1

JudicialEconomyandMovingBarsinInternationalInvestmentArbitration

LeslieJohnsAssociateProfessor,UCLA

CalvinThrall

PhDstudent,UniversityofTexasatAustin

RachelL.WellhausenAssociateProfessor,UniversityofTexasatAustin

February2019

Abstract

Historically, international investment law has centered on protecting foreign investorsfrom direct expropriation, but much of modern law includes legal standards that allowinvestors towin compensation for other kinds of investor-state disputes.Many scholarsandpolicyadvocatesworrythatmodernlegalprotectionsallowinvestorstopursuecaseswith low legal merit. We contend that this argument overlooks the impact of judicialeconomyandchanginglegalstandards:sinceforeigninvestorsonlyneedtoproveamainlegalviolationtosecurecompensation,arbitratorscananddoruleonlyonthosestandardsthataremosteasilyproven, inparticular, contemporary legalprotections. This suggeststhatmanyofthefearsabouttheabuseofinternationalinvestmentlawmaybeoverstated.

2

1 Introduction In2016,hundredsofthousandsturnedoutinGermanytoprotesttheTransatlantic

Trade and Investment Partnership (TTIP) thatwas then under negotiation. One divisive

issue was international investment arbitration, which The Guardian newspaper

summarizedas“plansforaspecialcourttohearcasesbycompaniesagainstgovernments

over breaches of regulatory issues.”1 In fact, since the 1990s, thousands of international

investmentagreements(IIAs)haveestablishedadefactointernationalinvestmentregime

wherebyforeigninvestorssuesovereignhoststatesoverpropertyrightsviolations, inad

hoc tribunals via Investor-State Dispute Settlement (ISDS). Modern international

investmentlawandISDSdevelopedasameanstoencourageinvestmentincountrieswith

weaklegalsystems,allowingforeigninvestorstoavoidpotentialpoliticalbiasindomestic

courts by using international arbitration instead (St John 2018). Critics wonder why

foreigninvestorsindevelopeddemocracieswithstrongruleof lawshouldhaveaccessto

ISDS, especially when domestic companies in those countries only have access to the

domesticlegalsystem.

Critics are further concerned about the scope of claims about property rights

violations that foreign investors canbring in ISDS.Thehistorical impetus for investment

protection was to stop developing country governments from directly expropriating

foreigninvestments,inwhichthegovernmentwouldforceachangeinownershipwithout

duecompensation.Inrecentdecadessuchactions,whilenoteliminated,areonthedecline

(Minor1994,Wellhausen2015).However, today’s IIAsallow foreign investors tosue for

1“TTIPprotestorstaketostreetsacrossGermany,”TheGuardian(UK),17September2016.Availableat:https://www.theguardian.com/business/2016/sep/17/ttip-protests-see-crowds-take-to-streets-of-seven-german-cities.

3

compensation over host state actions that they claimunlawfully infringe on the value of

theirproperty,ratherthanaffectingtheirownershipperse.AsTheGuardianputit,foreign

investorscanwinawardsforwhatare“breachesofregulatoryissues.”2Theconcernisone

oferodingsovereignty:foreign(andnotdomestic)investorscanreceivecompensationfor

adverseeffectsofgovernmentregulations,regulationsthatmayverywellbeseenbyother

actorsas legitimate.Conflictsbetweenhoststateregulatoryautonomyandobligations to

foreigninvestorshavealreadyroiledpoliticsandfomentedchallengestoIIAsindeveloping

countries(PeinhardtandWellhausen2016,HaftelandThompson2018).Suchconflictsare

spillingovertodevelopeddemocracies,too:investorshavesuedEuropeancountriesover

propertyrightsviolationstheyallegetohaveemanatedfromtheprovisionofgreenenergy

subsidies(Spain,Italy,andtheCzechRepublic),nuclearpowerregulation(Germany),and

banking regulation choices made in the wake of the financial crisis (Belgium), not to

mentionoveravarietyofenvironmentalandotherregulationsintheUnitedStates,Canada,

andMexicochallengedthroughISDSfacilitatedbyNAFTA.

In this context of widening claims of property rights violations, against both

developing and developed countries, more observers are calling for reforming (or

abandoning) modern international investment law (Waibel 2010). A key underlying

question is whether there are systematic biases in the system. Some interpret current

trends as evidence that the system is biased toward foreign investors rather than host

states (Simmons 2014, Van Harten 2010). Pelc (2017) focuses on low-merit, frivolous

claims, or ones for which a foreign investor may have little expectation of winning

compensation via ISDS arbitration but nonetheless files. A foreign investor may be

2Ibid.

4

motivatedtofilealow-meritcase,because,evenifitneverwinsanaward,thehoststateas

wellasothercountriesobservingthearbitrationmightdelayimplementingthequestioned

regulationuntilthedisputeisformallyresolved,aphenomenonknownasregulatorychill.

The foreign investor can thus win in terms of gaining time before having to change its

business practices in the host state or in other chilled countries. Moehlecke (2018)

providesthefirstsystematic,non-anecdotalevidenceofregulatorychill:afterPhilipMorris

sued Australia and Uruguay over their new, WHO-endorsed anti-smoking regulations,

multiplecountriesannouncedthattheywouldrefrainfromadoptinganalogousregulations

until the cases had been completed; the anti-smoking policies that were the subject of

litigationdiffusedsignificantlyslowerthanotherunchallengedanti-smokingpolicies.This

sortofuseof ISDS is indeed farafield fromtheconceptofprotecting foreign investment

fromunduegovernmentinterference.

We agree that even a single instance of regulatory chill challenges the notion of

equitybetweenforeigninvestorsandhoststatesinthedefactoregime.Wherewedisagree

iswhetherthereisatrendof“anincreaseinlow-meritclaimsovertime”(Pelc2017:561).

Pelc (2017) sees this trend in the concurrence of more ISDS cases being filed against

democraticcountries,thatarelesslikelytoresultinsettlementandlesslikelytoresultina

favorable ruling for the foreign investor. Pelc (2017) builds its argument around legal

claimsof“indirectexpropriation.”Wearguethattrendsinfilingsandrulingsonobservable

claims,andindirectexpropriationinparticular,donotestablishevidenceofatrendtoward

low-meritclaimsoverall.

Our core argument is that the qualities of observable claims and rulings in ISDS

cannotprovidedefinitiveevidenceofwhethercases themselvesareofhighor lowmerit.

5

We base this argument on two concepts overlooked by critics of modern international

investmentlaw:judicialeconomy,whereinarbitratorsdonotruleonallclaimsmadeina

given case, and moving bars, or changes in legal standards over time. We make our

argument through careful process-tracing of developments in jurisprudence, andweuse

noveldataonISDSclaimsandrulingstoillustrateobservableimplicationsofourargument.

Adjudicatingwhetherornotmore low-qualityclaimsarebeingmadeover time is

crucial for reformefforts.Theabsenceof evidenceof sucha trend implies that concerns

aboutlargenumbersofforeigninvestorsmanipulatingthesystemtocauseregulatorychill

areoverstated.Theresourcesput intoreformefforts intending tosolveproblemsofbias

shouldbemarshalledaccordingly.Broadly,ourargumentemphasizestheeffectsthatdata-

generating processes have on an analyst’s ability to draw inferences from observables.

Many social scientists have successfully used selection bias as a means of testing

hypotheses, and scholars in international lawand internationalpolitical economyareno

exception (see in particular Lupu 2013, von Stein 2005). Our intention in elucidating

selection issues here is to lay a firmer groundwork for scholarship that works to

characterizeISDSand,inturn,providesfodderforitscriticsandproponents.

2 JudicialEconomy

Inthissection,wereviewthelegalconceptofjudicialeconomyandexplainwhywe

expectittobeemployedinISDS.Insodoing,wearguethefollowing:first,arbitratorsoften

“economize” by making strategic decisions about which of an investor’s legal claims on

whichtoissuerulings.Second,thisstrategicbehaviorisorthogonaltothelegalmeritofthe

claims being ruled on or passed over. Finally, the strategic selection process of judicial

6

economy is problematic for scholars who attempt to measure the quality of individual

claimsbyexaminingtheirwin-ratesovertime.

Whenjudgesarepresentedwithcomplexcases, theyoftenmustconsidermultiple

alternative arguments or streamsof logic. Rather thandeciding on each and every issue

that ispresentedbeforeher,a judgeisencouragednottodecide legal issuesthatarenot

necessarytoresolvethecase;arulingshouldanswer“thequestionaskedinthemostdirect

manner,andvia theshortestavailable routeof legalanalysis” (Weller2011:130). In the

contextofdomesticlaw,Sunsteindefines"decisionalminimalism"as"thephenomenonof

saying no more than necessary to justify an outcome, and leaving as much as possible

undecided," and he calls it a legal virtue (1999: 3). Sunstein argues that "minimalism is

likely tomake judicial errors less frequent and (above all) less damaging. A court that

leaves thingsopenwillnot forecloseoptions inaway thatmaydoagreatdealofharm"

(1999:4).

The central premise that underlies the concept of judicial economy is that judges

should conserve their own and litigants’ resources by avoiding unnecessary decisions.

Posnerargues that "a failure toeconomizewhereverpossibleon…judicial resourcesmay

impose substantial social costs in the form of reduced judicial quality" (1983: 11).With

regard to international law in particular, Shany (2005: 919) argues that “since

international courts are todaybusier thanever, considerationsof judicial economyexert

growingpressuresoncourtstodelegatesomedecision-makingpowerstostateauthorities

andtoassumelessintrusive…standardsofreview.”

Yetinternationaljudgesmustnotonlyfocusonreachingthecorrectlegaloutcome;

theymust alsobeawareofmaintaining the supportof states for international law itself.

7

States can leave institutions and legal commitments atwill under the international legal

principleof consent (Helfer2005), and the threatof exit frombotha court’s jurisdiction

andfromsubstantivelegalobligationsisalwaysaconcern(Johns2014,2015).Thus,inthe

contextofinternationallaw,judicialeconomyisaboutmorethanconservingresources.As

Palombino argueswith reference to international law, “theprinciple of judicial economy

requiresthe judgetoobtainthebestresult in themanagementofacontroversywiththe

mostrationalandefficientusepossibleofhisorherpowers”(2010:909).

For example, when the International Court of Justice ruled on the legal status of

Kosovo in 2010, many observers criticized the Court for avoiding some legal issues.

However, inhisdefenseoftheKosovoopinion,WellerarguesthattheKosovocaseposed

“enormouslycontroversial” issues,andwas just the latest inastringofhighlypoliticized

disputestobepushedtotheCourt(2011:129).Itisunreasonable,heargues,toexpectthe

Courttouseitslegalcompetencetodecidepoliticalissuesthatcannotberesolvedbystates

themselves.PoliticalconcernsarealsoapparentinstudiesofjudicialeconomyattheWorld

TradeOrganization.Davey(2001)arguesthatjudicialeconomyisonewayinwhichWTO

judgesshowdeference tomember-governments,allowing theWTOtoavoidunnecessary

political conflicts. His insightshavebeenupheld in subsequent statistical studies.Busch

andPelc (2010) findevidence that judicialeconomy ismore likelyasWTOmembersare

dividedonimportantissues.Theyarguethatthisoccursbecause“judicialeconomylimits

the scope of the case law” (2010, 257). Similarly, Brutger and Morse (2015) find that

judicialeconomyismostlikelyinWTOrulingsagainsttheEUandUS,whichtheyinterpret

asevidencethatWTOjudgesmoderatetheiradverserulingsagainstmorepowerfulstates.

8

Several structural aspects of ISDS arbitration make judicial economy particularly

attractive in this setting. First,which legal claims an investorwinsmayhave little orno

impact on the amount of compensation in the award. Like international law in general,

international investment law is ultimately based on a remedial conception of justice

(Shelton 2002). When a foreign investor is harmed by a host state, it is entitled to

restitutionand/orcompensation,meaningthatitmustbemadewholefortheharmthatit

hassuffered.However,investorsarenotentitledtopunitivedamagesunderinternational

law.3Forexample,inMiculav.Romania,Swedishinvestorsbroughtnumerouslegalclaims

againsttheRomaniangovernment.Afterover200pagesofdensefactualfindingsandlegal

arguments about the investor’s claimof a “fair and equitable treatment” (FET) violation,

thearbitrationpanelabruptlywrote:

In lightoftheTribunal’sconclusionthat…theRespondentbreacheditsobligationtotreattheClaimants’investmentsfairlyandequitably,theTribunaldoesnotneedtoaddresstheClaimants’remainingclaims.Indeed,eachofthoseclaimsarisesfromthesamefactsasthefairandequitabletreatmentclaim,andtheClaimantsclaimthesame compensation in each instance…Thus, even if theTribunalwere to find infavor of the Claimants with respect to these claims, this would not impact theTribunal’s calculationof damages.As a result, any legal findings on thesemattersareunnecessary.

In the panel’s view, answering additional legal questions was unimportant, because the

Swedish investors had already “won” the right to compensation because of the FET

violation.

Similar examples of judicial economy abound. In Ascom and Others v. Kazahkstan, the

tribunalruledthatthestate’sactionsamountedtoanFETviolation;theychosenottorule

ontheclaimants’sevenadditionalclaims,onthegroundsthatadditionalviolationswould

3Insomeexpropriationcases,aninvestorcanbeawardedinterestand/orprojectedfutureearningsifitcanprovethatahoststatewasnotexpropriatingforapublicpurposeordidnotactingoodfaith.However,baselinedamagescannotexceedthevalueoftheinitialinvestment.

9

notaffectthedamagesowedtotheinvestors.4ArbitratorsinBoscav.Lithuaniareachedthe

sameconclusion,deciding thatanFETviolationmeans that"anyadditionalbreach isnot

relevantunlessitleadstoadditionaldamages,whichitwouldnotdohere.”5

A second structural reason judicial economy is attractive in international

investmentlawisthatdelaysareverycostly.Investmentarbitrationisanadhocprocess:

litigantschoosewhicharbitratorstohirefortheircase,andpaythemadailyfeefortheir

services. If an arbitrator needlessly prolongs a case by holding an excessive number of

hearings, or drafting long awards that dealwithunnecessary legal issues, then she risks

losing out on future employment opportunities. The set of peoplewho are qualified to

serveasinvestmentarbitratorsisrelativelysmall,andindividualsdevelopreputationsfor

theirabilitytoresolvecasesefficientlyandfairly(Tucker2018).

Third,hoststateshavevery limitedopportunities tochallengearbitrationawards,

whichlessenstheincentivesinvestorshavetotrytowinmoreclaimsasinsuranceagainst

appeals.IfthetribunalisconvenedatICSID,apartycanfileforannulmentoftheaward,but

this is explicitlynotanappealsprocess.Annulmentsmustbebasedonerrors inprocess

and not in reasoning. Annulment is rare regardless of whether it is sought by states or

investors;outofthehundredsofIIA-relatedawardsissuedbyICSIDtribunals,onlytwelve

havebeensuccessfullyannulledby2018(Tucker2018).Otherwise,theclosesthoststates

cancometochallengeawardsistheabilityofparties,undersomeconditions,tofileforset-

asideproceedingsindomesticcourtsattheseatofjurisdictionfollowingaruling.However,

there isnotaclear-cutconceptof“double jeopardy” inthesystem,especiallysincemany

4AnatolieStati,GabrielStati,AscomGroupSAandTerraRafTransTraidingLtdv.Kazakhstan,SCCCaseNo.V116/2010.5LuigiterzoBoscav.Lithuania,UNCITRAL,Award17May2013,para.244.

10

investorshaveaccesstoISDSunderavarietyofbilateraltreatiesduetotheirmultinational

holdings.Thus,aninvestorcouldbringnewproceedingsintheeventofanoverturnedpro-

investor ruling, making the number of claims ruled on in the initial proceeding less

relevant, althoughwenote that restarting litigation is costly. Themain exception to this

lineof reasoning is in the caseof enforcementproceedings. Investors seeking to enforce

arbitralawardscanfileenforcementproceedingsindomesticcourtsofstatesinwhichthe

respondenthoststatehasrecoverableassets.Asthelawaroundthesekindsofproceedings

is quite decentralized and in flux, the merits of a case may play more of a role in an

investor’sabilitytosucceed,suchthattheinvestormighthavemoreincentivetowinmore

claimsinitsinitialproceedings.

Fourth, litigants in ISDS arbitrationhave little incentive touse individual cases in

ordertoestablishfavorableprecedentsforthefuture.Strictlyspeaking,judicialrulingsin

priorcasesarenotabindingsourceof international law,although theyareconsidereda

“subsidiary means for the determination of rules of law.”6 In practice, this means that

international law operates according to what many scholars call de facto stare decisis

(Bhala2001;Schill2010).Pelc(2014)arguesthatthesedynamicsinfluencelegalclaimsin

internationaltradelaw:astatewilloftenbringrelativelysmalltestcasestotrytoestablish

favorable precedents that can be used in later cases. However, in ISDS arbitration, host

statesarealwaysplayingadefensivestrategysincetheyarealwaysthedefendant,never

theplaintiff. For a risk-aversehost state, it is farbetter tohaveno rulingon an alleged

violation than to risk losing a legal argument. For their part, it is possible that foreign

investors that foresee filingmultiple cases couldpress for extra legal rulings in order to

6StatuteoftheInternationalCourtofJustice,Article38,para.1(d).

11

bolsterfuturecases.YetbecauseISDSarbitrationsrelyonthousandsofdifferentIIAsand

contracts,favorableprecedentscreatedwithregardtoonetreatymaybeinapplicabletoan

investor’scaseagainstanotherhoststate.

Insum,tribunalsregularlyexercise judicialeconomyanddonotruleonallclaims

broughtinISDS.Further,arbitrators’decisionsaboutwhichclaimstoruleonandwhichto

pass over are not guided by considerations of the claims’ legal merits. This creates a

problem of classification: in previous literature using ISDS data, there is no distinction

betweenclaimsthatwerejudgedonthemeritsanddeclinedandclaimsthatwerenotruled

(andthusneverexaminedonthemerits)duetojudicialeconomy.Thismisclassificationof

nonruledclaimsbiasesthewin-ratesofindividualclaimsdownwards,andmaythusleadto

inaccurateinferencesaboutthequalityofclaims–thisisparticularlytruewhenonelooks

atlongitudinaltrendsinwin-rates,asjudicialeconomyhasbecomeincreasinglyprevalent

inISDSinrecentyearsaswedescribebelow.

3 MovingBars,orChangingLegalStandards

Inthissection,wepresentfurtherevidencethatwecannottaketrendsinrulingson

anyparticularclaiminISDSasanindicatorofquality.Wedosobyexplainingtheconcept

ofmovingbars: legalstandardschangeovertime.Ifthebarsaremoving,wecannotinfer

changesinthequalityofcasesfromchangesinthewin-ratesononeparticularclaimover

time.We argue that indirect expropriation claims have been judged against a relatively

stricter test over time, such that trends in indirect expropriation win-rates are not an

appropriatemeasureofquality, contraPelc (2017).Tomakeourargument,we illustrate

12

whack-a-moledynamicsbetweentwokeyclaimsinmoderninternationalinvestmentlaw:

indirectexpropriationandfairandequitabletreatment(FET).

When suing via ISDS, foreign investors draw from a menu of claims available to

them based on the treaty or investor-state contract invoked. Modern international

investmentlawlacksthestandardizationandcoherencethatisapparentinotherareasof

internationallaw,likeinternationaltradelawundertheGATT/WTO.However,undermost

instruments, investors have access to three sets of claims: (1) rules regarding

expropriation of foreign investors; (2) absolute treatment standards; and (3) relative

treatmentstandards.Thefirsttwocategoriesaremostpertinenttoourargument.7

3.1RisingBarforIndirectExpropriation

Many critics of ISDS focus on the first set of expropriation standards and in

particularon indirect expropriation. In contrast todirect expropriation,when the state’s

actionsdepriveafirmofownershipofitsinvestment,indirect(orcreeping)expropriation

occurswhenagovernment’sactionsviolatepre-existingcontractsorlawsandreducethe

valueofaforeignfirm’spropertywithoutnecessarilychangingownership.Apivotalruling

with regard to this claim came in 2000 inMetalclad v. Mexico, brought under NAFTA

Chapter 11. The US firm Metalclad argued that a local Mexican government’s actions

amountedtoindirectexpropriation,becausetherefusaltoissueanenvironmentalpermit

meantthatthelanditownedcouldnolongerbedevelopedintoahazardouswastelandfill.

Thearbitraltribunalagreed,writing,7Lesspertinenttoourargumentarethesetofstandardrelativetreatmentstandards.Nationaltreatmentspecifiesthataforeigninvestormustreceivetreatmentthatisatleastasfavorableasthetreatmentreceivedbyasimilardomesticinvestor(thatisusuallydefinedasaninvestorin“likecircumstances”).Foranillustration,seeforexampleCargill,Incorporatedv.RepublicofPoland,ICSIDCaseNo.ARB(AF)/04/2.Most-favorednationtreatmentspecifiesthataforeigninvestormustreceivetreatmentthatisatleastasfavorableasthetreatmentreceivedbyaforeignnationalfromanotherstate.Foranillustration,seeforexampleATAConstruction,IndustrialandTradingCompanyv.HashemiteKingdomofJordan,ICSIDCaseNo.ARB/08/2.

13

expropriationunderNAFTA includesnot onlyopen, deliberate and acknowledgedtakingsofproperty,suchasoutrightseizureorformalorobligatorytransferoftitleinfavorofthehostState,butalsocovertorincidentalinterferencewiththeuseofpropertywhichhastheeffectofdeprivingtheowner,inwholeorinsignificantpart,of the use or reasonably-to-be-expected economic benefit of property even if notnecessarilytotheobviousbenefitofthehostState.8

The2000Metalcladrulingprovidedaclearexampleofindirectexpropriationintheformof

regulatorytakingasthebasisforawardingcompensation.Theparticularlytroublingoptics

aroundahazardouswaste landfillandenvironmentalpollutionalsomade ita focalpoint

thatspurredearlyoutcry that international investment lawwasdangerous forhoststate

sovereignty.9

Legalscholarshavecometocharacterizetwocompetingdoctrinesinjurisprudence

on indirect expropriation. The first, typified byMetalclad, is the sole effects doctrine.10

DolzerandBloch(2003)definethisasarulingthat“restrictsitselftofocusingsolelyonthe

particulareffectthatagivenmeasurehasonthelegalpositionoftheinvestor"(158).The

second has come to be known as the police powers doctrine, which “considers, in

establishingwhetheraregulatorymeasureamountstoanexpropriation,thepurposeand

contextofthemeasure”(Brunetti2003:151).Thiscontextualapproachallowsfor“amore

elaborate weighing and balancing exercise” incorporating the government’s goals, in

contrasttothesoleeffectsdoctrine(DolzerandBloch2003:158).11Bothdoctrinesremain

8SeeMetalcladCorporationv.TheUnitedMexicanStates,ICSIDCaseNo.ARB(AF)/97/1,Awardof30August2000,para.103.9SeeforexamplethePBSdocumentary,“BillMoyersReports:TradingDemocracy”(premiered5February2002).10Brunetti(2003:151)arguesthatthisdoctrinedescendedfromtheIran-USClaimsTribunal,notingforexampletherulingintheTippettscase:"Theintentofthegovernmentislessimportantthantheeffectsofthemeasuresontheowner,andtheformofthemeasuresofcontrolorinterferenceislessimportantthantherealityoftheirimpact."Iran-USClaimsTribunal,Tippetts,Abbett,McCarthy,Strattonv.TAMS-AFFA,6IRAN-U.S.C.T.R.,at219etseq.SeealsoDolzer2002,Olynyk2012.11Formoreontheimplementationofpolicepowersdoctrine,seeWeiner2003,Heiskanen2003,Vicuña2003,Olynyk2012.

14

relevanttoindirectexpropriationjurisprudencetoday,whichmeansthatthesecompeting

doctrines have resulted in a “fragmented and frequently contradictory body of

jurisprudence”(Olynyk2012:254).

As summed up byKnahr (2007), “since there is no definition ofwhat constitutes

indirectexpropriationthescopeandmeaningofthisnotionhastobedeterminedthrough

arbitral practice” (85).12 This reality has created considerable uncertainty for investors

bringing indirect expropriation claims, not to mention the host states they are suing

(Mostafa 2008: 268). It also clearly influences tribunal decision-making. Knahr (2007)

observed a trend in arbitral practice against successful claims of indirect expropriation:

“Whatbecomesalsoapparentfromthemostrecentcasesisthefactthattribunalsremain

hesitant to actually reach findings of expropriation” (101).13 In short, the expansion of

indirect expropriation into a contradictory body of jurisprudence has made the bar for

indirectexpropriationrise.

3.2LoweringBarforFairandEquitableTreatment(FET)

Giventhehigherbarforindirectexpropriation,Knahr(2007)articulatedinvestors’

clearalternative:“claimantsarecertainlywelladvisedtopursueclaimsthattheyhadbeen

treatedunfairlyandinequitably–thethresholdforafindingofaviolationofthisstandard

seems to be comparatively lower and has so far promised a higher chance of success”

(102). The evolution of jurisprudence on indirect expropriation created incentives for

tribunals, and thus litigants, to turn to the absolute treatment standard called fair and12“Oncethejurisprudentialfactthatownershipitselfinvolvesabundleofintangiblerightsinrelationtopropertyisacknowledged,thenitfollowsthatitisnotonlytheoutrighttakingofthewholebundleofrightsbutalsotherestrictionoftheuseofanypartofthebundlethatamountstoatakingunderthelaw"(Sornarajah2004:368).13Today,theunderstandingamonglawyersinthisspaceisthattribunalsareincreasinglyunwillingtomakeafindingofindirectexpropriationifahostgovernment’sactionsareonlytemporaryorreducethevalueofaninvestmentonlyincrementally(Interview,LosAngeles,May2018).

15

equitable treatment (FET).14 Today, common causes of adverse host government action

that lead foreign firms to invoke the FET standard include: violations of due process in

domestic judicial and administrative proceedings; non-transparency about government

policies and procedures; arbitrary, unreasonable, or discriminatory treatment; policy

changesthatviolatelegitimateexpectationsabouttheregulatoryenvironment;andacting

inbadfaithtowardsforeignfirms.Evenacursoryglanceatthesecausessuggeststhatstate

actionsthattriggeranindirectexpropriationclaimmakealsobelikelytounderpinanFET

claim.

InearlyISDScases,thebarforFETwashigh:FETwasvague,withnoclearidentity

of its own, with one scholar noting that “the content of this standard has causedmuch

anxiety” (Sornarajah 2004: 235-236). Metalclad was again a pivotal case in its

development: the Metalclad finding that “Mexico failed to assure a transparent and

predictable framework” for theUS investor typifieshowFEThascome tobe interpreted

(citedinLowenfeld2007:557;seealsoSornarajah2004).Transparencyandpredictability

arecoretotheconceptof“legitimateexpectations”thathascometounderpinFETrulings.

Tribunals now consider FET claimsper a legitimate expectations test: at the time of the

claimant’s initial investment, could they reasonably and legitimately expect that the

offendingstateactionwouldnot takeplace?15Forexample, thetribunal inGrandRiverv.

USA found no FET violation even though the new US regulation at issue did significant

14Lesspertinenttoourargumentisthe“fullprotectionandsecurity”absolutetreatmentstandardthatrequiresthathoststatesrefrainfrommilitaryattacksagainstforeignfirmsandtheirproperty.Itisalsoofteninterpretedtorequirethathoststatesprovidebasicprotectionagainstattacksbythirdparties,likerebelgroupsandmilitias.Foranillustration,seeforexampleSeeAsianAgriculturalProductsLtd.v.RepublicofSriLanka,ICSIDCaseNo.ARB/87/3.15Dolzer(2002)calledfor“legitimateexpectations”tobeaconceptusedinassessingindirectexpropriation(78-79).Vicuña(2003)notedearlydevelopmentsof“legitimateexpectations”standardsaroundindirectexpropriationinUKcourts(193).YettheconcepttodayissquarelyassociatedwithFET.

16

harmtotheclaimant’sinvestment;thetribunal’sreasoningwasthattheclaimanthadbeen

exploitingwhatwas in trutha loophole inUS tobaccoregulation,and theclaimantcould

notreasonablyexpectthattheloopholewouldnoteventuallybeshut.16

As recounted by Sornarajah, "It was only when tribunals started including the

violations of legitimate expectations…that the fair and equitable standard dawned

suddenlyasthedrivingforcebehindinvestmentarbitration”(2017:416).17Given“thefact

that expropriation has become difficult to establish where there is no direct taking of

property" (Sornarajah 2017: 554), there was room for an alternative legal claim to

overtakethatofindirectexpropriation,andFETbasedonlegitimateexpectationshasdone

so.Inshort,theclearerjurisprudenceonFETcreatedincentivesfortribunalstoprioritize

FET claimsover claimsof indirect expropriation, and for litigants to adjust the thrust of

theirargumentsaccordingly.

Forat least forsomearbitrators, theshift toFETwasan intentionalwaytoavoid

continuing to expand the definition of expropriation into more nuanced judgements of

whatconstitutedanindirecttaking.Semanticslikelyplayedarole,too:severalarbitrators

interviewed in Tucker (2018) note that they appreciate that an FET violation is a less

loaded characterization of states’ unlawful behavior than “expropriation.” Perhaps

inevitably,asFEThasgainedtractionwithtribunals,investorshaverespondedaccordingly

byinvokingFETclaimsinresponsetoabroaderswathofevents.Forexample,in1999,the

UNConferenceonTradeandDevelopmentwrotethat“there is littleauthorityon[FET’s]

application,” but by 2012 provided definitions of many categories of FET violations

16GrandRiverEnterprisesSixNations,Ltd.,etal.v.UnitedStatesofAmerica,NAFTA,filedin2004andawardissuedin2011.17Sornarajah(2017)claimsthat“legitimateexpectations”isaconceptthatwas“pluckedfromtheair”(417).

17

(Sornarajah 2017: 417). Just as occurred in the development of indirect expropriation

jurisprudence, concerns over expanding the definition of FET are emerging. Sornarajah

(2017)worries that recent awardshavemadeFET “themost importantprovision in the

investmenttreaty,virtuallyabsorbingallotherclaimsthatcanbemadeunderthetreaties"

(241). Whatever the normative implications of the expansion of FET, this worry

underscoresourbiggerpointthatjurisprudenceisdynamic:indirectexpropriationusedto

virtuallyabsorballregulatory-relatedclaims,andnowFETmaybefillingasimilarrole.

For ourpurposes, the takeaway from this jurisprudential history is the following:

wecannotinfertrendsinthequalityofcasesfromchangesinindirectexpropriationwin-

rates over time, because tribunals and litigants changed their legal strategies as the

controversyaroundindirectexpropriationraisedthebaronthoserulings.Norcanweinfer

trendsinqualityfromFETrulings,astheirrelativeunimportancebeforetheemergenceof

legitimateexpectationstestsconditionedtribunals’willingnesstoruleonFETclaimsand

thusinvestors’interestsinmakingthem.PerhapsFETwillsomedaygothewayofindirect

expropriation,suchthattribunalsandlitigantsturntoanotherstandard.Lastly,itwouldbe

wrong to put much stock in inferences of case quality by comparing early indirect

expropriationwin-rateswithlateFETwin-rates,becausejudicialeconomyistakingplace

acrossthepopulationofclaimsmadeasthebarsarechanging.

4 EmpiricalEvidence

Tosupportourarguments rooted in jurisprudentialhistory,weusenoveldataon

theclaimsforeigninvestorshavemadewhenfilingforISDSarbitration,includingacareful

analysis thatdistinguishesbetween lost claimsandclaimsonwhich thearbitral tribunal

18

does not rule. This exercise further underscores that the observable evidence onwhich

criticshaverelieddoesnotdemonstratethattoday’sISDScaseshavelesslegalmerit.

4.1AnatomyofanInvestor-StateDisputeCase

Beforediscussingthedata,weprovideabriefoverviewoftheISDSprocess.Figure1

presentsthebasictimelineofanISDScase.First,theclaimantfilestheircaseagainstthe

hoststate,includingalistofallegedviolationsoftheprovisionsoftherelevantIIAor

contract.Next,arbitratorsexaminetheclaimant’scaseanddecidewhetherthecourthas

standingtoruleontheinvestor’sclaims.Ifjurisdictionisdeclined,thecaseendsandnone

oftheallegedclaimsreceivearuling.Jurisdictionisoftendeclinedforreasonsunrelatedto

anyoftheallegedclaims,suchaschallengestotheclaimant’sownershipofthe

investment18orafindingthattheinvestorhasfailedtocomplywiththedisputesettlement

proceduresoutlinedintherelevantIIA.19Ifarbitratorsfindthattheydohavejurisdiction

toruleonthecase,theproceedingsadvancetothemeritsstage.

18Forexample,seeRomakS.A.v.TheRepublicofUzbekistan,PCACaseno.AA280,decisionissued2009.19Forexample,seePacRimCaymanLLCv.RepublicofElSalvador,ICSIDCaseNo.ARB/09/12,decisionissued2012.

19

Figure1:TimelineofanISDScase

Once a case has advanced to the merits stage, the tribunal examines each claim

individually. For each claim, there are three possible ruling categories: it can be upheld,

declined,ornotruledonatall.Figure2showsthepossibleoutcomesforacaseinwhich

theinvestorallegestwoclaims,basedonthetribunal’sdecisionsoneachclaim.20Notethat

Figure2:PotentialoutcomesofISDScases,conditionalonreachingmeritsstage

theonlyrequirementforacasetobeclassifiedasaninvestorvictoryisthatasingleoneof

theinvestor’sclaimsbeupheld.Thismeansthat,conditionalononeofaninvestor’sclaims

20Itisalsopossiblethatoneclaimbedeclinedandtheothernotruled,resultinginastatewin.Aswillbediscussed,however,thisoutcomeeffectivelyneveroccursinpractice.

20

beingupheld,theoutcomeofthecaseisunchangedregardlessofwhetherthesecondclaim

receivesarulingornot.Itisproblematicnottodistinguishclaimsthatwerenotruledon

fromclaims thatwereruledonanddeclined; thishas led to theconflationofclaims that

werefoundtolackmeritandclaimswhosemeritwentunexamined.Aswillbediscussed,

wemake thedistinctionbetweennonruledanddeclined claims inournoveldata,which

allowsustohighlighttrendsofjudicialeconomyininternationalinvestmentarbitration.

4.2Data

Ourdatabaseofinvestor-statearbitrationcasesincludesthosebroughtattheWorldBank’s

ICSID, those available in the commonly used UNCTAD Investment Dispute Settlement

Navigator,andseveraldozenadditionalcasesnotpresentinUNCTAD.21Ourdataincludes

allcasesfiledthrough2017withpublicrulingdocumentsavailable,butmostrecentlyfiled

casesareeitherstillpendingortherulingdocumentshaveyettobemadepublic.22Dueto

ourconcernthatsuchsmallsamplesmaynotberepresentative,weonlypresentdatafrom

casesfiledfrom1987(thefirstmodernISDScase)through2012,withrulingsassessedas

ofNovember2018.Thiscovers317cases,ofwhich154(48.6percent)hadapro-investor

ruling,74(23.3percent)hadapro-staterulingonthemerits,and89(28.1percent)hada

pro-staterulingonjurisdiction.

Note that not all ISDS cases are public:while all cases heard at ICSID are known,

casesunderUNCITRALruleshavebeenabletobekeptprivate,particularlywhentriggered

by an investor-state contract rather than an international treaty. Further, legal texts are

regularlyredacted,especiallyinrulingsdatingfartherbackintime.Litigantsoftenrefuseto

21ExpandedfromWellhausen2016.SomecasesexcludedfromUNCTADarebroughtunderinvestor-statecontractsratherthantreaties.Thatdistinctionmeanslittleforourabilitytoaccuratelyjudgetrendsinthequalityofcasesbroughtininternationaltribunalsovertime.22Forexample,only8ofthe76casesfiledin2015currentlyhavepublicdocumentsavailable.

21

disclose thecontentof their initial filingsaswellas theoutcomeofarbitration, including

thetextofrulings,thedirectionofrulings,andthecompensationawarded(ifany)(Hafner-

Burton, Steinert-Threlkeld, andVictor2016). Litigants also can andoftendo settle cases

afterformallybeginningarbitrationbutbeforethetribunalreachesaruling;thetermsof

thesesettlementsarerarelypublic.

These realities generate biases in the set of cases on which analysts can make

inferencesaboutthequalityofinvestorclaimsovertime.Itcouldbethatinvestorstendto

redact information in cases in which their claims are weak or unsuccessful; this would

mean that thebodyof public information is biased towardhigher-quality cases. Existing

work has thus faced a high hurdle in showing declining quality among a non-random

sample thatmightbe comprisedofexactly the cases increasing inqualityover time.Our

point is that trends on filings and rulings cannot in themselves establish whether the

quality of cases over time increases or decreases, irrespective of transparency-related

biases. While we can only explore public filings and rulings, our arguments regarding

judicialeconomyandchanginglegalstandardswouldapplytoallcasesshouldtheyeverbe

known.

Ournovelempiricalcontribution is tocodeseveralaspectsof the legalclaimsand

rulingsmadeinISDScasesinordertoprovideevidenceconsistentwithourarguments.23

Ourdatasetcodes1,109legalclaimsin317ISDSarbitrationsfiledfrom1987through2012

thatwere ruled on by the end of 2018. Cases are coded using original case documents,

mainly the Notice of Arbitration, the Award, and all available separate opinions and

annulmentproceedingdocuments.Therefore,witha fewexceptions,nocodingwasdone

23ThefullcodebookisavailableintheOnlineAppendix.

22

forcaseswithoutpublicdocuments.Thoseexceptionsarewhenhighlydetailedcasenotes

are available through secondary legal sources, particularly the ICSIDReview.24Wecoded

the instanceofa legalclaiminan investor’s filings;whetherthatclaimwasruledon; the

directionof theruling;aswellanyrole itmighthaveplayed inannulmentorotherpost-

initialrulingproceedingsordissentingopinions.Further,wecaptureandcodetextinthe

rulingsconcerningthetribunal’schoicetoruleon(some)claim(s).Tocharacterizethistext

in general, sometimes the tribunal deems that it lacks jurisdiction to rule on the claim.

Sometimes,thetribunaldeemsitunnecessarytorule,usuallybecauseithasalreadyruled

thattheoffendingacthasviolatedaseparateclaim.Wealsocapturetheinstancesinwhich

the tribunal does not mention a claim in the analytical section of an award, and the

instancesinwhichthetribunaldoesnotruleforidiosyncraticreasons.25

Specialists may be interested to know that our decision to rely on original case

documents sets our claims coding apart from that in the UNCTAD Investment Dispute

SettlementNavigator, thedataonwhichmanyscholars inthisarearely.Whennoofficial

casedocumentsareavailable,UNCTADattemptstogatheralistofclaimsfromthird-party

sources such as news outlets; for cases filed through 2012, UNCTAD reports claims and

rulings on 21 cases for which we do not have official case documents. We believe that

UNCTAD’smethodofgatheringclaimsleadstounderreporting.Newsoutletsare likelyto

report on only the most salient claims (such as expropriation), or on only claims that

received rulings (either pro-investor or pro-state). Thus, cases coded from third-party

sources are likely to underestimate the total number of claims and overestimate the

24Linkstoexternalsourcesareprovidedinthereplicationfiles.25Specifically,theseincludethetribunalnotingspecialagreementsbetweenthepartiesandpartialsettlementsthatmakearulingunnecessary.

23

percentageofclaimsthatreceiverulings.Inordertoidentifyinstancesofjudicialeconomy,

itisimportantthatwehaveaccesstothefullsetoflegalclaimsineachcase.Becausecase

documentsremainprivateonlywhenneitherpartywishestomakethempublic,excluding

cases with private documents could bias our observed trends if such cases differ

systematically from cases with public documents. However, because both states and

investorscanchoosetounilaterallyreleaserulingdocuments,wearguethatprivatecases

areunlikelytobebiasedtowardseitherstateorinvestor;evenifaninvestorwouldprefer

nottoreleasedocumentsinapro-stateruling, forexample,thestatecoulddosowithout

theinvestor’spermission.

Aninvestorisconstrainedintheclaimsitmakesbasedontheclaimsavailableinthe

underlying treatyorcontract it invokes in its ISDS filing.While thebodyof thousandsof

IIAshavemanysimilarities,thereareconsiderabledifferencesintheextenttowhichthey

preservewhatBroude,Haftel,andThompson(2018)call“stateregulatoryspace.”Oneway

inwhich this variationmanifests is in the kinds of claims available to investors in ISDS.

Today,effectivelyall ISDS-enabling IIAsprovide investorsastandardmenuofsixclaims:

direct expropriation, indirect expropriation, FET, full protection and security, national

treatment (NT), andmost-favored nation treatment (MFN). Ourmain analyses focus on

trends in these claims, as investors’ choices to invoke other claims beyond these are

constrained based on the treaties and contracts to which they have access (and any

possible treaty-shoppingwithin thatset).However, inrobustness in theAppendixand in

ourreplicationdata,weincludecodesforallclaims.

Specialists may be interested to know how we code FET claims and minimum

standard of treatment (MST) claims, especially as our coding rule is different from that

24

employedbyUNCTAD.26Inbrief,thereisconsiderabledisagreementinlegalpracticeand

scholarshipover themeaningandeven theexistenceofMST(Guzman1998;Paparinskis

2013).FET is sometimesreferred toas themodern-dayversionofminimumstandardof

treatment (MST) (Blandford2017,Haeri2011, Jiminez2001).27UNCTAD treatsMSTand

FETclaimsasnecessarily implyingoneanother: ifaninvestorallegesanFETviolationor

anMSTviolation,UNCTADcodestheexistenceofbothanFETandanMSTviolation.This

results inUNCTADrecordingsignificantlymoreMSTandFETclaimsthanwedo,aseach

incidenceofoneoftheclaimsisdouble-countedasanincidenceoftheother.

Incontrast,weseeconsiderablesupportfortheviewthatFETshouldbeconsidered

as an autonomous standard (separate from MST) in international investment law. In

particular,Kalicki andMedeiros (2007)point out that thedevelopmentof the legitimate

expectations test hasmoved FET away from the objective tests of MST (investorsmust

have access to courts, for example) and towards its current status as a more context-

specificprotection(couldtheinvestorhaveforeseentheoffendingstateactionatthetime

of investment?). Dumberry (2016) argues that the independence of FET and MST is

evidencedbythefactthatFETclausesbegantobeimplementedinIIAsduringthe1960s

and1970s,anerainwhichtheMSTprotectionwashighlycontested.Haeri(2011)further

argues that separatingFET fromMST(andsometimesexcludingMSTaltogether)wasan

intentionalmove on the part of developing stateswhen designing their IIAs: developing

states wanted to prevent the expansion of MST, as such an expansion would

disproportionatelydisadvantagestatesthatdon'thavetheabilitytomeettherequirements

26SeeAppendixFigure3aformoreinformationcomparingourdatatoUNCTADdata.27NATFApartiesin2001clarifiedthat,underthetreaty,FETdoesnotrequiremorethanMST,andtheUShasincludedlanguageonthispointinitssubsequentIIAs(Lowenfeld2007:556-557).

25

setbydevelopedstates.Defaulting todouble-countingFETandMST ignores thepolitical

contextinwhichtheclaimsoriginated.IntheAppendixweshowthatouridentifiedtrends

arerobusttoconsideringMSTclaimsasaseventhclaiminadditiontothecoresetofsix.28

4.3 Patterns

To illustrate the trends of judicial economy and moving bars in international

investmentarbitration,wepresenta seriesof figuresgenerated fromouroriginal claims

codingdataset.

4.3.1JudicialEconomy

Toillustratearbitrators’usageof judicialeconomy,wedocumentthegapbetween

thenumberofclaimsallegedbyinvestorsandthenumberofclaimsthatreceiverulings.As

explainedabove,intheinterestofmaximizingcomparabilityacrossIIAs,wepresenthere

only the set of core claims that are common to most investment agreements. That is,

investors canmake up to six claims and tribunals canmake up to six rulings. Investors

allegedanaverageof2.32claimspercase,withastandarddeviationof1.22claims.Inthe

Appendix,wepresentfiguresthatincludeallallegedclaimsandrulings;unsurprisingly,the

averagenumberofallegedclaimsishigher(3.5percase,withastandarddeviationof1.64

claims),butthetrendsidentifiedherearerobust.29

28SeeAppendixFigures4aand5a.29SeeAppendixFigures1aand2a.

26

Figure1:Overtime,alowerpercentageofallegedclaimsreceiverulings(byyearofcasefiling,1987-2012)30

Figure1displaysthepercentageofallegedclaimsthatreceiverulingsovertime,and

the averagenumber of core claims allegedper case.With the exceptionof 1992 (during

whichtherewasonlyonecase,whichwasthrownoutonjurisdiction),allclaimsbrought

by investors received rulings from arbitrators until 1997. After 1997, the proportion of

claims receivinga ruling fell substantially, dropping to less than50percent in2006and

again in2009.While theproportionof claims receiving a rulinghas fallen steadily since

1997,theaveragenumberofclaimsallegedpercaseremainedfairlystable; thissuggests

thatthedecliningproportionofruledclaimscannotbeexplainedbyanupwardtrendinthe

numberofclaimsallegedpercase.

30Recallthatonlycoreclaimsarecountedhere;averageslowerthan1arepossible,asnotallcasescontaincoreclaims.

27

Figure 2: Evidence of judicial economy: Since 2000, a lower percentage of allegedclaimsreceiverulings,conditionaloninvestorvictory(byyearofcasefiling,1987-2012)

It is important to note that Figure 1 includes jurisdiction losses; as mentioned

previously, noneof the alleged claims receive a rulingwhenarbitratorsdecide that they

lack jurisdiction to trya case.Recall that, inorder for judicial economy tobepresent, at

leastoneoftheallegedclaimsmustreceiveapro-investorruling.Thus,inordertobesure

that thetrend inFigure1 isattributable to judicialeconomy,wemustmakesurethatan

increaseinjurisdictionlossesaloneisnotdrivingthetrend.Todoso,Figure2presentsthe

sameinformationasFigure1butrestrictsthedatatocasesinwhich(1)thetribunalissued

arulingonthemerits,and(2)inwhichtheinvestorwonatleastoneclaim.Thisrestriction

limits thenumberof cases from317 (as inFigure1) to154.Here, thepre-trend is even

stronger:whenjurisdictionlossesareexcluded,allallegedclaimsreceiverulingsthrough

28

2000.Beginningin2001,however,theproportionofclaimsthatreceiverulingsdecreases

throughoutthe2000s.

Because all of the cases included in Figure 2 were victories for the investor, the

declining proportion of claims that receive rulings can be solely attributed to judicial

economy;arbitratorsissuerulingsononlytheclaimsthatarerelevantforcalculatingthe

compensation owed to the investor. Indeed, pro-investor rulings are virtually the only

casesinwhichjudicialeconomyisexercised.Ofthe74caseswithpro-statepublicmerits

rulings filedbetween1994and2012,only twocontainclaimsthatwerenotruledon for

non-jurisdictional reasons. This finding makes sense when one considers the rationale

behindjudicialeconomy:arbitratorseconomizebypassingoverclaimsthat,evenifruled

in favor of the investor, would not affect the financial compensation owed to them.

However,statesarenotawardedcompensationevenwhenclaimsareruledintheirfavor;

thus,thelegalreasoningmotivatingtheuseof judicialeconomyinpro-investorrulingsis

notapplicabletopro-staterulings.

Figure2illustratestheproblemwithtreatingclaimsthatdonotreceiverulingsas

equivalent to losses for the investor: even in cases that investors win, arbitrators

commonlydonottoruleonmanyofaninvestor’sclaims.In2010,forexample,arbitrators

issued rulings on fewer than 65 percent of claims made in cases won by investors.

Crucially, the decision not to rule on a claim can tell us nothing about how the tribunal

would have ruled, and thus can tell us nothing about the quality of the claim. For this

reason,treatingnon-ruledclaimsasequivalenttoclaimsthatwereruledinthestate’sfavor

isinappropriate.Wecannotinferthequalityofthecasefromtheabsenceofaruling.

29

Whilewesetoutargumentsaboutthedriversof judicialeconomyin international

investmentlawabove,analternativedrivercouldbethepowerorstatusoftherespondent

state.Recall that in lookingatWTOdisputes,BrutgerandMorse(2015) findthat judicial

economyismostlikelyinrulingsagainsttheEUandUS,whichtheyinterpretasevidence

thatWTOjudgesmoderatetheiradverserulingsagainstmorepowerfulstates.Ouranalysis

suggests that this trend does not hold in ISDS arbitration. First, judicial economy is not

presentinanycasebroughtagainsttheUnitedStates,astheUShasneverlostacaseand,

again,thereisscantevidenceofjudicialeconomyinpro-staterulings.Second,whilesome

EUmemberstateshave lostcases, judicialeconomy ispresent inonlyahandfulof those

cases. Further, judicial economy is rarely exercised in cases brought against any OECD

memberstates:in13outofthe17yearsinwhichcaseswerebroughtagainstOECDstates,

100percentofinvestors’claimsreceivedrulings.Thus,thereisnotacorrelationbetween

powerfulrespondentstatesandjudicialeconomyinISDS.

4.3.2MovingBars

We have argued that trends in win-rates for individual claims can be partially

attributedtotheshiftinglegalstandardsbywhichtheclaimsarejudged.Inparticular,we

arguethatthebarforindirectexpropriationclaimshasrisenwhilethebarforFETclaims

has lowered, and that this realignment of legal standards has resulted in a substitution

effect. An observable implication is that, in cases in which both claims are alleged,

arbitratorsshouldincreasinglychoosetoruleon,andtoruleinfavorof,FETclaimsinstead

of indirect expropriation claims. To illustrate this, we present two trends in FET and

indirectexpropriationclaimsovertime:(1)thepercentageofcasesinwhichtheclaimsare

alleged,and(2)thepercentageofcasesinwhichtheclaimsareruledon.

30

Figure 3: Trends in the percentage of cases with FET and indirect expropriationclaims(jurisdictionlossesexcluded),byyearoffiling(1987-2012) Figure3shows thepercentageof ruledcaseswithFETand indirectexpropriation

claims. We exclude cases that were thrown out on jurisdiction, because a claim cannot

possiblyberuledonifthecasedoesnotadvancetothemeritsstage.Thegraphhighlights

threetrends.First,FETclaimsbecamemorecommonthanindirectexpropriationclaimsin

cases filed after 2003, which corresponds with the jurisdictional history we recounted

above.Second,despitetheincreasingfrequencyofFETclaims,Figure3alsodemonstrates

thatindirectexpropriationcontinuestobeacommonclaim.Inmostyearssince2003,for

example,indirectexpropriationclaimshavebeenpresentinwellover50percentofruled

cases.Whilethestandardsforindirectexpropriationhavebeenraised,investorshavenot

abandonedtheclaiminfavorofFET.Indeed,whyshouldthey?Themarginalcostofadding

an indirectexpropriationclaimtoa filing is likelyoften lowenoughtobeoutweighedby

the expected value of a favorable ruling on the claim, even if such a ruling is a remote

31

possibility. In context of the evolution of jurisprudence on indirect expropriation, we

interpret the continued presence of indirect expropriation claims as strategic, and not

clearlyindicativeofthequalityofthecasesfiled.

Third,Figure3showsthepercentageofcasesinwhichinvestorsallegedbothFET

andindirectexpropriationclaims.Here,weseeastrikingrealignmentovertime:until2003

almost every FET claimwas combinedwith an indirect expropriation claim, though the

converseisnottrue.Asrecountedabove,arbitratorsdidnottreatFETasanindependent

standarduntil the early 2000s, anduntil that time investors invokedFETprimarily as a

supplementary claim for expropriation. In the years since 2003, however, the trend has

reversed:almosteveryindirectexpropriationclaimispairedwithanFETclaim,thoughthe

converse is not true. In this period, the legal community began to treat indirect

expropriationmore as a subset of FET. As suggested by the empirical trends, today any

actionwhichconstitutesindirectexpropriationgenerallyalsoconstitutesanFETviolation,

butnotallFETviolationsconstituteindirectexpropriation.

32

Figure 4: Trends in the percentage of cases with ruled FET and indirectexpropriationclaims(jurisdictionlossesexcluded),byyearoffiling(1987-2012) While Figure 3 displays the percentage of ruled cases with FET and indirect

expropriation claims alleged, Figure 4 displays the percentage of cases with FET and

indirectexpropriationclaimsthatreceiverulings.First,notethatruledFETclaimsovertake

ruledindirectexpropriationclaimsinthemid-2000s,paralleltoFigure3.Takentogether,

Figures3and4showthatFETclaimsareallegedandruledoninagreaterproportionof

cases than indirect expropriation claims in cases filed since the realignment of themid-

2000s. If the rise of judicial economy were the only change occurring in international

investmentarbitrationduringthistimeperiod,thisresultwouldbesomewhatpuzzling;we

might expect an upswing in the frequency of alleged FET claims to be met with an

equalizingdownswing in theproportionofFETclaims that receiverulings.However, the

findingcanbeunderstoodwhenviewedinthecontextofshiftinglegalstandards.TheFET

standard has been broadened over time to include protection against state actions that

33

wouldconstituteindirectexpropriation,andarbitratorshavebeenattemptingtolimitthe

scope of the indirect expropriation standard. Thus, arbitrators have incentive to rule on

FETclaimsfirstbeforedecidingwhetheritisnecessarytoruleonadditionalclaims,asthe

findingthatastatehasviolatedtheFETstandardcanallowthemtoeconomizebyopting

nottoruleon(andpotentiallyexpandtheinterpretationof)indirectexpropriationclaims.

OnenotabledepartureinFigure4is inthepercentageofcaseswithbothFETand

indirectexpropriationrulingsovertime.Until2003,thesolidlinesinFigures3and4look

almost identical: FET claims are almost exclusively paired with indirect expropriation

claims,andalmosteveryclaimreceivesaruling.However,Figure4showsthatafter2003,

a gap began to form between the percentage of cases with ruled indirect expropriation

claimsandthepercentageofcasesinwhichbothFETandindirectexpropriationclaimsare

ruled on. This gap represents the substitution effect between FET and indirect

expropriation:indirectexpropriationclaimsarehighlylikelytobepairedwithFETclaims,

but they are increasingly less likely to be ruled on alongside FET claims. Again, the

substitution effect can be explained by shifting legal standards. As the FET standard is

broadenedtoprotectinvestorsfromstateactionsthatconstituteindirectexpropriation,we

would expect virtually all indirect expropriation claims to be pairedwith an FET claim.

However, for reasons previously mentioned, arbitrators have incentives to rule on FET

claims and economize on indirect expropriation claims, thus awarding the investor the

damages owed to them while avoiding further expansion of the indirect expropriation

standard.

5 Conclusion

34

Inthisarticle,wearguethatobservabletrendsinlegalfilingsandrulingscangiveus

little concrete evidence of overall bias in international investment law. Specifically, we

cannotusethisdatatoconcludewhetherlegalclaimsmadeinISDScaseshave,overtime,

trended toward lower (or higher)merit. In general, we advocate caution on the part of

thoseleaningonthequalityofcasesasthebasisofargumentstoreformorevenabandon

ISDS.Surely,thedefactointernational investmentregimeiscontroversialforavarietyof

reasons,andthepossibilityofregulatorychillandconstrainedhoststatesovereigntyare

core among those. Still, there remains a strong advocacy, diplomatic, and scholarly

constituency in favor of ISDS. These observers focus on the ability of the regime to

reinforceruleoflawacrossnationalboundariesthat,inturn,mayreinforcethewillingness

andabilityofeconomicactorstoengageininternationalcommerce.Aspoliticalscientists,

weaimtoprovidecompellingargumentsandaccurateinformationaboutwhatarguments,

oneitherside,aresubstantiatedbytheevidence.Withreferencetothequalityofcases,we

contendthatitisaproblemtooverlooktheroleofjudicialeconomy.Wealsocontendthat

itisaproblemtooverlookthefactthatlegalstandardschange,evenoverarelativelyshort

periodoftimesuchasthatatplayhere.Therealizedstrategiesofarbitratorsandlawyers

do not provide clear evidence of the actual egregiousness of adverse state action or the

extenttowhichinvestorsinflateitsegregiousness.

Withournewanddifferentevaluationoftheevidence,weseeweaksupportforthe

fear that ISDS has already massively chilled even democratic host states from setting

regulations that theydesire, or that suchmassive chilling is on thehorizon.Manyof the

fearsthatforeigninvestorscanabuseandareactingontheirabilitytoabuseinternational

investmentlawmaybeoverstated.Thisisnotatalltosaythattherearenobiasesinthe

35

system. It is also not to say that regulatory chill does not take place: again, Moehlecke

(2018) provides compelling, specific evidence of regulatory chill around tobacco

regulation,whichprovides the first statisticallycompellingsupport for the long-rumored

problem.Butcontraarguments intheveinofPelc(2017),wecannotknowthatfrivolous

casesareontherisewithoutchangingthefocusofouranalysisandresearchmethods.

Having established in this article the problem that current approaches cannot

isolatetrendsinthequalityofISDScases,welookforwardtocontributingtoareevaluation

oftrendsinthemeritofcasesinfuturework.Ifclaimsandrulingscannotprovideevidence

ofthequalityofcases,whatcan?Wesuggestaresearchagendathatfocusesondomestic

politicalandinternationalmarketreactionstocasesandrulings.Ourcontentionisthat,to

understandthepoliticaleconomyofthelawandsystematicbiasesinthesystem,weneed

tofocusonhowhoststatesandinvestorsmightchangetheirbehaviorasaresultofthese

cases.

We see considerable, untapped richness in the jurisprudential history of international

investmentlawforpoliticalscientistsinterestedinunderstandingtheeffectsofISDS.

36

ReferencesBhala,Raj."PowerofthePast:TowardsDeJureStareDecisisinWTOAdjudication(Part ThreeofaTrilogy)."Geo.Wash.Int'lL.Rev.33(2000):873.Blandford,AndrewC."TheHistoryofFairandEquitableTreatmentbeforetheSecond WorldWar."ICSIDReview-ForeignInvestmentLawJournal32.2(2017):287-303.Broude,Tomer,YoramHaftel,andAlexanderThompson."LegitimationThrough Renegotiation:DoStatesSeekMoreRegulatorySpaceinTheirBITs?."Hebrew UniversityofJerusalemLegalResearchPaper17-1.2018.Brunetti,Maurizio.2003.“IndirectExpropriationinInternationalLaw.”InternatioanlLaw FORUMdudroitinternational5:150-154.Brutger,Ryan,andJuliaC.Morse."Balancinglawandpolitics:JudicialincentivesinWTO disputesettlement."TheReviewofInternationalOrganizations10.2(2015):179- 205.Busch,MarcL.,andKrzysztofJ.Pelc."ThepoliticsofjudicialeconomyattheWorldTrade Organization."InternationalOrganization64.2(2010):257-279.Davey,WilliamJ."HastheWTOdisputesettlementsystemexceededitsauthority?A considerationofdeferenceshownbythesystemtomembergovernmentdecisions anditsuseofissue-avoidancetechniques."JournalofInternationalEconomic Law4.1(2001):79-110.Dolzer,Rudolf."Indirectexpropriations:newdevelopments."NYUEnvtlLJ11(2002):64.Dolzer,RudolfandFelixBloch.2003.“IndirectExpropriation:ConceptualRealignments.” InternationalLawFORUMdudroitinternational5:155-165.Dumberry,Patrick."HastheFairandEquitableTreatmentStandardBecomeaRuleof CustomaryInternationalLaw?."JournalofInternationalDisputeSettlement8.1 (2016):155-178.Guzman,AndrewT."Isinternationalantitrustpossible."NyULRev.73(1998):1501.Lupu,Yonatan."Theinformativepoweroftreatycommitment:usingthespatialmodelto addressselectioneffects."AmericanJournalofPoliticalScience57.4(2013):912- 925.Hafner-Burton,EmilieM.,ZacharyC.Steinert-Threlkeld,andDavidG.Victor."Predictability versusflexibility:Secrecyininternationalinvestmentarbitration."World Politics68.3(2016):413-453.

37

Haftel,YoramZ.andAlexanderThompson.2018.“Whendostatesrenegotiateinvestment agreements?Theimpactofarbitration.”TheReviewofInternationalOrganizations 13(1):25-48.Haeri,Hussein."ATaleofTwoStandards:‘FairandEquitableTreatment’andtheMinimum StandardinInternationalLaw:TheGillisWetterPrize."Arbitration International27.1(2011):27-46.Heiskanen,Veijo."TheContributionoftheIran-UnitedStatesClaimsTribunaltoThe DevelopmentoftheDoctrineofIndirectExpropriation."InternationalLawFORUM dudroitinternational.Vol.5.No.3.2003.Helfer,LaurenceR."Exitingtreaties."Va.L.Rev.91(2005):1579.Jiminez,MaximoRomero."ConsiderationsofNAFTAChapter11."Chi.J.Int'lL.2(2001): 243.Johns,Leslie.2014.“DepthversusRigidityintheDesignofInternationalTrade Agreements.”JournalofTheoreticalPolitics26(3):468-495.Johns,Leslie.2015.StrengtheningInternationalCourts:TheHiddenCostsofLegalization AnnArbor,MI:UniversityofMichiganPress.Kalicki,Jean,andSuzanaMedeiros."Fair,EquitableandAmbiguous:WhatisFairand EquitableTreatmentinInternationalInvestmentLaw?."ICSIDReview22.1(2007): 24-54.Knahr,Christina.2007.“IndirectExpropriationinRecentInvestmentArbitration.”Austrian ReviewofInternationalandEuropeanLaw12:85-102.Lowenfeld,AndreasF.Internationaleconomiclaw.OxfordUniversityPress,USA,2007 (secondedition).Minor,MichaelS."ThedemiseofexpropriationasaninstrumentofLDCpolicy1980– 1992."JournalofInternationalBusinessStudies25.1(1994):177-188.Moehlecke,Carolina.2018.“InternationalInvestmentLawandChallengestoGlobalPolicy Diffusion.”Workingpaper.Mostafa,Ben."Thesoleeffectsdoctrine,policepowersandindirectexpropriationunder internationallaw."Austl.Int'lLJ15(2008):267.Olynyk,Stephen."ABalancedApproachtoDistinguishingBetweenLegitimateRegulation andIndirectExpropriationinInvestor-StateArbitration."Int'lTrade&Bus.L. Rev.15(2012):254.

38

Palombino,FulvioMaria.2010.“JudicialEconomyandLimitationoftheScopeofthe DecisioninInternationalAdjudication.”LeidenJournalofInternationalLaw23:909- 932.Paparinskis,Martins.Theinternationalminimumstandardandfairandequitabletreatment. OxfordUniversityPress,2013.Peinhardt,ClintandRachelL.Wellhausen.2016.“WithdrawingfromInvestmentTreaties butProtectingInvestment.”GlobalPolicy7(4):571-576.Pelc,KrzysztofJ."WhatExplainstheLowSuccessRateofInvestor-State Disputes?"InternationalOrganization71.3(2017):559-583.Pelc,KrzysztofJ."Thepoliticsofprecedentininternationallaw:Asocialnetwork application."AmericanPoliticalScienceReview108.3(2014):547-564.Posner,RichardA."Themeaningofjudicialself-restraint."IndianaLawJournal59(1983): 1-24.Schill,StephanW."TheMultilateralizationofInternationalInvestmentLaw:Emergenceofa MultilateralSystemofInvestmentProtectiononBilateralGrounds."TradeL.& Dev.2(2010):59.Shany,Yuval.2006.“TowardaGeneralMarginofAppreciationDoctrineinInternational Law?”TheEuropeanJournalofInternationalLaw16(5):907-940.Shelton,Dinah."Rightingwrongs:Reparationsinthearticlesonstate responsibility."AmericanJournalofInternationalLaw96.4(2002):833-856.Simmons,Beth.2014.BargainingoverBITs,ArbitratingAwards:TheRegimeforProtection andPromotionofInternationalInvestment.WorldPolitics.66.12-46.Sornarajah,Muthucumaraswamy.Theinternationallawonforeigninvestment.Cambridge UniversityPress,2017,thirdedition.Sornarajah,Muthucumaraswamy.Theinternationallawonforeigninvestment.Cambridge UniversityPress,2004,secondedition.StJohn,Taylor.2018.TheRiseofInvestor-StateArbitration:Politics,Law,andUnintended Consequences.Oxford:OxfordUniversityPress.Sunstein,CassR.Onecaseatatime:JudicialminimalismontheSupremeCourt.Harvard UniversityPress,2001.Tucker,ToddN.JudgeKnot:PoliticsandDevelopmentinInternationalInvestmentLaw. AnthemPress,2018.

39

VanHarten,Gus.2010.FiveJustificationsforInvestmentTreaties:ACriticalDiscussion. Trade,Law,andDevelopment2(1):1-32.Vicuña,FranciscoOrrego."Regulatoryauthorityandlegitimateexpectations:balancingthe rightsofthestateandtheindividualunderinternationallawinaglobalsociety."InternationalLawFORUMdudroitinternational.Vol.5.No.3.2003.VonStein,Jana."Dotreatiesconstrainorscreen?Selectionbiasandtreaty compliance."AmericanPoliticalScienceReview99.4(2005):611-622.Waibel,Michael,AshaKaushal,Kyo-HwaChung,andClaireBalchin,editors.2010.The BacklashAgainstInvestmentArbitration:PerceptionsandReality.KluwerLaw International.Weiner,AllenS."IndirectExpropriations:TheNeedforaTaxonomyof"Legitimate" RegulatoryPurposes."InternationalLawFORUMdudroitinternational.Vol.5.No.3. BrillAcademicPublishers,2003.Weller,Marc.2011.“ModestyCanbeaVirtue:JudicialEconomyintheICJ.”LeidenJournal ofInternationalLaw24:127-147.Wellhausen,RachelL.2015.TheShieldofNationality:WhenGovernmentsBreakContracts withForeignFirms.Cambridge,MA:CambridgeUniversityPress.Wellhausen,RachelL."Recenttrendsininvestor–statedisputesettlement."Journalof InternationalDisputeSettlement7.1(2016):117-135.

40

Appendix

Figure1a:Overtime,alowerpercentageofallegedclaimsreceiverulings(byyearoffiling,allclaims,1987-2012)

41

Figure2a:Overtime,alowerpercentageofallegedclaimsreceiverulings(byyearoffiling,jurisdictionlossesremoved,allclaims,1987-2012)

42

Figure3a:ComparisonofouroriginalclaimscodingwithUNCTADclaimscoding

43

Figure4a:Overtime,alowerpercentageofallegedclaimsreceiverulings(byyearofcasefiling,coreclaimsincludingMST,1987-2012)(Onlycore+MSTclaimsarecountedhere;averages lower than1arepossible,asnotallcasescontaincore/MSTclaims)

44

Figure5a:Overtime,alowerpercentageofallegedclaimsreceiverulings(byyearoffiling,jurisdictionlossesremoved,coreclaimsincludingMST,1987-2012)(Onlycore+MSTclaimsarecountedhere;averages lower than1arepossible,asnotallcasescontaincore/MSTclaims)