 166 9


Delaware 440 207 5

Idaho 543 191 14

Illinois 388 181 8

Louisiana 457 98 7

Maine 433 204 45

Michigan 437 248 19

Missouri 336 222 22

Montana 525 256 24


Nebraska 435 166 17


Ohio 457 271 14





Texas 475 238 12

Vermont 461 258 36


Wyoming 352 169 21




397Agency Design


Variable


b se(b)

















399Agency Design








Conclusion





401Agency Design



NOTES

















403Agency Design





1ln( )

n

iE p pi i










REFERENCES



















405Agency Design











































Research Design



Dependent Variable




391Agency Design























393Agency Design



Statistical Model






Results





395Agency Design





21

34

58

92

70

87

4137

179 8

0

10

20

30

40

50

60

70

80

90

100

0 1 2 3 4 5 6 7 8 9 10


Num

ber

of R

espo

nden

ts




by State


Arkansas 412 252 17


Colorado 300 166 9


Delaware 440 207 5

Idaho 543 191 14

Illinois 388 181 8

Louisiana 457 98 7

Maine 433 204 45

Michigan 437 248 19

Missouri 336 222 22

Montana 525 256 24


Nebraska 435 166 17


Ohio 457 271 14





Texas 475 238 12

Vermont 461 258 36


Wyoming 352 169 21




397Agency Design


Variable


b se(b)

















399Agency Design








Conclusion





401Agency Design



NOTES

















403Agency Design





1ln( )

n

iE p pi i










REFERENCES



















405Agency Design










































391Agency Design























393Agency Design



Statistical Model






Results





395Agency Design





21

34

58

92

70

87

4137

179 8

0

10

20

30

40

50

60

70

80

90

100

0 1 2 3 4 5 6 7 8 9 10


Num

ber

of R

espo

nden

ts




by State


Arkansas 412 252 17


Colorado 300 166 9


Delaware 440 207 5

Idaho 543 191 14

Illinois 388 181 8

Louisiana 457 98 7

Maine 433 204 45

Michigan 437 248 19

Missouri 336 222 22

Montana 525 256 24


Nebraska 435 166 17


Ohio 457 271 14





Texas 475 238 12

Vermont 461 258 36


Wyoming 352 169 21




397Agency Design


Variable


b se(b)

















399Agency Design








Conclusion





401Agency Design



NOTES

















403Agency Design





1ln( )

n

iE p pi i










REFERENCES



















405Agency Design

















































393Agency Design



Statistical Model






Results





395Agency Design





21

34

58

92

70

87

4137

179 8

0

10

20

30

40

50

60

70

80

90

100

0 1 2 3 4 5 6 7 8 9 10


Num

ber

of R

espo

nden

ts




by State


Arkansas 412 252 17


Colorado 300 166 9


Delaware 440 207 5

Idaho 543 191 14

Illinois 388 181 8

Louisiana 457 98 7

Maine 433 204 45

Michigan 437 248 19

Missouri 336 222 22

Montana 525 256 24


Nebraska 435 166 17


Ohio 457 271 14





Texas 475 238 12

Vermont 461 258 36


Wyoming 352 169 21




397Agency Design


Variable


b se(b)

















399Agency Design








Conclusion





401Agency Design



NOTES

















403Agency Design





1ln( )

n

iE p pi i










REFERENCES



















405Agency Design












































Results





395Agency Design





21

34

58

92

70

87

4137

179 8

0

10

20

30

40

50

60

70

80

90

100

0 1 2 3 4 5 6 7 8 9 10


Num

ber

of R

espo

nden

ts




by State


Arkansas 412 252 17


Colorado 300 166 9


Delaware 440 207 5

Idaho 543 191 14

Illinois 388 181 8

Louisiana 457 98 7

Maine 433 204 45

Michigan 437 248 19

Missouri 336 222 22

Montana 525 256 24


Nebraska 435 166 17


Ohio 457 271 14





Texas 475 238 12

Vermont 461 258 36


Wyoming 352 169 21




397Agency Design


Variable


b se(b)

















399Agency Design








Conclusion





401Agency Design



NOTES

















403Agency Design





1ln( )

n

iE p pi i










REFERENCES



















405Agency Design










































395Agency Design





21

34

58

92

70

87

4137

179 8

0

10

20

30

40

50

60

70

80

90

100

0 1 2 3 4 5 6 7 8 9 10


Num

ber

of R

espo

nden

ts




by State


Arkansas 412 252 17


Colorado 300 166 9


Delaware 440 207 5

Idaho 543 191 14

Illinois 388 181 8

Louisiana 457 98 7

Maine 433 204 45

Michigan 437 248 19

Missouri 336 222 22

Montana 525 256 24


Nebraska 435 166 17


Ohio 457 271 14





Texas 475 238 12

Vermont 461 258 36


Wyoming 352 169 21




397Agency Design


Variable


b se(b)

















399Agency Design








Conclusion





401Agency Design



NOTES

















403Agency Design





1ln( )

n

iE p pi i










REFERENCES



















405Agency Design












































by State


Arkansas 412 252 17


Colorado 300 166 9


Delaware 440 207 5

Idaho 543 191 14

Illinois 388 181 8

Louisiana 457 98 7

Maine 433 204 45

Michigan 437 248 19

Missouri 336 222 22

Montana 525 256 24


Nebraska 435 166 17


Ohio 457 271 14





Texas 475 238 12

Vermont 461 258 36


Wyoming 352 169 21




397Agency Design


Variable


b se(b)

















399Agency Design








Conclusion





401Agency Design



NOTES

















403Agency Design





1ln( )

n

iE p pi i










REFERENCES



















405Agency Design










































397Agency Design


Variable


b se(b)

















399Agency Design








Conclusion





401Agency Design



NOTES

















403Agency Design





1ln( )

n

iE p pi i










REFERENCES



















405Agency Design















































399Agency Design








Conclusion





401Agency Design



NOTES

















403Agency Design





1ln( )

n

iE p pi i










REFERENCES



















405Agency Design












































Conclusion





401Agency Design



NOTES

















403Agency Design





1ln( )

n

iE p pi i










REFERENCES



















405Agency Design










































401Agency Design



NOTES

















403Agency Design





1ln( )

n

iE p pi i










REFERENCES



















405Agency Design



















































403Agency Design





1ln( )

n

iE p pi i










REFERENCES



















405Agency Design













































REFERENCES



















405Agency Design










































405Agency Design










































Agency Design as an Ongoing Tool of Bureaucratic Influenceagency s structure and procedures in an...

Documents

Transcript of Agency Design as an Ongoing Tool of Bureaucratic Influenceagency s structure and procedures in an...