You are on page 1of 16

DECONSTRUCTINGNONDELEGATION

CYNTHIAR.FARINA*

ThisEssaysuggeststhatthepersistenceofdebatesoverdelegation to agencies cannot persuasively be explained as a determination fi nallytogetconstitutionallawright,fornondelegationdoctrine atleastastraditionallystateddoesnotrestonaparticularlysound legal foundation. Rather, these debates continue because nondelega tion provides a vehicle for pursuing a number of different concerns about the modern regulatory state. Whether or not one shares these concerns,theyarenottrivial,andweshouldvoiceandengagethem directlyratherthancontinuetousenondelegationasastalkinghorse. If Academy Awards were given in constitutional jurispru dence,nondelegationclaimsagainstregulatorystatuteswould win the prize for Most Sympathetic Judicial Rhetoric in a Hopeless Case. At the end of the nineteenth century, Justice Harlan,writingforamajorityoftheSupremeCourt,declared: That congress cannot delegate legislative power...is a prin cipleuniversallyrecognizedasvitaltotheintegrityandmain tenanceofthesystemofgovernmentordainedbytheConstitu tion.1 At the beginning of the twentyfirst century, Justice Scalia, for a unanimous Court, insisted: Article I, 1, of the Constitution vests [a]lllegislative Power herein granted...in aCongressoftheUnitedStates...[and]permitsnodelegation of those powers.2 Yet in both these casesand in virtually every intervening delegation challengethe court sustained thestatuteatissue.3Indeed,JusticeScaliasopinion,whichfol
*ProfessorofLaw,CornellLawSchool;Director,CornelleRulemakingInitia tive(CeRI);Director,CornellLaw&PublicPolicyProgram. 1.Fieldv.Clark,143U.S.649,692(1892). 2.Whitmanv.Am.TruckingAssns,531U.S.457,472(2001)(firstalterationin original)(quotingU.S.CONST.art.I,1). 3.SeePETER L. STRAUSS, TODD D. RAKOFF & CYNTHIA R. FARINA, GELLHORN & BYSESADMINISTRATIVELAW6677(rev.10thed.2003).

88

HarvardJournalofLaw&PublicPolicy

[Vol.33

lowed an array of interesting and sophisticated D.C. Circuit opinions debating the constitutionality of delegating author itytotheEPAtosetnationalambientairqualitystandards,4is one of the blandest, most pedantic opinions the Justice has everpenned.Hestopsjustshortofconsigningthisostensibly vital and unqualified constitutional principle to the nether world of nonjusticiability.5 Like the Ninth Amendment and theGuarantyClause,nondelegationappearstobeaconstitu tionallostcause.6 The courts hesitance to enforce nondelegation, however, doesnotdeterchallengestothelegalityofregulatorydelega tions. Symposia devoted to the debate appear with regular ity,7 and nondelegation articles often appear in symposia on other topics as well.8 Earlier this year at a conference on presidential power, Judge Ginsburg reiterated his conviction thatdelegationssuchasthoseintheCleanAirActareconsti tutionally unacceptable and deprecated what he views as ju dicialabdicationtolegislativejudgment.9Inremarksconclud ing the conference, Professor Steven Calabresi proposed forcing Congress to reconsider the scope of hundreds of fed eral programs by treating the legislative veto provisions that
4.SeeAm.TruckingAssnsv.EPA,195F.3d4(D.C.Cir.1999)(percuriam);Am. TruckingAssnsv.EPA,175F.3d1027(D.C.Cir.1999)(percuriam). 5.Whitman, 531 U.S. at 47475 ([W]e have almost never felt qualified to sec ondguessCongressregardingthepermissibledegreeofpolicyjudgmentthatcan be left to those executing or applying the law. (quoting Mistretta v. United States,488U.S.361,416(1989)(Scalia,J.,dissenting))). 6.State constitutional opinions similarly tend to begin with a ringing declara tion of the constitutional centrality of an absolute nondelegation principle, after whichtheysustainthechallengedstatute.See,e.g.,Stateexrel.R.R.&Warehouse Commnv.Chicago,Milwaukee&St.PaulR.R.Co.,37N.W.782,78688(Minn. 1888); Cincinnati, Wilmington & Zanesville R.R. Co. v. Commrs of Clinton County,1OhioSt.77,8791(1852). 7.E.g.,Symposium,ThePhoenixRisesAgain:TheNondelegationDoctrinefromCon stitutionalandPolicyPerspectives,20CARDOZOL.REV.731(1999). 8.See, e.g., James Boyle, A Nondelegation Doctrine for the Digital Age?, 50 DUKE L.J.5(2000);JayS.Bybee&TuanN.Samahon,WilliamRehnquist,theSeparationof Powers, and the Riddle of the Sphinx, 58 STAN. L. REV. 1735, 175051 (2006); Jim Rossi,StatutoryNondelegation:LearningfromFloridasRecentExperienceinAdmin istrativeProcedureReform,8WIDENERJ.PUB.L.301(1999). 9.SeeDouglasH.Ginsburg&StevenMenashi,NondelegationandtheUnitaryEx ecutive,12U. PA. J. CONST. L.(forthcoming2010).JudgeGinsburgwasinthema jorityofthepanelthatinitiallyheldthedelegationunconstitutional,Am.Trucking Assns,175F.3dat1033,andthendeniedthegovernmentspetitionforrehearing, Am.TruckingAssns,195F.3dat6.

No.1]

DeconstructingNondelegation

89

commonlyaccompanied regulatory delegations before INS v. Chadha10asnonseverable.11 This Essay reflects on the remarkable durability of nondele gation arguments. It begins by exploring why the legal case against delegation is not sufficiently robust to account for the persistence of the nondelegation debate in the face of nearly two hundred years of rejection by the courts12 and political branches. It suggests that the controversy endures because nondelegation offers a vehicle for pursuing several serious concerns about federal regulation. Each of these concerns at least arguably implicates constitutional values, and all arise fromthecumulative effectofCongressspracticeofbroad dele gationovertimeratherthanfromtheactofdelegatingitself.As a simultaneously timehonored and perennially unpersuasive framing, nondelegation provides a constitutional home for these concerns but ultimately prevents the kind of direct en gagementonthemeritsthattheydeserve. I. The existence of a constitutional nondelegation principle is typically accepted as given, so the focus of debate moves im mediatelytowhethertheexistingdoctrinalapproachcorrectly operationalizes this principle. But consider, for a moment, the basis for assuming that the Constitution forbids Congress to givesignificantpolicymakingauthoritytoanotherentity,such asaregulatoryagency.TheConstitutionstextisoflittlehelp, foritsaysnothingexplicitaboutdelegatingthepowerArticleI confers.13 The early cases that vehemently pronounced the in
10.462U.S.919(1983). 11.SeeStevenG.Calabresi,ClosingRemarksattheUniversityofPennsylvania Journal of Constitutional Law Symposium: Presidential Power in Historical Per spective(Feb.7,2009). 12.The earliest federal nondelegation challengesand lossescame within decades of the ratification of the Constitution. See Wayman v. Southard, 23 U.S. (10Wheat.)1(1825);CargooftheBrigAurorav.UnitedStates,11U.S.(7Cranch) 382(1813). 13.SomescholarsdoinsistthattheArticleIvestingclauseitselfestablishesthe nondelegationprinciple.See,e.g.,MarciA.Hamilton,RepresentationandNondelega tion:BacktoBasics,20CARDOZOL.REV.807,807(1999)(ThelanguageoftheCon stitution would seem to prescribe a brightline doctrinal approach.). With re spect,thispositionseemsunsupportable.Withinourlegalsystem,asimplegrant ofpowerwithoutmoredoesnotconclusivelyresolvewhetherorwhenanagent

90

HarvardJournalofLaw&PublicPolicy

[Vol.33

dispensability of a nondelegation principle invariably spent their energy demonstrating why the statutes under attack did notviolatetheprinciple,ratherthanexplainingtheprinciples origins.14 Moreover, these demonstrations were largely func tionalassessmentsofthepragmaticneedsofgovernment.The conspicuousabsenceoftypicalconstitutionalinterpretivecon cernswithtext,intent,andpurpose,coupledwiththepresenta tion of delegation as selfevidently problematic and the focus onpracticaljustifiability,suggeststhatthesecourtsweredraw ingonbackgroundlegalunderstandingsneitherspecifictothe Constitutionnoropentoseriouscontention. Inotherwords,nondelegationcaseshistoricallylookedmore liketheordinaryapplicationofgeneralcommonlawprinciples thanlikeextraordinarymomentsofconstitutionalexegesis.For example,amidnineteenthcenturyOhioSupremeCourtopin ioneventuallyquoted approvinglybyJustice HarlaninField v. Clark15describes as too clear for argument the proposi tionthatthegeneralassemblycannotsurrenderanyportion ofthelegislativeauthoritywithwhichitisinvested,orauthor ize its exercise by any other person or body.16 The court ex plained:Thisinabilityarisesnolessfromthegeneralprinciple applicabletoeverydelegatedpowerrequiringknowledge,dis cretion, and rectitude, in its exercise, than from the positive provisions of the constitution itself.17 Similarly, when Chief Justice Taft established the modern intelligible principle standard in 1928, he placed nondelegation squarely on ordi narycontractualprinciples:
cansubdelegate.Seeinfratextaccompanyingnotes2143.Moreover,iftheArticle Ivestingclausepreventsdelegation,so,itwouldseem,musttheArticleIIandIII vesting clauses. Yet the Presidents ability to delegate executive power is well established. See infra text accompanying notes 2728. As for Article III, the rela tionship between Article III judges and nonArticle III decision makers, such as magistratesandbankruptcyjudges,isaperennialconceptualmaze.SeeRICHARD H.FALLON,JR.ETAL.,HARTANDWECHSLERSTHEFEDERALCOURTSANDTHEFED ERALSYSTEM363(6thed.2009)(statingthattheSupremeCourthasbroughtlittle butconfusiontothisarea).Itblinksreality,however,toassertthatnopartofthe judicialpoweroftheUnitedStateshasbeendelegatedtothoseofficials. 14.See,e.g.,J.W.Hampton,Jr.&Co.v.UnitedStates,276U.S.394,40511(1928); Cincinnati, Wilmington & Zanesville R.R. Co. v. Commrs of Clinton County, 1 Ohio.St.77,8788(1852). 15.143U.S.649,69394(1892). 16.Cincinnati,Wilmington&ZanesvilleR.R.Co.,1OhioSt.at87. 17.Id.

No.1]

DeconstructingNondelegation

91

The wellknown maxim Delegata potestas non potest dele gari[Nodelegatedpowerscanbefurtherdelegated],ap plicable to the law of agency in the general and common law,iswellunderstoodandhashadwiderapplicationinthe construction of our Federal and State Constitutions than it hasinprivatelaw.18

Modern explanations of nondelegation emphasize political theoryratherthanthecommonlaw,tracingtheprincipletoJohn Lockes Second Treatise of Government.19 Even in Locke, though, thecontractualparadigmisevident,withthelegislativepartof government conceptualized as receiving its power from the peoplebypositivevoluntaryGranttoactastheiragent.20 It is thus worth considering what agency law actually says about an agents authority to delegate further the power con ferred by the principal. According to the first Restatement of Agency, the general rule is indeed that, [u]nless otherwise agreed,authoritytoconductatransactiondoesnotincludeau thoritytodelegatetoanothertheperformanceofactsincidental thereto which involve discretion or the agents special skill.21 Yet this rulecaptured in the delegata potestas maximonly beginstheanalysis.Asecondgeneralruleisthatauthorityto conduct a transaction includes authority to do acts which are incidentaltoit,usuallyaccompanyit,orarereasonablyneces
18.J.W.Hampton,Jr.&Co.,276U.S.at40506. 19.See, e.g., Indus. Union Dept v. Am. Petroleum Inst., 448 U.S. 607, 67273 (1980)(Rehnquist,J.,concurring);LarryAlexander&SaikrishnaPrakash,Reports of the Nondelegation Doctrines Death Are Greatly Exaggerated, 70 U. CHI. L. REV. 1297, 1297 (2003); Eric A. Posner & Adrian Vermeule, Interring the Nondelegation Doctrine,69U.CHI.L.REV.1725,1729(2002). 20.JOHN LOCKE, SECOND TREATISE OF GOVERNMENT 141 (C.B. Macpherson ed.,HackettPublgCo.,Inc.1980)(1690).Lockewrote: The legislative cannot transfer the power of making laws to any other hands: for it being but a delegated power from the people, they, who haveit,cannotpassitovertoothers....Andwhenthepeoplehavesaid, We will submit to rules, and be governed by laws made by such men, and in such forms, no body else can say other men shall make laws for them;norcanthepeoplebeboundbyanylawsbutsuchasareenacted bythose,whomtheyhavechosen,andauthorizedtomakelawsforthem. Thepowerofthelegislative,beingderivedfromthepeoplebyapositive voluntary grant and institution, can be no other, than what the positive grant conveyed, which being only to make laws, and not to make legislators,thelegislativecanhavenopowertotransfertheirauthorityof makinglaws,andplaceitinotherhands. Id.at141(emphasisomitted). 21.RESTATEMENT(FIRST)OFAGENCY78(1933).

92

HarvardJournalofLaw&PublicPolicy

[Vol.33

sary to accomplish it.22 This authority includes the power to appoint other agents or subagents when such appointment is reasonablyrequire[d]23oritisimpracticablefortheagentto perform [the task] in person.24 The rationale for this second ruleisstraightforwardandsensible:Because[a]llauthorityis granted for the accomplishment of certain purposes of the principal,25itisinferredthattheprincipalisnotdoingavain thing, but intends to give a workable and effective consent26 whencreatingtheagencyrelationship. Although the delegata potestas maxim is more familiar, this second general rule of agency law is no stranger to structural constitutionalinterpretation.Indeed,itiscentraltohowschol arsunderstandtheauthorityconferredbythePeopleinArticle II:ThePresidentmaydelegatehispowertosubagentsbecause delegation is reasonably necessary to accomplish his constitu tional functions.27 Presidential delegation rests no more ex presslyonthetextthandoescongressionaldelegation.Indeed, the President is not explicitly authorized, as Congress is, to takeactionswhichshallbenecessaryandproperforcarrying into Execution constitutionally conferred powers.28 Particu larlyinlightoftheNecessaryandProperClause,itisdifficult to understand why background principles of agency law wouldoperatemorerestrictivelywhenitcomestoappointing subagentsnecessarytoaccomplishthepurposesoftheprinci pal29that is, the peoplein Article I than in Article II. Of course, broad grants of regulatory power to administrative

22.Id.35. 23.Id.79([A]nagentisauthorizedtoappointanotheragentfortheprincipal if,amongothercontingencies,theproperconductoftheprincipalsbusinessin thecontemplatedmannerreasonablyrequirestheemploymentofotheragents.). 24.Id. 80 ([A]uthority to appoint a subagent is inferred from authority to conductatransactionfortheprincipalif,amongotherthings,thebusinessisof suchanature...thatitisimpracticablefortheagenttoperformitinperson.). 25.Id.34cmt.d. 26.Id.35cmt.c. 27.E.g.,GaryLawson,TheRiseandRiseoftheAdministrativeState,107HARV. L. REV.1231,1242(1994)(Ofcourse,thePresidentcannotbeexpectedpersonallyto executealllaws.). 28.U.S.CONST.art.I,8,cl.18. 29.RESTATEMENT(FIRST)OFAGENCY34cmt.d.

No.1]

DeconstructingNondelegation

93

agencies can be justified as reasonably require[d]30 subdele gationsonlyifachievementofthestatutoryobjectivesiswithin the charge that the people, through the Constitution and the ordinarypoliticalprocess,havegivenCongresstoaccomplish. ThescopeofCongresssauthorityisthusanimportantissue,to whichthisEssaywillreturnshortly. Anotherbasiccomponentofagencylawisworthnoting:the impact of changed conditions. When a principal engages an agent to act for him over time, the scope of actual authority cannotbestaticorthecourseofexternaleventsmightleavethe agentunabletoachievethegoalsoftheprincipal.Thegeneral rule,therefore,isthatauthorizationisinterpretedasofthetime it is acted upon, in light of the conditions under which it was made and changes in conditions subsequent thereto.31 As a result, a change of circumstances may increase, diminish, or terminate [the agents] privilege to exercise a power for the principal.32Theprincipalneednothaveanticipatedtheconse quences of the changed conditions.33 To be sure, what ulti mately defines the agents actual authority is the principals manifestations of consent to him.34 But this inquiry asks whether the conduct of the principal..., reasonably inter preted, causes the agent to believe that the principal desires him so to act on the principals account.35 When the agent is actingundersubstantiallychangedcircumstances,theconsent oftheprincipalisreasonablyinferrediftheprincipalisaware of the change and its effect and is in a position to change his orders if he desires such change.36 More generally, an infer enceofauthorityreasonablyariseswhentheagentperformsa seriesofactsofasimilarnatureandtheprincipal,knowingof thoseacts,doesnotmanifestdisapproval.37
30.Id.79([A]nagentisauthorizedtoappointanotheragentfortheprincipal if...the proper conduct of the principals business in the contemplated manner reasonablyrequirestheemploymentofotheragents.). 31.Id.33(emphasisadded). 32.Id.33cmt.a. 33.Id.35cmt.c(Itisnotessentialtotheauthorizationofanactthattheprin cipalshouldhavecontemplatedthattheagentwouldperformitasincidentalto theauthorizedperformance.). 34.Id.7. 35.Id.26. 36.Id.33cmt.a. 37.Id.43cmt.b.

94

HarvardJournalofLaw&PublicPolicy

[Vol.33

All these principles of agency law go to the scope of the agents actual authority. They are not defining the apparent authority upon which a third party may reasonably rely but whichdoesnotaltertherelationshipbetweentheagentandhis principal,38noraretheyaddressingthedistinctquestionofwhen anagentsunauthorizedactscanbelegitimatedafterthefactby the principals affirmance or ratification.39 They comprise the primary framework for defining the bounds of the agents le gitimate authority to acta framework in which constraint on (sub)delegationisonlyoneofseveralinterrelatedcomponents. Within this framework, the historical course of nondelega tion jurisprudence is more explicable. Delegata potestas non potest delegari begins, but does not end, judicial analysis. The pragmatic assessment of government needs40 and the notice takenofchangingnationalandinternationalcircumstances41 inquiriesthatseemodd,perhapsevenirresponsible,aspartof enforcing a fundamental constitutional prohibitionare quite unexceptionalwaystodeterminewhetherageneralagentmay lawfully delegate power to others to perform tasks impracti cable for the agent to perform...in person.42 Indeed, in UnitedStatesv.Grimaud,afamousearlytwentiethcenturycase that rejects a nondelegation challenge to a criminal conviction forviolatinganadministrativeregulation,theCourtmadeex plicit the common law reasoning that informed the constitu tional inquiry: Congress might rightfully entrust to others thepowertomakeregulationsjustasanownermaydelegate

38.See,e.g.,id.49,159(discussingapparentauthorityandtherulesthatapply tounauthorizedactstakenunderapparentauthority). 39.Seeid.8284,93(definingratificationandaffirmance). 40.See,e.g.,UnitedStatesv.Grimaud,220U.S.506,516(1911)(notingthat[i]n thenatureofthingsitwasimpracticableforCongresstoprovidegeneralregula tions for these various and varying details of management of federal reserved lands). 41.See,e.g.,Mistrettav.UnitedStates,488U.S.361,372(1989)([I]nourincreas ingly complex society, replete with ever changing and more technical problems, Congresssimplycannotdoitsjobabsentanabilitytodelegatepowerunderbroad generaldirectives.). 42.RESTATEMENT(FIRST)OF AGENCY80([A]uthoritytoappointasubagentis inferredfromauthoritytoconductatransactionfortheprincipalif,amongother things,thebusinessisofsuchanature...thatitisimpracticablefortheagentto performitinperson.).

No.1]

DeconstructingNondelegation

95

tohisprincipalagenttherighttoemploysubordinates,giving tothemalimiteddiscretion.43 II. Insum,theindefatigablefervorwithwhichweclingtonon delegation arguments is difficult to justify based on either the language of the Constitution or the background understand ingsofagencylawfromwhichthedelegatapotestasmaximde rives. Rather, it appears to be anxiety about the consequences oftwocenturiesofstatutorydelegationtoagenciesthatkeeps thedelegationdebatealive.Thisdisquietaboutmodernregula torygovernmentcomprisesseveraldistinctconcerns. The first is that broad congressional delegation to agencies has made it too easy for the federal government to establish, over time, a huge regulatory regime reaching virtually every significant aspect of our social and economic lives. Lawyers and political scientists alike have charged that delegation en ables the legislature to punt the really tough policy choices.44 Because Congress can do more when it is allowed to do less, the result has been a proliferation of federal social and eco nomic regulation, much of which arguably should not exist. The constitutionally based version of this concern, most clearly articulatedbyGaryLawson,seesoverlybroadconstructionsof theCommerceClause,thespendingpower,andtheNecessary andProperClauseasworkinginconjunctionwithoverlynar row applications of the Due Process and Takings Clauses and theTenthAmendmenttoallowthenationalgovernmentafar greater scope of action than was originally intended.45 Uncon straineddelegationenablesCongresstoexploittheseinterpre tive errors to the fullest, producing a regime of unconstitu tional federal regulatory overreaching.46 The policy based version insists that, as a matter of good government, less is more.Delegationhasmadeitsoeasytocreatefederalregula toryprogramsthatthenationalgovernmentnowintervenesin
43.Grimaud,220U.S.at516(internalquotationmarksomitted). 44.See,e.g.,THEODOREJ.LOWI,THEENDOFLIBERALISM:THESECONDREPUBLICOF THE UNITED STATES 92126 (2d ed. 1979); DAVID SCHOENBROD, POWER WITHOUT RESPONSIBILITY:HOWCONGRESSABUSESTHEPEOPLETHROUGHDELEGATION(1993). 45.Lawson,supranote27,at123337. 46.Id.at123741.

96

HarvardJournalofLaw&PublicPolicy

[Vol.33

mattersthatthemarket,orlocalgovernments,wouldmoreef ficientlyandeffectivelymanage.47Arelatedworryisthatdele gationhasexpandedthefederalregulatoryagendatothepoint of irredeemable unmanageability.48 Were Congress forced to focusandprioritizemore,theregulationthatdoesresultcould bebettercoordinated,assessed,andadjusted. Eventhosewho(liketheAuthor)donotsharethisconcernat thewholesalelevelwouldfindithardtodenythatcarefulat tention to the whether, who, and how of regulation in specific circumstances is important, and often lacking. At tempting to remedy this deficit through nondelegation, how ever, merely diverts the focus to questions, such as the ade quacy of particular statutory standards, that are tendentious andultimatelytrivialincomparisontotheunderlyingissues. A second concern is that allowing Congress to avoid the kindsofdetailedpolicyspecificationthatwouldfounderinthe bicameralism and presentment process has increased the pro ductionoffederalstatutorylaw.InsightfularticlesbyJonathan Macey49andothers50havedemonstratedhowtheArticleI,Sec tion7,lawmakingrequirementsworkinpracticetorestrictthe productionoflegislationandpreservethestatusquo.Thehis tory of the Framing provides some basis for believing that an antistatute bias was deliberate. State constitutions adopted shortlyaftertheRevolutionreactedtoperceivedabusesofthe kingbycreatinggovernmentstructuresdominatedbyapopu larlyelected,oftenunicameral,legislature.51Manyoftheseleg islatures dealt with the economic chaos and hardship of the postwar period by passing statutes that cancelled private (as wellaspublic)debtandtookvariousotherstepsendangering
47.Cf. CASS R. SUNSTEIN, AFTER THE RIGHTS REVOLUTION: RECONCEIVING THE REGULATORY STATE10610(1990)(advocating,amongotherthings,apresump tioninfavorofflexible,marketoriented,incentivebased,anddecentralizedregu latorystrategies). 48.Cf. STEPHEN BREYER, BREAKING THE VICIOUS CIRCLE: TOWARD EFFECTIVE RISK REGULATION (1993) (describing the overlap, inconsistency, and irrationality ofhealthandsafetyregulatoryprograms). 49.Jonathan R. Macey, Promoting PublicRegarding Legislation Through Statutory Interpretation:AnInterestGroupModel,86COLUM.L.REV.223(1986). 50.See, e.g., William N. Eskridge, Jr. & John Ferejohn, The Article I, Section 7 Game,80GEO.L.J.523,52829(1992). 51.JACK N. RAKOVE, ORIGINAL MEANINGS: POLITICSAND IDEASINTHE MAKING OF THE CONSTITUTION 21421 (1996); GORDON S. WOOD, THE CREATION OF THE AMERICANREPUBLIC:17761787,at40313(2ded.1998).

No.1]

DeconstructingNondelegation

97

existing property rights.52 Hence, for many of the well propertied Framers, the course of events between 1779 and 1787demonstratedthatpopularlyelectedlegislatureswerein clined to be reckless and that legislation threatened property. Soitisnotimplausibletoarguethatlessfederalstatutorylaw isitselfanoriginalconstitutionalvalue.53 Evenifthisweretrue,ameaningfulquestionremains:What weight should we accord today to a Framingera distrust of legislation?Wenowrecognizethatthecommonlawisitselfa systemofgovernmentregulationnomorethenaturalorder of things than any other legal approach. With the post eighteenthcentury spread of democratic government struc tures, statutory law created by popularly elected officials has become the preeminent form of social ordering in industrial and postindustrial societies. It now seems quaint, if not actu ally undemocratic, to treat statutes as a suspect incursion on judgemade law.54 To interpret our Constitution as locking in one particular approach to regulating social and economic ac tivityparticularly, a courtcentered approachuncomfortably highlights the dead hand potential of a twohundredyear olddocument.ItthreatenstomaketheConstitutionirrelevant in a society whose more recent political history has included not only President Roosevelts New Deal and President John sons Great Society, but also at least three decades of well documented public opinion insisting that the federal govern ment should take responsibility for solving environmental, healthandsafety,educational,andothercoresocialproblems.55
52.RAKOVE,supranote51,at216;WOOD,supranote51,at40313. 53.See,e.g.,JohnO.McGinnis,PresidentialReviewasConstitutionalRestoration,51 DUKEL.J.901,91020(2001). 54.See, e.g., WILLIAM D. POPKIN, STATUTES IN COURT 201 (1999) ([T]he domi nance of statute law in modern government requires reconsideration of the pre sumption that...statutes in derogation of the common law should be narrowly construed.); PETER L. STRAUSS, LEGISLATION: UNDERSTANDING AND USING STAT UTES 14451 (2006) (noting that the early twentieth century was a time when statutes began to replace case decision (i.e. the common law) as the primary sourceoflawinAmericanjurisprudence). 55.SeeChristopherEllis&JamesA.Stimson,OperationalandSymbolicIde ology in the American Electorate: The Paradox Revisited 25, 37 (Apr. 7, 2005)(unpublishedmanuscript),availableathttp://www.allacademic.com/meta/ p85321_index.html (reviewing data from 19702002 showing that Americans, on average,preferpoliciesthroughwhichthegovernmentspendsanddoesmoreto solve social problems, and that this clear preference varies within a relatively smallrange,neverquitetouchingtheneutralpointevenatitsmostextremecon

98

HarvardJournalofLaw&PublicPolicy

[Vol.33

ItnowseemsmoreimportanttoconsiderhowArticleI,Sec tion7operateswithintheregulatorysystemthatAmericansof the twentieth and twentyfirst centuries have chosen. Stability inregulatoryprograms,asinlawmoregenerally,isvaluable;it enables private planning and protects expectations. Entrench mentisdangerousbecausesomedegreeofregulatoryfailure approaches that produce unintended negative consequences, imposedisproportionatesocialcosts,andsoon56isinevitable giventhecomplexityandintractabilityoftheproblemsAmeri cans now expect government to solve. The constitutionally specified process makes amending and repealing statutes as difficult as enacting them. Indeed, once regulatory programs havegivenrisetocommunitiesofinterestintheircontinuation, amendingorrepealingaregulatorystatutemaybemorediffi cult than enacting it in the first place.57 Ambitious regulatory agendas require adaptability if they are not to do more harm than good.Given the statusquofavoring bias of the Article I, Section 7, process, less specificity in regulatory statutes may, ironically,beavirtueratherthanavice.58 This observation leads to what is doubtless the most broadlyresonantsetofconcernsvoicedthroughtherubricof nondelegation: Delegating so much policymaking power to administrativeagencieshascreatedseriousproblemsofcontrol and accountability. Agencies, according to this critique, are makingpolicywithlittleexternaloversightandlessdemocratic accountabilitytothepeople. Wehavegoodreasontoworryaboutthesethings.Currently, closetotwohundreddistinctentitieshaverulemakingauthor itythatis,thepowertomakeregulationshavingtheforceof
servativemoments);seealsoCynthiaR.Farina,FalseComfort&ImpossiblePromises: Information Overload, Uncertainty and the Unitary Executive, U. PA. J. CONST. L. (forthcoming) (on file with author) (reviewing political science literature on the majoritysliberalregulatorypreferences). 56.SeeSUNSTEIN,supranote47,at10709. 57.Forexample,theemergenceofthelucrativeenvironmentalcleanupindustry is often blamed for difficulty in amending the Comprehensive Environmental Response, Compensation, and Liability Act of 1980. See Cynthia R. Farina, Faith, HopeandRationality:Or,PublicChoiceandthePerilsofOccamsRazor,28FLA.ST. U. L.REV.109,12324(2000). 58.See,e.g.,BREYER, supranote48,at3942(identifyingcongressionalreactionto perceivedriskandtoregulatoryproblems,whichtakestheformofdetailedstatu toryinstructions,asonesignificantcauseofinefficient,ineffectiveregulation).

No.1]

DeconstructingNondelegation

99

law.59Theymakeaboutfourthousandrulesperyear,60anav erageofseventyfiveperweek.Tobesure,manyoftheseare minor or uncontroversial. But, as administrative law profes sorsknowwell,rulesareonlyoneofthemanywaysinwhich agenciesmakeimportantregulatorypolicy.Theyissueguid ance documents that, as a practical matter, the regulated community ignores only at its peril;61 they grant permits and other authorizations; they require product recalls and under takeenforcementactions.Agenciesevenmakesignificantregu latory choices by not acting. Consider that Massachusetts v. EPA,62whichmanyknow(notquitecorrectly)asdirectingthe EPAtoregulategreenhousegases,arosefromtheagencysde cision not to begin a rulemaking.63 The extentand diversity of regulatory activity is so great that we actually do not know, empirically,howmanyoftheseactivitiesoccuracrossthefed eralgovernmentannually. For cognitive and psychological reasons, humans seek to simplify complexity.64 The nondelegation doctrine, as tradition allyarticulated,representsonesimplifyingresponsetothecon cernthatthemassoffederalregulatorypolicymakingpoweris uncontrolledandunaccountable:Congressitselfshouldexercise thepower.Amorerecenttwistpresentsadifferentsimplifying response: The President should exercise, or at least direct the exercise of, the power. This unconventional deployment of nondelegationreasoninggoeslikethis:AlthoughtheCourthas almost never felt qualified to secondguess Congress regard ing the permissible degree of policy judgment that can be left

59.COMM.ONTHE STATUSAND FUTUREOF FED.ERULEMAKING, ACHIEVINGTHE POTENTIAL: THE FUTURE OF FEDERAL ERULEMAKING 3 (2008), available at http:// ceri.law.cornell.edu/documents/reportwebversion.pdf [hereinafter ACHIEVING THEPOTENTIAL]. 60.JohnAshlinetal.,Regulations.govFederalRegulatoryPortal,30J. GOVT INFO. 81,82(2004). 61.See Peter L. Strauss, Publication Rules in the Rulemaking Spectrum: Assuring ProperRespectforanEssentialElement,53ADMIN.L.REV.803(2001). 62.549U.S.497(2007). 63.Seeid.at511.AlthoughthecasewaswidelyreportedasrequiringtheEPAto regulate, the majority held only that the EPA has authority to regulate carbon dioxideandothergreenhouseemissionsandmustprovidearational,legallyvalid explanationfornotdoingso.Seeid.at533. 64.SeeFarina,supranote55.

100

HarvardJournalofLaw&PublicPolicy

[Vol.33

toagenciesinregulatorystatutes,65theConstitutionpermits nodelegationof[ArticleI]powers.66Therefore,thepowerexer cisedbyagenciesinregulatoryprogramsthathavepassedcon stitutionalmuster bysatisfyingtheintelligibleprinciplestan dard67 cannot be legislative power. Rather, as the Constitution speaksofonlythreetypesof power,theauthority conferred in regulatorystatutesmustbeexecutivepower.Tobesure,some administrative agency actionrulemaking, for examplemay resemblelawmaking.68Butacertaindegreeofdiscretion,and thusoflawmaking,inheresinmostexecutiveorjudicialaction.69 BecauseArticleIIveststhePresidentwithallthefederalexecu tivepower,thePresidentsdutytotakecarethatfederallawbe faithfully executed entails the power to direct the decisions of regulatorydecisionmakers.70 Either simple solution to the concern about control and ac countabilityofregulatorypolicymakingthepowershouldbe wielded either by Congress or by the Presidentwill be, at best, inadequate and disappointing. The sheer size and com plexityofthefederalregulatoryenterprisedefeatsrational,co ordinated, democratically responsive decision making by any single entity, be it the 535 members of Congress or the 1,500 peopleintheCabinetandExecutiveOfficeofthePresidentwho are the Presidents eyes, ears, and often voice with respect to regulatory decisions.71 Even with respect to particular, highly salient regulatory policy choices, it is far from obvious that a congressional or presidential decision is a significant gain in democratic control and accountability. As the judiciary has elaborated on the basic procedural framework of the Adminis
65.Whitmanv.Am.TruckingAssns,531U.S.457,47475(2001)(quotingMis trettav.UnitedStates,488U.S.361,416(1989)(Scalia,J.,dissenting)). 66.Id.at472(emphasisadded). 67.SeeJ.W.Hampton,Jr.&Co.v.UnitedStates,276U.S.394,409(1928). 68.INSv.Chadha,462U.S.919,953n.16(1983). 69.Mistretta,488U.S.at417(Scalia,J.,dissenting);accord.Freytagv.Commr, 501U.S.868,912(1991)(Scalia,J.,concurringinpartandconcurringinthejudg ment) (It seems to me entirely obvious that the Tax Court, like the Internal RevenueService,theFCC,andtheNLRB,exercisesexecutivepower.);Chadha, 462U.S.at953n.16(Itisclear,therefore,thattheAttorneyGeneralactsinhis presumptively Art. II capacity when he administers the Immigration and Na tionalityAct.). 70.SeeSTEVEN G. CALABRESI & CHRISTOPHER S. YOO, THE UNITARY EXECUTIVE: PRESIDENTIALPOWERFROMWASHINGTONTOBUSH29(2008). 71.SeeLYNRAGSDALE,VITALSTATISTICSONTHEPRESIDENCY253,25762(1996).

No.1]

DeconstructingNondelegation

101

trativeProcedureAct,agencydecisionmakingisoftenfarmore broadlyparticipatory,transparent,andpubliclyjustifiedthanis congressional or presidential action. Moreover, the growing in terestinusingtheInternetandotherinformationandcommuni cation technologies in the regulatory processegovernment hasgreatpotentialtomakeagencydecisionmakingevenmore open,comprehensible,andaccessibletocitizens.72 AbsentafundamentalrevisioninAmericansexpectationsof whatthefederalgovernmentshouldaccomplish,wemustrely onmultipleentitiesandprocessestomeetthechallengeofde mocraticcontrolandaccountabilityintheregulatorystate:the House and the Senate through their overlapping, and often competing, oversight and appropriations committees;73 the multiple centers of executive influence in the Cabinet and the variousWhiteHouseofficesthatorbitthePresidentandoften compete to be his authentic voice in the administration;74 the courtsintheirroleasreviewers;andprivateindividuals,enti ties,andinterestgroupsintheirroleaslitigants,lobbyists,re peatplayers,andwatchdogs. III. Fornearlyacentury,Congressesandpresidentsofbothpar ties have responded to perceived economic and social prob lems by creating regulatory agencies that wield substantial policymakingauthority.Fordecades,publicopinionpollshave revealed solid and remarkably stable majority support for ac tive federal government engagement in environmental, health andsafety,andeconomicissues.Ifallthisisnottheauthentic workingoutofrepresentativedemocracy,thenitishardtosee what selfgovernment would mean for the people of a large, heterogeneous nation. Of course, legitimate government action inoursystemissubjecttotheprovisothatsimplemajoritarian preferencesmaynotoverriderequirementsandprohibitionsof
72.SeeACHIEVINGTHEPOTENTIAL,supranote59,at811. 73.In the 110th Congress, more than twentyfive committees and subcommit tees have jurisdiction over the EPA and its regulatory programs. See U.S. EPA, MAJOR CONGRESSIONAL COMMITTEES WITH JURISDICTION OVER EPA ISSUES 13, http://epa.gov/ocir/leglibrary/pdf/110housejuris.pdf. 74.See,e.g.,LisaSchultzBressman&MichaelP.Vandenbergh,InsidetheAdmin istrativeState:ACriticalLookatthePracticeofPresidentialControl,105MICH.L.REV. 47,6869(2006).

102

HarvardJournalofLaw&PublicPolicy

[Vol.33

the Constitution. But it is hardly surprising that courts would notdeployaprohibitionthatisneitherexplicitintheconstitu tionaltextnorabsoluteinthebackgroundcommonlawunder standingstoblocksociopoliticaldevelopmentsthathavebipar tisan,crossbranch,andenduringpopularsupport. The real problem with framing concerns about regulatory governmentasaquestionofpoweristhatwhentheinevitable confirmation of congressional authority comes, we tend to act asifthereisnothingmoretosay.DebatesaboutwhetherCon gress can delegate have crowded out debates about whether Congressoughttodelegate.Dowereallybelievethatthesum and substance of congressional and presidential responsibility is to avoid doing that which they are prohibited from doing? Surelythepowerthatwe,thepeople,havegiventhemthrough theConstitutioncomesimpressedwithanobligationtoreflect carefully upon whether what may be done should be done. Whether or not any of the various concerns that continue to impel nondelegation talk merit a systemic revision of U.S. regulatoryobjectivesandstructures,theyshouldbepartofse rious discussion about regulatory proposals in Congress, the WhiteHouse,andbroaderpublicdiscourse.Continuingtosub limatetheseconcernsinanultimatelyunproductiveargument aboutconstitutionalfirstprinciplesdisservesusall.