Handbook on E-Commerce and Competition in ASEAN · E-commerce in ASEAN E-commerce markets have grown significantly within ASEAN over recent years. Since 2015, the number of internet

Handbook on E-Commerce and Competition in ASEAN

1

ContentsEXECUTIVE SUMMARYGLOSSARYINTRODUCTION1. Introduction

PART A: OVERVIEW OFE-COMMERCE IN ASEAN

2. Introduction to E-commerce and its value chain

2.1. DefinitionofE-commerce2.2. Overviewofthebusinessmodels

associatedwithE-commerce2.3. Overviewofthevaluechain

3. Overview of the E-commerce landscape in ASEAN, the current state of E-commerce development in each of the AMS and its growth potential

3.1. Introduction3.2. Overviewofthecurrentretail

E-commerce markets in ASEAN andtheirlikelyevolution

3.3. ImpactofE-commerceon thevaluechaininASEAN

3.4. Discussiononfiveindustries disruptedbyE-commerce withinASEAN

4. Key competition and other regulatory challenges and/or barriers faced by businesses in the E-commerce sector for AMS and how they hinder competition and growth of the E-commerce sector in the region

4.1. Introduction4.2. Barriers to expansion4.3. Barrierstoentry

PART B: SECTION FOR COMPETITION AUTHORITIES

5. Introduction

6. Marketdefinition,multi-sided markets, and market power

6.1. Definingtherelevantmarket6.2. Multi-sidedmarkets6.3. Assessingmarketpowerand

theimpactofdynamiccompetition frominnovation

7. Vertical agreements 7.1. Introduction7.2. ChallengesfacedbyCompetition

Authoritiesintheassessment ofverticalrestraints

7.3. Selectivedistributionnetworksthatexcludeorrestrictonlinesales

7.4. ResalePriceMaintenance(RPM)7.5. Dualpricingsystems7.6. Geo-blocking7.7. Platformbans7.8. MostFavouredNation(MFN)

clauses7.9. Restrictions on price comparison

websites7.10. Exclusivepurchaserestrictions7.11. Practicalsteps/guidelinesor

recommendationstoidentifyand addresscompetitionpolicyand

lawissues

8. Horizontal coordination8.1. Introduction8.2. Pricemonitoringtoolsand

pricesettingalgorithms8.3. Onlineplatformsandcollusion8.4. Coordinateduseofvertical

restraintsbycompetitors8.5. Practicalsteps/guidelinesor

recommendationstoidentify andaddresscompetitionpolicy andlawissues

9. Unilateral conduct 9.1. Introduction9.2. Tying/bundling9.3. Predatorypricing9.4. Pricediscrimination9.5. Fidelityrebatesorloyaltydiscount

schemes

3

9

13

14

16

22

16

1618

22

22

28

30

34343439

44

464647

52

5656

56

6064666871

72

7577

79

8282

8385

87

89

9292939698

100

2

9.6. Imposingverticalconditions (e.g.quantityforcing)

9.7. Practicalsteps/guidelinesorrecommendationstoidentify andaddresscompetitionpolicy andlawissues

10. Mergers and acquisitions 10.1. Introduction10.2. Abilityofexistingcompetition

rulestocapturerelevant transactions

10.3. Innovationanddynamic competitioninmerger assessments

10.4. Networkeffectsinmerger assessments

10.5. Structuralandbehavioural remedieswherenetwork effectsarepresent

11. Recommendations on improving the design of competition policy and enforcement of competition law to proscribe anti-competitive conduct relating to E-commerce for AMS

11.1. Introduction11.2. Designofcompetition

policyandlaw11.3. Enforcementofcompetitionlaw

12. Competition policy and law compliance checklist for businesses engaged in E-commerce in ASEAN

12.1. Introduction12.2. Stagesofriskmanagementtoavoid

competitionlawinfringement12.3. Identifyrisks12.4. Assess risks12.5. Takeactiontoreducerisks12.6. Review processes

14. The impact of intellectual property rights (including its territorial nature) as a barrier to E-commerce in ASEAN and as an impediment to a single digital market in ASEAN

14.1. Introduction14.2. Intellectualpropertyrightsasa

barrier to E-commerce14.3. Intellectualpropertyrightsasa

barriertoasingledigitalmarket in ASEAN

14.4. Theroleofcompetitionauthorities

15. Recommendations on the strategies, tools or approaches AMS can adopt to help government bodies within their respective countries to understand the impact of their policies and initiatives on competition in the E-commerce sector

15.1. Introduction15.2. Roleofcompetitionauthorities

andregionalbodies15.3. Exanteevaluationsofpolicies15.4. Expostevaluationsofpolicies

CONCLUSIONS16. Conclusions

BIBLIOGRAPHYResources citedCompetition cases reviewed

ANNEXESAnnex 1: Technical InformationA1.1. Multi-sidedmarketsA1.2. Blockexemptionsandhardcore

restrictions

Annex 2: Government initiatives on E-commerceA2.1. BruneiDarussalamA2.2. CambodiaA2.3. IndonesiaA2.4. LaoPDRA2.5. MalaysiaA2.6. MyanmarA2.7. ThePhilippinesA2.8. SingaporeA2.9. ThailandA2.10. Vietnam

PART C: SECTION FOR COMPETITION ADVOCACY

13. Regulatory and legal barriers in ASEAN to E-commerce and as impediments to a single digital market

13.1. Introduction13.2. Access to E-commerce13.3. Cybersecurity

100

101

105105

105

110

111

112

113113

113119

122122

123123126126126

128128

128129

131131

132

132135

136136

137

138140

142

157

146

160

160

161

162163163164164165

167168168

166

141

145

159

162

3

ExecutiveSummary

4

1 Statista (2017n,o,p,q,r,s).

E-commerce in ASEAN

E-commerce markets have grown significantly within ASEAN overrecentyears.Since2015,thenumberofinternetusersinthesixlargesteconomiesinASEANhasrisenfrom244millionto283million1,andthisgrowthisprojectedtocontinueatanannualrateof17.7%until20202. Despite these high levels of growth, there remains room for furtherexpansioninE-commercemarketsacrossASEAN.Singapore,Malaysia,Thailand,Indonesia,VietnamandthePhilippinesallcurrentlygeneratelessthan4%oftheirretailsalesonline,amuchlowerproportionthanotherE-commercemarketssuchastheRepublicofKorea (16%)andChina(7%)3.

To make the full potential of the E-commerce market in ASEAN,improvementsarerequiredintermsoftechnologicalinfrastructure,andin the regulatory and legal environment in which E-commerce firmsoperateacrossASEAN.AbroadbanddividecurrentlyexistsbetweentherichermetropolitancitiessuchasBangkok,KualaLumpurandJakarta,andmorerurallocations.InsomeASEANMemberStates(AMS),broadbandremainsexpensiveincomparisontootherdevelopedcountries,therebyinhibitingaccesstoE-commercemarketsforsomeconsumers,althoughthegrowth inM-commerce istosomeextenthelpingtoaddressthisdisparity.Cyber-securityconcernsarealsocommonacrosstheregion,resulting in a lack of trust among consumerswhen asked to providebanking details online. Consequently, many consumers still have apreferenceforshoppinginbrick-and-mortarstores.

InordertosupportthedevelopmentofE-commercemarketsacrossASEAN, and facilitate cross-border trade, greater harmonisation ofregulationsacrossthe region is required,for instancewith regardstocustomsandtaxruleswheredisparitiesamongAMScurrentlyexist.

Impact of E-commerce on competition

Theeasewithwhichconsumerscancomparepricesacrossdifferentretailers has increased. Price comparison websites (PCWs) havegreatlyenhancedpricetransparencyforconsumers inmanymarkets.Competitors’ prices are also now more visible to firms, enablingretailerstoimplementmoreresponsivepricingstrategies.Thishasbeensupportedbythedevelopmentofnewtechnologiessuchasautomatedpricingalgorithmswhichallowfirmstoinstantlyrespondtocompetitors’price movements.

The variety of products available to consumers has also increased.E-commerceretailersarenowabletostockamoreextensiverangeofproductsincomparisontobrick-and-mortarstoresduetoareductioninphysical constraintsandan increase in theability toaccesswidergeographicmarkets.

Consumershavelargelybenefittedfrombothofthesedevelopments.Searchcostshavedecreasedsignificantly,both intermsoftimeandcost,andcompetitiononpricehas intensified.Consumersbenefitaslongaspricecompetitionisnotattheexpenseofquality,innovationordiversityofgoods/servicesonoffer.

For new entrants and smaller retailers, some barriers to entry andexpansion have diminished as a result of the emergence and growthof E-commerce. Economies of scale that large retailers may benefit

ExecutiveSummary

Retail sales online

The number of internet users in the six largest economies in ASEAN is

estimated to be

283million

16%

7%

<4%Singapore, Malaysia, Thailand, Indonesia, Vietnam and Philippines

China

Republic of Korea

2 Frost & Sullivan (2016b). 3 UNCTAD (2015), page 21.

5

frominbrick-and-mortarmarketshavefallenasthefixedcoststoretailersfromenteringnewmarketsandlocationshavedecreased.Othernewbarrierstoentryandexpansionhavehoweveremerged,orbecamemorepronouncedinE-commercemarketsincomparisontotraditionalbrick-and-mortarsaleschannels.Barrierstoentrymaybepresentinmulti-sidedmarketswherenetworkeffectsarepresent(i.e.thevalueoneuserplacesonaplatformincreasesasotherusersjointhatplatform).Asaplatformgrowsinsize,networkeffectsincrease,thereforeincreasingthecoststoconsumersfromswitchingtoanalternativeplatform.Asaresult,itisharderforsmallerplatformstoenterandgainmarketshare. Ifconsumersusemultipleplatforms(i.e.theymulti-home)however,networkeffectsposelessofabarriertonewentrants.Accesstosupportinginfrastructure,suchaslogistics,inventoryandpaymentsystemsmayalsoconstituteabarriertoentry,andverticalintegrationbyaplatformorsingle-sidedfirmmayaffectotherfirms’abilitytogainaccesstothesesystems.

E-commercehasalsoenabledfirmstocollectmoredetaileddataontheircustomers.Thishasmadeitpossibleforfirmstoofferproductsandservicesbettertailoredtoconsumers’preferences.Itiswidelydebatedwhetheraccesstothisdataconstitutesabarrierfornewentrants.Someconsiderdatatobeanassetthatnewfirmsareunabletoreplicate.However,inmanymarketssuchdatacanbeobtainedfromavarietyofsources,therebyreducingtheextenttowhichthedataanincumbentfirmholdscaninhibitthegrowthofsmallercompetitors.

Defining markets, multi-sided markets, and assessing market power

Manynewmulti-sidedonlinemarketshaveemergedasaresultofthegrowthinE-commerce,suchasonlinemarketplacesandPCWs.Inthesemarkets,existingapproachestodefinetherelevantmarket(s)maynolongerapplyduetotheinterrelationshipsandexternalitiesbetweendistinctsidesofthemarketwhichaffectthewayinwhichfirmssetprices.Ifthevaluefromusingaplatformincreasesononesideasaresultofmoreusersontheotherside,aplatformmaysetpricebelowcostononesideofthemarkettoattractusersontheotherside.Thetraditionaltests4usedbycompetitionauthoritiesfordefiningamarketarethereforetypicallynotapplicable.Whenconductingamarketdefinitionassessmentinmulti-sidedmarkets,insomeinstancesthetotalpricechargedtoallsidesofamarketshouldbeconsidered(i.e.thesumofthepricechargedtoallsidesofamarket),asopposedtoconsideringthepricechargedtoeachsideinisolation.Inreality,however,insteadoftechnicallydefiningarelevantmarket,competitionauthoritiesmaybebetterplacedtopursueamoreholisticassessmentofthemarket,byconsideringmorebroadlythecompetitiveconstraintsthatafirmfacesonallsidesofthemarketandtheabilityofconsumerstosubstitutetoanalternativeprovider.Also,whenassessingmarketpowerinmulti-sidedmarkets,thenatureofcompetitionshouldbeassessed,andinparticulartherelationshipsbetweenallsidesofthemarketshouldbeconsidered,focusingonnetworkeffectsandanyadditionalfeedbackeffects.

Whenassessingmarketpower,competitionauthoritiesmayalsowanttoassessthedatathatafirmholdsonitscustomers,andtheaccessthatcompetitorshavetosimilarinformation.Thisiscurrentlyanareaofdebateinthefieldofcompetitionpolicy.Ononehand,firmsmaybeabletopurchasesuchdatafromothersources,butontheotherhand,insomeinstancesthisalternativedatamaynotbeofequivalentqualitytothedatapossessedbythemarketleadingfirm.

Competitionauthoritiesmayneedtoadapttheirapproachfortheassessmentofallegedanti-competitiveconductinmulti-sidedmarkets.Thepresenceofexternalitiesbetweendifferentsidesofmarketsmakesthestandardanalyticalframework,foundedonassumptionsfromsingle-sidedmarkets,ill-suitedtoinvestigatingallegedanti-competitiveconduct.Forexample,ininstancesofpotentialpredatorypricing,thecostsincurred,andpriceschargedtoallsidesofthemarketmayneedtobeconsideredtogetherratherthanfocusingonthepriceandcostoneachsideofthemarket in isolation. Inassessingharminmulti-sidedmarkets,theinterrelationshipsbetweendifferentsidesof themarketshouldalsobeconsidered, thoughthisdoesnotnecessarilymeanthatharmononesideofthemarketcanbeoffsetbybenefitsonanotherside.

Onlinemarketsoftenevolverapidlyascompetitorssuccessfully innovateanddisplace leading incumbentfirmsfromtheirpositioninthemarket.Therefore,whenassessingmarketpower,bothinsingle-andmulti-sidedmarkets,competitionauthoritiesshouldlookbeyondthestaticmarketshareofafirm,andalsoconsiderthedynamiccompetitionfrompotentialfutureentrantstoamarket.Thisisparticularlyrelevantinmergerassessmentswherethemergermayresultintheremovalofapotentialfutureentranttoamarket,evenifthereisnooverlapintheproductsorservicescurrentlyprovidedbythemergingparties.

Vertical agreements

TheemergenceandgrowthofE-commercehasresultedinanincreaseintheadoptionofverticalrestraintsbyfirms,duetoconcernsoffree-ridingbyonlineretailersonthepre-orpost-salesservicesprovidedbybrick-and-mortarstoresorotheronlineretailersandplatforms.

4 The ‘small but significant and non-transitory increase in price’ test, or SSNIP test, is a typical example which consists of identifying the smallest possible market (in terms of products and geographic scope) that a hypothetical monopolist could sustainably and profitably increase price.

6

5 European Commission (2010), para. 100.6 A hardcore restriction is one that is so serious that consideration of any pro-competitive effects is highly unlikely and rare. For example, resale price

maintenance, RPM, is widely treated in this manner. 7 Passive sales are where a consumer independently reaches out to a retailer. Conversely, active sales are where a retailer directly targets a consumer e.g. through advertising.8 It is noted that in some jurisdictions, such as Singapore, vertical restraints are per se exempt therefore the recommended approach would also apply to those

vertical restraints regarded as hardcore restrictions in other jurisdictions.9 A wide MFN is a vertical restraint that ensures that no other competitor will be given more favourable terms by a supplier/customer/platform – for instance

being able to sell at a lower price. A narrow MFN restricts a firm from setting a lower price in its own store, but it is free to agree to a lower price with a competing store e.g. a hotel that enters a narrow MFN agreement with a hotel booking platform cannot set a price on its own website lower than the price on the booking platform, but it can agree to lower prices on competing platforms.

10 See for example: CE/9320-10 (CMA), B 9-121/13 (Bundeskartellamt).11 An international working group including ten competition authorities (Belgium, Czech Republic, France, Germany, Hungary, Ireland, Italy, Netherlands, Sweden

and the UK) was set up to coordinate actions for a possible harmonisation of approach on wide and narrow MFN clauses across jurisdictions.12 Active sales refer to cases in which a firm reaches out to consumers (for example through targeted advertising); whereas passive sales consist of cases in which a

consumer independently reaches out to a retailer to make a purchase. 13 European Commission (2017b), para. 52. 14 CMA, 50223, Online sale of posters and frames (2016); US Department of Justice, Press release number 15-1488 (2015).15 A comprehensive questionnaire on competition in E-commerce in ASEAN was designed for the purpose of this handbook. The competition authorities of

Singapore, Malaysia, the Philippines, Vietnam and Indonesia completed this questionnaire in April 2017.

Agreementsbetweenfirmsatdifferentstagesofproductionaregenerallybenignastheygenerateefficiencies,forexample,improvingtheavailabilityorqualityofservicethatconsumersreceivebyovercomingissuesoffree-riding,reducingpricebyovercomingdoublemarginalisation,and/orresolvingpotentialspecificinvestmenthold-uprisks.However,insomeinstances,verticalrestraintscanposechallengestocompetitionauthoritiesinthattheycanalsogive risetoanti-competitiveeffects.Vertical restrictionsgenerally inhibit intra-brandcompetition.Thismayfacilitatecollusioninsomeinstanceswheninter-brandcompetitionislimited.5

SomeoftherestraintsusedbyfirmsinE-commercemarketshavebeenregardedashardcorerestrictionsofcompetitioninEurope,6onthebasisthattheanti-competitiveeffectshavebeendeemedtogreatlyexceedany efficiency benefits to consumers. Restrictions that unjustifiably prevent all sales via the internet, ordiscriminatebetweenonlinestoresandbrick-and-mortarretailers (e.g.onthewholesalepricechargedtoaretailer),areregardedashardcorerestrictionsandthereforenotallowedintheEU.Restrictionsoncross-borderpassivesales intheEU7 arealsotreated inthisway,for instancewhereconsumersareunabletoaccessaforeignwebsiteorunabletocompletetransactionsonaforeignwebsite.

AsaresultofthenoveltyoftheapplicationofverticalrestraintstoE-commerce,andoftheensuinguncertaintyas towhether thesevertical restraints are to the benefit or to the detriment of consumers, a clear andconsistentpositionhasnotyetbeenreachedbycompetitionauthoritiesaroundtheworldonallformsofverticalrestraints.Withtheexceptionofhardcorerestrictionswhichareunderstoodtobeessentiallyharmfulto competition,when assessing any such vertical restraints a case-by-case approach is recommended.8 ThisappliestoMostFavouredNation (MFN)clauses,9alsoknownasbest-priceguaranteesorpriceparityclauses.Currently,competitionauthoritiesaroundtheworldhavetakencontrastingpositionsontheuseofMFNs.Thiswasthecase,forexample,inthehotelbookingmarket,10where,despiteinternationalattemptsforcoordination,11competitionauthoritieshavereacheddifferentconclusions.InternationalconsensushasalsonotyetbeenreachedontheuseofagreementsthatpreventaretailerfromsellingviaonlinemarketplacesoradvertisingonPCWs.

As the ASEAN region continues to pursue its objective of becoming a more integrated market, ASEANcompetitionauthoritiesmayalsobeconcernedwithverticalrestraintsthatrestrictcross-bordertrade.Geo-blockingstrategiesemployedbyfirmsmayinhibitthedevelopmentofE-commercemarketsacrosswiderregions such asASEAN. In Europe,where digitalmarket integration amongMember States is also a keyobjective,animportantdistinctionismadebetweenverticalrestraintsthatrestrictactiveandpassivesalestoaparticularcountry.12Theformerispermittedifitconcernssalesintoanexclusiveterritory,whereasbothactiveandpassivesalesrestrictionsareprohibited if implementedwithinaselectivedistributionsystem.13 Blockingpaymentfromothercountriesorredirectingweb-browserstoalocalwebsitemaybeconsideredpassivesalesrestrictions.

Horizontal coordination

Greaterpricetransparency,andthedevelopmentofadvancedpricesettingalgorithmshavemadeestablishingandenforcingpricecoordinationeasierforfirmsinsomemarkets.EvidencefromcasesintheUSandtheUKhaveshownthatexistingcompetitionpolicyand laware largelysufficienttodealwiththechallengesraisedbypricealgorithmsatthisstage.14NoequivalentcaseshavebeeninvestigatedinASEAN,andonlyonequestionnairerespondent15currentlyconsidersprice-settingalgorithmstobeacompetitionconcernwithinitsjurisdiction.However,asE-commercemarketscontinuetogrow,thischallengemaybecomemoreprevalentintheregion.Thechallengefacedbycompetitionauthoritiesaroundtheworldinthisareahasbeenamorepracticalone.Theneedtoinvestigatethenatureofpricealgorithmsandtheirfunctionshasmadeitessentialtorecruitthenecessaryexpertisewiththeabilitytoundertakesuchinvestigations.

7

Competition authorities should however closely monitor the development of price-setting algorithms.Concernshavebeenraisedthatassuchtechnologiesbecomemoresophisticated,theymayself-learnthatcoordinationamongcompetitorsisoptimal.Weresuchdevelopmentstooccur,legalclaritywouldberequired,forinstanceexplainingwheretheliabilityfalls.Theissuehasonlyjustemergedasaquestionforcompetitionpolicyandlawthereforeaconclusiononthisquestionhasnotyetbeenreached.

Thedevelopmentofonlineplatformsinmulti-sidedmarketshasalsomadecompetitors’pricingmoretransparenttofirms.Tworecentcaseshighlighthowcoordinationcanoccurbetweencompetingfirmsonthepriceschargedonplatformsbothwithandwithoutfacilitationfromtheplatformitself.16Inbothoftheseinstances,existingcompetitionpolicyandlawwassufficienttoidentifyandinvestigatetheallegedanti-competitiveconduct.

AsevidencedbytheE-bookscaseinvestigatedbycompetitionauthoritiesaroundtheworld,17 firmsoperatingin E-commercemarketsmay also implement vertical restraints in a coordinatedmanner, leading to anincreaseinpricesinamarket.Thereisnogeneralruleastowhenanetworkofverticalagreementsconstituteshorizontalcoordination,however intheE-bookscase, theUSauthoritieshighlightedthe integral rolethatAppleplayedinensuringthatfiveleadingpublishersalladoptedthenewstructureofverticalagreementwithAmazonasanimportantfactor.

Unilateral conduct

Formsofconductthatcompetitionauthoritiesmaydeemtobeanti-competitivebyafirminadominantpositionareanalogousinE-commercemarketstoexclusionaryorexploitativetypesofconductobservedinbrick-and-mortarmarkets,forexample:settingunreasonablyhighprices,sellingatartificiallylowpricestoforeclosecompetitorsfromthemarket,orobstructingcompetitorsinthemarketthroughtyingorbundling.

ThegrowthofE-commercehashoweverincreasedtheprevalenceofsomeofthesetypesofconduct.Manymulti-sidedplatformsthatoffer a rangeof relatedserviceshaveemployedtyingandbundling strategies,attractingtheattentionofcompetitionauthoritiesaroundtheworld.Forexample,Googlehasbeeninvestigatedforaseriesofallegedinstancesoffavouringitsownservices.18TworelevantcaseshavealsobeeninvestigatedinASEAN,notably relatingtothetyingandbundlingofonlineservices19, andthe impositionofexclusivityagreementsbyanonlineticketingplatform20.

Experience to date has indicated that the legal framework for abuse of dominance in brick-and-mortarmarketsisbroadlysufficienttodealwithanalogousconductinE-commercemarkets.Animportantfactorinsuchassessmentsshouldbetheextenttowhichtheconductisharmingcompetition,orwhetheradominantfirmissimplymoreefficientorinnovatingatafasterpacethanitsrivals.

SomeconsiderBigDatatobeasourceofmarketpower,thereforewhenassessingwhetherafirmisdominant,thedata(orabsenceofdata)thataleadingfirmanditscompetitorspossessmaybeanimportantfactortoconsider.However,todate,nocompanyhasbeenfoundtohaveinfringedcompetitionlawasaresultofabusingapositionofdominancethroughtheuseofBigData,anddoubtshavebeenraisedinseveralforaastowhetherBigDatacouldpossiblyberegardedasanessentialfacilitygivenitsnature,whichallowsittobereplicated.Thedebateonthisissueisongoingatthetimeoffinalisingthishandbook.

Mergers and acquisitions

Giventhe rapidpaceofchange inmanyE-commercemarkets, and relatively lowbarrierstoentry,whenassessingwhether existingmerger control regimes are suitable for capturing potentially harmfulmergersinE-commercemarkets, it is importanttoconsiderdynamiccompetition.Competitionauthoritiesshouldconsiderwhethertheirexistingregimeincludesruleswhicharesufficientlybroadsothatcasesofpotentiallessening of dynamic competition can be assessed, even if there is limited or no current overlap in theproductsandservicesofferedbytheparties,orwhenturnoverthresholdsarenotmet.

AMSarecurrentlyatdifferentstages indevelopingtheirmergercontrols,withCambodiaannouncingtheirdraftlawin2016.Wheretherearemergercontrolrulesinplace,thesemayfailtocapturemergersthatcouldremoveapotentialfutureentranttothemarket,forexampleifrevenuesfallbelowtherelevantthreshold,despiteahightransactionvalueonthedeal.ThisiscommoninE-commercemarketswheretheacquiringfirmmayplaceahighvalueonthetechnologyoftheacquiredfirmbasedontheprospectoffuturerevenue,

16 CCS 500/003/13; and Lithuanian Competition Council (LCC), Case C-74/14, Eturas (2016).

17 See for example European Commission, COMP/39.847 (2012); and Case 13-3741, United States v. Apple Inc. et al. (2016).

18 European Commission, 40099 Google Android, 39740 Google comparison shopping; and UK High Court Streetmap.EU Limited v Google Inc., Google Ireland Limited and Google UK Limited [2016] EWHC 253 (Ch).In the Google comparison shopping case, the European Commission determined that Google had abused its position of dominance as a search engine by favouring its own comparison shopping service in search results ahead of competing comparison shopping providers.

19 My E.G. Services Berhad (24/06/16); Malaysia Competition Commission.

20 Abuse of a Dominant Position by SISTIC.com Pte Ltd (CCS/600/008/07).

8

and,potentially, internalisingthevalueofhigherfutureprofitsduetoareductioninexpectedcompetition.Thisissuehasledsomecompetitionauthoritiesaroundtheworldtoreviewtheirtestsformergernotification.Someauthorities(e.g.Germany)areintheprocessofupdatingtheirtestsfornotificationsoastoincludeatestbasedonthevalueofthetransaction.CurrentlynoAMShasatransactionvaluethresholdinplace.

Inassessingproposedmergers inmulti-sidedonlinemarkets,thepresenceandextentofnetworkeffectsshouldbeconsidered.Authoritiesshouldevaluatewhetheratippingpointismorelikelytooccurasaresultofthemerger.Suchassessmentshouldconsidertheextenttowhichconsumersmulti-home,theswitchingcoststhatconsumersencounter,theinteroperabilitybetweencompetingplatforms,andthebarrierstoentryandexpansionsmallerfirmsface.Ifitisdeemedthatremediesarerequired,theyshouldbedesignedtofocusonmaintainingorimprovingthesemarketcharacteristics.

More generally, in E-commercemarkets, competition authoritiesmay identify potential issues inmergersbetweenfirmsatdifferentstagesintheverticalchainifthemergergivesrisetotheincentiveandtheabilityforthemergedentitytopursueforeclosurestrategiesaimedatexcludingormarginalisingcompetitor/s,orwhenamavericknewentrantisbeingacquiredbyalargerincumbentfirm.Themergerreviewshouldalsoconsiderwhetherthemergermaygiverisetomarketpowerasaresultofthepoolingofconsumerdataheldbythemergingparties.However,thismaybemitigatedifcompetitorsareabletosourceequivalentdatafromothersources.

Sufficiency of existing competition policy and law to protect and promote effective competition in E-commerce markets

To date, competition authorities around theworld have found the legal framework provided by existingcompetitionpolicyandlawtobelargelysufficienttodealwithvirtuallyallcompetitionchallengesbroughtabout by the emergence and growth of E-commerce. Case reviewspresented throughout this handbookillustrate this.However, themore technical nature of some formsof alleged anti-competitive conduct inE-commercemarketshasgivenrisetoabroadneedtodevelopspecificresourceswhichareabletoexploreandassesstheseissues,suchasdealingwithpotentialcoordinationviapricingalgorithms.

ThegrowthofE-commercehasgivenrisetoasignificantincreaseintheadoptionofverticalrestraints.Thisisparticularlydue to the growthofonlineplatforms.As a result,vertical restraintshavebeen theobjectofwidedebate, andonoccasion, different competitionauthoritieshave takencontrastingpositions, thusposing a challenge for firms operating internationally. For example, the different conclusions reached bycompetitionauthorities investigatingtheuseofMFNclauses inthehotelbookingsmarkethavemadetheneedforinternationalcoordinationquiteapparent.Whilsttheremightnotbeabroadneedtocreatenewrulesinordertodealwiththesetypesofissues,internationalcoordinationcouldhelptoharmonisetheapproachindealingwithcompetitionchallengesinE-commercemarkets,thoughattemptssofarinthehotelbookingmarkethaveprovenunsuccessful.

In order to create a stable and consistent policy environment for firms to operate within, cooperationamongcompetitionauthoritiesacrossAMSontheapproachesusedtoinvestigateinstancesofallegedanti-competitiveconductinE-commercemarketsisparticularlyimportant.

Competition advocacy role of authorities

CompetitionauthoritiescanalsofacilitatethegrowthanddevelopmentofE-commercemarketsinASEANthroughsupporttobusinessesandgovernmentbodies intheformofadvocacy.Thechecklistprovided inSection12ofthishandbookprovidesguidanceandsupporttobusinessesengagedinE-commerceacrossASEANincomplyingwithcompetitionlaw.

Byencouraginggovernmentbodiestoharmonisethelegalandregulatoryenvironmentinwhichbusinessesoperate,cross-bordertradewillbeencouraged.Ensuringcoordinatedandeffectivesystemsof intellectualproperty(IP)rightsallocationandenforcementacrossAMSwillprovidebusinesseswithsufficientconfidencethat the returns fromtheir investmentswill beprotected.Thiswould increasefirms’ incentives to investwhichwouldinturn,facilitateinvestmentin,andthedevelopmentof,E-commercemarkets.Greaterregionalcoordination to tackledataprotection, cybersecurity, andaccess tobroadband issueswouldalso furtherfacilitatethedevelopmentofasingledigitalmarketinASEAN. Competition authorities can provide further support to government bodies through offering guidance onconducting assessmentsof the impact of proposedpolicies on competition inE-commercemarkets. Byundertakingexanteassessmentsofthelikely impactofapolicy,anyunwelcomeanti-competitiveeffectscanbepreventedormitigated. Inconductingsuchassessmentsgovernmentbodiescanadoptarangeofqualitativeandquantitativeapproaches,suchaslookingattheeffectofsimilarpoliciesinrelatedproductorgeographicmarkets.Toevaluatehowsuccessfulapolicyhasbeen,ortodecidewhethertoexpandapolicywider,anexpostevaluationcanbeadoptedusingsimilartechniques.

9

Glossary

10

AustralianCompetitionandConsumerCommission

ASEANExpertsGrouponCompetition

ASEAN Member State

TheAssociationofSoutheastAsianNations.

ThesixlargesteconomiesinASEAN:Singapore,Vietnam,Thailand,Malaysia,Indonesia,andthePhilippines.21

ASEANWorkingGrouponIntellectualPropertyCooperation

Businesstobusiness.

Businesstobusinesstobusiness.

Businesstobusinesstoconsumer.

Businesstoconsumer.

Afirmthatdoesnotconductbusinessonline,butonlythrough‘traditional’offlinechannels(e.g.inphysicalstores).

Consumertobusiness.

Consumertoconsumer.

CompoundAnnualGrowthRate.

CompetitionCommissionofSingapore

Afirmthatconductsbusinessonlineandalsothrough‘traditional’offlinebrick-and-mortarchannels.

CompetitionandMarketsAuthority(UK)

Theindividualorentitythatusestheproductorservice

Theindividualorentitythatpurchasestheproductorservice

ThepartoftheoveralllegalsystemthatdealswiththeInternetandcyberspace.

DigitalComparisonTool

Theuseofemergingdigitaltechnologytodriveefficienciesacrossdifferentbusinessprocesses.

DepartmentofJustice(USA)

Instanceswhenanonlineretailerpassesanorderdirectlytothewholesale/manufacturer,thereforeremovingtheneedtohaveaphysicalwarehousetostoretheproductstheysell.

EuropeanCommission

EuropeanCourtofJustice

EuropeanCompetitionNetwork

Thebuyingandsellingofgoodsandservicesovertheinternet.

Glossary

21 GDP (IMF, October 2016).

3 DefinitionTerm

ACCC

AEGC

AMS

ASEAN

ASEAN6

AWGIPC

B2B

B2B2B

B2B2C

B2C

Brick-and-mortar firm

C2B

C2C

CAGR

CCS

Click-and-mortar firm

CMA

Consumer

Customer

Cyberlaw

DCT

Digital adoption

DoJ

Drop shipping

EC

ECJ

ECN

E-commerce

11

The people, processes and technology required to deliver anonlineordertoaconsumer.

Fair,reasonableandnon-discriminatory

FairTradeCommission(USA)

Governmenttobusiness.

Governmenttoconsumer.

Measurestorestricttheaccesstoproductsorservicesthroughtheinternetbasedonthegeographiclocationoftheuser.

Agreements between competing businesses operating at thesame level inthemarkettocollectivelyagreeonsomeactivity(e.g.setaspecificlevelofpricesorproduction).

IntellectualProperty

JapanFairTradeCommission

KomisiPengawasPersainganUsaha(IndonesiaCompetitionAuthority)

LithuanianCompetitionCouncil

MonetaryAuthorityofSingapore

AnE-commerceactivityconductedviaamobiledevice

Multi-lateralInterchangeFee

A wide MFN is a vertical restraint that ensures that no othercompetitorwill be givenmore favourable terms by a supplier/customer/platform– for instancebeing able to sell at a lowerprice.AnarrowMFN restrictsafirmfromsettinga lowerpricein itsownstore,but it is free toagree toa lowerpricewithacompetingstoree.g.ahotelthatentersanarrowMFNagreementwith a hotel booking platform, cannot set a price on its ownwebsitelowerthanthepriceonthebookingplatform,butitcanagreetolowerpricesoncompetingplatforms.

Atwo-ormulti-sidedmarketisoneinwhichdistinctbutrelatedcustomer groups are connected by a common platform. Eachsideofamulti-sidedmarkettypicallygives risetoexternalitieswhichimpacttheotherside,andthiscanaffectthewayinwhichfirmssettheirpricingstructures.

A two- or multi-sided platform is a firm which facilitatestransactions between different types of users in amulti-sidedmarket. Suchplatforms typically have the feature that at leastonetypeofuservaluetheplatformmorewhentherearemoreusersofanothertypeusingthesameplatform.Forexample,anewspaper connects readers and advertisers; a hotel bookingwebsiteconnectshotelswithtravellers.Theremaybemorethanonemulti-sidedplatforminaparticularmulti-sidedmarkete.g.multiplenewspapersavailableinaparticularlocation.

MalaysiaCompetitionCommission

Term

E-Fulfilment

FRAND

FTC

G2B

G2C

Geo-blocking

Horizontal agreement

IP

JFTC

KPPU

LCC

MAS

M-commerce

MIF

Most Favoured Nation clause (MFN)

Multi-sided market

Multi-sided platform

MyCC

3 Definition

12

Theutilitythatagivenuserderivesfromthegooddependsuponthenumberofotheruserswhoareinthesame"network"asisheorshe.22

OrganisationforEconomicCo-operationandDevelopment

OfficeofFairTrading(UK,nowknownastheCMA)

Across-channelbusinessmodelthatcompaniesusetoenhancecustomerexperience.

A platform whereby buyers and sellers are connected, andtransactionsareprocessed(e.g.Amazonmarketplace,eBay).

OnlineTravelAgent

Apubliclytradedcompanyfocusedononlyoneindustryorproduct.

PersonalComputer

PatentCooperationTreaty

A service enabling consumers to compare between differentprovidersofagoodorservice.Usersaretypicallyabletofilterorrankofferingsbasedoncriteriasuchasprice,availabilityofcertainfeatures,or reviewscores.Userscan followa link topurchaseagoodorservicefromthewebsiteoftheirselectedprovider.

Aformofverticalrestraintbroadlydefinedasanyrestrictiononthepricethatresellerscansellaproductat.

RecommendedRetailPrice

Averticalrestraintwherebyafirmonlyallowssomeretailerswhoadheretocertaincriteriatosellitsproducts.

SubstantialLesseningofCompetition

A non-subsidiary, independentfirmwhich employs fewer than agivennumberofemployees(thenumbervariesacrosscountries).

SmallbutSignificantNon-transitoryIncreaseinPrice

TreatyontheFunctioningoftheEuropeanUnion

Trade-RelatedAspectsofIntellectualPropertyRights

UnitedNationsConferenceonTradeandDevelopment

Vertical Agreement Block Exemption Regulation, also known asVBER(VerticalBlockExemptionRegulation)

Verticalagreements,alsoknownasverticalrestraints,arebroadlydefined as instances of coordination between firms at differentstagesofthesupplychainthatrestrictorlimitinsomewayoneofthefirms’activityinthemarket.Mostcommonly,verticalrestraintsimposerestrictionsonretailerssellingamanufacturer’sproduct.

WorldIntellectualPropertyOrganization

WorldTradeOrganization

Term

Network effects

OECD

OFT

Omni-channel strategy

Online marketplace

OTA

Pure-play

PC

PCT

Price Comparison Website (PCW)

Resale Price Maintenance (RPM)

RRP

Selective distribution

SLC

Small and Medium-sized Enterprises (SMEs)

SSNIP

TFEU

TRIPS

UNCTAD

VABER

Vertical restraint

WIPO

WTO

22 Katz, M. and Shapiro, C. (1985); page 424.

3 Definition

13

Introduction

14

23 http://asean.org/asean-economic-community/.

1.1. The rapid emergence and growth of E-commerce markets has brought significant benefits toconsumers and businesses worldwide. Consumers benefit from increased price transparency,reducedsearchcostsandaccesstoagreatervarietyofgoodsandservices.Firmsbenefitfromaccesstonewmarkets,reducedbarrierstoentry,andoperationalcostsavings.

Motivation for handbook

1.2. TherapidchangeinthecharacteristicsandcompetitivedynamicsofsomemarketsasaresultofthegrowthofE-commercehasraisedanumberofchallengesforcompetitionauthorities.Theneedtodealwithcasesinvolvingnewerissues,suchasonlinepriceparityagreements,hasledcompetitionauthoritiesaroundtheworldtoquestionwhetherexistingcompetitionpolicyandlawareabletodealwithantitrustissuesarisinginE-commercemarkets.Inthecontextofthisdebate,theCompetitionCommission of Singapore (CCS) commissioned PricewaterhouseCoopers (PwC) to assist them inpreparingahandbookforcompetitionauthoritieswithinASEANMemberStates(AMS).ThishandbookaimstoincreasetheunderstandingofthecurrentlevelofdevelopmentofE-commerceinASEAN,andofthechallengesemergingforcompetitionauthoritiesintheregion.Thishandbookalsoaimstoenhanceauthorities’understandingofhowbesttorespondtoanysuchchallengeswhentheyariseso that any anti-competitive behaviour can be identified and addressed appropriately,whilst stillpromotingthedevelopmentofE-commerceforthebenefitofconsumersandbusinesses.

Supporting materials

1.3. ThishandbookshouldbeusedinconjunctionwiththeguidelinesandstrategiessetoutbyASEANtopromotethesustainablegrowthofE-commerceintheregionoverthecomingyears,notablytheASEANEconomicCommunityBlueprint2025,23theASEANCompetitionActionPlan(2016-2025)24 and theASEANICTMasterplan2020.25

Research sources

1.4. Thishandbookdrawsuponthe latestdevelopments inthedebateonE-commerce incompetitionlawand economics, aswell as case law from jurisdictions around theworld, and insights fromacomprehensivequestionnaireonE-commerceinASEANdesignedforthepurposeofthishandbook.ThecompetitionauthoritiesofSingapore,Malaysia,thePhilippines,VietnamandIndonesiacompletedthisquestionnaireinApril2017.InterviewswithexpertsfromacrossthePwCnetworkspecialisinginindustriesdisruptedbyE-commercehavefurtherinformedthecontentsofthishandbook.

Content of handbook

1.5. It is importanttonotethatcompetitionpolicyonE-commerce isanareaofcurrentdebateamongcompetitionauthoritiesandpractitioners.Someoftheemergingchallengesarenotyetfullyunderstood,andcontrastingpositionshavesometimesbeentakenbycompetitionauthoritiesindifferentjurisdictions.Thishandbookaimstosummarisethelatestdevelopmentsbasedoncurrentliterature,however,giventhedynamicnatureofthissubject,itshouldnotbeseenaspresentinganultimatesetofprinciples,butprovidingguidancebasedoncurrentunderstandingandexperienceinstead.

Structure of handbook

1.6. Thishandbookconsistsofthreemainparts: a. Part A introduces E-commerce and its value chain, and presents the current E-commerce landscape inASEAN,withaparticularfocusonthebarriersfacedbybusinesses inE-commerce marketswithinASEAN; b. PartBoutlinesthechallengesfacedbycompetitionauthoritiesinE-commercemarkets,andprovides recommendationsonhowbesttorespondtothesechallengesshouldtheneedarise.Examples ofrealcasesfromdifferentjurisdictionsaroundtheworldarepresentedthroughouttoillustrate theissuesdiscussed.AcompetitioncompliancechecklistforbusinessesengagedinE-commerce inASEANisalsoprovided;and c. Part C considers the advocacy role of competition authorities, looking at regulatory and legal barrierstoE-commerce inASEANsuchas intellectualproperty (IP) rights.Theextenttowhich thesebarriersareimpedimentstoasingledigitalmarketinASEANisalsooutlined.PartCconcludes bypresentingrecommendationsforcompetitionauthoritiesinsupportinggovernmentbodiesto evaluatetheimpactoftheirpoliciesoncompetitioninE-commercemarkets.

Introduction01

24 http://www.asean-competition.org/read-publication-asean-competition-action-plan-acap-2016-2025.

25 http://www.asean.org/storage/images/2015/November/ICT/15b%20--%20AIM%202020_Publication_Final.pdf.

15

Overview of E-commerce in ASEAN

PartA:

16

26 ADBI (2016), page 1.

27 Statcounter (2016).

IntroductiontoE-commerce and itsvaluechain

02

2.1. Definition of E-commerce

2.1.1. Therearevariousdefinitionsofelectroniccommerce,orE-commerce.Themostwidelyuseddefinitionis the sale and purchase of goods and services through electronic networks and the internet,encompassingabroad rangeofcommercialactivity.26 It is importanttonotethatthedefinitionofE-commerce in this handbook includesmobile commerce (M-Commerce) which is the sale andpurchase of goods and services usingmobile (smart) phones. This is an important considerationindevelopingcountriesasthegrowthinsmartphoneusageisoutstrippingaccesstoconventionalcomputers/laptops.27ThetermE-commercealsocoversactivitiesthroughouttheentirevaluechainofthetransactionprocess,andincludesactivitiessuchasthedeliveryofthegoodtotheconsumer’spreferredlocation.

2.2. Overview of the business models associated with E-commerce

2.2.1. ThereareavarietyofbusinessmodelsthatfallunderthebroadbannerofE-commerce.Table1providesasummaryofthesedifferentmodels.Thesearereferredtothroughoutthehandbook.

Table 1: E-commerce business models

Consumer of good or service

Source:PwCAnalysis.

Consumer Business

Consumer-to-Consumer (C2C) e.g.eBay,Carousell

Business-to-Consumer (B2C) e.g.retailertoendconsumer

(Aliexpress,Lazada)

Government-to-Consumer (G2C) e.g.governmentwebsite (www.eCitizen.gov.sg)

Consumer-to-Business (C2B) e.g.freelancerstobusinesses (Freelancing.my,Upwork)

Business-to-Business (B2B) e.g.manufacturertowholesaleror wholesalertoretailer(Alibaba)

Government-to-Business (G2B) e.g.eProcurement

(www.philgeps.gov.ph)

Consumer

Business

Government

Prov

ider

of g

ood

or s

ervi

ce

17

2.2.2. Definitions of the businessmodels listed above are providedbelow:

a. B2B: describestransactionsthatexistbetweenbusinesses, suchasone involvingamanufacturerandwholesaler,ora wholesalerandaretailer;28

b. B2C: refers to transactions that are from a business to a consumer. Businesses might exclusively trade with consumersthroughelectronicmeans,conductsalesthrough traditional physical brick-and-mortar stores or sell both onlineandinphysicalstores;

c. C2C: referstocommercialtransactionsbetweenconsumers through a third party (i.e. an online platform provider). An auction,wheremultiple consumers can bid for the same productorservice,isacommonmethodusedtocompletea transaction in this instance.Third party providers, such as eBay,benefitbychargingaflatfeeoracommissiononthe purchaseprice;29

d. C2B: refers to commercial transactionswhere consumers (individuals)offerproductsandservicestobusinesses.The simplestexampleofthisistheemerginggigeconomywhere potentialemployeesoffertheir skillsandtimetopotential employers;

e. G2B: refers to commercial transactions between a governmentandtheprivatesector;30 and

f. G2C: refers to commercial transactions between a governmentandaprivateindividual.

Scale of business models

2.2.3. TheB2BandB2Cbusinessmodelsarethetwomostsignificantintermsofmarketvalue.AccordingtoUNCTAD,B2BE-commercemarkets are valued at around US$19.9 trillion globally.31 B2C markets are significantly smaller, totalling US$2.2 trillionglobally.32WhilsttheB2BmarketconstitutesthelargestshareofglobalE-commercemarkets,theB2Csegmentisexpandingquickly,withmostofthefuturegrowthexpectedtocomefromtheAsiaPacificregionasaresultoftherapidlyexpandingmiddleclassintheregion.33

Emerging business models

2.2.4. Recently, new generations of businessmodels have emerged,including brokerage systems that have increased the numberof tierswithinthedifferentbusinessmodelsdescribedabove.Onetypeofbrokeragesystem isanaggregatorthatdisplaysarangeofrelatedcontent,suchasRakutenwhichsellsavarietyofproducts,fromfashiontoelectronics.SuchanaggregatorbasedbusinessmodelisclassifiedashavingathreetierarchitectureastheplatformisintermediatingthemoreconventionalB2Cmodel,and,assuch,extendingthearchitecturetoaB2B2Cmodel.

28 Investopedia (2017).

29 World Applied Programming (2011), page 102.

30 World Applied Programming (2011), page 102.

31 UNCTAD (2016c).

32 UNCTAD (2016c).

33 UNCTAD (2015), page 12.

B2B E-commerce markets are valued

at around

US$19.9

B2C E-commerce markets are valued

at around

US$2.2trillion

trillion

18

2.2.5. Somebrokerage systems are a little harder to classify.An example isAirbnb. It could be arguedthat Airbnb is a C2C model with a facilitated platform in the middle that takes a commission.However,hotelsandsmallbusinessesalsousetheAirbnbplatform,thereforethemarketcouldalsobeclassifiedasB2B2C.Theadvantageofintermediariestoconsumersisthattheyprovideawidernumberofmarketofferings,allowingconsumerstooptimisetheirsearchtime.Independentprovidersalsobenefitfromtheplatform’smarketreachandthepoweroftheplatformbrand(suchasAirbnb)whichhelpstofostertrustbetweentheproviderandthefinalconsumer.

2.2.6. BothB2B2CandB2B2Barebasedontheideaofautomation.Thetheoryisthatinefficienciesintheprevioustwo-tierarchitecturecanbeovercomebyreplacingtheprocessofmanuallyselectingindividualpreferences (B2C)withanalgorithmthat automatically comparespricesandproduct informationacrossvariouswebsites.B2B2Capplicationsarecommoninthetravelandaccommodationsectors,(e.g.Skyscanner,ExpediaandTrivago).Businessesinthecentreofthisthree-tierarchitectureareoftenreferredtoas‘platforms’operatingin‘two-ormulti-sidedmarkets’.Thesetermsareusedthroughoutthishandbook,andarediscussedindetailinPartB,consideringthecompetitionchallengesemergingfromthesemarketstructures.

2.3. Overview of the value chain

2.3.1. Eachbusinessmodeldescribedabovehasaspecificvaluechain(i.e.theend-to-endprocessfromwherethetransactioncommencestowhereitfinishes).ThekeyelementsoftheB2Cvaluechainare:

a.ProductSourcing;

b.CustomerInterface;

c.Delivery;and

d.Aftersalesservice.

2.3.2. Figure1belowdepictsthevaluechainfromstarttofinishwithintheB2Cbusinessmodel.

Figure 1: B2C E-commerce value chain

Source:PwCAnalysis.

1.ProductSourcing

2.ConsumerInterface

4.After Sales

Service3.

Delivery

Information Agreement Transaction

Select product supplier

Place and confirm order

Order delivered to warehouse (if needed)

Compare prices between suppliers

Visit review websites

Consumer decides on the supplier based on all information

Visit the supplier website

Select the item to purchase

Indicate preferences

Agree with the terms and conditions of the purchaseBroker /

Intermediaries such as comparison websites (e.g. Skyscanner, Expedia, Trivago)

Select mode of delivery (e.g. self-collections, e-delivery, physical delivery) Proceed with the payment

Choose mode of payment (e.g. credit card, cash on delivery, bank transfer) Complete Transaction

Alternative payment providers (e.g. Visa, Mastercard, Apple pay, Paypal)

Depending on the indicated mode of delivery, goods are either:

- Collected from the supplier;

- Delivered in digital form; or

- Delivered to a specified location.

Logistics or delivery companies (e.g. DHL, Fedex)

Ware-housing

ShippingLogistic

LastMile

Delivery

Return of products

Handle queries and complaints

Consumer

Business

KEY

Key

Act

ivit

ies

NOTE: In a 3 tier model the intermediate platform would perform the end to end consumer interface.

19

2.3.3. Thefollowingsub-sectionsdiscussthisprocessingreaterdetail,workingfromlefttorightofFigure1.

Product sourcing

2.3.4. An E-commerce business, just like a traditional brick-and-mortar business, must initially sourceitsproducts.Partofthisprocess includesmanaging itssupplychain intermsof inboundlogisticsand inventories.E-commercehas,however,presentednewopportunitiesforproductsourcing,ascompanies can potentially avoidwarehousing and storage costs by acting purely as the conduitbetween themanufacturer and the final customer. Assuming the E-commerce retailer is tradingphysicalgoods,thereisanopportunityforthefirmtoenhanceefficiencybyplacinganorderwiththemanufacturertobedeliveredonlywhentheproductneedstobeshippedtothefinalcustomer,thussavingstorageandwarehousingrequirements.Alternatively,tomaximiseefficiency,anE-commerceretailermayallowthemanufacturertousetheirown logisticscapabilitiestodeliverdirecttothecustomer,thereforeminimisinganystorageorhandlingtimebytheE-commerceretailer.

Customer Interface

2.3.5. The customer interface links the consumer with the seller’s products and services. Customerscanaccess informationonwhat isbeing traded, choosetheir selected itemsandcomplete theirtransaction.Thecustomerinterfacemaytaketheformoftwointegratedsystemsbetweenbusinesses(inthecaseofB2Btransactions,businessescandirectlylinktheirsystemstocommunicatewithoneanothersothattheydonotneedtouseapublicplatform),oralternativelyathirdparty interface,suchasawebsiteorappthatcustomerscandirectlyaccess,canbeused(inthecaseofB2C,wherepubliclyavailableinterfacesareused).

2.3.6. Businessesmaydecide to develop their ownwebsites to sell direct to customers, or sell via athirdpartyplatform,suchasAmazonorQoo10.Transactionsthroughthirdpartyplatformsare(asmentionedabove)referredtoasB2B2CandB2B2B,reflectingthefactthattheplatformservesasalinkbetweenthecustomerandtheseller(e.g.Zalora).

2.3.7. The decision to use a third party platform as the customer interface presents challenges andopportunities.It ischeaper(atleastintheshortterm)comparedtocreatingabespokeplatformanditislikelytoprovideaccesstoawidercustomerbase.Consumersarealsomorelikelytotrustan establishedplatformas opposed to the newwebsite of an independent retailer.Third partyplatformscan therefore reducebarriers toentry forbusinesses (whicharediscussed in greaterdetail inSection4), potentially leading to increasedcompetition.However, onceabusinesshasestablished itselfonathirdpartyplatform, itmaybedifficulttotradeoutsideofthatplatform,whichcouldlimitfuturegrowth.

20

Delivery

2.3.8. Delivery remainsoneof thekeychallengesforE-commerce.Onlineplatformscanenableaccesstoglobalmarkets,butthephysicalchallengeofdeliveringproductstofinalcustomersstillremains.Deliverycan involve interactionsbetweendifferenttypesoffirms, suchas logisticscompaniesorpostalservices.Deliveryalsorequiresreliableinfrastructuretobeinplace.Thecostsassociatedwithdeliveryandthetimeittakesforconsumerstoreceivegoodspresentsakeychallengeforbusinessestoovercomeastheycompeteoncustomerexperience.

2.3.9. ThegrowthofE-fulfilmentservicesinrecentyearshasenabledE-commercecompaniestodeliveramorecompellingend-to-endcustomerexperience.E-fulfilmentisdefinedasthepeople,processesandtechnologyrequiredtodeliveranonlineordertoacustomer.34Dedicatedcompanieshavebeensetuptoservicethisneed,offeringorganisationsparticipating inE-commercetheopportunitytooutsourcethiscriticalpartoftheirvaluechain.

2.3.10. Within thedelivery phase, three sub-stages form theE-fulfilmentvalue chain.This is shown inFigure2below.

Figure 2: E-fulfilment Value Chain

34 nchannel (2016).

Warehousing ShippingLogistics

Last MileDelivery

Delivery

Inventorymanagement

Qualitycontrol

Track and trace

Pickandpack

Shippinglabel

Pickup

Sorting

Customs

Transportation

Track and trace

Delivery

Returnsmanagement

Track and trace

Deliveryandpaymentflexibility

Source:ATKearney(2016),PwCAnalysis.

21

2.3.11. There are three key areas where E-fulfilment services could provide substantial advantages forE-commerceplayers:35

a. End-to-end capabilities:Thiscouldincludeflexiblepick-uptimings,packingsolutions,inventory management,andfulfilmentsolutions(definedastheprocessofreceiving,packagingandshipping orders for goods.36). These elements fit within the warehousing and shipping section of the E-fulfilmentvaluechain;and

b. Enabling cross-border E-commerce transactions: E-fulfilment can help small andmedium sizedenterprises (SMEs)toextendtheir reach intonewmarkets.Byoutsourcingtoafocussed logisticsorE-fulfilmentcompany,SMEsaremorelikelytohaveaccesstointernationalpartnerships andnetworksthatcanassistwithcross-bordertransactionsanddeliveries.

c. Last mile delivery:Theintroductionofautomatedlockershasledtoprogresswithinthelastmile deliveryphase.In2016,SingaporePostintroducedSingapore’sfirstislandwideopenparcellocker service.Thisallowsretailersandconsumersto renta lockertodeliverandcollecttheirgoods securely.Thisprocesscantakeplaceatanytimeduringtheday37overcomingtheissueofhaving tohavesomeoneavailabletocollectagood.Italsomeansreturningaproductiseasierasgoods cansimplybeleftinthelockerreadyforcollection.Theintroductionof‘FederatedLockers’has alsobegunwithinSingapore,whichhastheaimofcreatinganationwidecommonparcellocker system.Thiswill be thefirst of its kind in theworld.38Theprojectwill involve the large scale deploymentofparcellockerswithinSingapore,aimingtoeasethelastmiledeliverychallenges currentlybeingfaced.Itisacentralisedsystemthatcanbeusedbyalllogisticscompanies,rather thaneachcompanyhavingtosetuplockersthemselves.Thetheorybehindthismethodisthatit willbecheaperforbusinessesandconsumerstohaveoneproviderforlockersystems.

After sales service

2.3.12. As well as competing on price, E-commerce firms compete on customer service by providingadditionalofferingssuchasonlinecustomerqueryresolutionandfreereturnofproducts.

2.3.13. Thereturnofproductsisoneofthebiggestchallengesforonlineretailers.Manycompaniesofferafreereturnsservicetoreducetheburdenonthecustomer,whilstothersstillchargeafeetocovertheassociatedcosts.

35 AT Kearney (2016).

36 Entrepreneur (2016).

37 Singapore Post (2016).

38 Prime Minister’s Office Singapore. See: http://www.pmo.gov.sg/newsroom/dpm-tharman-shammugaratnam-opening-ceremony-singapore-post-regional-ecommerce-logistics.

22

03 OverviewoftheE-commercelandscapeinASEAN,thecurrentstateofE-commercedevelopmentineachoftheAMSanditsgrowthpotential

3.1. Introduction

3.1.1. ThissectionprovidesanoverviewofthecurrentE-commercelandscapeinASEAN,andisstructuredasfollows:

a. Firstly,thekeycurrentbusinessmodelsinASEANareoutlined,aswellastheexpectedgrowth trajectorieswithintheregion;

b. Secondly,thecurrentstateofE-commerceinallAMSisexploredingreaterdetail;and

c. Finally,theimpactofE-commerceonthevaluechaininASEANispresented,beforelookingmore closelyatthechangesinfiveheavilydisruptedindustries.

3.2. Overview of the current retail E-commerce markets in ASEAN and their likely evolution

Current scale of E-commerce markets in ASEAN

3.2.1. Since the opening of the internet for commercial use in the early 1990s, E-commerce has grownsignificantlybothintermsofrevenueandthenumberofmarketswhereitisoperational.InthesixlargesteconomieswithinASEAN(Indonesia,Malaysia,thePhilippines,Singapore,ThailandandVietnam,hereafterreferredtoastheASEAN6),retailE-commercehasatotalmarketsizeofUS$7billion.39

3.2.2. Table2belowshowsthemarketsizewithintheseAMS.

Table 2: The retail E-commerce markets in ASEAN (2015)

ASEAN member state

BruneiDarussalamCambodiaIndonesiaLaoPDRMalaysiaMyanmarThePhilippinesSingaporeThailandVietnamASEAN total

Source:ATKearney(2015);Statista(2017u,2017v).Note:Adash(-)indicatesunavailabledata.

Market size (US$bn)40

--1.3-1.3-1.01.70.90.8

Population (millions)41

0.415.6257.66.830.353.9100.75.568.091.7630.5

Market size per capita (US$ per capita)

--5.05-42.90-9.93309.0913.248.72

Internet users per capita (%)42

71.219.022.018.271.121.840.782.139.352.734.1

Internet users (millions)43

0.33.056.61.221.511.841.04.526.748.3214.9

39 AT Kearney (2015), page 2.


41 Statista (2017u).

42 Statista (2017v).

43 Ibid.

23

3.2.3. Table2showsthatin2015themarketsizepercapitawashighestinSingapore,andlowestinIndonesia.ThisdemonstratesthatE-commercehaspenetratedfurtherintoSingaporethanotherAMS,butthepotentialforE-commercegrowthisgreatestinIndonesia,especiallygivenitssignificantpopulation.Internetuserspercapita is alsohighest inSingapore, and lowest in IndonesiaandThailand, thusshowingacorrelationbetweeninternetusageandthesizeoftheE-commercemarketpercapita.

3.2.4. The level of market penetration of E-commerce varies significantly across nations. To assessE-commerceadoptionacrossASEAN,internetretailsalesasashareofbrick-and-mortarbasedretailsalescanbeconsidered.UNCTADdatashowsthatSingapore,Malaysia,Thailand,Indonesia,VietnamandthePhilippinesallcurrentlygeneratelessthan4%oftheirretailsalesonline.44ThecountrywiththehighestproportionofretailsalesfromE-commercewithinthestudyof42selectedcountriesisTheRepublicofKorea,at16%.TheequivalentfigureinChinais7%.

3.2.5. Table3,highlightskeycharacteristicsofE-commercemarketsinASEAN,outliningthecurrentstateof the sector for theASEAN6. Each of these countries is then discussed in greater detail in thefollowingsub-sections.FurtherdetailsofthegovernmentinitiativestoenhanceE-commercemarketsare provided in Annex 2.

Table 3: The retail E-commerce market in the ASEAN6

44 UNCTAD (2015), page 21.

45 Statista (2017c).

46 DBS (2015).

47 DBS (2015).

48 Digital News Asia (2016). See https://www.digitalnewsasia.com/digital-economy/indonesia-announces-e-commerce-roadmap-part-jokowi%E2%80%99s-newest-economic-reform-package.

ASEAN member state

Indonesia

Malaysia

ThePhilippines

Singapore

Thailand

Vietnam

Source:Various(seefootnotes).

Market size (US$bn), 2015

1.3

1.3

1.0

1.7

0.9

0.8

Market Size per cap (US$), 2015 5.05

42.90

9.93

309.09

13.24

8.72

Expected annual growth rate 2017-2021 (%) 20.145

23.249

17.353

11.257

15.961

16.565

Key sectors impacted by E-Commerce Entertainment media(books,video,games),consumerelectronics,fashion,travel46

Travel,entertainmentmedia,consumerelectronics,fashion50

Consumerelectronics,food&grocery54

Travel,fashionandbeauty,entertainment &lifestyle,ITandelectronics,insurance58

Travel,fashion,electronics,media62

Fashion,electronics,media,food,homeappliances66

Main impediments to growth Cyber-security,productreliability47

Cybersecurity,consumerspreferenceforbrick-and-mortar shopping51

Credit card penetration55

Brick-and-mortar shoppingculture59

Cybersecurity,poorlogistics,infrastructure63

Productreliability,cybersecurity67,poorlogistics68

Government initiatives E-commerce Roadmap48

NationalE-commerce StrategicRoadmap52

E-commerce Roadmap56

RetailIndustryTransformationMap,60

SMEsGoDigitalProgramme,initiativestospeedupcustomsclearance

Latest economic growthplan,Thailand4.0,includesE-commerce initiatives64

E-commerceGrowthPlan2016-2069

49 Statista (2017e).

50 DBS (2015).

51 DBS (2015).

52 Malaysian Digital Economy Corporation, (2016). See https://www.mdec.my/digital-innovation-ecosystem/ecommerce/nesr.

53 Statista (2017d).

54 DBS (2015).

55 DBS (2015).

56 Philippine E-commerce roadmap, (2017). See http://www.dti.gov.ph/roadmap.

57 Statista (2017b).

58 DotEcon (2015), page 24.

59 DBS (2015).

24

60 SPRING Singapore (2016). See https://www.spring.gov.sg/NewsEvents/PR/Pages/Retail-Industry-Transformation-Map-to-Drive-E-Commerce--Omni-Channel-Formats-to-Enhance-Growth-and-Competitiveness-20160915.aspx.

61 Statista (2017a).

62 DBS (2015).

63 DBS (2015).

64 Asia Today (2017). See http://www.asiatoday.com/pressrelease/thailand-drafts-roadmap-digital-economy-0

65 Statista (2017f ).

66 Qandme.net. (2016).

67 Vietnam Net (2016). See http://english.vietnamnet.vn/fms/business/162619/vn-government-sets-targets-for-e-commerce-in-2016-2020.html.

68 Qandme.net (2016).

69 DBS (2015).

70 Statista (2017c).

71 DBS (2015).

72 DBS (2015).

73 International Business Times (2014).

74 Statista (2017e).

75 DBS (2015).


77 World Bank (2016).

78 DBS (2015), page 20-22.

79 DBS (2015), page 12.

Indonesia

3.2.6. Thegrowthprojection for the IndonesianE-commercesectoris a compound annual growth rate (CAGR) of 20.1% (2017-21).70The largest sectorswithinE-commerce in Indonesia areentertainmentmedia(e.g.books,videosandgames),consumerelectronics, fashion and travel.71 Logistical infrastructure andinternet penetration are relatively weak compared to othermembers of the ASEAN6, making it harder for E-commerceretailerstoreachconsumers.72Factorsthataresupportingthedevelopment of E-commerce include a growingmiddle class(expectedtobe140millionby2020,from74millionin2014)73,andayoungpopulation(70%ofthepopulationisundertheageof40).

Malaysia

3.2.7. ThecurrentgrowthprojectionfortheMalaysianE-commercesector is a 23.2% CAGR from 2017-21.74 The largest sectorswithin E-commerce are travel, followed by entertainmentmedia,consumerelectronicsandfashion.75Keydriversoftheexpected growth in E-commerce markets within Malaysiaincludehighinternetpenetration,at67%,76 higherthanaveragecredit card usage for the region,which is at 9%,77 and goodtransport infrastructure for product sourcing, logistics anddelivery.Despitethesefactors,theonlineretailsegmentisstilllessthan1%oftotal retailsales.78Reasonsforthis includealackoftrust inonlineretailers, intermsofproductreliabilityandsafetyofpaymentmechanisms,andpoor local logisticsinfrastructure.79 Questionnaire responses indicate that anumber of successful platforms have emerged in MalaysianE-commercemarkets,namelyLazada,ZaloraandLelong.

The largest E-commerce sector in Malaysia is

TRAVEL

The largest sectors within E-commerce in Indonesia are

ENTERTAINMENT MEDIA

CONSUMER ELECTRONICS

FASHION

TRAVEL

25

The Philippines

3.2.8. ThecurrentgrowthprojectionforthePhilippinesE-commercesectorisaCAGRof17.3%between2017and2021.80Accordingto the Singapore Post, the largest online retail sectorwithinthePhilippinesisconsumerelectronics,followedbyfoodandgroceries.81 Euromonitor (2017), however, found that mediaproducts,suchas in-gamepurchases, isthe largestsector.82 Setagainstthis,however,accordingtothePhilippinesRetailersAssociation, only approximately 3%of the total retailmarketisbasedonline.83Onedriverofthe lowlevelofE-commerceadoptioninthecountryisthesmallproportionofpeoplewhoownacreditcard;specifically,thereareonly2.5millioncreditcardsinapopulationofaround100million.84ThefragmentedgeographyofthePhilippinesalsomakesitachallengetohavereliable courier services, particularly serving the more ruralareas.85

3.2.9. LikemanycountriesinASEAN,thePhilippineshasmanycitizensworkingasmigrantsoverseas.Thishascreatedopportunitiesforthesecitizenstobuydomesticproductsonlinewhilstoverseastodelivertofamilyandfriendsstillresiding inthePhilippines.Onlinestorestargetingsuchcustomershaveemerged.86IslandRose, as an example, is an online flower retailer that allowsconsumers from all over the world to purchase gifts to bedeliveredwithinthePhilippines.

Singapore

3.2.10. Singapore has a high online penetration rate (78%) and apopulationwhich spends a large amount of timeonline (5.3hoursadaythroughdesktop,and2.4hoursadaythroughmobiledevices).87Also,theexistingexportandimportinfrastructureisstrong.Singaporehascomparativelylowmarketentrybarrierscompared tootherAMS, evidenced through its rating as thesecondmostfreeeconomyintheworldinthe2014indexofEconomicFreedom,behindonlyHongKong.88

3.2.11. The growth projection for the Singaporean E-commercesectorisaCAGRof11.2%(2017-21).89ThelargestsectorswithinE-commerce are travel, fashion and beauty, entertainmentand lifestyle, IT and electronics and general insurance.90 Keydriversofgrowth inthemarketarehigh internetpenetrationandsmartphoneadoption,strongfinancialinfrastructure,andgoodlogisticalfacilities.91

3.2.12. Overtheperiodfrom2009to2014,B2ConlinebusinesssuchasReebonz,Qoo10,Luxola,Groupon,Deal.com.sg,NoQStore,Bellabox, VanityTrove, Kwerkee, Zalora, Food Panda, Taobao,HipVan,Omigo,RakutenandLazadahaveenteredthemarket.There have also been new entrants in terms of C2C onlinebusinesses, for example Clozette and Carousell, as well asplatformslikeUberandGrab(GraboffersbothC2CandB2CservicesthroughGrabCar,GrabHitchandGrabTaxi).

80 Statista (2017d).

81 DBS (2015).

82 Euromonitor (2017).

83 DBS (2015), page 22.

84 DBS (2015), pages 22-24.

85 Philippine Competition Commission (2017).

The largest online retail sector in the Philippines is

CONSUMER ELECTRONICS

The largest E-commerce sectors in Singapore are

TRAVEL

FASHION AND BEAUTY

ENTERTAINMENT AND LIFE STYLE

IT AND ELECTRONICS

GENERAL INSURANCE

86 DBS (2015), pages 22-24.

87 Singapore Post (2014), page 14.

88 Ibid.

89 Statista (2017b).

90 DBS (2015).

91 DBS (2015), page 13.

26

92 IMDA (2014).

93 DBS (2015), page 13.

94 Ibid.

95 Singapore Post (2015)

96 Business Times (2017).

3.2.13. DespiteSingaporepossessingtherightenablersforE-commercemarkets to flourish, retail E-commerce adoption rates arenotashighasJapanorSouthKorea.Thismaybeduetotheconvenienceofshoppingmalls,andthecultureofshoppingintraditionalbrick-and-mortaroutlets.AsurveybyIMDArevealedthat one of the top reasons for not shopping online was a“preferencetoshopinpersonordealpersonallywithaserviceprovider”.92 Questionnaire responses also highlighted this is abarrierinSingapore.

3.2.14. Online retail adoption could, however, increase due to thefollowingreasons:

a. Recent labour policy changes have reduced the supply of labour in themarket, prompting retailers to look again at E-commerce, as trading online tends to be less labour intensivethansellingviabrick-and-mortarstores;93

b. Recently,therehasbeenanemergenceofstrongE-commerce playersintheregion,suchasAlibaba,whichhasledtocost reductionsandincreasedqualityofserviceforcustomers;94

c. The new federated locker system will improve last mile delivery. In addition, the introduction of a ‘shopping mall’ bySingPost,whichoffersacompletesuiteofE-commerce logisticssolutions,willalsodriveonlineretailsales.Shopping throughonlineretailerswillincludein-shoponlineordering andflexibilityindeliveryandpickuptimings;95 and

d. Changi Airport’s E-commerce AirHub facility, which is designed to speed up the processing of parcels flown in fromabroadwilldecreasethetimetakenforonlinepurchases to be delivered to final customers. This will be done by increasing mail-sorting capability by three times and reducing processing time by half,96 and driven by the introductionof a fully automatedmail-sorting systemthat will increase mailbag processing capability from 500 per hourtomorethan1,800perhour.Thefacilityisexpectedto bereadyduringthesecondhalfof2017.

Thailand

3.2.15. The growth projection for Thailand’s E-commerce sectoris a CAGR of 15.9% (2017-21).97 The largest sectors withinE-commercearetravel,fashion,electronics&media.98Thekeyclick-and-mortar99 players includeTesco Lotus,7-Eleven andCP-ALL.

3.2.16. The challenges within Thailand are similar to those ofIndonesia,mainly surrounding a lack of trust. 62% of onlineshoppers inthecountryarereluctanttogiveouttheircreditcard informationonline.100Other issues includethehighcostof E-payment and logistics, expensive telecommunicationsand internet access, and a lack of capital to assist start-upcompanies.101

The largest E-commerce sectors in Thailand are

TRAVEL, FASHION, ELECTRONICS &

MEDIA

Singapore has a high online penetration rate of

78%

Through mobile devices: 2.4 hours a day

Through desktop: 5.3 hours a day

Amount of time online

97 Statista (2017a).

98 DBS (2015).

99 Firms that conducts business online and also through ‘traditional’ offline brick-and-mortar channels.

100 The Paypers, (2017).

101 Electronic Development Transactions Agency, (2017).

27

Vietnam

3.2.17. ThegrowthprojectionforVietnam’sE-commercesector isaCAGRof16.5%(2017-21).ThelargestsectorswithinE-commercearefashion,electronics&media,food&homeappliances.102

3.2.18. Asof2016therewere45millioninternetusersinthecountry,andover34millionsmartphonedevicessold.Themostpopularmethodof payment for E-commerce transactions is cash ondelivery,followedbybanktransferandpaymentcard.103

3.2.19. Thetop5B2CVietnamesewebsitesaccordingtoquestionnaireresponsesareThegioididong.com.vn;Nguyenkim.com;Fptshop.com.vn;Dienmayxanh.com;andVienthongA.vn.Thetop5C2CVietnamese websites are Vatgia.com; Chotot.com; 5giay.vn;Chodientu.vn;andWebmuaban.vn.

3.2.20.ReasonsforpooradoptionofE-commerceinthecountryincludeconcerns over the quality of products, security worries, highprices,anddeliverycosts.ManyoftheseissuesstemfromthelackoflogisticsinfrastructurewithinthecountryandinefficientE-commerce practices.104 Trust is a big issue for consumersonline, who often prefer to purchase from brick-and-mortarcompanies.105

Brunei Darussalam, Lao PDR, Cambodia, Myanmar

3.2.21. The uptake of E-commerce has increased within BruneiDarussalam in recent years, as access to mobile andinternet capabilities has increased. However, themajority ofE-commerce is limitedtotheaccommodationandtransportbooking sectors. For example, Royal Brunei Airlines has anonline reservation system. Hoteliers also have e-bookingservices.106

3.2.22.A lack of investment in telecommunications infrastructure,andthelowrateofformalbankingandcreditcardusearetworeasonswhyE-commerceisnotwidelyusedinLaoPDR.Whilebroadbandaccessiswidelyavailableinthecapital,Vientiane,country-wideaccesstotheinternetismainlythroughmobiledevices.107

3.2.23.ThereisagrowingnumberofwebsitesbeingsetupinCambodiawhichmainlycaterforasmallnumberofconsumers inthemajorcitieswithbetteraccesstothe internet. Impedimentstogrowthareinadequateinfrastructureandlowlevelsofcreditcard penetration.108

3.2.24.As of 2015,Myanmar’s internet penetrationwas around 22%ofthepopulation,afigurewhichhasgrownfromunder3%in2013.109Internetaccesshas,however,historicallybeenunreliableandslow.Therehavebeenattemptstoestablishapresenceonline within the real estate and automotive industries inrecentyears.Thegrowthinsmartphonepenetration(at45%as of November 2015) bodes well for further E-commercedevelopmentgoingforward.110

102 Qandme.net. (2016).

103 Foreign Trade University (2017).

104 Qandme.net (2016).

105 Vietnam Net (2016).

106 Export.gov (2016a).

The largest E-commerce sectors in Brunei Darussalam are

The largest E-commerce sectors in Vietnam are

FASHION

ELECTRONICS AND MEDIA

FOOD

HOME APPLIANCES

ACCOMMODATION & TRANSPORT

BOOKING

107 Export.gov (2016b).

108 Export.gov (2016c).

109 World Bank (2017).

110 Export.gov (2016d).

Myanmar's internet penetration was around

22%of the population as of 2015

28

111 Nielsen (2014). Middle class defined as people that “have the financial means to make purchase decisions based on their level of disposable income.”

112 DBS (2015), pages 12-14.

113 ADBI (2016), page 2.

114 Competition Commission of Singapore (2015), page 4.


Likely evolution of E-Commerce in ASEAN

3.2.25.Keydriversofgrowth inASEANincludearapidlyexpandingpopulationandarisingmiddleclass(expectedtogrowfrom190millionpeoplewithinSoutheastAsiain2012to400millionby2020).111 Thisgrowthisalsobeingsupportedbyahighpenetrationofsmartphones(seeFigure3below), anincreaseinsalesthroughM-commerce,morepaymentandshippingoptions,andmajorbrandsenteringlocalE-commercemarkets.112

Figure 3: Smartphone penetration in the ASEAN6

3.2.26.ThereareseveralchallengesthatE-commercemarkets inASEANarefacing.These includepoorE-commerceinfrastructure(suchasbankinginfrastructure)andalackofE-commerceregulations.ThesearediscussedingreaterdetailinSection4.

3.3. Impact of E-commerce on the value chain in ASEAN

3.3.1. ThissectionconsiderstheimpactthatE-commercehashadonthevaluechaininAMSincomparisontotraditionalbrick-and-mortarsaleschannels. It looksatthedifferences incoststructures,theavailabilityof information,andthesupplychainand logisticsfunctionsoffirms.Finally,thenewbusinessmodelsthathaveemergedintheE-commercespacewithintheregionareoutlined.

Differences in cost structures

3.3.2. Onlineretailersarenotasphysicallyconstrainedastheirbrick-and-mortarcounterparts.Theycanofferawidervarietyandquantityofgoodswithouttheneedforaphysicalshopfloortoshowcasetheirproductsandservices.113Businessesareabletosaveonbothfixedandvariablecosts,suchasrent,labourandotheroverheadsrelatedtomaintainingaphysicalpresenceonthehighstreet.

3.3.3. Manyofthecostsassociatedwithcross-bordertradearereducedasthephysicalpresencerequiredtotradeisdiminished.Asaresult,morenewandexistingcompaniesareexpandingtheirsalesintonewmarketsandgeographies.114Forexample,Rakuten,aJapanesefirm,hassetuptheirregionalheadquartersinSingaporetoreachotherASEANmarkets.115

Indonesia Malaysia The Phillipines Singapore Thailand Vietnam

Source:Statista(2017g),Statista(2017h),Statista(2017i),Statista(2017j),Statista(2017k),Statista(2017l).

80%

70%

60%

50%

40%

30%

20%

10%

0%2015 2016 2017

29

Availability of information

3.3.4. E-commercehas increasedtheavailabilityof information,bothtoconsumersandbusinessesatallstagesinthevaluechain.Providersindigitalmarketscollectandmakeuseoflargequantitiesof data and information on consumer preferences. This can be used to themutual benefit ofproducers and customers by better meeting consumers’ needs through tailoring products toindividuals’preferences.

3.3.5. Competitors’pricemovementsarealsomorevisible indigitalmarkets.Theavailabilityofonlinealgorithmsusedtoidentifyandrespondtopricemovementsmeanscompaniesareabletoreactautomatically to changes in their competitors’ prices.Companies, such asZalora, havevisibilityofanypricechangestoproductsmadebyonlinecompetitors, andcan react tothesechangesalmostinstantly.PricingalgorithmsarediscussedingreaterdetailinSection8,withafocusontheimplicationsforcompetition.

Supply chain/logistics

3.3.6. A supply chain is defined as an “entire network of entities, directly or indirectly interlinked andinterdependentinservingthesameconsumerorcustomer”.116Thisincludeslogistics,manufacturingandprocurement.ThegrowthofE-commercehasledtochangesinthesupplychain.Shorterdeliverytimesarebeingdemandedbyonlineshoppers,andcompanieswanttodifferentiatethemselvesinthemarket.117

3.3.7. Someretailershaveimplementeda‘drop-shippingsupplychain’,whereanorderispasseddirectlyonto awholesaler/manufacturer, removing the need to have a physicalwarehouse to store theproducts,thereforedecreasingcosts.Cleo-catfashionandBlogshophaveadoptedsuchprocessesinSingapore.118

3.3.8. E-commerce companieshave also started to acquire or develop elements of the supply chain,ratherthanrelyonothercompaniestocompletethesepartsofthecustomerjourney.Thisisoftenusedtogaingreatercontrolovertheentirecustomerexperience.Forexample,JindongMall,hasrecentlybeengivenalicencefor its logisticssubsidiary,allowing itdevelopanin-houselogisticsframeworkratherthanrelyonthirdpartyinfrastructure.119

New business models

3.3.9. E-commerceiscreatingnewcustomer-centricbusinessmodels.Advancementsindataanalyticsallowbetter targetingofproductsandmarketingviathemosteffectivedistributionchannelstoconsumers,whoaredemandingamoreuniqueandefficientcustomerexperience.

3.3.10. PriceComparisonWebsites(PCWs)whichallowconsumerstoeasilycompareandfilterdifferentsuppliersofgoodsorservices,havebecomeprevalentacrossASEAN.PCWsmaketheirmoneyfromadvertising,andalsocommissionfromthecompanythecustomerpurchasesfrom.CompareXpress.cominSingapore,CompareHero.myinMalaysia,andWebsosanhinVietnam,alladoptthisbusinessmodel.InonlinemarketplacessuchasAmazon,actualsalesaremade,whereasonPCWs,consumersaredirectedtoretailers’websites.

3.3.11. TheimpactsofthesenewbusinessmodelsoncompetitionareconsideredindetailinSection7ofthishandbook.

116 BusinessDictionary.com (2017).

117 Nomura (2016).

118 Competition Commission of Singapore (2015), page 11.


30

3.4. Discussion on five industries disrupted by E-commerce within ASEAN

3.4.1. This section considers the changes discussed in Section 3.3 in detail for five industrieswithinASEAN.Theseindustrieshavebeenselectedonthebasisoftheextenttowhichchallengeshavearisenforcompetitionauthorities inE-commercemarkets inthese industriesaroundtheworld(asdiscussedindetail inPartBofthishandbook).Giventhistwo-sidedselectionapproach, it isimportanttonotethatthesearenotthefivelargestE-commercesectorsinASEAN,butthosethathavebeensignificantlydisruptedbyE-commerceandposedchallengestocompetitionauthoritiesaroundtheworld.Thefiveindustriesconsideredinthissectionarelisted(innoparticularorder)inTable4below.

Table 4: Five disrupted industries by E-commerce within ASEAN

3.4.2. Table5,below,providesasummaryofthekeyimpactsfromthedisruptioncausedbytheemergenceand growth of E-commerce in ASEANwithin the five industries discussed. The following sub-sectionsthendiscussthedisruptionineachoftheseindustriesinASEANingreaterdetail.

Table 5: Summary of key impacts from E-commerce in the five disrupted industries within ASEAN

Industry

Accommodationbooking Flightbooking Land transport Cosmeticsandbeautyproducts Fashion

Source:PwCAnalysis.

Key impacts and consequences

Increaseinindependentproviders,increasedpricecompetition,newaudiencesfortraditionalhoteliers.

Increasedpricecompetitiveness,lowerpricedispersion,demiseofindependenttravelagents. Increasedinnovation,reshapingofmarkets,moresensitivepricingstrategies. Increased price competition and competitiononproductselection,demiseofbrick-and-mortarcompanies. Increasedproductchoiceforconsumers,growthinconsumerexpectations.

Examples of significant players in ASEAN

Agoda,Trivago,ExpediaBooking.com,Hilton,Intercontinental,Hyatt,Airbnb

Expedia,Skyscanner,AsiaTravel,FlightCentre,FlightWorld,HelloWorld Uber,ComfortDelGro, GoJek,Grab Luxola,Hermo,Bellabox Zalora,Clozette,VanityTrove, Lelong.my,FashionValet,ASOS,Shopbop.com

Industry

Accommodation booking

Flight booking

Land transport

Cosmetics and beauty products

Fashion

Source:PwCAnalysis.

31

Accommodation booking

3.4.3. Brick-and-mortarcompanieswithinASEANintheaccommodationbookingsector(includingtraditionaltravelagents)haveseenfiercecompetitionfromaggregatorsandothersitessuchasAgoda,TrivagoandBooking.com,whichhavebecomekeyplayersintheindustry.

3.4.4. Whilstmanytravellersintheregion,especiallyoldergroups,stillrelyonbrick-and-mortaroperators,80%oftravellersinSingapore120saythattheygathertheirtravelinformationonline.Ithasbeenarguedthatthesurvivalofbrick-and-mortarprovidersinSingaporeisinjeopardy.121Thisisparticularlytrueforbusinesstravelagentsasitisnoweasierforfirmstomanagetheseservicesin-houseorthroughsingleonlinevendors.

3.4.5. TheriseofPCWsanddigitalplatformsforaccommodationprovidershasledtoalargeincreaseinindependentaccommodationproviderscomingtothemarket,usingdigitalplatformstoreachtheircustomers.Airbnb,asanexample,actsasanintermediaryconnectingaccommodationprovidersandtravellers.Independentprovidersaregiventheopportunitytoimprovetheirbrandsinceconsumersrelyonthereputationoftheplatformtoensuretheirexpectationsaremet,alongwithreviewsthatconsumersandvendorsprovide.OnplatformssuchasAirbnb,Tripping,andHomeAway,themajorityofvendorsareindependentproviders.Onotherplatforms,suchasBooking.comandTrivago,traditionalhotelchainssuchasHilton,IntercontinentalandHyattsellroomvacanciesinordertoreachawiderrangeofcustomers.

3.4.6. TheimpactofthedisruptionbroughtaboutbyE-commerceonthehotelindustryvariesbythesizeofthebusiness.The largest internationalbrandshaveavery loyalconsumerbase, incentivisedbyloyaltyprogrammes,suchasHilton,Intercontinental,andHyatt.ManyofthesebusinesseswerequicktodeveloptheirownB2Cwebsite,which,combinedwiththeir loyalconsumerbase,allowsthemtomanagerealtimedemandandadoptflexiblepricingmodels.Smallerhoteliersaremorelikelytobenefitfromintermediaries’platformstoaccessawiderconsumerbase.

3.4.7. There has also been a shift in theway the largest hotel chains structure themselves to combatthe downward pressure on prices arising from greater competition. Many large hotels now havesister brands allowing them to offer both premium accommodation and budget offerings underdifferentbrandswithinthesamegroupenablingthemtoservetwodistinctcustomersegments.Anexampleofthis isHilton,whichhastheHiltonhotelbrandforpremiumaccommodation,andtheTrubrand,competingwitheconomybrands.Similarly,IntercontinentalHotelsGroupofferspremiumaccommodationthroughitsIntercontinentalbrand,andstandardaccommodationthroughitsHolidayInnhotels.

Flight booking

3.4.8. E-commercehasheavilydisruptedthissectorwithcompaniessuchasExpedia,WegoandSkyscannerprovidingaplatformforconsumerstocomparepricesacrossairlines.Traditionaltravelagentsarethebrick-and-mortarcompaniesdisruptedbyE-commerceintheflightbookingsmarket.Flightsarenowwidelysoldthroughairlines’ownwebportalsandnewonlinetravelagenciessuchasAsiaTravel.InthelattercaseAsiaTravelactsasanintermediary,andtakesacommissionorfeeforthebookingserviceitprovides.Airlinesarealsonowsellingticketsviaaggregatorsites.ExamplesofaggregatorsinthissectorincludeMomondoandWebjet.

3.4.9. Theimpactofthesecompaniesonflightbookingshasbeenevidentinpricecompetition.Studieshaveshownanincreaseinpricecompetitionasthesizeoftheinternettravelsearchpopulationgrows.122 Thereisalsolowerpricedispersiononlinecomparedtobrick-and-mortaragencies.123

3.4.10. Therearethreekeydevelopmentswhichhaveledthissectortoitscurrentmodel.Firstly,theadventoftheinternetandthesubsequentvastglobaluptakehasbeenakeyenablertothegrowthoftoday’smarket.Secondly,airlineshavesetupdirectwebsites,whereticketsaresolddirectlytocustomers.Thethird,andperhapsmostimportant,changewastheadventofanE-ticket;firstcreatedbytheInternationalAirTransportAssociation(IATA)on1June2008.124Theresultofthesethreedevelopmentsisthatflightscannowbeboughtandsoldanywhereintheworldwithouttheneedforapaperticket

120 GFK Singapore, (2014).

121 DotEcon (2015), page 132.

122 Verlinda and Lane (2004), page 8.

123 Sengupta and Wiggins (2007), page 1.

124 IATA (2008).

32

tobeissued.Thishashadaprofoundeffectonthedevelopmentofthemarket.Oneconsequenceisthedemiseoflocalindependenttravelagents,whichhavebeenunabletocompetewiththegrowthoflargeE-commercecompanies.

3.4.11. Some largerbrick-and-mortarentitieshavehoweversurvived, suchasFlightCentre,FlightWorld,andHelloWorld.Thesuccessoftheseoperatorshasbeenachievedbyprovidingahigher levelofcustomercarewhilstofferingabundledservice,includingflights,accommodation,andothertripsontheholiday.InthecaseofFlightCentre,95%ofitssalesarestilldoneinstoreratherthanonline.125

3.4.12. E-commerceisnowentrenchedintheflightbookingindustry,withaggregatorsitesprovidingmoreinformationandcomparabilityforcustomersthantheirbrick-and-mortarcounterparts.Thesesitesarealsoabletoperformsophisticatedpricediscrimination,withthelikesofSkyscannerchargingahigherpricebasedonanalyticsofhowmanytimesconsumershavelookedataparticularflightorroute.Asaresult,theyareabletoestimateconsumers’levelofdemandfortheservice.

Land transport

3.4.13. Brick-and-mortarcompanieswithinASEANthatoperateinlandtransportincludelocaltaxioperatorswhichhavelicences,aswellasnationalised(orfranchised)railandbusservices.NewonlinecompaniessuchasUber,GoJekandGrabhavedevelopedtoadifferingextentwithinASEAN.Forexample,Grabclaimstohavemorethan95percentofthethird-party-taxi-hailingmarketwithinSoutheastAsiaasawhole.126Private-hirecareserviceshavealsoemerged,suchasSmoveinSingapore.

3.4.14. ThegrowthofE-commerceviaM-CommercehasmeantthatinnovativeappsarebeingdevelopedbyfirmssuchasUber,whereGPStechnologyon lowercostsmartphonescanbeusedtomatchcustomers anddrivers. Customers can alsofindother customers that are travelling in the samegeneraldirectionand“sharetheirride”.GPStechnologyenablesfare-seekingdriverstobematchedwith journey-seeking customers, rather than themore conventional system of customers havingtofindanavailable taxi. ByusingGPStechnology customers get a smoother experienceand thedriversbenefitastheapp identifiescustomersclosesttothatdriver,thereforeminimisingdrivers’non-feepayingjourneys.Theappscanalsobeusedglobally,providingevengreaterconvenienceandfamiliaritytoconsumers.

3.4.15. Datagatheredbythesecompanies’meansthatfirmsarealsoabletoimplementrealtimepricing,matchingsupplytodemandandenablingpriceincreasesduringpeaktimes.Uberisagoodexampleofafirmusingthesecapabilities.

3.4.16. Consumerchoicehasincreasedasnewalternativetransportoptionsarise.Driversarealsomoreabletochoosewhenandwheretheywork.

3.4.17. Thesenewdisruptiveorganisations(e.g.Grab/Uber)arenowdiversifyingtheiroffertologisticssuchaslastmiledelivery(e.g.UberEats,whichprovidesadeliveryserviceforrestaurantstodelivertheirproducttotheirfinalcustomers).

3.4.18. Transportmarketsarealsobecomingincreasinglycomplex.Forinstance,taxidriversarestartingtousetheGrabapptofindcustomers,andcarownerscanalsouseGrabtoearnextraincomefromtheirjourneysbyusing‘Grabhitch’.

Cosmetics and beauty products

3.4.19. Traditionalcosmeticsandbeautyfirms(brick-and-mortarcompanies)withinASEANhavestruggledtocompetewithnewonlineretailerssuchasLuxola.Hermo,amarketplaceforcosmeticsandbeautyproductsbasedinMalaysia,hasmonthlytrafficof870,000hits.127

3.4.20.Traditionally,thecosmeticsindustryhasreliedontheabilityforconsumerstosmellandtouchproductspriortopurchase,buttheonlinealternativeisprovingtobeanattractiveoptionduetoconvenientdelivery,wider product selection and competitive pricing.Online retailers are able to offervalue-addedoptionssuchasrememberingaconsumer’sallergiesandpreferences,helpingconsumerstomakequicker,moreinformeddecisions.Digitaldisruptionhasledtoinnovativesubscriptionservicesandothernewwaysforfirmstoattractcustomers,manyofwhichhavenowbeenfollowedbybrick-and-mortarsellers.

125 News.com.au (2016).

126 Wired (2016).

127 ASEANUP (2017).

33

3.4.21. A survey of online cosmetics andbeauty products shoppers has shown that the key issues thatinfluence online purchases are site security, product availability and free shipping.128 This isunderstandableasmostshoppersarerepeatpurchasers,andthereforeworrylessaboutbeingabletosee,touchorsmellproducts.

3.4.22. Someof the large trustedbrands in this industry aremakinguseofonlineplatforms inorder toreachabroadercustomerbaseandtogetcloseraccesstotheircustomers.BenefitUS,forexample,hasusedFacebooktoreachitscustomers,whilstPinterestandVineareusedbymanycosmeticscompaniesforasimilarpurpose.129Otherfirms,suchasCotyandPierreFabre inEurope,are lessaccommodatingtothesaleoftheirproductsviaonlinechannels.TherestrictivepracticesadoptedbythesefirmsarediscussedindetailinSection7.

3.4.23.Manybrandsarealsotryingtousetheirownwebsitestobuildloyaltyandtoimpartinformationabouttheirproducts,ratherthanasapointofsalestool.L’OréalParislauncheditsMakeupGeniusappin2014whichusesfacialmappingtechnologytoturnthefront-facingcameraofasmartphoneintoavirtualmirror,allowinguserstofeelliketheyaretryingonproducts.130

Fashion

3.4.24. SincetheemergenceofE-commerce,globalonlineretailerssuchasASOShaveregularlyshippedtoASEAN.MorerecentlyASEANbasedretailers,suchasZalora,haveenteredthemarket,disruptingboththetraditionalbrickand-mortarfirmsandearlierE-commercesites.Formedin2012byRocketInternet,thecompanynowhasapresence inalloftheASEAN6,with1.85millionuniquehitspermonthasofFebruary2017.131AccordingtoEuromonitor,in2007internetretailingwasaround3%offashionsalesglobally,butby2012thissharehaddoubledandshowedthefastestgrowthrateofallretailchannels.132

3.4.25. The typical model of expansion employed by fashion retailers is a three-step process. Firstly, awebsiteissetupinthehomecountry,providingaglobalservicebasedoutofthatcountry.Oncethecompanyhasstartedtogenerateinterestinothernations,itwillthensetupalocalsiteinthatregion,withproductsstillbeingsuppliedfromthehomejurisdiction.Finally,oncethelevelofdemandhasreachedagivenlevelofmaturityinthenewlocation,adistributioncentrewillbesetuptoservicethedemandmoreeasilyandefficiently.

3.4.26.TheimpactofE-commerceonfashionhasbeenconsiderable.Theindustryhasalongandcomplexsupplychainwhich,sincethe1990sand2000s,hasseenstructuralanddisruptivechangesduetoglobalisation,sustainabilityconcerns,andE-commerce.133Incontrasttobeautyproducts,wherepricecompetitionhasincreased,competitioninthefashiondomainisbasedmoreonrangeandchoice.

3.4.27. The role of socialmedia is onewhich is often overlooked by industry observers andfirms alike.Picturesofcelebritiessometimesleadtohugespikesindemandforfashionproducts.Whilstsomeofthesepromotionalactivitiesareplanned,otherunplannedinstancesmaytakeretailersbysurprise.Retailersarerequiredtobefarmoreagilethaneverbefore.ThegrowthofM-commerceisleadingtoamoreimmediategratificationmodel,wherebyconsumerssurftheinternetfromtheirsmartphone,often lookingat celebrities andwhat theywear.Consumersareable topurchasetheseproductswithinafewminutesfromtheirphoneandhavethemdeliveredpotentiallybytheendoftheday.

3.4.28.ConsumerexpectationshavegrownsincetheadventofE-commerce.Onlineretailershaverespondedto this inASEANbyoffering click-and-collect services,making agreementswithdeliverypartnerswhichenableconsumerstocollectproductsfromspecifiedlocations.Despitethisprogress,thereis still a longwaytogo incomparisontotheserviceexperienceavailable toconsumers inothercountriesoutsideofASEAN.Lastminutedeliveryismoredifficultinsomepartsoftheregionduetopoorlogistics.Freereturnsarealsooftennotoffered.

3.4.29.The reduced need for inventory in shops, and greater dependence on distribution networks hasimpacted supply chains.Therehasbeen a shift in labour needs frombrick-and-mortar shops todistributioncentrestoassistwithpackaginganddeliveryofproducts.Consumersarealsonowmoredemanding,requestingnextdaydeliverytospecificlocations,whichmeansindividualitemsarenowbeingdeliveredontheirowntohouseholds,asopposedtointhepastwherebigdeliverieswentfromoneplacetoanother.

128 AT Kearney (2014), pages 1-11.

129 Econsultancy (2016).

130 Translate Media (2017).

131 ASEANUP (2017).

132 Euromonitor (2013), pages 1-51.

133 Credit Suisse (2016).

34

04 Keycompetitionandotherregulatorychallengesand/orbarriersfacedbybusinessesintheE-commercesectorforAMSandhowtheyhindercompetitionandgrowthoftheE-commercesectorintheregion

4.1. Introduction

4.1.1. Businesses competing in E-commerce markets may face twodistinct types of barriers. Firstly, there may be barriers to thegrowth and development of E-commerce markets affecting allfirms in the market (barriers to expansion), such as restrictiveformsofregulation,orbroadtechnologicaldelays.Thequalityofconnectivity infrastructuremay also be considered a barrier tothegrowthanddevelopmentofE-commerceinASEAN.Secondly,theremaybebarrierstoentrythatapplytopotentialentrantsorsmallfirms(e.g.economiesofscaleornetworkeffects).Thesetwotypesofbarriersareexploredinthefollowingsub-sections.

4.2. Barriers to expansion

4.2.1. Abarriertoexpansionisdefinedas“somethingthatpreventsafirmalreadyinthemarketfrombeingablequicklyandcheaplytoincreaseitsoutput”.134 ThepresenceofthesebarriersinhibitsthedevelopmentofE-commercemarketsinASEANandaffectsall firms, including incumbents. In order for E-commercemarketstoflourish,theserviceprovidedtocustomersmustbetrustworthy,andsuitablyefficient,suchthat it isanattractivealternative to brick-and-mortar transactions. ThroughoutASEAN,however,thereisalackoftrustamongcustomerswhencompletingtransactionsonline,forinstancewithregardstodataprotection,bankingfraud,unfulfilleddeliveries,andtheinabilityto returnproducts.20%ofMalaysianSMEs reportE-paymentconcernsasoneofthemainobstaclestothedevelopmentofE-commerce.135This lackoftrust,ultimatelystemmingfromalack of technological infrastructure in the region, and aweakregulatory environment has prevented E-commerce marketsfromgrowingtotheirfullpotentialandhasinhibitedthegrowthoffirmstradingacrossborders.Thesetwocategoriesofbarriersareexploredingreaterdetailinthefollowingsub-sections.

134 Bishop, S. and Walker, M. (2010), page 81.

135 ACCCIM (2012), page 9.

E-payment concerns as one of the main obstacles

20%of Malaysian SMEs report

35

Level of technological infrastructure in ASEAN

4.2.2. Despitesignificantinvestmentintheirtechnologicalinfrastructure,manyAMSstilllagbehindinglobalrankings in terms of speed, efficiency and reliability of internet services.136 Multiple questionnairerespondentshighlightthecurrentleveloftechnologicalinfrastructureasanemergingissueorbarriertothedevelopmentofE-commerceintheir jurisdiction.Theseissuesareseparatedintothreekeyareas:InformationandCommunicationTechnology(ICT),broadbandandmobileinternet,andlogisticsanddelivery.

4.2.3. ICT:TheICTDevelopmentIndex(IDI)137publishedbytheInternationalTelecommunicationUnion(ITU)scorescountriesandranksthembasedon11benchmarkscoveringthreekeyareas:ICTaccess,ICTuseandICTskills.

4.2.4. RankingsofAMSonthebasisofthis indexvarysignificantly.OftheAMS,Singapore is rankedthehighestin20thplacewithascoreof7.95,whereasLaoPDRisrankedlowest,in144thplacewithascoreof2.45.138TherankingsofallAMSarepresentedinTable6below.

Table 6: ASEAN rankings of IDI

136 International Telecommunication Union (2016).

137 http://www.itu.int/net4/ITU-D/idi/2016/.

138 There are 175 countries in the IDI rankings published by the ITU.

139 A high IDI score (maximum of 10) indicates an advanced level of ICT development with respect to access, use and skills.

4.2.5. TheaverageIDIvalueforAMSin2016was4.5,marginallybelowtheglobalaverageof4.9.However,thereareclearsignsofimprovementinCambodia,MyanmarandLaoPDR(thethreelowestrankedAMS),whichachievedthehighestyear-on-yearchangeamongAMS.

ASEAN ranking

12345678910

Source:InternationalTelecommunicationUnion(2016).

World ranking

20617782105107115125140144

Country

SingaporeMalaysiaBruneiDarussalamThailandVietnamThePhilippinesIndonesiaCambodiaMyanmarLaoPDR

IDI 2016 value139

7.956.225.335.184.294.283.863.122.542.45

IDI 2015 value

7.885.645.255.054.023.973.632.781.952.21

% change year-on-year

1%10%2%3%7%8%6%12%30%11%

36

4.2.6. Broadband and mobile internet:AbroadbanddivideexistsinmanynationswithinASEANbetweenthe richermetropolitancities thathavestrongandstable internetcoverage, andthepoorer ruralregions that havevery limited connectivity.140The connectivity infrastructure inASEAN alsovariessignificantly between AMS, affecting businesses’ ability to sell online, and consumers’ access toE-commercemarkets.AverageinternetspeedsandthecostofaccessingtheinternetintheASEAN6arepresentedinTable7below.FiguresfortheUKandUSAareprovidedforcomparison.

Table 7: Connectivity infrastructure across ASEAN

4.2.7. Table7highlightsthedifferencesinthespeedandcostofaccesstotheinternetacrossAMS.WhilstthePhilippineshastheslowestaveragebroadbandspeedoftheASEAN6,itisalsothemostexpensiveboth in nominal terms and as a percentage of average income. Research suggests that fasterbroadbandisavailableinthePhilippines,offeringspeedsof20,50and100Mbps,buttheassociatedcostsmakeitevenmoreunaffordablethanthecurrentsituation.141SingaporeistheonlyAMSthathascomparablespeedsandcoststotheestablishednetworksintheUKandUSA.

4.2.8. TheUNESCAP (2013)outlineshowthere isanopportunityforapan-regionalterrestrialfibreopticnetworkwhichcouldprovidefastbroadbandconnectivitytotheentireregion,allowingAMStorealisethefullpotentialofE-commerceonadomesticandinternationallevel.Foraprojectsuchasthistobesuccessful,thecooperationofgovernmentsandotherinternationalorganisationsacrossASEANwouldbeessential,andaccesstosignificantfundingwouldberequired.

Country

IndonesiaMalaysiaThePhilippinesSingaporeThailandVietnamUKUSA

Source:Averagebroadbandspeed:Akamaistateoftheinternet(Q22016–Q12017);Averagemobileinternetspeed:Opensignal.comFebruary2017(DatafromNovember2016–January2017);Connectivitycost:Numbeo.com(2017);Averageincome:Worlddata.info(2015).Note:Averagebroadbandspeedcalculatedasanaverageoffourquartersofdata.DataonlyprovidedforthoseAMSwhereavailable.

Average broadband speed (Mbps)

6.67.94.619.013.77.315.816.9

Average mobile internet speed (Mbps)

4.77.93.330.16.1-15.112.5

Connectivity cost (USD) (60 Mbps, Unlimited Data, Cable/ADSL) (Utilities (Monthly))

25.737.442.430.418.410.528.352.7

Connectivity cost as a % of average income

9.04.214.30.73.96.30.81.1

140 UNESCAP (2013), page 5.

141 Philstar (2016).

37

4.2.9. Logistics and delivery:ToenhancethespeedandreliabilityofE-commerceprocesses,andreducedeliverycosts,improvementsinlogisticsanddeliverysystemsareneededacrossASEAN.Fornearlyhalf of Singaporeans, the primary reason for not buying online is delivery concerns.142 However,researchsuggeststhatthenumberofparcelsdeliveredwithinASEANwillgrowatanannualgrowthrateof23%between2016and2020,143largelydrivenbythegrowthofE-commercewithintheregion.GiventhatonlySingaporeandthePhilippineshaveliberalisedtheirpostalindustries,144theabilitytomanagetheincreaseinthevolumeofdeliveriesarisingfromE-commercewillbeachallengeforthestateownedpostaloperatorstoovercome.However,themarketforcourierservicesislessregulatedsothere isanopportunityfor last-miledeliverytobeperformedviatheseproviderstomeetthisincrease in demand.

4.2.10. ThefactthatsomeAMSarearchipelagosalsocausessignificantlogisticalrestraintsforthesaleofgoodsviatheinternet.Forexample,withabout2,000inhabitedislandsinthePhilippines,thedeliveryofgoods isbothexpensiveandtimeconsuming.Asaresult,E-commerce iscurrently largelyonlyavailabletowealthyconsumers,orthoselivingonbetterconnectedislands.

Regulatory and legal barriers inhibiting E-commerce transactions and cross-border trade

4.2.11. Cybersecurity: Aswellasthecurrentleveloftechnologicalinfrastructureintheregion,theregulatoryandlegalenvironmentacrosstheregionhasalsofailedtoprotecttransactionsfromcyber-threats.

4.2.12. Sophos(2013)foundthatfourofthetopfiveworldwideriskiestcountriesforcyber-attacksareinASEANandthatAsiahasthemostspamsourcesbycontinent.145Itisthereforeapparentthatworkisrequiredon regulationstotacklecybersecurity issues, inordertobuildtrustamongconsumersandallowE-commercetoflourishintheregion.UNCTAD(2013)highlightscoordinationinregulationstacklingcybercrime,consumerprotectionandrecognitionofelectronicsignaturesascriticalrequirements,inadditiontotheestablishmentofaregionalonlinedisputeresolutionfacility.

4.2.13. AMS are focusing on cybersecurity as an area for development, with S$10 million set aside tofundworkinthissectoroverthenextfiveyears.ThisASEANCyberCapacityProgrammehasbeendesignedtodevelopthetechnical,policyandstrategy-buildingcapabilitiesrequiredwithinAMSthatwillallowbusinessestooperateconfidentlywithinE-commercemarkets.146SeniorofficialsacrossAMSrecognisethatasecureand resilientcyberspace isacriticalenablerforAMStoharnesstheopportunitiesfromdigitaltechnologiesandE-commercetoachieveeconomicgrowthandimprovelivingstandardsthroughoutASEAN.147


143 Nomura (2016).

144 Postal services is, however, currently a government monopoly in the Philippines.

145 Sophos (2013), page 29.

146 NATO CCDCOE (2017).

147 Cyber Security Agency of Singapore (CSA) (2016).

38

4.2.14. Customs and taxes:Nationaltaxpolicieswereraisedinquestionnaireresponsesasanissueorbarrierto E-commercewithinASEAN. Specifically, responses suggested that consumers and businessesarediscouragedfrompurchasinggoodsfromoverseasfirmsbecauseofuncertaintyanda lackofawarenessofcustomsandtaxrules.TherearealsovariationsintheimportdutiesandtaxespayablewhenpurchasinggoodsfromanotherAMSinASEAN.Importdutiesandtaxesona$50handbagrangefrom$0to$19.55.148ThebreakdownofthesedutiesandtaxesineachcountryispresentedinTable8below.

Table 8: Import duties and taxes on a US$50 handbag

148 Duty Calculator (2017).

149 UNCTAD (2016), page 24.

150 Price control measures are “those implemented to control or affect the prices of imported goods in order to, inter alia, support the domestic price of certain products when the import prices of these goods are lower; establish the domestic price of certain products because of price fluctuation in domestic markets, or price instability in a foreign market; or to increase or preserve tax revenue. This category also includes measures, other than tariff measures, that increase the cost of imports in a similar manner (para-tariff measures).” UNCTAD (2016), page 5.

AMS

BruneiDarussalamCambodiaIndonesia*LaoPDRMalaysiaMyanmarThePhilippines*SingaporeThailandVietnam

Source:Dutycalculator.com(2017).Note:CalculationmadeonbasisofacottonhandbagmanufacturedinItalyandsoldinSingapore(withtheexceptionofSingaporewhichwasassumedtobesoldinMalaysia).*Noduty/taxesasthegood’svaluedoesnotexceedUS$50inthecaseofIndonesia,andPhP10000inthecaseofthePhilippines.

Duty (US$)

0.003.500.005.000.003.750.000.0015.0012.50

Taxes (US$)

2.505.350.005.503.002.690.000.004.556.25

Total (US$)

2.508.850.0010.503.006.440.000.0019.5518.75

4.2.15. Thedegreetowhichforeigncompaniesareabletocompetewithdomesticplayersthereforevariesacrosstheregion.Asaresultofimportduties,firmsexportingtoanothercountryinASEANareatadisadvantageincomparisontodomesticfirms.

4.2.16. Non-tariffbarriersalsorestrictfirmsfromabroad,ashighlightedbyUNCTAD(2016).Althoughsuchbarriers may have a main objective unrelated to trade, such as protecting public health or theenvironment,theymayhavetheadverseeffectofinhibitingcross-bordertrade.Complextechnical,sanitaryandphytosanitarymeasuresareparticularlyprevalentbarriersacrossASEAN.149Pre-shipmentinspectionandpricecontrolmeasures150havealsobeen identifiedasbarriers regularly restrictingcross-border trade in ASEAN.

39

4.2.17. Application of competition policy and law: Differences in approaches to the application ofcompetitionpolicyandlawinAMS,assetout inPartBofthishandbook,alsoposechallengestofirms looking to operate internationally acrossASEAN. This is particularly importantwith regardstotheuseofvertical restraintsbyfirmsoperating inonlinemarkets (i.e.whenarestraintmaybedeemedanti-competitivebyauthorities-seeSection7foramoredetaileddiscussion).Internationaldifferencesinapproachestoapplyingcompetitionpolicyandlawgivesrisetoanadditionalburdenforthefirms,astheymayneedtoadapttheirconductdependingonthedifferentapproachadoptedinthedifferentterritorieswheretheywishtoconducttheirbusiness.TherecentBooking.comcase(seeCasereview17)isagoodexampleofsuchacase,withdifferentcompetitionauthoritiesreachingdifferentconclusionsontheuseofwideandnarrowMFNclauses151 inthehotelbooking industry.Onlinebookingplatformshadtoadapttheirconductindifferentjurisdictionsasaresult.

4.2.18. AMSarehoweverworkinghardtoovercomeproblemsandimprovetheharmonisationofregulations,forexampleby introducingtheASEANCompetitionActionPlan (ACAP)2016-2025,152whichaimstoimprovetheconsistencyofregulationsandbuildtrustforconsumerslookingtocompletetransactionsinother jurisdictions.The increase inglobalisation,fuelledbyE-commerce,has ledtoarise inthechallenge to identify and combat anti-competitive conduct, and mergers which may lead to alesseningofcompetitionacross internationalborders.AMSarealreadyconductingtrainingonhowbesttoapproachsituationslikethis,whichisanessentialfirststepincreatingeffectivecross-borderenforcement.153UNCTADhighlightedthatmembershipoftheInternationalConsumerProtectionandEnforcementNetworkwouldbeanotherbeneficialmovetoimproveregionalcooperation.Atpresent,onlythePhilippinesandVietnamaremembers.154

4.3. Barriers to entry

4.3.1. Barrierstoentrycanbedefinedas“acostofproducingwhichmustbebornebyafirmthatseekstoenteranindustrybutisnotbornebyfirmsalreadyinthemarket”. 155Barrierstoentrylimittheabilityofnewentrantstoenterandexpandoutputinagivenmarket.Thesebarrierscanbeconsideredunderfourbroadcategories:economicadvantagesenjoyedbyincumbents;costsandnetworkeffectsthatinhibit consumersfromswitchingsuppliers; legalbarriers; andtheconductof incumbentfirms.156

Thesebarriersarepresentbothinbrick-and-mortarandonlinemarkets,buttherearedifferencesintheprevalenceandmagnitudeofsomeofthesebarriersbetweenthetwosaleschannels.Thissectionconsiders eachof the four categories in turn, highlighting any important featuresofE-commercemarketsthroughout.

Economicadvantagesenjoyedbyincumbentfirms

4.3.2. Incumbentfirmsinamarketmaybenefitfromcertaineconomicadvantagesthatnewfirmsorsmallerplayersareunabletoachieve,byvirtueoftheirsize.Economiesofscaleandscope,privilegedaccesstoessential inputs, technologiesor information,andanestablishedsalesnetworkallputsmallerfirmsandnewentrantsatadisadvantage.157This sub-sectionfocusesontwopotentialeconomicadvantagesthathavebeenimpactedbythegrowthanddevelopmentofE-commerce:economiesofscale,andprivilegedaccesstoinputs,technologiesorinformation.

4.3.3. Economies of scale: Economiesofscalearisewhentheaveragecostperunitofoutputdecreaseswiththe increase inthescaleoftheoutputproduced,andeconomiesofscopeoccurwhen it ischeapertoproducetwoproductstogetherthantoproducethemseparately.Insuchinstances,newentrantsorsmallerfirmsareunabletoproduceasefficientlyaslargerfirms,orfirmsproducingarangeofrelatedproducts.

151 A wide MFN is a vertical restraint that ensures that no other competitor will be given more favourable terms by a supplier/customer/platform – for instance being able to sell at a lower price. A narrow MFN restricts a firm from setting a lower price in its own store, but it is free to agree to a lower price with a competing store e.g. a hotel that enters a narrow MFN agreement with a hotel booking platform, cannot set a price on its own website lower than the price on the booking platform, but it can agree to lower prices on competing platforms.

152 ASEAN Competition Policy and Law (2016).

153 ASEAN (2017).



156 Whish, R. and Bailey, D. (2015), page 194.

157 Ibid. page 46.

40

4.3.4. Whilsteconomiesofscaleapplytobrick-and-mortarfirmsaswellastoonlineretailers,theyappeartorepresentlessofabarriertoentryinE-commercemarketsasthefixedcostsofenteringanewlocationormarketviatheinternetaresignificantlylower.InE-commercemarketsthereisnoneedtobuildorrentaphysicalretailspacetosellgoods.Thecostsofmakingawebsiteaccessibleinanewlocationarerelativelylow,forexamplethecostoftranslatingthewebsiteintothelocallanguageascomparedtothecostofestablishingabrick-and-mortarretailpresenceinothercountries.

4.3.5. IncumbentfirmsdostillbenefitfromsomeeconomiesofscaleinE-commercemarkets,thereforesomebarriersfornewentrantsremain.Theabilitytospreadmarketingcostsoveralargerquantityof goods sold remains a constraint for online retailers seeking to grow or enter newmarkets incomparisontolargerincumbentfirms.

4.3.6. Privileged access to inputs, technologies or information:Accesstosupportinginfrastructure,suchaslogistics,inventoryandpaymentsystemsmayalsoconstituteabarriertoentry.Verticalintegrationbyanincumbentplatformorsingle-sidedfirmmayaffectotherfirms’abilitytogainaccesstothesesystems.

4.3.7. Some also consider the data that afirmholds on its customers to be an asset that incumbentfirmshaveprivilegedaccessto.The rise inthequantityofdatathatsomefirmsarecollecting inE-commerce markets is under close consideration by some competition authorities around theworld.Thequestionfacingauthoritiesiswhetherthisdata,oftenreferredtoasBigData,constitutesabarriertoentryandthereforeislikelytoenhanceapotentialpositionofmarketpower.

4.3.8. BigDataisdefinedas:“theuseoflargescalecomputingpowerandtechnologicallyadvancedsoftwareinordertocollect,processandanalysedatacharacterisedbyalargevolume,velocity,varietyandvalue.” 158

4.3.9. ThepresenceofBigDatahasgrownsignificantlyoverrecentyearsthroughtheautomatedcollectionofinformationononlineactivity,includingfromsocialnetworkingsites.Firmsareabletousecomplexalgorithmsautomaticallytosievethroughthisdatatoidentifythepatternsandtrendsinconsumers’behaviour.Consumersbenefitiffirmspassonanyefficiencygainsfromtheuseofthisdata,improvethequalityandscopeoftheirgoods/services,and/oroffermoretargetedadvertising.159

4.3.10. Ontheotherhand,thereareconcernsthattheadditionalinsightsthatfirmshaveoftheircustomersmaybeanassetthatsmallerfirmsornewentrantsareunabletoreplicate,andthereforeincreaseapotentialfirm’spositionofmarketpower.However,generalconsensusonthisissuehasyettobereached.Owning largedatasetsdoesnotnecessarily leadtomarketpower,oractasabarriertoentryperse,especiallyinE-commercemarketswherecompetitionisdynamic.160Inmanymarkets,datacanbecollectedfrommultiplesources,andsuchcustomerinsightsarenotexpensive,evenforsmallcompaniesandnewentrants,togainaccessto.When,ontheotherhand,suchdatacannotbereplicated,itisimportanttounderstandwhethersuchdataconstitutesanessentialfacilitywithoutwhichcompetitorsareunabletooperate.Thisisnotanewissueincompetitionpolicy.

4.3.11. Competitionauthorities’approachestodealingwithE-commerce relatedcaseswhereBigData isa factor shouldbenodifferent to those inofflinemarkets.TheOECD (2017) states thatalthoughfurther research is needed in this area, traditional antitrust tools can be adapted and applied totackledata-relatedanti-competitivepractices.Nonetheless,BigDataremainsawidelydebatedareaofcompetitionpolicyatthetimeoffinalisingthishandbook.

158 OECD (2017).

159 Skadden (2017).

160 OECD (2017).

41

Costs and network effects that obstruct consumers from switching suppliers

4.3.12. Switchingcostsforconsumersalsomakeitharderfornewentrantsandsmallerfirmstocompetewithlargeincumbentplayers.Switchingcostsmakeitmoreexpensiveforconsumerstopurchaseagoodorservicefromanalternativesupplierbeyondthedirectpricecharged.Thesecostsmaybemonetaryornon-monetary.Forexample,thetimespentincreatinganaccountwithanewprovideris consideredanon-monetary cost of switching. Somecostsmaynotmaterialise, suchas costsarisingfromtheriskofonlinefraud,but intheseinstancestheriskthatanadditionalcostwillbeincurredmaydeterconsumersfromswitchingproviders.Switchingcostsmayarisenaturally,ormaybecreatedorincreasedasaresultoftheactionsofincumbentfirmsinordertorestricttheentryandexpansionofsmallerfirms.Forexample,loyaltyrewardschemesaredesignedbyincumbentfirmstoincreaseswitchingcostsforconsumerstoalternativeproviders.

4.3.13. Theestablishedreputationofanincumbentrepresentsabarrierfornewonlineretailers.Forconsumersconsideringswitchingthereisariskthatthenewretailermaynotbereliable.Thequalityoftheservice(e.g.reliabilityofdeliverytimes,returnpolicy,etc.)andtheproductitself(e.g.ifitisacounterfeit)arebothuntested.Aconsumeristhereforemorelikelytopurchasefromaretailerithasusedbefore,andtrusts.

4.3.14. Switching costs arepresentboth inbrick-and-mortar andonline sales channels. InE-commercemarketssomeswitchingcostshaveemergedforconsumers,makingitharderfornewentrantsandsmallerfirmstocompetewithincumbentonlineretailers.Asindicatedin4.3.12,therearemorerisksinvolvedinswitchingtoanalternativeretailerintheonlinespacethaninbrick-and-mortarmarkets.Thisisbecauseinanonlineenvironmentconsumersarelessabletoassesstheriskstheyfaceintermsofthereliabilityoftheservice,thequalityoftheproducts,thetreatmentoftheirpersonaldata,andthesafetyofsharingtheirpaymentdetails.Thesepotentialrisksconstituteaswitchingcostthatmakesconsumersmorelikelytouseanincumbentonlineretailerwhomtheyhavepurchasedfrombefore,andthattheytrust,asopposedtoanewonlineretailerofferingthesameproduct,therebycreatingbarrierstoentry.Accreditationfromindependentconsumerbodies,aswellastestimonialsandreviewsfromcustomers,cangosomewaytowardsreducingtheseswitchingcosts,althoughconsumersmaynotnecessarilyknowwhethertotrustsuchendorsements.

4.3.15. Networkeffectscanalsocreateswitchingcostsforconsumers.AsexplainedinTechnicalExplanation1inAnnex1,networkeffectsarepresentwhenthevaluethatoneuserplacesonagoodorserviceincreasesasthenumberofotherusersofthatgoodorservicerises(thatisthescaleofthenetwork).Ifanindividual,andalargeproportionofthatindividual’snetwork,areusingagoodorserviceprovidedbyonefirm,thereisacosttothatindividualfromswitchingtoanalternativeproviderinthatfewerpeopleareusingtheotherservice.Thevaluethattheindividualderivesfromconsumingthegoodorserviceprovidedbyasmallerfirmislower(asthevaluedependsonthesizeofthenetwork,whichissmaller).Thenetworkeffectsthereforeconstituteabarrierfornewentrantsandsmallerfirms.

4.3.16. While network effects are present in both brick-and-mortar and onlinemarkets, the emergenceandgrowthofE-commercehasresultedinthedevelopmentofmanynewplatformsinmulti-sidedmarketswherenetworkeffectsarehighlyprevalent,suchasonlinemarketplaces,PCWsandsocialmedia sites.

4.3.17. Networkeffectsare lessof abarrier toentry if individualsmulti-home; that is, theyusemultipleprovidersofagoodorservice.Consumersmayprefertheuseofaplatformwhichprovidesaccesstoalargenumberofproductsorservices,butiftheycaneasilysourcetheproductsorservicesfromotherplatforms (i.e.multi-home) the larger scaleof the incumbentnetworkdoesnotnecessarilyconstituteabarrier.Bycontrast,ifthereisacosttomulti-homingthenthebarriertonewentrantsandsmallerfirmsisgreater.

42

4.3.18. In online markets multi-homing is common, therefore network effects do not always representa significant barrier to entry for new entrants and smaller firms. Moreover, even in caseswhenconsumerssingle-home,theadventofabetterproductorservicecaninducethoseconsumerstoswitch.FacebookisagoodexampleofafirmovercomingnetworkeffectswhenitdisplacedMySpaceasmarket leader insocialmedia.161Similarly,Taobao’sdisplacementofeBay intheChineseonlinemarketplacesectoralsohighlightsthis.162

Legal barriers

4.3.19. “Legaladvantagessuchasregulatoryrulesthat limitthenumberofmarketparticipants”canalsoconstitutebarrierstoentry,163inparticularwithregardstoIPrights.Thisiswithregardstoindustrialproperty, namelypatents for inventions, andcopyright laws,164wherebynewentrants and smallerfirmsmaynotbeabletoaccesspatentedtechnologyorcopyrightedcontent.MostacademicsagreethatIPrightsarecrucialforcertainmarketstofunctioneffectively,165thoughtheycanhavetheeffectofrestrictingentry.ThistopicisdiscussedindetailinrelationtoE-commerceinASEANinSection14.Otherlegalbarriersmayderivefrom“governmentlicensingrequirementsandplanningregulations,statutorymonopolypowerandtariffandnon-tariffbarriers”.166

Theconductofincumbentfirms

4.3.20.Finally,theconductoflargeincumbentfirmsmayrestrictentrytoaparticularmarketwhentheyareabletoexercisemarketpowerandtherebyexcludeormarginalisecompetitors.TheconductoffirmsinE-commercemarkets,eitherunilaterallyorincoordination,isdiscussedindetailinPartBofthishandbook,whichpresentsadiscussionofwhensuchconduct islikelytoleadtoanti-competitiveeffects.

4.3.21. Additionally,asdiscussedin4.3.12,switchingcostsforconsumersmaybeincreasedasaresultoftheactionsofincumbentfirms.

161 Bloomberg (2011).

162 The Economist (2006).

163 Whish, R. and Bailey, D (2015), page 920.

164 WIPO (2011).

165 Center on Law and Information Policy (2011).


43

Section for Competition Authorities

PartB:

44

5.1. ThegrowthofE-commercehasbroughtaboutanumberofchangestothewaymarketswork.Pricetransparencyhasincreased,onlineplatformssuchasmarketplacesandPCWshaveemergedandrapidlygrown, andvertical restraints havebeenusedmore frequently asmanufacturers seek to gain bettercontroloverdistributionnetworks.167

BenefitstoconsumersfromE-commerce

5.2. Overall, consumers have benefited from the rapid innovation brought about by the internet andE-commerce.Pricecomparisonhasbecomesignificantlyeasierforconsumers,whichreducessearchcostsforconsumers,bothintermsoftimeandcost.Consumersarealsoabletoswitcheasilyfromone channel (online/offline) to another.168 The increase in price transparency has intensified pricecompetitioninmanymarkets,tothebenefitofconsumersaslongasthiscompetitionisnotattheexpense of product quality. For themost part, product diversity has increased for consumers asaresultoftheemergenceandgrowthofE-commerce,asfirmsareabletostockawiderrangeofproductsduetoareductioninphysicalconstraints.

Barrierstoconsumerbenefits

5.3. Tofullyrealisethebenefitsfromthesedevelopments,itisimportantthatcompetitionisnotimpairedbyanyformofunilateralorcoordinatedconductbycompaniesactiveinE-commerce.

5.4. Thispartofthehandbooklooksatthechallengesfacedbycompetitionauthorities inestablishingwhetherthebehaviouroffirmsinE-commercemarketsisanti-competitiveorefficiencyenhancing,andconsiderswhetherexistingcompetitionpolicyandlawissufficienttodealwithsuchchallenges.TheapproachtakeninthissectionfocusesonspecificcompetitionissuesarisingasaresultoftheemergenceandgrowthofE-commerce.

Current stage of debate on competition policy and law in E-commerce markets

5.5. ItisimportanttonoteattheoutsetthatcompetitionpolicyinE-commerceisanareaunderconstantdevelopmentaroundtheworld.Someoftheemergingchallengesarestillnotfullyunderstood,andonoccasion,competitionauthoritieshavetakencontrastingviewpointsonsimilarissues,illustratingthe complexities surrounding this area of competition policy.This part of the handbook providesguidancebasedoncurrentdevelopmentsandtrends.However,duetotheongoingdevelopmentofthedebateinthisareaofcompetitionpolicy,itshouldnotbeseenaspresentinganultimatesetofprinciples.

167 European Commission (2017b), para. 15.


Introduction05

45

Research sources

5.6. Thispartofthehandbookdrawsuponcasesfromjurisdictionsaroundtheworld,andinsightsfromacomprehensivequestionnaireonE-commerceinASEANcompletedbythecompetitionauthoritiesofSingapore,Malaysia,thePhilippines,VietnamandIndonesia.WhilstcompetitioncasesrelatedtoE-commercearestillrelativelyfewinASEAN,onlyoneofthequestionnairerespondentshasnotdealtwithanyE-commercerelatedcases.Itisalsoclearthatauthoritiesexpecttodealwithmorecasesinthenearfuture,as80%ofrespondentsseeE-commerceasoneofthefocusareasoftheirauthority’sworkoverthenextthreetofiveyears.

Structure of Part B

5.7. Thispartofthehandbookisorganisedasfollows:

• Section6looksatthechallengesthatcompetitionauthoritiesfacewhendefiningmarketsand assessingmarket power in E-commercemarkets, including a consideration of adaptations to approachesthatmayberequiredwheninvestigatingmulti-sidedmarkets;

• Section7exploreschallengesfacedbycompetitionauthoritieswithregardstotheuseofvertical restraintsbyfirmsinE-commercemarkets;

• Section8considersissuesrelatingtohorizontalcoordinationinE-commercemarkets;

• Section9looksatunilateralanti-competitiveconduct(i.e.abuseofdominance),andissuesarising asaresultoftheemergenceandgrowthofE-commerce;

• Section10considersmergersandacquisitionsinE-commercemarkets,withaparticularfocuson issuesarisingasaresultofnetworkeffects,inadditiontolookingatwhetherexistingtestsareable tocaptureallrelevantcases;

• Section11outlineswaysinwhichAMScanaddressthechallengesdiscussedinSections6-10 throughthedesignandenforcementofcompetitionpolicyandlaw;and

• Section12concludesthispartofthehandbookbypresentingastand-alonecompetitionpolicy andlawcompliancechecklistforbusinessesengagedinE-commerceinAMS.

46

6.1 Defining the relevant market

Traditional approach

6.1.1. Typically,thefirststepinacompetitioninvestigationistodefinetherelevantmarket,bothintermsoftherelevantproductsand/orservices,andofthegeographicareacovered.169

Relevant product market(s) in E-commerce

6.1.2. Thetraditionalmethodofestablishingtherelevantmarketistoconsidertheproductsand/orservicesthatconsumersregardassubstitutable.Supplysidesubstitutionandpotentialcompetitionfromnewentrantsandsmallerfirmsarealsoimportantfactors.Thisapproachhasbeendevelopedtoapplytotraditionalsingle-sidedmarketsand,assuch, itappliesequallytosingle-sidedonlinemarkets.Formany competition authorities, defining the relevantmarket involves implementing the SSNIPtest (small but significant and non-transitory increase in price) or hypothetical monopolist test.Thistestinvolvesestablishingwhetherahypotheticalmonopolistinamarketcouldprofitablyraisepricesby5-10%forasustainedperiodoftime.However,inpractice,itmaynotbepossibletoobtainthe necessary data to conduct a SSNIP test, inwhich case themarketmay be defined throughlesstechnicalmeans,suchasconductingathoroughassessmentofthedemandandsupplysidesubstitutesinordertoidentifyalltherelevantcompetitiveconstraints.

6.1.3. ApplyingaSSNIPtesttoonlinemarketsoftenmeansthatonlineandofflinesalesarefoundtobeincludedinthesamerelevantmarketi.e.onewouldexpectthatformostproducts,ifpricesweretoriseby5-10%inbrick-and-mortarstores,consumerswouldswitchtoonlinechannels,andviceversa.Ahypotheticalmonopolistinonlythebrick-and-mortarmarketcouldnotsustainaSSNIP.Theremay,however,beinstanceswhenonlineandofflinesalesarenotfoundtobepartofthesamerelevantmarketandthereforeshouldbeconsideredseparately.AnalogousconsiderationsapplyincaseswhentherelevantmarketisdefinedonthebasisofinterchangeabilityandsubstitutabilityratherthanbymeansofaSSNIPtest.

Relevant geographic market(s) in E-commerce

6.1.4. Followingthedefinitionoftherelevantproductmarket,thenextstepshouldleadontodefinethegeographicboundariesoftherelevantmarketunderconsideration.170Onewouldexpectthisdimensionoftherelevantmarkettobewiderinonlinemarketsgiventheabilitytosubstitutepurchaseswithonlinestoresfromremoteareas inawaythatwassimplynotpossible inatraditionalbrick-and-mortarstore.However,otherfactors,suchasgeographicrestrictionsonaccesstothewebsitemayaffectthis.Hence,as for traditionalmarketanalysis, thegeographicscopeof the relevantmarketshouldbedeterminedonacase-by-casebasis.

169 For assessments of alleged cartels, the relevant market may not need to be defined; but rather the affected goods determined.

170 The relevant geographic market is the region in which “the conditions of competition are sufficiently homogeneous and which can be distinguished from neighbouring areas because the conditions of competition are appreciably different in those area” (European Commission, 1997). Determining the relevant geographic market effectively involves starting from a narrow geographic market area and assessing whether customers would switch to providers in a wider geographic area following a SSNIP by an incumbent hypothetical monopolist (demand side substitution), or whether suppliers from further afield would enter the market following a SSNIP (supply side substitution).

Marketdefinition,multi-sidedmarkets,and market power

06

47

Challenges in applying this approach

6.1.5. Although traditional assessments of substitutability still appear fit to tackle online single-sidedmarkets,theemergenceandgrowthofE-commercehasposedchallengestocompetitionauthoritiesin defining markets in instances where the interaction between manufacturers/distributors andconsumersdeviatesfromtraditionalmodels.Thisishighlightedbythefactthat80%ofquestionnairerespondentswithinASEANpointtomarketdefinitionasoneofthebiggest issuestheyfacewhendealingwithE-commercecases.

6.1.6. Inmany instances, onlinemarkets aremulti-sided in nature, andoftendue to the rapidpaceofinnovation and flexibility of different technologies, firms may face competitive constraints fromoutsidetherelevantmarket.Thesechallengesarediscussedindetailinthefollowingsub-sections.

6.2. Multi-sided markets

Definitionofmulti-sidedmarkets

6.2.1. Traditionalbrick-and-mortarretailenvironmentsareone-sidedinnature,namely,throughtheretailingoperation a store acquires the ownership of products through a one-sided relationship with itssuppliers,andsellsthemontoitscustomersthroughaseparateone-sidedretailingarrangementwithconsumers.However,manyonlinemarketsaremulti-sidedinnature. Inotherwords,relationshipsbetweenthetwosidesofthemarket,e.g.thesupplierandthecustomer,areinterdependent.Althoughnot a unique feature of onlinemarkets,multi-sidedmarkets aremore prevalent in E-commercemarkets.TechnicalExplanation1inAnnex1definesandexplainsmulti-sidedmarketsindetail,anddiscussesimportantcharacteristicsofsuchmarketsrelevanttocompetitionauthorities.

Marketdefinitioninmulti-sidedmarkets

6.2.2. When defining relevantmarkets in competition investigations relating to E-commerce, traditionalmethodsofassessingdemandandsupplysubstitutabilityarewellpositionedtocontinuetoserveauthorities in one-sidedmarkets. If, however, amarket ismulti-sided sided in nature, a numberof challenges arisewhen defining relevantmarkets using existing approaches. One questionnairerespondenthighlightedhowitiscurrentlyfacingthischallengewhendefiningonlinemarketsthataremulti-sidedinnature.

6.2.3. Intraditionalsingle-sidedmarkets,afirm’smarketsharewilltypicallybetheproportionofsalesofaparticulargoodorservice. Inmulti-sidedmarkets,aswellasthemarketfortheparticulargoodorservice,theprocessoffacilitatingtransactionsbetweendistinctsidesmayitselfbeconsideredarelevantmarket.Onemayconsideraplatform’smarketsharetobetheproportionoftransactionsinamarketthatitfacilitates,orthenumberofplatformusersthatitserves(e.g.buyersandsellers),assuggestedbyonequestionnairerespondent.

6.2.4. Competition authorities therefore face the challenge in decidingwhether to define one relevantmarketcomprisingallsides,orseparaterelevantmarketsoneachside.Thereisnogeneralruleonthis,howeversomeconsiderdefiningonesingle relevantmarketmoreappropriatewhenamulti-sidedplatformfacilitatestransactionsbetweensides(as,forexample,isthecaseforcreditcards),ratherthanjustprovidingaccessforusers.171Nevertheless,consensushasyettobereachedonthecorrectapproach172andamorethoroughapproachconsideringboththecaseofseparaterelevantmarketsandofonecombinedmarketincludingallsidesshouldbepursuedandevaluatedcase-by-case.

6.2.5. Beforedefiningamarket,competitionauthoritiesshoulddeterminewhetheramarketistrulymulti-sidedinnature.Thereisnostandarddefinitionofamulti-sidedmarket,howeveroftenineconomicliterature,amulti-sidedmarket isdefinedasonewherean intermediaryservesmultipledifferentcustomergroups,andthereareindirectnetworkeffectsbetweenthesesidesofthemarketwhichaffectsthepricethatissetonbothsides.173Inpractice,determiningwhetheramarketismulti-sidedinnaturemayrequireacross-elasticityanalysistoshowthatpriceononesideofthemarketaffectsdemandonanotherside.

171 Filistrucchi (2017).

172 As evidenced by the OECD Hearing on “Rethinking the Use of Traditional Antitrust Enforcement Tools in Multi-Sided Markets” held in Paris on 22nd June 2017.

173 See, for example, Hagiu and Wright (2015).

48

6.2.6. Ifamarket ismulti-sided innature,thestandardSSNIPtest isunlikelytoapply,as it isbasedonassumptionsoriginatingfromone-sidedmarketswhichdonotapplytomulti-sidedmarkets.

6.2.7. Inone-sidedmarkets,afirmtypicallyprofitmaximisesbysettingpriceabovemarginalcost.However,inmulti-sidedmarketspricingbelowcostononesideofamarketisacommonstrategyemployedbyfirmsinordertoattractuserstoanothersideofthemarket.Anincreaseinpriceabovemarginalcostinaone-sidedmarkettypicallydecreaseswelfareassomeconsumersarenolongerserved.Bycontrast,inmulti-sidedmarketsanincreaseinpriceononesideofthemarketdoesnotnecessarilymeanthattotalwelfarehasfallenorthatquantitysolddeclinesbecauseoftheinterdependencyofdemandbetweenthetwosides.Forexample,whenconsideringatwo-sidedmarket:ifatthesametimeaspriceisincreasedonsideA,thepriceonsideBisreduced,consumerwelfareonsideBmayriseandpotentiallyoffsetorexceedanylossinwelfareonsideA.Furthermore,consumerwelfaredoesnotnecessarilyfallonsideA,asthefallinpriceonsideBmayincreasethevaluetousersonsideA,therebyincreasingdemand,andpossiblyoffsettingorexceedingthedeclineindemandfromthedirectincreaseinprice.

6.2.8. Asaresultofthedifferingcharacteristicsofmulti-sidedmarkets,whenapplyingtheSSNIPtestEvans(2003)proposedthatinsomeinstancesthetotalpriceshouldbeconsideredinsteadoftheseparatepriceschargedtoeachsideindividually(i.e.thesumofthepricechargedtoallsidesofthemarket).HesseandSoven(2006)adoptedthisapproachwhendefiningelectronicpaymentnetworkmarketsintheUS.

6.2.9. Intheory,aSSNIPtestcanbeappliedtomulti-sidedmarketsusingtotalprice,howeverinrealitythisisnotalwayspossible.Multi-sidedmarketsareoftenhighlycomplexinthataplatformmaycompetewithsingle-sidedfirmsonsomesidesofthemarketand/orplatformsthathavemoreorfewersidesthanitself.Additionally,inapplyingaSSNIPtestauthoritiesmustaskthemselvesonwhichside(s)ofthemarkettheincreaseintotalpriceshouldbeapplied,andwhattherelativedistributionofthispriceincreaseshouldbe.Also,inmanyinstancesusersofaplatformdonotpayapricepertransaction,butratherapricetoaccessaplatform,thereforemakingacalculationofapricepertransactionverychallenging.

6.2.10. Whereamarketcannotbedefinedusingthetraditionalmarketdefinitiontestsbasedonone-sidesubstitutability,asaresultofthesecomplexities,authoritiesmaybebetterplacedadoptingalessrigidmethodologyforthedefinitionofthemarket.Whenadoptingsuchanapproach,allcompetitiveconstraintsthatafirmfacesoneachsideofthemarketshouldbeconsidered;forexamplefromotherplatforms,one-sidedfirms,and/orpotentialentrants.Authoritiesshouldconsidertherelationshipsbetween all sides of themarket, specifically focusing on the presence and direction of networkeffectsonbothsidesofthemarket,andanysubsequentfeedbackeffects.Thepricelevelshouldbeconsideredforeachsideofthemarket,aswellasthetotalpriceforallsidesanditsstructure.

6.2.11. Whendefiningtherelevantgeographicmarketinonlinemulti-sidedmarkets,equivalentadaptationstothestandardapproachmayberequiredaswhendeterminingtherelevantproductmarkets.Asinthegeneralcaseforsingle-sidedmarkets, inonlinemulti-sidedmarkets,onewouldexpecttherelevantgeographicmarkettobewiderthaninequivalentbrick-and-mortarmarketsbecauseoftheirwidergeographicreach.

6.2.12. In summary, in somemulti-sided markets distinct sides transact directly with one another. Forexample,foranonlinemarketplaceonetransactionwouldbethesaleofagoodfromaretailertoaconsumer. Inothermulti-sidedmarketstherelationshipbetweendistinctsides is lessclear.Forinstance, anadvertiserdoesnotdirectly interactwithan individualnewspaper reader; insteadanadvertiserischargedafeetoaccessallreaders.Althoughtherelationshipbetweensidesisarguablyclearerwhendistinctsidesofamarkettransactonaone-to-onebasis, it isnotnecessarilyeasiertodefinethesemarkets.Forexample,anonlinemarketplacefacescompetitiveconstraintsnotonlyfromotheronlineplatforms,butalsofromsingle-sidedonlineretailersandbrick-and-mortarshops,aswellasmulti-sidedbrick-and-mortarshoppingmalls.

6.2.13. There isthereforenogeneral ruleastohowtodefineamulti-sidedmarket. Instead,competitionauthoritiesshouldcarefullyconsiderallaspectsofcompetitionthataplatformfaces.Practicallythisinvolvesassessingthemarketforfacilitatingtransactionsbetweeneachdistinctside,thecompetitionthattheplatformfacesoneachside,therelationshipsbetweenallsides,andanyfeedbackeffects.

49

Assessing market power in multi-sided markets

6.2.14. QuestionnaireresponseshighlightthechallengesthatcompetitionauthoritiesinASEANarecurrentlyfacingwhenassessingmarketpowerinmulti-sidedmarkets,particularlywhereaccesstoaplatformisfreeforonedistinctgroupofusers.Whenassessingmarketpowerinmulti-sidedmarkets,allsidesofthemarketmayneedtobeconsideredtogether,aswellasinisolation.Inassessingwhetherafirmisabletosustainsuper-normalprofits,insomeinstancesthecostsincurredandpriceschargedtoallsidesofthemarketshouldbetakenintoaccount.Thisismorelikelytobethecasewhenamulti-sidedplatformfacilitatestransactionsbetweensides,insteadofprovidingaccessforusers.Moreover,thedegreeofdemandandsupplysubstitution,andthelevelofactualandpotentialcompetitionthatafirmfacesshouldbeconsideredonallsidesofthemarket.

6.2.15. Duetothe importanceofexternalitiesbetweensides, itmaybesufficient foraplatformtohavemarket power on only one side of amarket for it to be in a strong position in themarket as awhole, throughthecreationofacompetitivebottleneck.Thedirectionandmagnitudeofnetworkeffectsbetweensidesofaparticularmarketshouldthereforebeanimportantconsiderationintheassessmentofmarketpower.AsdiscussedinSection4,networkeffectsmayconstituteabarriertoentryandexpansioninsomemarketsbyimposingaswitchingcostonusers,howeverthisbarriermaybemitigatedifmulti-homingiscommon.

6.2.16. Section6.3providesamoregeneraldiscussionontheassessmentofmarketpowerinonlinemarkets,withaparticularfocusontheroleofdynamiccompetitionfrominnovation.First,theassessmentofharminmulti-sidedmarketsisconsidered.

Assessing harm in multi-sided markets

6.2.17. Theoverallneteffectfromcertainformsofconductbyfirmsinmulti-sidedmarketsisoftenunclear.Inaone-sidedmarket,anyincreaseinpriceabovemarginalcosttypicallyreducestotalwelfarebycreatingadead-weightloss. Inmulti-sidedmarkets,however, it ismorecomplex.Consideracasewhereaplatformincreasespriceforonesideofamarket,andreducespricefortheothersideofthemarket.Assumingtherearepositiveexternalitiesbetweensides(i.e.moreusersononesideisbeneficialfortheotherside),theoveralleffectonwelfareforbothsidesisunclear.

6.2.18. Ifoneextendsthisanalysistoconsidermoreintricateformsofconduct,suchastyingandbundling,thewelfareanalysisbecomesevenmorecomplex.

6.2.19. Competitionauthoritiesalsofacethechallengeofdecidingwhetherharmtoonesideofamarket(i.e.afallinwelfare)canbeoffsetbybenefitstoanotherside(i.e.ariseinwelfare).Ifwelfareeffectstobothsidesareconsideredtogether,furtherquestionsemerge,suchas:Areallsidesweightedequally?Ifonesideconsistsofbusinessesandtheothersideisconsumers,shouldtheybetreatedequally?Typicallycompetitionauthoritieswouldpursueconsumerwelfare,however,somejurisdictions,suchas Singapore, adopt a totalwelfare test. If one side of themarket is represented by businesses,whatweightshouldbeputontheirwelfareincrease?174Furthermore,thequestionisnotalwaysassimpleaswhetherharmtoonesideofthemarketcanbeoffsetbybenefitstoanotherside.Insomeinstances,withoutcertainconductbyfirms,amarketmaynotexistatall.Asanexample,itmaybehardforaplatformtoattractusersononesideofthemarketunlessapricebelowmarginalcostischarged.

6.2.20.Considertheexampleofaplatformsettingpriceabovemarginalcostforonesideofthemarketandbelowmarginalcostfortheotherasaresultofexternalitiesbetweenthetwosides.Ifeachsideisevaluatedseparately,aplatformcouldbeaccusedofpredatorypricingononeside,andexcessivepricingontheotherside.However,thisbusinessmodel iscommoninonlinemulti-sidedmarketssuchasonlinesearchandsocialmedia,andiswidelyacceptedbycompetitionauthorities.Thispointdemonstratesthat,whenassessingallegedharm,competitionauthoritiesshouldatleastconsiderthe network effects between distinct sides of amulti-sidedmarket and any additional feedbackeffects.

174 This is especially true if the rise in welfare translates into increased consumer welfare for the business’ final customers.

50

6.2.21. Insightsfrompreviouscasesinmulti-sidedmarketsarelimited.Sofar,internationalconsensushasyettobereachedonwhetherharmononesideofamarketcanbeoffsetbythebenefittoanotherside,asoutlinedinthefollowingparagraphs.

6.2.22. In the US, in the recent investigation of non-discrimination provisions implemented byAmericanExpress(seeCasereview2),175theSecondCourtofAppealsfoundthat“thetwosidesoftheplatform[merchantsandconsumers]cannotbeconsideredinisolation”.ItwasalsoconcludedthattheDistrictCourt’spreviousanalysis:

“erroneously elevated the interests of merchants above those of cardholders” and that “the market as a whole includes both cardholders and merchants, who comprise distinct yet equally important and interdependent sets of consumers sitting on either side of the payment-card platform”.

6.2.23.Bycontrast,intheEUinvestigationofMasterCard’smultilateralinterchangefees(MIF)(seeCasereview1),176theEuropeanCourtofJustice (ECJ)concluded in itsreviewoftheGeneralCourt’sdecisiontoprohibittheconductthat:

“the General Court took into account the two-sided nature of the system, since it analysed the role of the MIF in balancing the ‘issuing’ and ‘acquiring’ sides of the MasterCard system, while recognising that there was interaction between those two sides. Furthermore, in the absence of any proof of the existence of appreciable objective advantages attributable to the MIF in the acquiring market and enjoyed by merchants, the General Court did not need to examinetheadvantagesflowingfromtheMIFforcardholders,sincesuchadvantagescannot,by themselves, be of such a character as to compensate for the disadvantages resulting from those fees.” 177

6.2.24. InSingapore,MIFhasalsobeenassessedaspartofanotificationfordecisionfiledbyVisaWorldwidePteLtd.InSeptember2013,followinganin-depthreviewCCSapprovedtheuseofVisaWorldwidePteLtd.’sMIFsystem(seeCasereview3).178Initsdecision,CCShighlightedhow:

“in conducting the analysis of the Visa Group’s MIF system, it is important to recognise how the separate markets are inter-related in the context of a two-sided platform, and how the actions in one market can directly affect the other markets and vice versa.”

175 United States et al. v. American Express Company et al., No. 15-1672 (2d Cir. 2016).

176 C-382/12 P - MasterCard and Others v Commission.

177 Judgment in Case C-382/12 P MasterCard Inc. and Others v Commission, Press release No 122/14.

178 CCS 400/001/06 (2013).

51

CASE REVIEW 1 – MASTERCARD MIF

Industry: Payment systems

Country / Union of countries: EU

Court / Competition Authority: European Commission, ECJ

Case name and citation: C-382/12 P

Date of decision: 11th September 2014

Type of alleged infringement: Horizontal price coordination

3 Case summary

InDecember2007,theEuropeanCommissionfoundMasterCardtohaveimplementedclausesinitsMIFsystemthatineffectsetapriceflooronretailersandthereforerestrictedcompetitiononpricebetweenpaymentsystems.ConsequentlyMasterCardwasrequiredtoremovetheMIFwithin6months.MasterCardinitiallyappealedtotheGeneralCourtbeforethenappealingtotheECJ.Inbothcasestheappealwasdismissed.

Importantly,indismissingtheappealtheECJruledthat:“theGeneralCourttookintoaccountthetwo-sidednatureofthesystem,sinceitanalysedtheroleoftheMIFinbalancingthe‘issuing’and‘acquiring’sidesoftheMasterCardsystem,whilerecognisingthattherewas interactionbetweenthosetwosides.Furthermore,intheabsenceofanyproofoftheexistenceofappreciableobjectiveadvantagesattributabletotheMIFintheacquiringmarketandenjoyedbymerchants,theGeneralCourtdidnotneedtoexaminetheadvantagesflowingfromtheMIFforcardholders,sincesuchadvantagescannot,bythemselves,beofsuchacharacterastocompensateforthedisadvantagesresultingfromthosefees.”

CASE REVIEW 2 – AMERICAN EXPRESS MIF


Country / Union of countries: US

Court / Competition Authority: US DoJ, US Court of Appeals for the Second Court

Case name and citation: No. 15-1672


Type of alleged infringement: Restriction of price competition through a vertical restraint

3 Case summary

AmericanExpresswas investigated in theUSfor itsuseofnon-discriminationprovisionswhichpreventedretailersfromofferingdiscountstocustomerswhouseothercards(whichincurlowerfeesfortheretailers).

InitiallytheUSDoJdeemedtheseagreementstobeanti-competitiveonthebasisoftheeconomicharmcausedtoretailers.However,followinganappeal,theUSCourtofAppealsfortheSecondCourt reversedthe initialdecisiononthebasisthatthe lowercourt’sanalysisfailedtotake intoaccountthemulti-sidednatureof themarketandtheeffectof theconductoncustomersandretailers.

52

CASE REVIEW 3 – VISA MIF


Country / Union of countries: Singapore

Court / Competition Authority: CCS

Case name and citation: CCS 400/001/06

Date of decision: 3rd September 2013

Type of case: Notification

3 Case summary

VisaWorldwidesoughtapprovalfromCCStoimplementitsMIFsysteminSingapore.

InSeptember2013,followinganin-depthreview,CCSapprovedVisa’sMIF.CCSassessedtheeffectofthepricingsystemonthreerelevantmarkets;namely,theissuingmarket,theacquiringmarketandthecardschememarket.

CCSfoundthatinboththeissuingmarketandthecardschememarketitwasunlikelytherewouldbemorecompetition intheabsenceofVisa’sMIF;and, intheacquiringmarket itwasnotclearwhethercompetitionwouldbesignificantlygreaterintheabsenceofVisa’sMIF.

ItwasalsofoundthatintheabsenceofVisa’sMIF,barrierstoentryandexpansionwouldlikelybehigherforsmall,ornew,acquirers.

InitsdecisionCCShighlightedhow:“inconductingtheanalysisoftheVisaGroup’sMIFsystem,itisimportanttorecognisehowtheseparatemarketsareinter-relatedinthecontextofatwo-sidedplatform,andhowtheactionsinonemarketcandirectlyaffecttheothermarketsandviceversa.”

6.3. Assessing market power and the impact of dynamic competition from innovation

Impact of E-commerce in assessment of market power

6.3.1. TheriseofE-commercehasnotsubstantiallychangedtheassessmentofmarketpowerinsingle-sidedmarkets.Typically,marketpowerisdefinedas“theabilityoffirmstoincreasepricesprofitablyabovethecompetitivepriceforasustainedperiod”,179regardlesswhetherthisisanonlineorabrick-and-mortarsetting.Asthecompetitivepriceisoftendifficulttoidentifyinpractice,anindirectassessmentisgenerallyrequiredtodeterminewhetherafirmenjoysmarketpower.Calculatingmarketsharesofthecompanyinquestionandofrivalcompaniessupplyingsubstitutablegoodsorservicesremainsacommonapproachforexaminingafirm’spositioninamarket,andisapplicabletobothonlineandtraditionalbrick-and-mortarmarkets.

6.3.2. As in brick-and-mortar markets, other factors beyondmarket share should be consideredwhenassessingmarketpower,inparticular,theabilityofsmallerfirmstoexpand,andnewentrantstojointhemarket.Barrierstoentry,suchasnetworkeffectsandswitchingcostsforconsumers(asdiscussedinSection4),shouldalsobeconsidered,aswellasthecountervailingbuyerpowerofcustomersshouldalsobetakenintoaccount.

6.3.3. SomeconsiderBigDataasafactorthatshouldbetakenintoaccountwhenassessingdominance,asitmayrepresentanassetthatcompetitorsareunabletoreplicate,andthereforecompriseabarriertoentry(asdiscussedinSection4).


53

180 European Commission (2009), para. 13.

181 Malaysia Competition Commission (2010).

182 Commission for the Supervision of Business Competition of the Republic of Indonesia (1999).

183 ASEAN (2013), page 9.

Assessing dominance in ASEAN

6.3.4. Marketsharesarewidelyregardedasausefulfirstindicationofthestructureofagivenmarket,andtheimportanceofthevariousoperatorsactiveinit.180Itisalsousefulforauthoritiestolookathowmarketshareshaveevolvedovertime,ratherthantakingastaticview.Differentcompetitionauthoritiesapplydifferentmarketsharethresholdswhenassessingafirm’smarketpower.Themarketsharesofmorefirmsinamarketmayalsobeconsideredwhenlookingintowhetherthereisapositionofcollectivedominanceinaparticularmarket.

6.3.5. Table 9, below, provides a breakdown of themarket share thresholds in Europewhen assessingmarketpower,emergingfromcaselaw:

Table 9: Market shares thresholds for the assessment of dominance in Europe

Market share

100%

80%

50%

40%

Source:Whish,R.andBailey,D.(2015),pages48-51.

Description

Completemonopolist.

Afirmwithamarketshareabove80%isinapositionof‘super-dominance’,wheretheirconductislikelytohaveastrongadverseeffectonthemarket,andthereforeislikelytofallunderArticle102oftheTreatyoftheFunctioningoftheEuropeanUnion(TFEU).

There is a legal presumption that afirmwith amarket shareof 50%ormore is in adominantposition.Thispresumptionappliesinthecaseofcollectivedominance,aswellassingle-firmdominance.

Afirmwithamarketshareof40%maybeconsidereddominantunderArticle102oftheTFEU. IntheUK,theCompetitionandMarketsAuthority(CMA)considersafirmwithamarketsharebelow40%‘unlikely’tobeinadominantposition.

6.3.6. TheassessmentofdominanceundercompetitionlawvariesacrossAMSandisnotalwaysasprescriptiveasTable9above.ForexampleinMalaysia,Section10(4)oftheCompetitionAct2010 181states:

“the fact that the market share of any enterprise is above or below any particular level shall not in itself be regarded as conclusive as to whether that enterprise occupies, or does not occupy, a dominant position in the market’’.

6.3.7. Incontrast,theCommissionfortheSupervisionofBusinessCompetitionoftheRepublicofIndonesia(KPPU)states:

“Business actors shall be reasonably suspected or deemed to control the production and or marketing of goods and or services… if one business actor or a group of business actors controls more than 50% of the market share of a certain type of goods or services”. 182

6.3.8. Table10presentsadetailedbreakdownofthedefinitionsandmarketsharethresholds(whererelevant)fordominanceinAMS.ThedifferencesacrossASEANintheassessmentofdominancemayraisesomedifficultiesinensuringaconsistenttreatmentofE-commerceoperatorswhentheiroperationsspanmultiple jurisdictions.Competitionauthorities shouldbewaryof thedifferingcriteriaofdominancethatmaybeinplaceifassessinganti-competitivebehaviouracrosstwoormoreAMS.ThiswouldavoidsituationswheresimilarcasesreachcontrastingjudgementsindifferentAMS,causinguncertaintyandriskforfirmsoperatingwithinASEAN.Atpresent,thegeneralviewofdominanceacrossASEANis:

“a situation where the business operator has enough economic strength to act in the market without regard to what its competitors (actual or potential) do.” 183

54

Table 10: Market shares and dominance in AMS

ASEAN Member State

BruneiDarussalam

Cambodia

Indonesia

LaoPDR

Malaysia

Myanmar

ThePhilippines

Singapore

Thailand

Vietnam

Description

Dominant position means a situation in which one or moreundertakingspossess such significantpower in amarket to adjustpricesoroutputsortradingterms,withouteffectiveconstraintfromcompetitors or potential competitorswithin Brunei Darussalam orelsewhere.

Dominant position means a situation in which a person, eitherindividuallyortogetherwithotherpersons,isinapositioninamarketto act without effective constraint from competitors or potentialcompetitors

Businessactorsshallhaveadominantpositioninthefollowingevents:

a.Onebusinessactororagroupofbusinessactorscontrolsmorethan50%ofthemarketshareofacertaintypeofgoodsorservices;or

b.Twoorthreebusinessactorsoragroupofbusinessactorscontrolmorethan75%ofthemarketshareofacertaintypeofgoodsorservices

-

Dominantpositionmeansasituationinwhichoneormoreenterprisespossesssuchsignificantpowerinamarkettoadjustpricesoroutputsor trading terms,without effective constraint from competitors orpotentialcompetitors.

Thefactthatthemarketshareofanyenterprise isaboveorbelowanyparticularlevelshallnotinitselfberegardedasconclusiveastowhether thatenterpriseoccupies, ordoesnotoccupy, adominantpositioninthemarket.

No market share thresholds have been provided for a dominantpositionbasedonMyanmar’sCompetitionLaw(2015).

Dominant position refers to a position of economic strength thatanentityorentitiesholdwhichmakes itcapableofcontrollingtherelevant market independently from any or a combination of thefollowing:competitors,customers,suppliers,orconsumers.

Thereshallbearebuttablepresumptionofmarketdominantpositionif themarket share of an entity in the relevantmarket is at least50%, unless a new market share threshold is determined by theCommissionforthatparticularsector.

ThereisnodefinitionofdominancegiveninSingapore’sCompetitionAct.However,CCSconsidersamarketshareinexcessof60%aslikelytoindicatethatanundertakingisdominantintherelevantmarket.184

The Commission shall have the power to issue a written orderrequiringabusinessoperatorwhohasmarketdomination,withthemarketshareofmorethan75%,tosuspend,ceaseorvarythemarketshare.

UnderThailand’smergerregimeafirmisregardedasdominantiftheirshareofthemarketisgreaterthan50%.185

Anenterpriseshallbedeemedtobeinadominantpositionifsuchenterprisehasamarketshareof30%ofmoreintherelevantmarketoriscapableofsubstantiallyrestrainingcompetition.

Market share

-

-

50%

-

-

-

50%

60%

75%

30%

184 Getting The Deal Through (2017).

185 Allen & Overy (2017).

Source:InformationobtainedfromtherespectiveAMS’scompetitionlawunlessstatedotherwise.

55

186 GCR (2017), page 36.

187 The Economist (2006).

6.3.9. Theharmonisationof competition lawwithinASEANwashighlighted in theMarch2017GCRLiveAsia-PacificLawLeadersForum.Specifically,itwasoutlinedhow,together,AMSareidentifyingthedifferencesandsimilaritiesofeachAMS’scompetitionpolicyandlaw.186Giventhedifferentrulesfortheassessmentofdominance,thisisapotentialareathatmaybenefitfromaharmonisedapproach.A standardised approach would avoid situations in which competition authorities in AMS reachinconsistentjudgementsinsimilarcasesdealingwiththesameoperatorandthesameconduct.

Importance of dynamic competition in E-commerce markets

6.3.10. Intheassessmentofafirm’smarketpower,manyfactorsbeyondafirm’smarketshareareimportant,such as the extent towhich barriers to entry and expansion restrict or prevent new or smallerfirmsfromenteringorexpandinginthemarket(asdiscussedinSection4ofthishandbook).ThisisparticularlythecaseinE-commercemarketswheresuccessfulinnovationcanresultinanincumbentfirm rapidly losing or gainingmarket share. Taobao’s displacement of eBay as the leading onlinemarketplaceinChinademonstratesthispoint.187InassessingmarketpowerinE-commercemarkets,competitionauthoritiesshouldthereforealsoconsiderthelongrundynamicsofmarketsinadditiontostaticmarketshareanalyses,asdiscussedbyAffusoandHall(2016).

6.3.11. Many online markets are interconnected through so-called ‘digital eco-systems’. Due to theinfrastructurethatlargeonlinefirmshavedeveloped,enteringneworadjacentmarketscanbemucheasierthaninofflinemarkets,astechnologiesmaybeeasilyadaptedtoserveasimilarpurposeinarelatedonlinemarket.Forexample,Google,Amazon,Apple,andFacebookallofferarangeofservicesbasedonasetofadaptabletechnologiesandcapabilities.Indeed,entrybythesefirmsintomarketsthatarebeingledbyanotheroftheseglobalplayersiscommon;forexample,GoogleenteredthesocialmediamarketwithGooglePlus,andAmazonexpandedfromitscorecompetencyasaretailmarketplacetoproducedevicessuchastheKindleandFireincompetitionwithApple’siPad.

6.3.12. Asaresultoftheadaptabilityofonlinefirms,playersoutsideofwhatmaybeconsideredtherelevantmarket can still impose a competitive constraint on activemarket players.Afirmwhichmaybedefinedasoutsideoftherelevantmarketcanthereforestillconstrainanoperatorwithasignificantpresencein(andshareof)themarket.

6.3.13. Alibaba’s presence in online marketplaces, electronic payment services, and cloud computingdemonstrateshowfirmscansuccessfullyexpandintoonlineservicesoutsideoftherelevantmarketfromwhichtheyoriginate.Consequently,ifanonlinefirmweretotrytotakeadvantageofitsstrongmarketshareinaparticularmarket,forexamplebychargingexcessivelyhighpricesorreducingqualityofservicetoconsumers,otherfirmsoperatinginrelatedmarketsbasedonsimilartechnologiesmayenterandquicklydisplacetheincumbent.

6.3.14. Neighbouring online platforms, operating in a different market, can be thought of as potentialinnovatorswhomaybeabletodevelopalternative,betterproductsorservices inthefuture.Thisdynamiccompetitionincentivisesincumbentfirmstocontinuallyinnovatetoensuretheymaintaintheirpositioninthemarket.Intheabsenceofapotentialinnovator,forexamplefollowingamerger,the incumbentfirmmay face reduced incentive to innovate, leading toa reduction in thequalityreceivedbyconsumersinthelongrun.

6.3.15. As a result, dynamic competition, and the role of firms outside the relevantmarket should alsobeconsideredascompetitiveconstraintsincompetitioninvestigationswhenassessingthemarketpowerheldbyafirminonlinemarkets.

56

7.1. Introduction

7.1.1. Verticalagreements,orverticalrestraints,arebroadlydefinedasagreementsbetweenfirmsatdifferentlevelsinthesupplychain.Mostcommonly,verticalrestraintsimposerestrictionsonretailerssellingamanufacturer’sproduct;forexample:whethertheyareallowedtoselltheproduct;conditionsthatmustbemetforthemtobeabletoselltheproduct;whotheycansellto;thepricetheycanselltheproductat;orthequantityoftheproducttheymustbuy/sell.

7.1.2. Firstly, this section looks at the challenges faced by competition authorities in analysing verticalrestraints in E-commerce markets and, following that, outlines recommendations on how besttoassesstheserestraints.Examplesofcasesfromacrosstheworldarepresentedthroughouttosupportthediscussion.

7.2. Challenges faced by Competition Authorities in the assessment of vertical restraints

7.2.1. Verticalrestraintsaregenerallyconsideredasabenignbusinesspracticeastheygiverisetoanumberofefficiencies.However,verticalrestraintscanposeachallengetocompetitionauthoritiesiftherearebothpro-andanti-competitiveeffectsfromsuchagreements.Thissectionlooksattheseeffectsinturn,beforeevaluatingtheimpactthatE-commercehashadontheuseofsuchrestraints.

Potential pro-competitive effects of vertical restraints

7.2.2. Verticalrestraintscanhavepro-competitiveeffects,posingachallengetocompetitionauthoritiesinassessingwhethertoallowsuchagreements.

Vertical restraints to overcome free-riding problems

7.2.3. Verticalagreementsareoftenneededtoovercomeissuesoffree-ridingwhichresultintheunder-provisionofimportantpre-orpost-salesservices.EvidencefromtheUKandEuropeindicatesthatthisisacommonreasonamongfirmsfortheuseofverticalrestraintsinonlinemarkets.188Consider,forexample,themarketforcontactlenses:consumersmaymakeuseofthepre-saleserviceinabrick-and-mortarstore,tryingvarioustypesofcontact lenses,andassessingwhichsuitthemthebest.Theymaythenpurchasetheselectedcontactlensesonlineatacheaperprice.Inthisexample,theonline store is free-riding on thepre-sales service of thebrick-and-mortar store.Conversely,consumersmaymakeuseofonlinepre-salesservices(e.g.pricecomparisonandcustomerreviews),beforepurchasinginstore.Onlinestoresmaybeabletosellatalowerpricethanbrick-and-mortarstoresastheydonotincurthecostsrelatingtopre-saleservicesthatbrick-and-mortarstoresface.In theabsenceofvertical restraints, inorder forbrick-and-mortar stores to competewithonlinechannelsonprice,theymayreducepre-salesservicequalityinordertoreducecosts.Consumersmaythereforebeworseoffduetotheunder-provisionofpre-salesservices.Selectivedistributionsystemsareoftenusedtoovercomesuchconcerns–forinstanceamanufacturermayonlyallowretailerswhoofferacertainlevelofpre-orpost-salesservicetoselltheirproducts.

188 UK - Oxera and Accent (2016) – commissioned by the CMA, and European Commission (2016).

Vertical agreements

07

57

189 Lee, I. (2016), page 2319.

190 Case C-230/16, Coty Germany GmbH vs. Parfümerie Akzente GmbH (ongoing).

191 Case C-439/09, Pierre Fabre Dermo-Cosmétique SAS v Président de l’Autorité de la concurrence and Ministre de l’Économie, de l’Industrie et de l’Emploi (October 13, 2011).



7.2.4. Anotherissuewhichverticalrestraintscanhelptoovercomeisthehold-upproblem.Insomeinstances,thereisaneedforafirmtomakearelationship-specificinvestmentinaverticalrelationship(i.e.withaspecificfirm,up-ordown-stream)beforemakinganysales.Forexample,amanufacturermayneedtoinvestinnewmachineryinordertofulfilanorderwithaparticularretailer,oraretailermayneedtoinvestintrainingitsstaffpriortosellingaspecificmanufacturer’sproduct.Intheabsenceofaverticalagreement,theseinvestmentsmaynotbemadeas“oncerelationshipspecificcostshavebeensunkbyoneparty,anotherpartymayopportunisticallyseektorenegotiatetermsinitsfavour.”189

Vertical restraints to protect a product's image

7.2.5. Overcomingissuesoffree-ridingisgenerallyacceptedbycompetitionauthoritiesasapro-competitivebenefitofverticalrestraints.However,caselawislessclearonwhetherthisreasonisalegitimatejustificationfortheuseofverticalrestraintsbyfirms.Specifically,afirmmaytrytoincreaseusers’valuationofaproductbydevelopinga ‘luxury’brand imagesothatowningthatgoodbecomesasignalofthatconsumer’sstatusinsociety.Designerhandbagsareoftenconsideredinthiscategoryofgoods.Todevelopastatusimage,amanufacturermayuseaselectivedistributionsystem,onlyallowingpremiumretailerstostockitsproducts,therebyincreasingthevaluethatconsumersplaceonthegood.AlandmarkcaseonthisissueiscurrentlyunderreviewinEuropeconcerningthebeautyproductmanufacturer,Coty190 (seeCasereview16).TheECJisconsideringwhetherensuringaluxuryimageisavalidreasonforpreventingsalesthroughonlinemarketplaces.Interestingly,theECJhaspreviouslyruledonasimilarcaseconcerningPierreFabre191(seeCasereview4),amanufacturerofluxurycosmeticproducts.In2011,theECJagreedwiththeParisCourtofAppealthatPierreFabre’srestrictionsonretailerstoonlysellthroughphysicalstoresinthepresenceofaqualifiedpharmacistrestrictedcompetitionasallonlinesaleswereprevented.

7.2.6. Theremayalsobesignallingbenefitsfromverticalrestraints,whereby:

“certain retailers have a reputation for stocking only "quality" products. In such a case, selling through these retailers may be vital for the introduction of a new product. If the manufacturer cannot initially limit his sales to the premium stores, he runs the risk of being delisted and the product introduction may fail.” 192

7.2.7. Intheabsenceofaverticalrestraintallowingonly‘quality’retailerstoselltheproduct,consumersmaybeworseoffinthattheproductmaybediscontinuedfollowingafailedintroductoryperiod.

Vertical restraints to avoid double marginalisation

7.2.8. Verticalrestraintsmayalsobebeneficialforconsumersastheycanleadtolowerpricesinmarketsbyovercomingissuesofdoublemarginalisation.Doublemarginalisationariseswhenfirmsinaverticalrelationship both havemarket power (i.e. they can both set prices abovemarginal cost). In thissituation,theupstreamfirmsetsitspricesabovethemarginalcostofproduction,inadditiontothedownstreamfirmsettingasubsequentmark-upaboveitsinputprice,bothunilaterallymaximisingprofit. As a result the retail price has beenmarked up twice.193 If however, through the use of averticalagreement,amanufacturerandretaileragreedtocoordinatepriceorquantitysoldinordertomaximisejointprofits,theretailpriceforconsumerswouldfallasonlyasinglemark-upwouldbeappliedtothegood,andindoingsoincreasetotalwelfare.

58

Potential anti-competitive effects of vertical restraints

7.2.9. Vertical restraintscanhavetheeffectoffacilitatingcollusion, limiting inter-brandcompetition (i.e.competition between different brands) and/or limiting intra-brand competition (i.e. competitionbetweenproductsofthesamebrandsold indifferentoutlets).Whenthere isadegreeofmarketpoweratthelevelofthesupplier,buyer,orboth,anti-competitiveeffectsarelikelytobegreater.194

7.2.10. Forexample,anagreementbetweenamanufacturerandretailersthataspecificproductcannotbesoldbelowaspecifiedpricewouldlimitintra-brandcompetitionandmayfacilitatecollusionbetweenretailers.Asanotherexample,anagreementbetweenamanufacturerandaretailerspecifyingthattheretailermuststockatleastacertainquantityofaproductmaylimittheabilityoftheretailertostockcompetingmanufacturers’products,therebyharminginter-brandcompetition.

7.2.11. If there isstrong inter-brandcompetition,a reduction in intra-brandcompetition isunlikelytobeharmfultoconsumersasretailerssellingcompetingproductsimposeacompetitiveconstraintonanyretailerthatsellsahighproportionofaparticularmanufacturer’sproducts.Forthisreasonrestrictionslimitinginter-brandcompetitionaremoreofaconcerntocompetitionauthoritiesthanrestrictionslimiting intra-brand competition. For example, theremay be an extreme scenariowhere there isonlyasingleretailersellingamanufacturer’sgood,assumingfornowthatinter-brandcompetitionisstrong(i.e.thereisanabundanceofretailerssellingothermanufacturers’products).Ifthisretailertriedtoexploititspositionofpowerbyraisingitspriceand/orreducingservicequality,consumerswouldswitchtoanalternativeretailersellingasimilargoodproducedbyadifferentmanufacturer.Bycontrast, if inter-brandcompetition isweak,marketoutcomesaremore likelytobeadverselyaffected.As a result, in bothonlineor offlinemarkets, inter-brand competition isvital toprotectandpromoteconsumerinterests.TheUSantitrustauthoritiesapplythislogicintheirrule-of-reasonapproachtotheassessmentofverticalagreements.195

7.2.12. Someformsofverticalrestraintsrestricttheabilityofconsumerstoengageintransactionsinotherterritories.ThisisespeciallythecaseinE-commercemarketswherecross-bordertransactionsarecommon.Ininstanceswhereseveralcountriesagreetopromotefreemovementofgoodswiththeaimofpromotingcross-bordertrade,asforexampleintheEU,thistypeofrestrictionmayviolatethefreemovementruleandharmconsumers.Instanceswhereabusinessimposesrestrictionsinhibitingcross-bordertradebetweenMemberStatesarereferredtoasgeo-blockingstrategies.TheevidenceinEuropesuggeststhatthisiswidespread.196SuchmeasuresmaybeofconcerntoASEANasitmovesclosertowardsanintegratedmarket,throughtheimplementationofitsASEANEconomicCommunityBlueprint2025,theASEANCompetitionActionPlan(2016-2025)andICTMasterplan2020.

Comparing the pro- and anti-competitive effects of vertical restraints

7.2.13. Inthoseinstanceswhenverticalrestraintsarenotunequivocallybeneficial,competitionauthoritiesfacethechallengeofassessingwhetherverticalrestraintsareanti-competitiveandharmconsumers,orwhetherthebenefitstheygenerateforconsumersoutweightheharm.Generallyspeakingverticalrestraints are considered beneficial, and are typically permitted unless there are potential anti-competitiveeffects.

7.2.14. Insomejurisdictions,thereareinstanceswhenaspecificverticalrestraintisidentifiedasarestrictionofcompetitionwhichisnotlikelytogiverisetoanypotentialconsumerbenefits,and,asaresult,itisprohibitedwithoutneedforamoredetailedassessment–socalledhardcorerestrictionsasexplainedinTechnicalExplanation2inAnnex1(suchisgenerallythecaseforResalePriceMaintenance,RPM).Overall,competitionauthoritieswouldneedtoperformacase-by-caseanalysisinordertoassesswhether the benefits outweigh any potential consumer detriment. In other jurisdictions however,verticalrestraintswhicharenotdefinedashardcorerestrictions,maybepermittedwithoutneedforanassessment.Theseareoftenidentifiedinsomeformofblockexemptionregulation,suchastheEuropeanCommission’sVerticalAgreementBlockExemptionRegulation (VABER),197asexplained ingreaterdetailinTechnicalExplanation2inAnnex1.ThesegeneralconsiderationsapplytoE-commerceinthesamewayastheyhavebeenappliedtothebrick-and-mortar retailingmodeloftraditionalmarkets.


195 See, for example, Rosch, J. (2012).

196 European Commission (2017d).

197 Slaughter and May (2016).

59

198 European Commission (2016) and Oxera and Accent (2016).

199 This is a non-exhaustive list. There are other vertical restraints used by firms which competition authorities may investigate under certain circumstances, such as franchising arrangements, however these are not considered here as evidence has shown they are not prevalent in online markets and/or have not raised challenges to competition authorities around the world.

200 See, for example, Oxera and Accent (2016).

201 See, for example, European Commission (2016).

202 The following vertical restraints have been encountered by questionnaire respondents: Resale price maintenance, selective distribution, geo-blocking, bans on PCWs and MFN clauses. However, as it stands, the following restraints have not been encountered: dual pricing systems, bans on online marketplaces, or exclusionary practices between physical stores and online shops.

Effect of E-commerce on the use of vertical restraints

7.2.15. Whilevertical restraints arepresent in bothbrick-and-mortar andonline channels, experience todatedemonstratesthattheuseofverticalrestraintsisparticularlyprevalentinE-commercemarkets,largelydrivenbymanufacturers’concernsthatonlineretailersmayfree-rideontheservicesprovidedbytheirbrick-and-mortarcounterparts.198Free-ridingmayalsooccurbetweendifferentonlineretailersandplatformwebsites.

7.2.16. Someonlineretailersmayfreerideontheservicesandfeaturesofferedonotherwebsites,suchasindependentreviewsorpricecomparisontools.Takingtheexampleofahotelbookingplatformwherealeadingplatformmayinvestincertainfeatures,suchasadvancedfilteringcapabilitiesortailoredmappingservices.Inordertofundthesefeaturesthehotelbookingplatformmayagreeapriceparityclausesothatitsofferscannotbeundercutbyotherwebsites/platforms,orthehotelsontheirownwebsites.Otherplatformsthatdonotofferthesefeaturesmaythenagreeonlowerrateswithhotelsandfree rideonthe investmentthatthesuperiorbookingplatformhadmade.Consumerswouldselectthehoteltheywanttobookbymakinguseofthefeaturesavailableonthesuperiorplatformbeforeeventuallypurchasingthehotelroomonanalternative,cheaperplatform.Similarly,ahotelwhichhas itsownwebsitemayfree-rideonaplatform’sfreeadvertisement,butofferroomsatacheaperrateonitsownsite.Asaresultofthisfree-riding,investmentindevelopingsuchfeaturesonplatformsmaybecompromisedintheabsenceofverticalrestraints,ultimatelyresultinginpoorerservicesforconsumers.

Overview of the following sub-sections

7.2.17. Tosupportcompetitionauthoritiesinassessingtheuseofverticalrestraintsinonlinemarketsthefollowingsub-sectionsconsiderinturndifferenttypesofverticalrestraints199thathavebeenfoundtobeprevalent inonlinemarkets200andhaveposedchallengestocompetitionauthoritiesaroundtheworld.201AlthoughquestionnaireresponsesindicatethatsomeofthesechallengeshavenotyetemergedinASEAN,onewouldexpectthattheuseofsuchrestrictionswillincreaseasE-commercemarkets continue to expand in the region. Currently only two out of five AMS authorities haveencounteredtheseverticalrestraintsinonlinemarkets.Additionally,someoftheverticalrestraintsconsideredinthishandbook,specificallydualpricingsystemsandmarketplacebans,202havenotyetbeenencounteredinanyofthefivejurisdictions.

7.2.18. Relevant cases from jurisdictions inASEAN and around theworld are presented to illustrate theissuesdiscussed,and,whererelevant,theapproachfollowedbycompetitionauthoritiesinpursuingtheirinvestigationsisoutlined.

60

7.3. Selective distribution networks that exclude or restrict online sales

Overview of restraint

7.3.1. Selectivedistributionagreementsaredefinedasinstanceswhere:

“[a] producer establishes a system in which the products can be bought and resold only by authorised distributors and retailers. Non-authorised dealers will not be able to obtain the

products, and the authorised dealers will be told they can resell to other members of the systemortothefinalconsumer.”203

7.3.2. Selectivedistributioniscommonplaceinmanymarketsasmanufacturersseektoensurethattheirproductsaresoldinanappropriatemanner.Toovercomefree-ridingconcerns,amanufacturermayonlyformdistributionagreementswithretailerswhoagreetoacertainlevelofpre-orpost-salesservice.

7.3.3. OnequestionnairerespondentindicatedthattheyhadencounteredselectivedistributionagreementsinE-commercemarketswithintheirjurisdiction.

7.3.4. Selectivedistribution agreements limit intra-brandcompetition; butunless themanufacturerhasastrongposition inthemarket, inter-brandcompetitionshouldbesufficienttopreventanyanti-competitiveeffects frombeing toosevere.204Nevertheless, a case-by-caseapproachto reviewingselective distribution agreements is recommended so as to assess any anti-competitive effectsarisingfromtheagreementsandanycountervailingefficiencieswhichmayjustifytheiradoption.

Insights from cases

7.3.5. InEurope,competitionauthoritieshavefocusedtheirattentiononselectivedistributionagreementsthatexcludeawholechannel (suchasthe internet).ThePierreFabre ruling205 (seeCase review4below)determinedthatdistributionagreementswhichpreventedfirmsfromsellingontheinternetamountedtoanti-competitiveconduct,astheyrestrictedfirmsfrompassivesalesacrossbordersandpreventedthebenefitsfromtheinternetbeingrealised.Agreementsthatpreventsalesviaacertainchannel,suchastheinternet,aretreatedasahardcorerestrictionbytheEuropeanCommission,asexplainedingreaterdetailinTechnicalExplanation2inAnnex1.

7.3.6. Amanufacturer ishowevergenerallypermittedtorequireretailerstohaveat leastonebrick-and-mortarstoreinordertoensurethatcertainqualitystandardsaremet,though:

“while acknowledging that brick and mortar requirements are generally covered by the VBER [Vertical Block Exemption Regulation], certain requirements to operate at least one brick and

mortar shop without any apparent link to distribution quality and/or other potential efficienciesmayrequirefurtherscrutinyinindividualcases.”206

Indeed,suchformofrestraintmightbeusedinordertopreventpureonlineretailersfromaccesstothedistributionofcertainproducts.



205 C-439/09, Pierre Fabre Dermo-Cosmétique SAS v Président de l’Autorité de la concurrence and Ministre de l’Économie, de l’Industrie et de l’Emploi (2011).


61

207 CE/9578-12, Pride/Roma (2014).

CASE REVIEW 4 – PIERRE FABRE

Industry: Cosmetics and beauty products

Country / Union of countries: European Union

Court / Competition Authority: Paris Court of Appeal, ECJ

Case name and citation: Pierre Fabre Dermo-Cosmétique SAS v Président de l’Autorité de la concurrence and Ministre de l’Économie, de l’Industrie et de l’Emploi. (C-439/09)

Date of decision: 13th October 2011

Type of alleged infringement: Selective distribution

3 Case summary

PierreFabreisamanufacturerofluxurycosmeticproductssoldprimarilythroughpharmacies.In2011PierreFabrewas investigatedfor itsuseofvertical restraints.Specifically,distributorswererestrictedtosellPierreFabre’sproductsonlyfromaphysicallocationwithaqualifiedpharmacistpresent,despitetheproductsnotbeingmedicines.Retailerswerethereforepreventedfromsellingonline.

Similarconcernsrelatingtoselectivedistributionagreementsusedbycompetitorswerealsoraised,but in 2007, theFrenchConseil de laConcurrence (theFrenchcompetitionauthority) acceptedcommitmentsfromthesefirmstoamendtheirselectivedistributionarrangementstoallowinternetsales.However,PierreFabre,whocontrolledroughly20%ofthemarket, refused,arguingthat itsproductsrequiredaqualifiedpharmacistpresentatthepointofsaletoprovidespecialistadvice.

TheParisCourt ofAppeal approachedtheECJ for adviceon thecase.TheECJconfirmedthatthe ban restricted competition as it reduced the ability of a distributor to sell the products tocustomersoutsideitsterritory.Importantly,theECJruledthatablockexemptionunderVABER(i.e.anexemptionfromcompetition lawbasedonPierreFabre’smarketsharebeing lessthan30%)couldnotbeappliedtothisagreement.InEurope,thiscasehasledtoanunderstandingthatfirmscannotblocksalesthroughaspecificchannel,suchastheinternet.

7.3.7. IntheUK,thePierreFabrerulingwasrelieduponbytheOFT(nowtheCMA)inits investigationofthemobilityscootersector,rulingthattheprohibitionofonlinesalesamountedtoanti-competitiveconduct,inadditiontootherinfringementsinthemarket(seeCasereview5).207

62

CASE REVIEW 5 – MOBILITY SCOOTERS

Industry: Mobility scooters

Country / Union of countries: UK

Court / Competition Authority: OFT, now the CMA

Case name and citation: Mobility scooters prohibitions on online sales and online price advertising, CE/9578-12

Date of decision: 27th March 2014

Type of alleged infringement: Exclusion of online sales and RPM

3 Case summary

TheOFTnowknownasCMA,conductedamarketstudyofthemobilityscootermarket in2011.Followingthisreview,aninvestigationwasopenedintoonlineverticalrestraintsbeingusedinthemarket.

TheinvestigationfoundthatRomahadprohibitedonlinesalesforsevenretailersbetweenJuly2011andApril2012.TheOFTalsofoundthatRomahadprohibitedonlineadvertisingofanypricesforsomeretailersinthesameperiod.

Additionally,theinvestigationfoundthatanotherproducerofmobilityscooters,Pride,hadenteredintoagreementswitheightofitsretailers,preventingthemfromadvertisingonlinepricesbelowtheRRP (RecommendedRetailPrice).TheOFT ruledthat theseagreementsprevented, restrictedordistortedcompetitioninthesupplyofmobilityscooters.

TheOFTdirectedbothpartiestoremoveanyformofpricerestrictionrelatedtotheabovefindingswithin20daysofthedateofdecision,andtowritetotheaffectedretailersinformingthemthatsuchrestraintswerenolongerinplace.

63

CASE REVIEW 6 – PING

Industry: Sports equipment


Court / Competition Authority: CMA

Case name and citation: Ping Europe Limited (Ping)

Date of decision: Ongoing

Type of alleged infringement: Exclusion of online sales

3 Case summary

TheCMAisinvestigatingPingEuropeLimited(Ping),agolfclubmanufacturer,fortheuseofbanswhichpreventretailersfromsellingPinggolfclubsonline.TheCMA’sfindingsare,however,provisionalatthisstage,andPinghasbeeninvitedtorespondtotheCMA’sconcerns.TheCMAhasarguedthatonlinesalesareanincreasinglyimportantdistributionchannel,andthatretailers’abilitytosupplythroughthischannelshouldnotbeundulyrestricted.TheinvestigationisbeingconductedunderChapter1oftheCompetitionAct1998,andArticle101oftheTFEU.

CASE REVIEW 7 – BMW

Industry: Automotive



Case name and citation: “BMW changes policy on car comparison sites following CMA action” (https://www.gov.uk/government/news/bmw-changes-policy-on-car- comparison-sites-following-cma-action)

Date of decision: 24th January 2017

Type of alleged infringement: Exclusion of online sales

3 Case summary

InJanuary 2017, BMWUKchanged its policy following the threat of aCMA investigation into anallegedexclusionofonlinesales.Inparticular,carwow,aproviderofanonlinecomparisontool,hadcomplainedtotheCMAthatBMWUKhadpreventeddealersfromlistingBMWandMINIcarsonitsportal.Followingdiscussionsbetweencarwow,theCMAandBMW,BMWagreedtoallowdealerstolistBMWandMINIcarsoncarwowandotherinternetplatforms.

7.3.8. IntheUK,theCMAiscurrentlyinvestigatingthegolfclubmanufacturer,Ping,foranoutrightbanoninternetsales(seeCasereview6below).208

7.3.9. ThePierreFabrerulingwasalsofollowedinaninvestigationofBMW’sagreementswithdealersofitscarsthatpreventedlistingsononlineplatforms.209However,followingcommitmentsmadebyBMWtowithdrawtherelevantclausesfromitsagreements,theinvestigationwassubsequentlyterminated(seeCasereview7below).

208 Press release: CMA alleges breach of competition law by Ping; https://www.gov.uk/government/news/cma-alleges-breach-of-competition-law-by-ping (2016).

209 News story: BMW changes policy on car comparison sites following CMA action; available at: https://www.gov.uk/government/news/bmw-changes-policy-on-car-comparison-sites-following-cma-action (2017).

64

7.4. Resale Price Maintenance (RPM)


7.4.1. RPM consists of “agreements or concerted practices having as their direct or indirect object theestablishmentofafixedorminimumresalepriceorafixedorminimumpriceleveltobeobservedbythebuyer”.210Pricefloorsmaybeexplicitlyspecifiedinacontract,orimplicitlyenforcedthroughthreatsbyamanufacturertopunisharetailerifapricedifferenttowhatthemanufacturersuggestsisset.RPMiswidelyregardedasahardcorerestrictionasitreducesintra-brandcompetition,andmayfacilitatecollusion,raiseprices,andreduceinter-brandcompetitionifimplementedbymultiplemanufacturers.IntheUS,however,atfederallevel,minimumresalepricerestrictionsareanalysedunderarule-of-reasonapproach.211

7.4.2. One questionnaire respondent indicated that they have encountered RPM within E-commercemarketsintheirjurisdiction.

7.4.3. Whilearecommendedretailprice(RRP)istypicallynotdeemedbycompetitionauthoritiesaroundtheworldtobeanti-competitive,anyattempttoenforceanRRP,forexamplebythreateningtopunishanyretailerswhodeviatefromsuchrecommendations(e.g.byremovingdiscountsorlimiting/endingsupply),isregardedasahardcorerestriction.

7.4.4. By the same logic that is applied in cases of RPM in brick-and-mortarmarkets, in E-commercemarkets RPM is deemed likely to be harmful to consumers and is therefore treated in a similarmanner.

Insights from cases

7.4.5. TheapproachtoRPMandRRPsdiscussedabovehasbeenobservedincasestodate.Forexample,theOFTfoundthatrestrictionspreventingdealersfromdisplayingadvertisedpricesbelowanRRPamountedtoRPM inthemobilityscootersector (Casereview5).212 EnforcedRRPshavealsobeendeemed anti-competitive. Case review 8 highlights how theGerman competition authority foundLegoguiltyofRPMthroughthreatstoremovewholesalediscounts,213andCasereview9presentstheUKCMA’sinvestigationsinthecateringequipmentandbathroomfittingssectors.214

7.4.6. Onrareoccasions,RPMmaybepermittedifitisdeemedthatefficiencybenefitsoutweighanyanti-competitiveeffects.AlthoughtheAustralianCompetition&ConsumerCommission(ACCC)typicallyregardsRPMasperse illegal, suchconductcanbepermitted if itcanbedemonstratedthattheefficiency benefits outweigh any costs.Thiswas the case in 2014whenTooltechnicwas grantedpermissiontoimplementaminimumresalepriceforFestoolpowertools(seeCasereview10).Inthisinstanceitwasdeemedthatduetothetechnicalnatureoftheproduct,andtheimportanceofpre-andpost-salesservices, intheabsenceofRPM,free-ridingbyretailerswouldhavebeenaseriousconcern.GiventhatTooltechnichadasmallmarketshareinthesupplyofpowertools,significantanti-competitiveeffectsweredeemedunlikely.


211 Rosch, J. (2012). In some states, such as California, RPM is still considered per se illegal (Lindsay, M. (2017)).

212 CE/9578-12, Pride (2014).

213 Press release: Bundeskartellamt fines LEGO for vertical resale price maintenance, bundeskartellamt.de (2016).

214 CE/9856-14, Commercial catering equipment sector: investigation into anti-competitive practices (2016); and CE/9857-14, Bathroom fittings sector: investigation into anti-competitive practices (2016).

65

CASE REVIEW 8 – LEGO

Industry: Children’s toys

Country / Union of countries: Germany

Court / Competition Authority: Bundeskartellamt

Case name and citation::“BundeskartellamtfinesLEGOforverticalresalepricemaintenance” (http://www.bundeskartellamt.de/SharedDocs/Meldung/EN/ Pressemitteilungen/2016/12_01_2016_Lego.html) “LEGO changes its discount system - Fairer conditions for online sales” (https://www.bundeskartellamt.de/SharedDocs/Meldung/EN/ Pressemitteilungen/2016/18_07_2016_Lego.html)

Date of decision: 12th January 2016

Type of alleged infringement: RPM

3 Case summary

InJanuary2016theGermancompetitionauthorityfinedLEGO€130,000forRPMstrategies.Thefirmwasfoundtohaveenforcedrecommendedpricesbythreateningtopunishdeviatorswiththeremovalofdiscountsonwholesaleprices.

MorerecentlyinJuly2016,theBundeskartellamtalsoinvestigatedLEGOforanotherinfringementinhibitingonlinesales,specificallythroughofferingdifferinglevelsofdiscountstoonlineandofflineretailersonthewholesaleprice.ToallaytheseconcernsLEGOcommittedtochanging itsonlinepricingstructuressothatbrick-and-mortarandonlineretailersweretreatedequally.

CASE REVIEW 9 – CATERING EQUIPMENT AND BATHROOM FITTINGS

Industry:Cateringequipment/bathroomfittings



Case name and citation: Commercial catering equipment sector: investigation into anti- competitive practices (CE/9856-14); and Bathroomfittingssector:investigationintoanti-competitivepractices (CE/9857-14)

Date of decision: 24th May 2016; 26th April 2016


3 Case summary

Afridgesupplier,ITWLtd,wasfinedover£2minJune2016forusingRPMstrategiesforonlinesales.Specifically, ITW implemented aminimum advertised price and threatened dealerswith higherwholesalepricesorthewithdrawalofsupplyifthesuggestedpricingstructureswerenotfollowed.

TheCMApursuedasimilarcase inMay2016relatingtoRPM inthebathroomfittingsmarket. Inthis instanceUltraFinishingLtdwasfoundtohaveenforcedrecommendedretailpricesthroughthreatstochargeretailershigherprices,withdrawingrightstousethesupplier’simagesonline,orwithdrawingsupplyofproducts.Afineof£786,668wasimposed.

66

7.5. Dual pricing systems


7.5.1. Somefirmsmaychargedifferentwholesalepricesdependingonthechannelthroughwhichretailersselltheirfinalproducts.TypicallydualpricinginE-commercemarketsinvolvesfirmssettingahigherwholesalepriceforgoodssoldviaonlinechannelsincomparisontosalesmadeinbrick-and-mortarstores.Althoughsuchpracticesmaycompensatemanufacturersfordifferences incostsbetweenbrick-and-mortarstoresandonlineretailers,theymayalsobeusedasastrategyto inhibitonlinesales.

7.5.2. DualpricingsystemsareregardedasahardcorerestrictioninEurope.However, in itsFinalReportontheE-commerceSectorInquiry,theEuropeanCommission(2017b)215outlineshowthishardcorerestrictiononly applies todualpricing systems for click-and-mortar retailers (hybrid retailers); i.e.a retailerthatsellsbothonlineandviabrick-and-mortarstorescannotbeunjustifiablychargedadifferentwholesalepricedependingonthechannelthroughwhichtheproductissold.Bycontrast,“chargingdifferentwholesalepricestodifferentretailersisgenerallyconsideredanormalpartofthecompetitive process”.216

215 European Commission (2017b), page 10.

216 European Commission (2017b), page 10. Note: “Unless different wholesale prices to (online) retailers have the object of restricting exports or partitioning markets.”

CASE REVIEW 10 – TOOLTECHNIC

Industry: Power tools

Country / Union of countries: Australia

Court / Competition Authority: ACCC

Case name and citation: : Tooltechnic Systems (Aust) Pty Ltd - Authorisation - A91433

Date of decision: 5th December 2014


3 Case summary

InJune2014,TooltechnicsoughtapprovalfromtheACCCtoamenditscontractswithdealerstoallowittoimposeaminimumresalepriceinthesupplyofFestoolpowertools,wheretheyweretheexclusiveimporterandwholesaler.

TheACCCstatesthat“undertheCompetitionandConsumerAct2010(CCA),resalepricemaintenanceisprohibitedperse.However,theACCCcanauthoriseresalepricemaintenancewhereitissatisfiedthatinallthecircumstancestheconductislikelytoresultinpublicbenefitswhichoutweighthepublicdetrimentslikelytoresultfromtheconduct.”(ACCCDetermination,2014,pageii.)

On5thDecember2014theACCCgrantedTooltechnicpermissiontoimplementtheseclausesuntil31December2018onthebasisthattheextentofanydetrimentaleffectwaslikelytobelowgiventhewiderangeofalternativepowertools,thesmallmarketshareofFestoolproducts,ahistoryofentryandexpansioninthemarket,andthehighlyinnovativeanddifferentiatednatureofproductsinthemarket.Itwasdeemedthatovercomingissuesoffree-ridingbyretailersonthepre-andpost-salesservicesprovidedbyotherretailersoutweighedanycoststhatwouldarise.Investmentinpre-salesserviceswasdeemedimportantinthisinstanceduetothecomplexnatureoftheproductsbeingsold.Furthermore,initsdecision,theACCCspecificallycitedonlineretailersaspotentialfree-riders on brick-and-mortar stores services.

TheACCCalsocommittedtomonitortheimpactoftheRPMonanannualbasis.

67

7.5.3. Dualpricingsystemsforclick-and-mortarfirmsmaybepermitted inEuropeifpricingdifferentialscan be justified by differences in costs incurred by the manufacturer from retail sales in onechannelcomparedtoanother.217Forexample,whereagoodbenefitsfromprofessionalinstallation,amanufacturermayfacehighercostsfromsalesmadeonline,forinstanceifcustomercomplaintsandwarrantyclaimsarehigherwhentheproductisnotappropriatelyinstalled.218

Insights from cases

7.5.4. AdualpricingsystemwasobservedintherecentinvestigationofLegoinGermany(Casereview8).Specifically,Legowasfoundtohaveoffereddifferentlevelsofdiscountsforonlineandofflinesales.219 Similarcases,alsoinGermany,arepresentedinCasereview11andCasereview12below.Interestingly,inGermany,theBundeskartellamt’sinterventionappearstogobeyondtheEuropeanCommission’scurrentthinking.Inadditiontowholesalepricediscriminationforclick-and-mortarretailersdependingonthesalechannel,theGermanauthority isalsoconcernedbywholesalepricingstructuresthatdiscriminatebetweenretailersthatonlysellonlineandretailersthatonlysell inbrick-and-mortarstores,andbetweenclick-and-mortarandbrick-and-mortarretailers.ThisdifferenceininterpretationofthelawposesachallengeforbusinessesoperatingacrossdifferentjurisdictionswithinEurope.



219 Press release: LEGO changes its discount system - Fairer conditions for online sales, bundeskartellamt.de (2016).

CASE REVIEW 11 – BOSCH

Industry: Home appliances



Case name and citation: Bosch Siemens Hausgeräte; B7-11/13

Date of decision: 23rd December 2013

Type of alleged infringement: Dual pricing

3 Case summary

InDecember2013BoschSiemensHausgeräte(BSH)agreedtowithdrawrebateofferswhichfavouredofflineonlyretailersasopposedtoretailerssellingbothonlineandinbrick-and-mortarstores.Thehighertheturnoverso-calledhybridretailersmadefromonlinechannels,thelowertheirrebateswouldbe.ThecommitmentsmadebyBSHwereenoughtoputahalttotheBundeskartellamt’sinvestigationfollowingcomplaintsfromretailers.

Inthisinstance,theBundeskartellamtacknowledgedthatamanufacturerisallowedtoagreewithhisdealersonqualityrequirementsforthesaleofhisproducts,butinthisspecificinstanceretailerswererestrictedintheirchoiceofsaleschannelandincentivisedtolimitonlinesales.

68

CASE REVIEW 12 – GARDENA

Industry: Gardening tools



Case name and citation: Gardena, B5-144/13


Type of alleged infringement: Dual pricing

3 Case summary

In 2013 the Bundeskartellamt investigatedGardena for implementing dual pricing practices thatlimitedtheonlinedistributionofitsproducts.OnlinedistributorsofGardena’sproductsclaimedthatGardena’sdiscountsystemfavouredtraditionalbrick-and-mortarstoresoveronlineretailers.

TheBundeskartellamtruledthatGardenamustremovetheirdualpricingsystems,therebygivingequallevelsofdiscounttobothbrick-and-mortarandonlineretailers.ThoughtheBundeskartellamtacknowledgedthat“amanufacturermaywelltakeaccountofthedifferentconditionsinthedifferentdistributionchannels”(Bundeskartellamtpressrelease:28/11/2013),itwasdeemedinthisinstancethatthesystemdiscriminatedagainstonlinesales.

7.6. Geo-blocking


7.6.1. InmeetingASEAN’sobjectivetobecomeamoreintegratedmarket,competitionauthoritiesmayalsobeconcernedbyfirms’attemptstorestrictcross-bordersales–atypeofconductreferredtoasgeo-blocking.Onequestionnairerespondentindicatedthattheyhadencounteredgeo-blockingstrategiesinE-commercemarketswithintheirjurisdiction.

7.6.2. TheviewinEuropeonthismatteristhatamanufacturermaybeallowedtorestrictretailers’abilitytoactivelyselltoaparticularregionorcountrythatisexclusivetoanotherdistributor,forexamplethroughadvertisingbans.220Thisisbecauseexclusiveterritoriescangenerateefficiencies,forexampleovercomingfree-ridingissuesinhibitingtheincentivestoinvest.However,restrictionsonpassivesalestoothermemberstates(i.e.preventingcustomerswhoindependentlyreachouttoaretailer’swebsiteinaforeigncountryfrompurchasingagoodfromthatstore)areregardedasahardcorerestrictionastheypreventthebenefitsofthesinglemarketfrombeingrealised.221Passivesalesmayberestrictedby automatically re-routing customers to their domesticwebsite, or refusing payment to foreigncustomers.Outrightbansonsalestoforeigncustomersinothermemberstatesarethereforealsonotallowed.Additionally, inEurope,anagreementwithretailersinaselectivedistributionnetworkmustnot“haveasitsobjecttorestrictactiveorpassivesalestoendusers”orbetweenauthoriseddealers.222


221 These restrictions are regarded as a hardcore restriction if they are part of an agreement between a manufacturer and retailer. Unilateral decisions of non-dominant companies are permitted.


69

Insights from cases

7.6.3. InthecontextofE-commerce,inlinewithitsstrategytopromoteasingledigitalmarket,theapproachfollowedbytheEuropeanCommissionregardsgeo-blockingasaninfringementofcompetitionlaw.Asaresult,theEuropeanCommissionhasimposedanumberoffinesinseveralinstancesofgeo-blockingpractices.In2005Peugeotwasfoundtohaveengagedingeo-blockingmeasuresthroughitsagreementstoonlypaybonusestodealersforcarssoldtoDutchcitizens(seeCasereview13).223 Similarly, through restrictions in its agreementswith retailers, Yamahawas also deemed to haverestricted cross-border trade through geographic limits on product guarantees and requirementsforretailerstonotifyYamahaiftheyweretosellabroad(Casereview14).224Additionally,followingthepublicationofthepreliminaryfindingsonitsE-commercesectorinquiry,theEuropeanCommissioninitiatedaninvestigationintothehotelbookingsmarket,lookingatwhetherpricingsystemsadoptedbyhotelsandtouroperatorsdiscriminatebetweencustomersbasedonwheretheyarelocated,andthereforeinhibitcross-bordertrade(Casereview15).225

223 37275 SEP et autres / Automobiles Peugeot SA (2005).

224 37975 PO/Yamaha (2003).

225 Press release: Antitrust: Commission opens three investigations into suspected anti-competitive practices in e-commerce (2017).

CASE REVIEW 13 – PEUGEOT

Industry: Automotive


Court / Competition Authority: European Commission

Case name and citation: 37275 SEP et autres / Automobiles Peugeot SA


Type of alleged infringement: Geo-blocking

3 Case summary

InOctober2005,theEuropeanCommissionannouncedthat ithad imposedafineof€49.5monPeugeotforobstructingnewcarexportsfromtheNetherlandstootherEUMemberStatesbetween1997 and 2003.

Dealerswereonlypaidabonus ifacarwasregistered intheDutchmarket,andthosewhosoldcross-borderwerepressuredintonotdoingso,forexamplebythreateningtolimitthequantityofcarssupplied.

70

CASE REVIEW 14 – YAMAHA

Industry: Musical instruments



Case name and citation: 37975 PO/Yamaha

Date of decision: 16th July 2003


3 Case summary

InJune2003theEuropeanCommissionannouncedthatithadimposeda€2.56mfineonYamaha,themanufacturerofmusicalinstruments,forrestrictionsoftrade(geo-blocking)andRPM.

TheEuropeanCommissiondeterminedthatasaresultofrequirementstonotifyYamahaifaretailerwanted to export via online channels, “dealers were clearly discouraged from exporting”. “TheCommissionseesnoreasonstojustifysuchanobligationtoconsultYamahabeforeexportingviatheinternetandinterpretsthisclauseasdeterringexportsviatheinternet.”

Initsagreementswithretailers,Yamahahadincludedrestrictionstoonlyselltofinalcustomersandnotdealers,anddealerswereobligedtoonlybuyfromYamaha’snationalsubsidiaryandnotfromforeigndealers.Additionally, insomecountriesproductguaranteeswereonlyvalid inthecountryoforigin.OncetheEuropeanCommissionopened itsproceedings,Yamaha removedthe relevantconditionsfromitscontracts.

CASE REVIEW 15 – HOTEL ACCOMMODATION

Industry: Accommodation booking



Case name and citation: AT.40308 – Hotel pricing



3 Case summary

InFebruary2017,followingthepublicationoftheEuropeanCommission’sE-commercesectorinquirypreliminaryfindings,theEuropeanCommissionopenedaninvestigationintothehotelaccommodationmarketinrelationtoallegedagreementsbetweenahotelgroup(Melia)andfourtouroperatorswhichdiscriminatebetweenconsumersbasedontheirlocationinEurope.

Itisallegedthatpricingmechanismsdesignedtomaximiseroomusagemaydiscriminatebetweencustomers based on where they are located within Europe. Consumers in some countries arethereforenotabletoseethebestpricesavailable;conductwhichmaybedeemedanti-competitiveasaresultofpartitioningtheSingleMarket.

71

7.7. Platform bans


7.7.1. Platform bans occurwhenmanufacturers prevent sales through certain onlinemarketplaces viaselectivedistributionagreements.Thistypeofagreementisdifferentfromanoutrightbanofonlinesales,assalesthroughsomeonlineretailersarepermitted,butsalesthroughonlineplatformssuchasAmazonmarketplacearerestricted.

7.7.2. Suchagreementsmaycombatthesaleofcounterfeitgoods,ensuresufficientpre-andpost-salesservice,protectthestatusimageofaproduct,and/orsignalthatagoodispremiumbyonlysellingitthroughhigher-end,onlineretailers.226Ontheotherhand,intra-brandcompetitionisrestricted.Giventhattherearebothpro-andanti-competitiveeffectsofsuchrestraints,acase-by-caseapproachtoassessingplatformbansisrecommended.

Insights from cases

7.7.3. AlandmarkcaseiscurrentlyunderinvestigationinEuropeconcerningthebeautyproductmanufacturer,Coty(Casereview16below).227Specifically,theECJisconsideringwhetherstatusreasonsarevalidjustifications for the use of vertical restraints to prevent sales through onlinemarketplaces. Theoutcomeof thiscase is likely tohavefar-reaching implicationsfor the luxury-goods industry,butalsoforplatformssuchasAmazonmarketplace,asplatformbansofthistypehinderthegrowthofE-commerce.Asimilarcasewasheard in2011 inthesportswearmarket (seeCase review20).FollowingdifferentassessmentsofplatformbansbytheGermancourts,itwaseventuallydecidedbytheHigherCourtofFrankfurttorefertheCotycasetotheECJtoseekclarityonthematter.

7.7.4. InitsfinalreportontheE-commerceSectorInquiry,theEuropeanCommission(2017d)hasindicatedthatitdoesnotconsidermarketplacebansahardcorerestrictionasthey“donothaveastheirobject(i)arestrictionoftheterritoryorthecustomerstowhomtheretailerinquestionmaysellor(ii)therestrictionofactiveorpassivesalestoendusers.”228CompetitionauthoritiesshouldhoweverreviewtheECJ’srulingontheCotycaseonceitisissuedasthiscouldsupersedetheEuropeanCommission’scurrentthinkingifadifferentviewistakenbytheCourt.

226 European Commission (2016), page 148.

227 Coty Germany GmbH v Parfümerie Akzente GmbH, Case C-230/16 (2016).

228 European Commission (2017d), page 152.

CASE REVIEW 16 – COTY

Industry: Cosmetic and beauty products


Court / Competition Authority: Higher Regional Court of Frankfurt, ECJ

Case name and citation: Coty Germany GmbH v Parfümerie Akzente GmbH, Case C-230/16

Date of decision: Pending (lodged on 25th April 2016)

Type of alleged infringement: Selective distribution/marketplace bans

3 Case summary

TheHigherRegionalCourtofFrankfurtiscurrentlyreviewingthelegalityofrestrictionsimposedbythebeautyproductmanufacturer,Coty,preventingitsdistributor,ParfümerieAkzente,fromsellingproducts via third party online platforms such as Amazonmarketplace due to fears that suchplatformsweakenthestatusimageassociatedwithitsproducts.

In2014,theRegionalCourtofFrankfurtdismissedCoty’sclaim, insteadarguingthatthesetermsinfringedGermanantitrustrules.Cotyhasappealedthisdecision,andtheECJhasbeenaskedtoprovideguidanceonwhethertheseselectivedistributionagreements,intheformofonlineplatformsalesbans,infringeEuropeancompetitionlaw.

72

7.8. Most Favoured Nation (MFN) clauses


7.8.1. AnotherformofverticalrestraintoftenusedbyonlineplatformsareMostFavouredNation(MFN)orpriceparityclauses.Specifically,firmsmayincluderestrictionsincontractsthatensurethatnoothercompetitorwillreceivemorefavourableterms–forinstancebeingabletosellatalowerprice.ThisiscommonlyreferredtoasawideMFN.Incontrast,anarrowMFNpreventsafirmfrombeingabletosetalowerpriceonitsownwebsite,butitisfreetoagreelowerpriceswithotherplatforms.

7.8.2. One questionnaire respondent indicated that they have encountered the use of MFN clauses inE-commercemarketswithintheirjurisdiction.

7.8.3. MFN clauses pose a challenge to competition authorities in that they have both pro- and anti-competitive effects. MFNs restrict intra-brand competition and can facilitate collusion betweensellersinthemarketbyenforcinguniformprices.However,MFNclausescanhelptoovercomeissuesoffree-riding.Consideringthehotelbookingmarket,MFNclausespreventotherplatforms (inthecaseofwideMFNs),orhotelsthemselves(underbothwideandnarrowMFNclauses),fromfree-ridingontheserviceprovidedbytheplatform,andofferingacheaperpricethemselves(forexamplefree-ridingonthesuperiorplatform’sfunctionalitiesincludinghotelreviews,pricecomparisonand/orfreeadvertisementofthehotel).MFNclausesmayalsoresult inreducedsearchcostsforbuyers,andavoidpricediscriminationbetweenbuyers.Toeffectivelyweighupthesepro-andanti-competitiveeffects,acase-by-caseapproachissensible.Suchanapproachenablesthecompetitionauthoritytoanalysethepotentialharmand/orbenefitdependingonthemarketstructure,typeofMFNclauseused,andthecharacteristicsoftheproductmarketandofitsbuyersandsellers.229

Insights from cases

7.8.4. MFNclausesarecommoninthehotelbookingmarket,andhavebeeninvestigatedbytheUKCMA,theGermanBundeskartellamt,theParisCommercialCourtandsevenothercompetitionauthoritiesacrossEurope(seeCasereview17below).Thiscaseisofparticularinterestbecauseithasresultedindifferentconclusionsbeingreachedbydifferentcompetitionauthorities.Onereasonforthedifferentapproachestakenbyauthoritiesmaybetheinfluenceofgovernments insomecountrieswhoarelooking to pursue their own industrial strategy objectives. In general, competition policy and lawshouldfocusonassessingthepro-andanti-competitiveeffectsofvariousformsofconductbyfirms,andremainindependentfromindustrialstrategyconsiderations.

7.8.5. TheUKCMAhasprohibitedhotelbookingplatformsfromusingwideMFNs,buthasbeenmorelenientontheuseofnarrowMFNsonthebasisthatnarrowMFNshelptoovercomeissuesoffree-ridingbyhotelsontheplatform’sservice.Germany’sBundeskartellamt,however,hasprohibitedbothwideandnarrowMFNsinthehotelbookingmarket,arguingthatnarrowMFNsrestrictpricecompetitionacrossthemarket,asevenunderthelessrestrictivenarrowMFNsthereislittleincentiveforahoteltoallowoneonlinebookingplatformtosetalowerpriceifithastodisplayhigherpricesonitsownwebsiteduetoanMFNclauseithasagreedwithanotherplatform.230Analogously,Italy,AustriaandFrancehaverecentlyintroducedlegislationtobanbothnarrowandwideMFNsinthehotelbookingmarket.Althoughtherearenopublicdetailsofthecase, ithasbeenreportedthattheChineseauthoritieshavealsoinvestigatedonlinehotelbookingplatformsregardingtheuseofMFNclauses.231Bycontrast,intheUS,authoritieshavenotpursuedtheuseofMFNsbyhotelbookingplatforms,andclassactionshavefailedastherewasnoevidenceofaconcertedpractice.232

7.8.6. TheACCChasalsofoundnarrowMFNstobeanti-competitiveinitsinvestigationintoFlightCenter’sallegedpressureonairlinestonotsellflightsatapricelowerthanwhatwasavailableonitswebsite(seeCasereview19).233InthiscaseacriticalfactorinthedecisionwasthedeterminationofwhetherFlightCenter,asabookingplatform,wasincompetitionwiththeairlinesthemselves,Followinganinitialdecisionandtwoappeals,itwaseventuallyestablishedthatthiswasindeedthecase,thereforeinthis instancetheuseofMFNclauseswasconsideredahorizontalagreementasopposedtoaverticalrestraint.

229 CCS (2015).

230 Andreas Mundt, Bundeskartellamt (2015).

231 Freshfields (2017).

232 Ibid.

233 Case B15/2016 Flight Center (2016).

73

7.8.7. Amazon has also been investigated for its use of MFN clauses. In 2012, the OFT234, opened aninvestigation into the price parity policy that Amazon implemented on its onlinemarketplace.235 Amazonagreedtoremovethispolicy,resulting intheOFTclosing its investigation.Similarly,wheninvestigatedbytheEuropeanCommissionforitsuseofMFNclausesintheE-booksmarketin2017,thecasewasclosedfollowingcommitmentsmadebyAmazon.Japan’sFairTradeCommission(JFTC)alsorecentlyinvestigatedAmazonJapanforuseofwideMFNclauses,followingdawnraidsinAugust2016(seeCasereview18).236Bycontrast,intheUS,sofar,authoritieshavechosennottoinvestigateAmazonfortheuseofMFNclauses.

7.8.8. The difference in competition authorities’ opinions on MFN clauses across the world has raisedconcerns.TheBooking.comcase illustrates thechallenges facedbybusinesseswhencompetitionauthoritiestakecontrastingviewsonissuessuchastheuseofnarrowMFNs.Whilsttheirpracticesmaybeperfectlylegalinonejurisdiction,theymaynotbesoinaneighbouringjurisdiction,thuslimitingtheirabilitytoexpandinternationallyviathesameplatformandaccompanyingbusinessmodel.

7.8.9. TennationalcompetitionauthoritieshavebeenassessingtheeffectofthevariousapproachestoMFNclausesinthehotelsectorviatheEuropeanCompetitionNetwork(ECN).Thefindings237showthatasaresultofonlinetravelagents(OTAs)suchasBooking.comandExpediaswitchingfromusingwidetonarrowMFNclauses(whichallowthehotelstoofferdifferentpricestodifferentOTAsaslongasthehotel'swebsiteratesarenolower),therehasbeenarecognisableincreaseinpricedifferentiationviaOTAs.TheCMAhasthereforeconcludedthat itwillnotprioritisefurther investigationofpricingpracticesinthesectorbutwillseektoraisefurtherawarenessofthisrecentchangeinpricingclauses.WithinaregionsuchasASEAN,wherebusinessesoperate internationally,aconsistentapproach isrecommendedinordertofacilitategrowthinE-commercemarkets.Ifpossible,aconcertedapproachcouldbefacilitatedviatheASEANExpertsGrouponCompetition.

234 Now known as the UK CMA.

235 CE/9692/12 Online retail sector (2012).

236 JFTC (2017), Press release: “The JFTC closed the investigation on the suspected violation by Amazon G.K."

237 European Commission (2017e).

CASE REVIEW 17 – ONLINE HOTEL BOOKING

Industry: Accommodation booking

Country / Union of countries: UK, Germany, France, Belgium, Hungary, Ireland, Italy, the Netherlands, Czech Republic and Sweden

Court / Competition Authority: CMA (UK), Bundeskartellamt (Germany)

Case name and citation: CE/9320-10 (CMA), B 9-121/13 (Bundeskartellamt)

Date of decision: Varying (2015 - 2016)

Type of alleged infringement: MFN clauses

3 Case summary

InDecember2015, theGermancompetitionauthority (theBundeskartellamt)prohibitedBooking.comfromapplyingits‘bestprice’(orMFN)clauses.TheBundeskartellamtprohibitedclauseswhichpreventedhotelsfromofferinglowerpricesonplatformscompetingwithBooking.comaswellastheirownwebsite(wideMFNs).TheGermancompetitionauthorityalsoprohibitedtheuseofnarrowMFNs,preventingtravelwebsitesfromimplementingclausesrestrictinghotelsfromofferinglowerroomratesontheirownonlinebookingsystem,butallowinghotelstoagreelowerrateswithotherplatforms.Booking.comhasarguedthatnarrowMFNsarerequiredtopreventhotelsfromfree-ridingbyusingBooking.comtopromotetheirhotelsbutofferingacheaperpriceontheirownwebsite.TheFrench,ItalianandAustrianauthoritiesarefollowingtheGermanposition,andareimplementingnewlegislationtoprohibitallMFNsinthehotelbookingmarket.

TheCMAintheUKhashoweverdecidedthatBooking.commustremoveitswideMFNs,butpermittedtheuseofnarrowMFNs.Initsview,narrowMFNsdonothaveasignificanteffectoncompetitionandarelikelytobenecessarytoensurethebenefitsthatonlineplatformsofferconsumers,suchastheeaseofcomparingpricesandswitchingbetweenproviders.ThedecisionfromtheBundeskartellamtisthemoreconservativejudgement,prohibitingbothnarrowandwideMFNs.ThedividednatureofthesedecisionshasledtotheabsenceofanEU-wideposition.

74

CASE REVIEW 18 – AMAZON JAPAN

Industry: Online marketplaces

Country / Union of countries: Japan

Court / Competition Authority: JFTC

Case name and citation: Press release – “The JFTC closed the investigation on the suspected violation by Amazon Japan”

Date of decision: 1st June 2017


3 Case summary

In June 2017 the JFTC announced that it had closed its investigation into Amazon Japan. TheinvestigationhadfocusedonAmazon’suseofMFNclauseswhichrestrictedthepriceretailerscouldselltheirgoodsforoncompetingsites.

The investigationwas closed following voluntary commitments by Amazon to remove the MFNclausesfromtheircontracts,andtoreportannuallyontheimplementationstatusofthesecontractualchanges.

CASE REVIEW 19 – FLIGHT CENTER

Industry: Flight bookings

Country / Union of countries: Australia

Court / Competition Authority: ACCC, Full Court of the Federal Court of Australia, High Court of Australia

Case name and citation: Australian Competition & Consumer Commission v. Flight Centre Travel Group Limited (Case B15/2016)



3 Case summary

BetweenAugust2005andMay2009FlightCenterwereallegedtohaveattemptedtoforcethreeairlines(Emirates,MalaysiaAirlines,andSingaporeAirlines)tonotsellflightsatapricecheaperontheirownwebsitethanonFlightCenter.In2012theACCCcommencedproceedingsagainstFlightCenterforproposingtheseprovisions,thatintheirviewhadthepurposeoreffectof“fixingorcontrollingormaintainingpricesforthesupplyofserviceswhichitandthey[theairlines]wereselling”(HighCourtDetermination,2016,para.4).AkeyfactorintheinvestigationwaswhetherornottheairlinesweretobedeemedcompetitorsofFlightCenter.TheACCCruledthatthiswasthecaseandthereforedeemedthepricingpracticetobeanti-competitive,quashingtheclaimthatFlightCenterwasactingasanagenttotheairlines.

ThoughtherulingwasinitiallychangedfollowinganappealtotheFullCourt,theACCC’sinitialdecisionwasreinstatedfollowingaHighCourtappeal,thoughsomeadjustmentsweremade.Forinstance,themarketwasrewordedas‘internationalairlinetickets’insteadof‘distributionandbookingservicesforinternationalpassengerairtravel’.Theleveloffinancialpenaltieshasyettobeconfirmed.

75

238 See, for example, European Commission (2016).

239 ASICS (B2-98/11) (2015).


7.9. Restrictions on price comparison websites


7.9.1. Some manufacturers may prevent retailers from using price comparison tools. Indeed, onequestionnairerespondentindicatedthattheyhadencounteredtheuseofbansonPCWswithintheirjurisdictioninE-commercemarkets.

7.9.2. Competitionauthoritiesmaybeconcernedthatthese restrictionsarebeingusedto restrictpricecompetitiononlinebyreducingpricetransparency.Theremay,however,bepro-competitivebenefitsofrestrictionsonPCWs,forinstanceincreasingcompetitiononproductquality.PCWsfocusmainlyonpricecompetition,andoftendonotcomparethequalityofproductsorservicesofferedbyfirms.PCWsthereforeencouragefirmstocompeteintensivelyonprice,butreducetheincentivestocompeteonproductqualityasfirmsseektokeepcostsandthereforepricetoaminimum.RestrictionsonPCWsmaythereforeencouragecompetitiononproductqualityandinvestmentsininnovation.

7.9.3. Brand image arguments may also be given by firms for restricting retailers from using PCWs.238 However,statusreasonsarenotyetwidelyregardedasapro-competitivejustificationfortheuseofverticalrestraintsbyfirms.TheCotyrulinginEuropeononlinemarketplacerestrictionswillprovideinsightsonthis.However,itisimportanttonotethatalthoughparallelscanbedrawnbetweencasesrelating tomarketplace bans and PCW restrictions, there are important differences between thebusinessmodelsofthetwotypesofplatformsthatmustbeconsidered.Forexample,onmarketplaceplatforms,actualsalesaremade,whereasonpricecomparisonsitesconsumersareinsteaddirectedtoretailers’websites.

Insights from cases

7.9.4. RestrictionsonPCWswereconsideredintherecentASICScase(Casereview20).239Inthisinstance,therestrictionsweredeemedtobehardcorerestrictions,howeveritisworthnotingthatthisviewmaychangefollowingtheECJ’sguidanceontheCotycaseinvolvingplatformbans.

7.9.5. InitsFinalReportontheE-commerceSectorInquiry,theEuropeanCommission(2017d)240outlineditscurrentviewthat:

“Absolute price comparison tool bans which are not linked to quality criteria therefore potentially restrict the effective use of the internet as a sales channel and may amount to

a hardcore restriction of passive sales under Article 4 b) and 4 c) of the VBER [Vertical Block Exemption Regulation]. Restrictions on the usage of price comparison tools based on objective qualitative criteria are generally covered by the VBER.”

76

CASE REVIEW 20 – ASICS

Industry: Athletics and sportswear



Case name and citation: ASICS (B2-98/11)

Date of decision: 26th August 2015

Type of alleged infringement: Selective distribution

3 Case summary

In2012,ASICS,aproducerofathleticandsportswearintroducedanumberofrestrictionsonretailers.Specifically,thefollowingwereprohibited:theuseoftheASICSbrandnamebyretailersonthirdpartywebsites(i.e. inadvertsforthatretailer); linksfromPCWs;andsalesviaonlinemarketplaces.TheBundeskartellamtlaunchedaninvestigationinSeptember2011,followingcomplaintsfromvariousdistributors.Initsview,ASICSimposedrestrictionswhichconstitutedarestrictionofcompetitionbyobject,andthereforeviolatedArticle101(1)oftheTFEU.

IntheeyesoftheBundeskartellamt,theprohibitionoftheuseofbrandnames(ASICS),andrestrictionsontheuseofPCWsconstitutedhardcorerestrictions,andthereforecouldnotbeexemptundertheVABERTheBundeskartellamtalsoruledthattheprohibitionofsalesviaonlinemarketplaceswasahardcorerestrictiononcompetition,andagaincouldnotbedeemedexemptunderVABER.Asthismeanttherestrictionswereanti-competitivebyobject,therewasnofurtherinquiryintoefficiencyconsiderations.

241 CMA (2017), update paper 28/03/2017.

7.9.6. Inlate2016,theUKCMAlaunchedastudyondigitalcomparisontools(DCTs),lookingattheeffectsthattoolssuchasPCWshavehadonindustriessuchasmotorinsurance,energysuppliersandretailbanking.ThestudyaimstoexplorethebenefitsfromDCTstoconsumers,andhelpunderstandandaddressanypotentialissuesorbarriersinordertomaximisethesebenefits.ThefinalreportisdueinSeptember 2017.241

7.9.7. Therearebothpro-andanti-competitiveeffectsfromtheuseofrestrictionsonPCWs.Additionally,giventhefactthatthesepracticesarerelativelynew,thereislimitedcaselawandalackofinternationalprecedent.Therefore,anyrestrictionsonPCWswhichhavenotbeendeemedahardcorerestrictionarebestdealtwithonacase-by-casebasis.

77


243 Garuda Abacus Case; KPPU Decision No. 01/KPPU-L/2003; Central Jakarta District Court Decision No. 001/KPPU/2003/PN.Jkt.Pst; Supreme Court Decision No. 01 K/KPPU/2004 In 28 August 2000.

7.10. Exclusive purchase restrictions


7.10.1. Asintraditionalbrick-and-mortarmarkets,exclusivepurchaserestrictions,whichpreventacustomerfrompurchasingaparticularproduct,orgroupofproductsfromanyotheralternativesupplier,canbedeemedtobeananti-competitiveagreement.Suchagreementsareknownundervariousterms,forexampleexclusivepurchasing,singlebranding,requirementscontracts,andnon-competeobligations.Despitethedifferingnames,theunderlyingconcept isthesame:thepurchaser ispreventedfrompurchasing competingproducts fromanyoneother than themanufacturer it hasentered into anagreementwith.242

7.10.2. Suchclausescanhelptoovercome issuesoffree-ridingbetweensuppliers,forexamplewhereamanufacturer has to invest in training a retailer or inproviding special equipment to support thesalesprocess.Thisisparticularlyrelevantforhighlytechnicalproducts.However,exclusivepurchaseprovisionsremoveinter-brandcompetitiononthewebsite,orinthestoresoftheretailerthatagreesto the clause.Therefore, theremay be both pro- and anti-competitive effects of such practices.Consequently,acase-by-caseapproachissensible.Insuchassessmentsthepositionsofboththeretailerselling,andthemanufacturerproducingthegoodarehighlyimportant.Ifeitherisinapositionofdominance,thepracticeishighlylikelytobeanti-competitive.Bycontrast,ifneithertheretailernorthemanufacturerareinapositionofmarketpower,significantanti-competitiveeffectsarelesslikely.

Insights from cases

7.10.3. This approachwasevident in theCCS’s recent reviewof theonline fooddelivery industry,whereexclusivity clauseswere identified. It was deemed that, at present, such practices are not anti-competitiveintheonlinefooddeliverysector;howevertheCCScommittedtocloselymonitorthemarketgoingforward,onthebasisthatsuchagreementscouldbeproblematic inthefuture if aparticularfirmusingsuchrestraintsbecamedominant(seeCasereview21below).

7.10.4. A similar case involving exclusivity provisionswas investigated in Indonesia in theflight bookingsmarket(seeCasereview22below).243ItwasdeemedthatprovisionsthatrestrictedtravelagenciesmakingGarudaticket reservations fromusingsystemsother thananAbacusterminal,wereanti-competitiveunderIndonesianverticalintegrationprohibitions.

78

CASE REVIEW 21 - ONLINE FOOD DELIVERY

Industry: Online food delivery



Case name and citation:Mediarelease–25/08/16“CCSinvestigationfindsonlinefooddelivery industry to be currently competitive but exclusive agreements could be problematic in future”

Date of decision: 25th August 2016

Type of alleged infringement: Exclusive purchase restrictions

3 Case summary

Following complaints, the CCS investigated an online food delivery provider for alleged anti-competitiveconductrelatingtotheuseofexclusivepurchasingprovisionswithcertainrestaurants.

Inthisinstanceitwasdeemedthatcompetitionwasnotharmedbytheagreements.However,theauthoritycommittedtomonitoringthesectorgoingforward,asintheinstancethatanonlinefooddelivererbecamedominant,suchagreementsmaybedeemedanti-competitive.

CASE REVIEW 22 - GARUDA / ABACUS

Industry: Flight booking

Country / Union of countries: Indonesia

Court / Competition Authority: KPPU; Central Jakarta District Court; Supreme Court of Indonesia

Case name and citation: Garuda Abacus Case; KPPU Decision No. 01/KPPU-L/2003; Central Jakarta District Court Decision No. 001/KPPU/2003/PN.Jkt.Pst; Supreme Court Decision No. 01 K/KPPU/2004


Type of alleged infringement: Restriction of competition through vertical integration

3 Case summary

InAugust2000,GarudaandAbacusformedanagreementthattravelagentsmustuseanAbacusterminalwhenmakingGarudaflightbookings,thusimposingabarrieronotherprovidersofsimilarsystems.

TheKPPUdeemedthatthisagreementconstitutedabreachofArticle14ofIndonesianCompetitionLaw(Vertical Integration). ItwashighlightedhowGarudaownedasignificantnumberofshares inAbacusandthatsomeindividualssatontheboardofdirectorsforbothfirms.

GarudaappealedtotheCentralJakartaDistrictCourtonefficiencygrounds,eventuallyresultingintheinitialdecisionbeingoverturned.However,followinganappealbytheKPPU,theinitialdecisionwasreinstatedbytheSupremeCourt.

79

7.11. Practical steps/guidelines or recommendations to identify and address competition policy and law issues

7.11.1. Ineachofthecasesoutlinedabove,existingcompetitionpolicyandlawcoveringverticalrestraintsfortraditionalbrick-and-mortarmarketshasbeendeemedsufficientlybroadandflexibleenoughtoallowcompetitionauthoritiestocaptureverticalrestraintsusedbyfirmsinE-commercemarkets.

7.11.2. Someauthoritiesare,however,introducingnewlegislationtobanMFNclauses,notablyItaly,AustriaandFranceinthehotelbookingmarket.Ontheotherhand,othershavedeemedtheexistinglegalframeworksufficienttodealwiththeissuesarising,suchastheUKandGermany,thelatterbanningMFNclausesinthehotelbookingmarketunderitsexistinglaw.

7.11.3. Inenforcingcompetition law,giventhefactthatvertical restraintscangive risetobothpro-andanti-competitiveeffects,asdiscussedthusfar,competitionauthoritieshaveappliedthegeneralsetofprinciplesderivingfromtheirrulesonverticalagreements,andconsideredeachcaseindividuallybyweighingupthepro-competitiveandanti-competitiveeffects(e.g.onpriceandqualityofgoodsbothintheshortandlongrun,onthelevelofpre-andpost-salesserviceprovidedinthemarket,aswellastheincentivestoinvest).Tosupportauthorities inASEANinconductingsuchassessmentsinE-commercemarkets the following typesofquestions andaccompanying guidance shouldbeconsidered:

Firstly, it should be determined whether the agreement is horizontal or vertical in nature.

• Does the agreement involve coordination between competing firms?

Ifso,consideralsotheguidanceprovidedinSection8.5onhorizontalcoordination.Ifthefirmsareatdifferentstagesofproduction,proceedontostage2below.

1.

The market share of the parties involved in the vertical agreement should be considered.

• Does any party in the vertical agreement have a large market share in the buying or selling of the good?

Ifyes,theverticalrestraint ismorelikelytohaveanti-competitiveeffects,ceterisparibus.However,dependingonthejurisdiction,somehardcorerestrictionsmaybeprohibitedregardlessofthesizeoftheparties,suchasRPM.

• Is the market multi-sided in nature?

If yes, consider not just themarket for goods/services but also themarket forproviding the platform service e.g. the share of transactions facilitated by theplatform.Additionally,goingforwardwiththeinvestigation,ensurethatallsidesofthemarketareconsideredandnetworkeffects,betweenandwithinsides,aretakenintoaccountaswellasanyfeedbackeffects.

2.

80

If it is deemed that the firms involved in the vertical restraint are not small enough to limit any anti-competitive effects from arising, and no hardcore restrictions have been breached, a more in depth assessment of the vertical restraint should be conducted. Firstly, the extent of the anti-competitive effects resulting from the restraint should be evaluated. The following questions should be considered:

• Is inter-brand competition harmed by the vertical restraint?

Ifyes,theverticalrestraintismorelikelytobeanti-competitive,ceterisparibus.

• Is intra-brand competition harmed by the vertical restraint?

Ifyes, thevertical restraint ismore likely tobeanti-competitive, ceterisparibus.However, restrictions of inter-brand competition are typically more problematicthanrestrictionsofintra-brandcompetitionthereforethenextstepshouldconsiderwhetherinter-brandcompetitionissufficient.

• Does the vertical restraint create or increase barriers to entry or expansion?

Ifyes,thehigherthebarrierstoentryinthemarket,themorelikelyistheverticalrestrainttohaveanti-competitiveeffects,ceterisparibus.

For all of the questions below, the vertical restraint is more likely to be anti-competitiveincaseofapositiveanswer.

• Is price competition inhibited as a result of the vertical restraint?

• Are retailers restricted in any way in the price that they can set, either explicitly or implicitly through threats of punishment?

• Does the vertical restraint facilitate collusion among competing firms at any stage of production?

• Are sales through an entire channel unjustifiably restricted? E.g. all online sales.

• Are online retailers unjustifiably treated differently to brick-and-mortar retailers? E.g. charged a different wholesale price, despite the costs to the manufacturer being the same.

• Does the vertical restraint reduce the diversity or quality of goods available to consumers, or the level of pre- or post-sales service in any way?

• Are any MFN clauses wide in nature (as opposed to narrow)? Note – this question depends on the competition authorities position on the use of MFN clauses

3.

81

Once the extent of the anti-competitive effects has been evaluated, any pro-competitive effects should also be considered by asking the following questions:

For all of the below: If yes, the vertical restraint may have pro-competitive benefits which may justify use of the restraint, if the benefits outweigh any anti-competitive effects.

• Doestheverticalrestrainthelptoovercomeissuesoffree-ridingbyretailers, manufacturersorotherplatforms?

• Doestheverticalrestraintreducepricetoconsumers?E.g.byovercoming doublemarginalisation.

• Areincentivestoinvestorinnovateincreasedasaresultofthevertical restraint?

4.

If it is determined that efficiency benefits may justify the use of the vertical restraint, the following factors should also be considered:

• Do consumers receive a fair share of any efficiency gains? I.e. at least compensating for the anti-competitive effects resulting from the vertical restraint.

If no, any pro-competitive effects are less likely to justify the use of theverticalrestraint.

• Are alternative vertical restraints (or other options) available to firms which are more beneficial/less costly to consumers? I.e. have less anti-competitive effects and/or greater pro-competitive effects.

Ifyes,explorethepossibilitythatthesecouldbeimplementedbyfirmsinsteadoftheexistingverticalrestraint.

5.

82

8.1. Introduction

Defininghorizontalcoordination

8.1.1. Competingfirmsatthesamestageofproductioncanhorizontallycoordinate,forexample,toincreasepricesabovetheprevailingcompetitive level inorderto increasetheirprofits.Explicitagreementsbetweenfirmsareoftenreferredtoashardcorecartels.Ahardcorecartelisdefinedas:

“an anti-competitive agreement, anti-competitive concerted practice, or anti-competitive arrangementbycompetitorstofixprices,makeriggedbids(collusivetenders),establishoutput restrictions or quotas, or share or divide markets by allocating customers, suppliers, territories, or lines of commerce”;244

and “hardcore cartels are prohibited by virtually all systems of competition law and are the

subject of ever more draconian penalties”. 245

8.1.2. Horizontalcoordinationdoesnotnecessarilyrequireanexplicitagreement.Indeedthesameoutcomecanbeachievedbymeansoftacitcollusion.Broadlyspeaking:

“a problem for competition policy arises in markets in which there are only a few operators who are able, by virtue of the characteristics of the market, to behave in a parallel manner and toderivebenefitsfromtheircollectivemarketpower,without,orwithoutnecessarily,enteringinto an agreement or concerted practice” 246;

8.1.3. Whereasexplicitagreementsincartelsarewidelyregardedashardcorerestrictions,tacitcollusionistypicallynotcaughtbycompetitionlaw.Ifamarkethasoligopolisticmarketcharacteristics,pricecompetitionwillnaturallynotbeasintenseasinamorecompetitivemarket,andfirmsoftenunilaterallyreacttootherfirms’conduct.WhishandBailey(2015),however,highlightedhowtacitcoordinationmayleadtoanallegedabuseofdominanceiffirmsareinapositionofcollectivedominance:

“adistinctissueiswhethercollectivelydominantfirmsmayabusetheirpositionbychargingexcessively high prices: here the abuse would lie not in the parallelism of the prices, but in their level.”

8.1.4. Theauthorsdo,however,notethatcasessuchasthisareveryrare,andtherehavebeenveryfewinvestigationsofthisnature,andnonetoourknowledgeinE-commercemarkets.Additionally,thosethathavebeeninvestigatedfocusedondifferentissues,forinstancerestrictingparallelimports.


245 Ibid. page 513.

246 Ibid. page 559.

Horizontalcoordination

08

83

247 Currie, D. (2017).

248 OECD (2016b), para. 81.

Effect of E-commerce on horizontal coordination

8.1.5. Horizontal coordination is a phenomenonwhich has long existed in traditional brick-and-mortarmarkets.TheemergenceandgrowthofE-commercehascreatednewchallengesforcompetitionauthoritiesindealingwithsuchcases.E-commercehasincreasedpricetransparencyinmarkets,anddigitaltoolsthatautomaticallymonitorcompetitors’priceshavemadeiteasierforfirmstoengageincollusivebehaviour,either implicitly,bymeansoftheirbest responsetocompetitors’prices,orintheimplementationofanexplicitagreement.Onlineplatformsmightalsobeusedasvehiclestoimplementhorizontalagreements.

Structure of this section

8.1.6. This section considers the challenges arising fromthe growthandemergenceofE-commerce infacilitatingcoordinationamongcompetitors,anddiscussesthecurrentstatusofthedebate,aswellaswaysinwhichcompetitionauthoritiesaroundtheworldhavedealtwithsuchissuesinpreviouscases.

8.2. Price monitoring tools and price setting algorithms

Overview of conduct

8.2.1. WiththedevelopmentandgrowthofE-commercemarkets,theeasewithwhichfirmscanmonitorcompetitors’prices,andadjusttheirpriceinresponsetoanyobservedmovements,hassignificantlyincreased.IntheearlyphasesofthedevelopmentandgrowthofE-commercemarkets,thiswouldhavebeensimplythroughmonitoringarival’swebsite.Morerecently,digitalpricemonitoringtoolsallowthis tobedoneautomatically. Such tools canmake it easier for cartels tooperate, as anydeviationsfromagreedpricesareeasiertoidentifyandreactto.

8.2.2. Toolshavealsobeendevelopedthatenablefirmstoautomaticallyadjusttheirpricesinresponsetocompetitors'pricemovementsusingalgorithmicsoftware.Whilstfirmsmayusesuchtoolsunilaterallytomaximiseprofits,concernshavebeenraisedthatsuchsoftwarehasfacilitatedcoordinationamongfirmsbyenablingthe implementationofexplicitagreements.Tacitcoordinationmayalsobecomemorecommonasaresultoftheemergenceandincreasedprevalenceofprice-settingalgorithmsinmarkets,thoughnocasesofthisnaturehavebeeninvestigatedtodate.

8.2.3. Alternatively,firmsmaycolludethroughso-calledhub-and-spokesystems,wheremultiplefirmsinamarketalloutsourceautomatedpricingtothesamethirdparty.Thisisarare,butpotentiallyharmfulsituationwherebytheoutsourcedfirm(whichcodesthepricingalgorithm)hassoldthealgorithmtoseveralcompetitors.Inthisinstance,theoutsourcedfirmhascompletepoweroverpricesthatarechargedwithinamarket,andmaypotentiallyhaveincentivestomaximiseindustryprofits(asacartelwould)asopposedtoeachfirms’ownprofitsinamorecompetitivesituation.

8.2.4. Thedevelopmentofalgorithmicsoftwarebasedon ‘machine learning’toolsmay leadtofurthercompetitionconcernsinfuture.AsDavidCurrie,theCMAChairman,recentlyremarked:“Machinelearningmeansthealgorithmsmaythemselves learncoordination is thebestwaytomaximiselong-term business objectives”.247 Were such developments in technology to arise, it is unclearwheretheliabilitywouldfall.TheOECDrecentlyindicatedthat“thereisnolegalbasistoattributeliabilitytoacomputerengineerforhavingprogrammedamachinethateventually‘self-learned’tocoordinatepriceswithothermachines.”248Thistechnologyisperhapsspeculativeatthisstage,butitissomethingthatcompetitionauthoritiesshouldbeawareofgoingforward,andcloselymonitoranydevelopmentsin,bothintheirownjurisdictionandinternationally.

84

Insights from cases

8.2.5. Giventhatthesetechnologicaldevelopmentsarerelativelynew, internationalcase law is limitedinthefield.Asitstands,therearenocasesinvolvingpricingalgorithmsthathavebeenassessedwithinASEAN,andonlyonequestionnairerespondentconsidersitasacompetitionconcernwithintheir jurisdiction at present. However, in the US, in December 2015 the Department of Justice(DoJ)prosecutedacompanyanditsfounderforengaginginaconcertedpracticewithcompetitorsregardingthesaleofpostersontheAmazonmarketplacebetweenSeptember2013andJanuary2014(seeCasereview23below).249Price-fixingalgorithmswereusedtoimplementthecollusiveagreements.AsimilarcasewasalsoinvestigatedintheUK.Tworetailerssellingpostersandframeswerefoundtohavebehavedanti-competitivelybyusingautomaticpricingsoftwaretoenforceaprice-fixingcartel.250 Inconductingits investigations,theUKCMAandUSauthoritiescoordinatedclosely.

249 US Department of Justice, Press release number 15-1488 (2015).

250 CMA, 50223, Online sale of posters and frames (2016).

CASE REVIEW 23 – TROD/GB EYE

Industry: Posters/frames

Country / Union of countries: US, UK

Court / Competition Authority: DoJ, UK CMA

Case name and citation: DoJ, Press Release: Number 15-1488; UK - Online sales of posters and frames (50223);

Date of decision: 4th December 2015 (US); 12th August 2016 (UK)

Type of alleged infringement:Pricefixingcartelimplementedthroughprice-fixingalgorithms

3 Case summary

InDecember2015,theUSDoJprosecutedacompany(TrodLtd)anditsfounderforfixingpricesinthesaleofpostersandframesontheAmazonmarketplacebetweenSeptember2013andJanuary2014.Price-fixingalgorithmswereusedtoautomaticallyimplementthisagreementontheplatform.

Similarly,intheUK,inAugust2016,TrodLtdandGBeyeLtd(GBE)werefoundtohavebeeninvolvedinanillegalpricefixingcartel,wherebyneitheragreedtoundercuteachother’spricesforpostersandframesontheAmazonmarketplacefromMarch2011toJuly2015.

AsintheUS,thepartieswerefoundtohaveusedonlineautomatedrepricingtoolstoimplementtheagreement.Inthisinstance,bothpartieswereusingdifferentre-pricingsoftwaresystems,butwerestillabletocollude.GBEimplementedaruleinitssoftwarethatifTrodLtdhadapriceset,andtherewasnootherselleronAmazonwithalowerprice,GBEwouldmatchTrodLtd'sprice,solongasthispricewasnotbelowGBE'sindependentlysetminimumpricefortheproduct.Throughtheimplementationofrulessuchasthis,thetwofirmswereabletosell99%oftheirproductsatthesamepriceatoneparticularpointintime.

IntheUK,afinancialpenaltyof£163,371wasimposedonTrodLtd,andGBEwasnotpunishedasaresultofnotifyingtheCMAofthecartelundertheCMA’sleniencypolicy.

85

251 Case C-74/14, Eturas (2016).

8.3. Online platforms and collusion

Overview of conduct

8.3.1. Competitionauthoritiesmayalsobeconcernedwithcontractualtermsregardingonlineplatformsthatmayfacilitatecollusion.Aplatformthatrestrictsinsomewaythepricethatfirmsareabletosellatviathesystemmaybedeemedtobefacilitatingcollusionasaresultofreducingcompetitiononprice.Insuchaninstance,boththeplatformandthefirmssellingthroughtheplatformmaybedeemedtohavebehavedanti-competitively.

8.3.2. Furthermore,questionnaireresponsesfromAMShighlightthedifficultiesauthoritiesfaceduetothenewandadvancedtechnicalskillsrequiredtoinvestigateandgatherevidenceontheinformationexchangedthroughonlinesystems,whichmayfacilitatecoordination.

Insights from cases

8.3.3. IntherecentEturascase (Casereview24),theLithuanianCompetitionCouncil (LCC)foundthata common cap on price discounts on hotels on the Eturas online booking system amountedto horizontal coordination among the travel agents.251 In this instance, itwas the platform thatimplemented the price cap, and travel agents were deemed to have engaged in horizontalcoordinationasaresultofacceptingtherestraintimposed.

CASE REVIEW 24 – ETURAS

Industry: Online travel booking

Country / Union of countries: Lithuania

Court / Competition Authority: LCC; Lithuanian Supreme Administrative Court (LSAC); ECJ

Case name and citation: Eturas (Case C-74/14)

Date of decision: 21st January 2016

Type of alleged infringement: Horizontal coordination on price

3 Case summary

The LCC imposedfines on Eturas (an online travel booking platform) and 30 travel agencies forapplyingacommoncapondiscountsforservicesofferedthroughtheEturasonlinebookingplatform.Thediscountcapof3%wascommunicatedtothetravelagentsviaaninternalmessagingsystem.ThisdecisionwasthenappealedtotheLSAC,whorequestedapreliminaryrulingfromtheECJ,inparticularastowhetherawarenessofthecapamountedtotacitparticipationintheagreement,andifnot,whatfactorsshouldbeconsideredindeterminingifafirmwasengagedintheagreement.

TheECJtookthepositionthatiftravelagenciesusedtheplatform,hadknowledgeofthecontentofthe internalmessage,anddidnotobjecttothediscountcaporreport ittotheadministrativeauthorities,thenitmaybepresumedthattheyhadparticipatedinthehorizontalagreement.TheLSACwasconsistentwiththeECJ’sguidance,andfoundthattheagencieswhichknewoftherestrictionanddidnotopposeitshouldbeheldtohaveparticipatedintheanti-competitiveconduct.

86

8.3.4. Firms using a platform may also collude on the prices available, or promotions offered on aplatformamongthemselves,withouttheneedforcoordinationbytheplatformitself.InarecentcaseinthefinancialadvisoryindustryinSingapore(Casereview25),acompetitorwaspressuredbyothercompetitorfirmsintoremovingalifeinsuranceofferonaplatformwebsite.252Absentthiscollectivepressure,thediscount (throughcommission rebates)wouldhaveputthecompetitorsundercompetitivepressuretofollowsuitandprovidesimilarofferstoconsumers.Thedisruptiontothefinancialadvisoryindustrywouldthereforehaveledtolowerpricesforconsumers.However,competitors were not happy with this disruption as they faced a competitive threat from aninnovativeoffer.Hence,theycolludedtopressurisethediscountingfirmintoremovingitsoffer.Asaresultofthiscollusion,improvedoutcomesforconsumerswerenotrealised.

8.3.5. Theemergenceofonlineplatformshasincreasedthetransparencyofpricesbothforcompetingfirms and for consumers, therebymaking higher pricesmore obvious to consumers, and pricematchingmorelikely.Competitionauthoritiesarelesslikelytohaveconcernsifpriceparallelismis reachedthroughunilateraldecisionsoffirms.Bycontrast,authoritiesaremore likelytoopeninvestigationsifthisparallelismisreachedasaresultofcoordinationbetweenfirmsnottoundercuteachotheronaparticularplatform.

252 CCS 500/003/13 Infringement of the section 34 prohibition in relation to the distribution of individual life insurance products in Singapore (2016).

CASE REVIEW 25 – iFAST

Industry: Financial advisory services



Case name and citation: Financial Advisers Penalised by CCS for Pressurising a Competitor to Withdraw Offer from the Life Insurance Market (CCS 500/003/13)

Date of decision: 17th March 2016

Type of alleged infringement: Horizontal coordination on price

3 Case summary

InMarch2016,tenfinancialadviserswerefoundtohaveengagedinananti-competitiveagreementinthefinancialadvisoryindustry.Thetenfirmswereadjudgedtohavepressuredacompetitorintoremovinganofferonanonlineinvestmentplatform(fundsupermart.com).Specifically,iFASTFinancialPte.Ltd.(iFAST)hadoffereda50%commissionrebateonlifeinsuranceproductsonfundsupermart.com,passingontoconsumersdistributioncostsavingsfromusingtheonlineplatform.

iFAST implemented the offer on 30th April 2013. On 3rd May 2013 the offer was withdrawn. AninvestigationintothiswithdrawalwasinitiatedfollowingmediareportsthatiFASTwithdrewtheofferduetounhappinessintheindustry.Initsinvestigation,CCSfoundthatthe10financialadvisersmeton 2ndMay2013todiscusstheoffer,whereitwasagreedthatasinglefirmwouldrepresentthegroupandputpressureon iFASTtoremovethediscount. iFASTdidnot introduceanotherofferonthiswebsiteuntilAugust2015.

CCSdeemedthatthispressurehadanadverseeffectoncompetitionduetothecontentoftheagreement, andcombinedmarketshareoftheparties.Thequantityof trafficonfundsupermart.commeantthathadiFAST’sofferremainedinplace,otherfinancialadviserswouldhavebeenundercompetitive pressure to also introduce similar incentives for customers. Ultimately, the parties’actionswerefoundtohavepreventedthemarketfrommovingtoamorecompetitivestate.

Followingthisdecision,financialpenaltieswereimposedonalltenparties.

87

253 In June 2015, Amazon itself was investigated for an alleged abuse of dominance in the market for E-books. In particular, the European Commission had concerns that MFN clauses in contracts with publishers made it harder for smaller firms in the market to compete. This case is ongoing.

254 Whish, R. and Bailey, D. (2012), page 621, explain how “the function of a sales agent is to negotiate business and to enter contracts on the producer’s behalf. In this case the agent may be paid a commission for the business it transacts or it may be paid a salary.”

255 COMP/39.847 E-Books (2012).

256 Case 13-3741, United States v. Apple Inc. et al. (2016).

8.4. Coordinated use of vertical restraints by competitors

Overview of conduct

8.4.1. Thevertical restraintsdiscussed inSection7ofthishandbookcanalsobedeemedtofacilitatehorizontalcoordinationifimplementedinaconcertedmanneramongcompetitors.Suchcoordinationcanlimitinter-brandcompetition,whichmayleadtoincreasesinpriceand/orreductionsinqualitytothedetrimentofconsumers.Inadditiontothecolludingcompetitors,otherpartiesintheverticalagreementsmay also be found to have participated in the concerted practice if they facilitatecoordination among competitors down-or upstream.Asdiscussed in detail in Section7, someverticalrestraintsdohavepro-competitiveeffects,for instanceovercoming issuesoffree-riding.Theseeffectsshouldalsobeconsideredinanycompetitionassessmentrelatingtothecoordinateduseofverticalrestraints.

Insights from cases

8.4.2. IntheE-bookscaseinEuropeandtheUS(seeCasereview26),itwasfoundthatinresponsetodecreasing prices of E-books on theAmazonplatform253, publishers collectively switched to anagencymodel254 (wherethepublishersetsprices)fromawholesalepricingstructure (wheretheretailerisfreetosetretailprices).255Inaddition,publishersimplementedMFNclauseswithApple,havingtheeffectofraisingpricesthroughoutthemarketbyeffectivelyforcingotherfirms,suchasAmazon,toadoptasimilarchangeincontractualmodel.TheUSauthoritieseventuallyfinedApple$450millionforviolatingfederalantitrust laws,highlightingthecritical rolethatAppleplayed inincreasinge-bookpricesfrom$9.99to$12.99or$14.99.256

88

CASE REVIEW 26 – APPLE AND E-BOOK PUBLISHERS

Industry: E-books



Case name and citation: COMP/39.847 — E-BOOKS


Type of alleged infringement: Horizontal coordination through switch to agency pricing model and implementation of MFN clauses

3 Case summary

In response to decreasing retail prices for E-books, for example on theAmazon platform, fivepublishers(Simon&Schuster,HarperCollins,HachetteLivre,VerlagsgruppeGeorgvonHoltzbrinck,andPenguin)enteredintocontractswithApplethataimedtoincreasethepriceofE-booksabovethosesetbyAmazon.Thecontractscomprisedaswitchfromthe incumbentwholesalemodel(wheretheretailerisfreetosetretailprices)toanagencymodel(wherethepublishersetsprices).Additionally,anMFNclausewasputinplacewhichmeantthatthepricethatApplepaidhadtobeatleastaslowasthepriceofferedtootheronlineretailers.Amazonwastherefore‘forced’toalsoadoptanagencymodelandthereforeincreaseitsprices.

ItwasdeemedthatApplesoughttocoordinatehigherpriceswithpublishers,whilstalsoensuringthatthesepricesmatchedthoseavailableonAmazon.Overall,thisconcertedpracticehadtheeffectofraisingretailpricesofE-booksacrossthemarket.

ThefivepublishersandApplehavesinceundertakencommitmentswiththeEC.ItwasagreedthattheagencyagreementswithApplewouldbeterminated,andthatotherretailerswouldbeofferedtheopportunitytoterminatetheiragencyagreements.Publisherswerealsonotallowedtorestrictretailers’abilitytosetpricesforaperiodoftwoyears,andwerenotallowedtosetMFNclausesforaperiodextendingthreefurtheryears.FourpublishersagreedtothesecommitmentsinDecember2012,whereasthefifthpublisher,Penguin,didnotfinaliseitscommitmentsuntilJuly2013.

IntheUS,asaresultoftheswitchtoanagencymodel,andthesimultaneousimplementationofMFNclauses,pricesintheE-booksmarketrosefrom$9.99to$12.99or$14.99.ItwasfoundthatAppleplayedanintegralroleinthismarketshift.InFebruary2016theAppealscourtupheldtheinitialdecisiontofineApple$450millionforbreachingantitrustlaws.Thefivepublishersalsosettledearlierinproceedings.

89

257 See, for example, Ezrachi, A. and Stucke, M. (2016).


8.5.1. AsevidencedinSections8.2-8.4above,existingcompetitionpolicyandlawappearstobeabletodealwithmostcasesinvolvinghorizontalcoordination.Therefore,thereappearstobenoneedforanoverhaulofcompetitionpolicyandlawtodealwithissuesofhorizontalcoordinationarisinginE-commercemarketscurrently.

8.5.2. Competition authorities should, however, monitor the development of pricing algorithms veryclosely.Ifpricingalgorithmsweretoself-learnthatcoordinationisoptimalduetobuilt-inmachinelearningcapabilities,itisunclearunderexistingcompetitionpolicyandlawifandwheretheliabilitywouldfall.AlthoughthisisnotcurrentlyaprobleminE-commercemarketsbecausethetechnologyhasnotbeendeveloped,debateontheissueisalreadydeveloping,thoughnoclearinternationalconsensushasyetbeenreached.257

8.5.3. Inapplyingandenforcingcompetitionpolicyandlaw,competitionauthoritiesmaywishtoconsiderthe following types of question and accompanying guidance to determine when coordinationamongcompetingfirmsisanti-competitive:

Firstly, it should be determined whether the conduct by firms amounts to a hardcore cartel or explicit collusive agreement by asking the following types of question:

• Are firms explicitly agreeing to fix prices, share markets or limit output?

Ifyes,theagreementishighlylikelytobedeemedtoconstituteacartel.

• Do a platform’s terms of use restrict in any way the price that firms can sell at on that platform?

Ifyes,collusionmaybefacilitatedbytheplatform(asintheEturascase–seeCasereview24)inparticularifcompetingfirmsareawareof/agreetothesametermse.g.alimitonpricediscounts.

• Are firms coordinating in any way to collectively implement a vertical restraint, such as an MFN clause? E.g. as in the E-books case (Case review 26).

Ifyes,collusiveoutcomesmaybereachedbysuchmeans.

1.

90

If horizontal coordination between firms is not deemed to be a hardcore cartel, a more in-depth evaluation should be conducted by comparing the anti-competitive effects of such coordination with any efficiency benefits. To determine the extent of the anti-competitive effects resulting from the coordination, the following questions should be asked:

• Does the horizontal coordination result in higher prices, a reduction in quality of goods/service, a decrease in the level of investment/innovation, and/or a decrease in consumer choice?

Ifyes,anti-competitiveeffectsfromhorizontalcoordinationaremorelikely.

To support competition authorities in determining the effect of horizontalcoordinationonthesemarketoutcomes,thefollowingquestionscanbeasked:

Forallofthequestionsbelow:Iftheanswerisyes,theextentofanyanti-competitiveeffectsislikelytobegreater.

• Do the parties of an agreement have a large market share, individually and/or collectively?

• Are the firms close competitors?

• Is it hard for customers to switch to an alternative provider?

• Are there high barriers to entry?

• Is the market transparent, concentrated, non-complex, stable and/or symmetric?

All of these factors increase the likelihood of collusive outcomes being reachedfromhorizontalcoordination(EuropeanCommission(2011),para.77).

2.

Once the extent of the anti-competitive effects has been evaluated, any pro-competitive effects should also be considered by asking questions such as:

• Are there any efficiencies arising from the horizontal agreement, for example resulting from the sharing of complementary skills/assets between the firms, risk sharing, and/or knowledge and innovation sharing?

Ifyes,thehorizontalcoordinationmayhavepro-competitivebenefitswhichmayjustifyuseofcoordination,ifthebenefitsoutweighanyanti-competitiveeffects.

3.

91

If it is determined that efficiency benefits may justify the horizontal coordination, the following factors should also be considered:

• Do consumers receive a fair share of any efficiency gains? I.e. at least compensating for the anti-competitive effects resulting from the coordination. (Note: This is not a requirement in all jurisdictions).

Ifno,anypro-competitiveeffectsarelesslikelytojustifythehorizontalcoordination.

• Are alternative less restrictive agreements available to firms which are more beneficial/less costly to consumers? I.e. have greater pro-competitive effects and/or fewer anti-competitive effects.

Ifyes,explorethepossibilitythatthesecouldbeimplementedbyfirmsinsteadoftheexistinghorizontalcoordination.

4.

92

9.1. Introduction

Definingdominance

9.1.1. OperatorswithmarketpowerinE-commercemarketsmayhavetheabilitytoengageinunilateralanti-competitiveconductbyabusingadominantpositioninthemarket.Adominantpositionmaybedefinedas:

“a position of economic strength enjoyed by an undertaking which enables it to prevent effective competition being maintained on the relevant market by affording it the power to behave to an appreciable extent independently of its competitors, customers and ultimately its consumers”. 258

9.1.2. The general definition of dominancewithinAMS is “a situationwhere the business operator hasenougheconomicstrengthtoact inthemarketwithoutregardtowhat itscompetitors (actualorpotential)do”. 259

9.1.3. AsnotedinSection6,someAMSrelyonamarketsharethresholdtodefinedominance,whilstothersarenotequallyprescriptive.

Assessing dominance in E-commerce markets

9.1.4. Whenassessingmarketpoweranddeterminingwhetherafirmisdominant,otherfactorsbeyondmarketshareshouldbeconsidered,asdiscussedinSection6.3.Inparticular,thisshouldtakeintoaccountthepresenceandextentofanycountervailingbuyerpowerofcustomers,aswellastheabilityofsmallerfirmstoexpandinthemarket,andnewfirmstoenter.Barrierstoentry,suchasnetworkeffectsandswitchingcostsforconsumers(asdiscussedinSection4)shouldthereforebeconsidered.

9.1.5. Itshouldbenoted,however,thatduetothenatureofnetworkeffects,onlineplatformsoftenhavetheabilityto increasetheirmarketshareoverashortperiodoftime.Facebook’sentry,and rapiddisplacementofMySpaceasthemarketleaderinonlinesocialmedia260demonstrateshowonlinefirmscanrapidlygainorlosemarketshare.

9.1.6. AsdiscussedinSection6,thereisanongoingdebateastowhetheraccesstodataisasourceofmarketpower.Someseeitasanassetthatsmallerfirmsareunabletoreplicate,however,dataisoftenreplicable,andcanbepurchasedfromanumberofsources,thereforemitigatingthisconcern.Todate,onlytheBundeskartellamt’sinvestigationintoFacebookhascentredaroundallegedabuseofdominancefrominfringingdataprotectionrules.Inthisinstance,FacebookdoesnothavesignificantmarketpowerbecauseoftheBigDatathat itholds,butrathertheBundeskartellamtallegedthatFacebookhasabuseditsdominantpositioninthesocialmediamarketbyimposingunfaircontractualdata terms and conditions on its users. Nevertheless, the Bundeskartellamt has indicated thatFacebookwillnotbefinedforthisconductfollowingtheinvestigation.261Internationalconsensushasyettobereachedastowhethercompetitionlawordataprotectionlawarethebesttoolstodealwiththeseissues,andwhetherdataownershipgivesrisetomarketpower,andtherebytheabilitytoexploitconsumersandexclude(ormarginalise)competitors.


259 ASEAN (2013), page 9.

260 See Section 4 for further information on a new entrant displacing an incumbent, e.g. MySpace and Facebook, and Taobao and eBay.

261 Whilst the investigation is ongoing, Andreas Mundt has made it clear there is no risk of a fine for Facebook, and the issue is being dealt with by the antitrust authority (rather than the German privacy agency) because they are seen to have a broader impact on privacy issues.

Unilateralconduct09

93


263 Ibid.

264 39740 Google comparison shopping (2017).

265 40411 Google Adsense (2017).

266 Streetmap.EU Limited v Google Inc., Google Ireland Limited and Google UK Limited [2016] EWHC 253 (Ch).

Abuse of dominance in E-commerce markets

9.1.7. Adominantfirmmayabuseitspositioninmanyways,forexamplebysettingunreasonablyhighprices,sellingatartificially lowpricessoastoforeclose itscompetitorsfromthemarket, imposingunfaircontractualterms,orforeclosingcompetitorsinthemarketthroughotherpracticessuchasbundling.

9.1.8. Thissectionlooksatsomerelevantcasesfromacrosstheworld,focusingonthetypesofconductthataremostcommonlyobservedinE-commercemarketsandhaveposedchallengestocompetitionauthoritiesaroundtheworld.ThereareotherformsofunilateralconductthatmaybedeemedtobeanabuseofdominancewhicharenotdiscussedhereastheydonotraiseanyspecialconsiderationswithregardtoE-commerce.

9.2. Tying/bundling

Overview of conduct

9.2.1. Aformofunilateralconductthat is relevantforE-commerce istheabuseofdominancethroughforeclosureofcompetitorsbytyingorbundling.Tying(underunilateralconduct)isasituationwherecustomerspurchasingagood/servicefromadominantfirmarealsorequiredtopurchaseanotherproductfromthesamefirm.262Bundlingcanbeeitherpureormixed.Purebundlingoccurswhenproductsare sold jointly infixedproportions, andmixedbundling (sometimesknownasamulti-productrebate)occurswhenproductsareavailableseparatelyinadditiontoasabundle,butthesumofthepriceswhenpurchasingproductsseparatelyishigherthanthebundleprice.263

9.2.2. Typically,bundlingandtyingarenotanti-competitiveperse.However,competitionconcernsmayariseifthemandatorysecondarypurchaseisforanunrelatedproductorservice.Thereareeconomicbenefitsthatcanberealisedfromengaginginsuchpractices.Forexample,afirmmayusetyingorbundlingtosaveinproduction,distributionandtransactioncosts.However,tyingorbundlingproductscanextendadominantfirm’spositionintoanothermarketthatmayhavepreviouslybeencompetitive.Undersuchcircumstances,competitionauthoritiesmaydeemtyingorbundlingtobeanti-competitivesincesuchconductfavoursthedominantfirm’sgood/serviceoverotherfirms’offerings.Giventhattherearebothpro-andanti-competitiveeffectsassociatedwiththebundlingandtyingofgoods, it issensibleforcompetitionauthoritiestoconductafullanalysisoftheeffectstoassesssuchpractices.(Notethatalthoughinthishandbooktheapplicationofaneffects-basedapproachtotheassessmentofunilateralconductisadvocated,establishedcaselawintheareastillreliesheavilyona‘form-based’approach,whichfocusesontheformoftheconductratherthanitseffect.)

Insights from cases

9.2.3. Bundlingandtyingstrategieshavelongbeenemployedbyfirmsinbrick-and-mortarmarkets.However,theuseofsuchstrategiesisalsoprevalentinE-commercemarkets,particularlyinmulti-sidedmarketswhereplatformssuchasGoogleofferavarietyofrelatedservicesforinternetusers.Googlehasbeeninvestigatedforaseriesofalleged instancesof favouring itsownservicesovercompetitors,withanumberofongoinginvestigationswhichhaveyettoreachaconclusiontodate.

9.2.4. Oneof these investigations involves theEuropeanCommissionexploringwhetherGoogleabusedadominantpositioninonlinesearchbyfavouringitsownonlinecomparisonshoppingserviceoveritscompetitors’ (seeCase review 27).264 InJune 2017, the EuropeanCommission determined that thisconduct amounted to an abuseofdominance, and therefore issuedafineof€2.42billion, thoughGooglemaydecidetoappeal. Inaddition,theEuropeanCommissionhadconcernswiththeway inwhichGooglerestrictswebsitesfromdisplayingsearchadvertsfromGoogle’scompetitors.265

9.2.5. Googlehaspreviouslybeeninvolvedinsimilarcasesofallegedlyabusingitsmarketpositioninonlinesearchtofavouritsownservices.In2016,along-runningdisputebroughtagainstGooglebyStreetmapreacheditsconclusion(seeCasereview28).266Specifically,itwasallegedthatGooglehadaimedtoextenditspositionofdominanceinonlinesearchbyfavouringitsownmappingserviceinthewayitdisplayedsearchresults,therebyforeclosingStreetmapfromthemarket.However,theUK’sHighCourtruledthatGoogle’sactionswereinsteadtheresultofpro-competitiveinnovationratherthananti-competitiveconduct.

94

9.2.6. Thiscaseillustratesthatnotallinstancesofbundlingareanti-competitiveassuchbehaviourcanbetheresultofinnovationleadingtohigherqualitygoodsandservices,andtherebygeneratingbenefitsforconsumers.Competitionauthorities,aswellascourts,shouldexploretheeffectsofspecificformsofconductonthemarket.

CASE REVIEW 27 – GOOGLE SEARCH

Industry: Online search



Case name and citation: 39740 Google comparison shopping; and 40411 Google Adsense

Date of decision: 27th June 2017; Ongoing

Type of alleged infringement: Abuse of dominance

3 Case summary

Ina long-running investigation, theEuropeanCommissionhas investigatedGoogle foranallegedabuse of its dominant position as a search engine; specifically relating to Google systematicallyfavouring its own comparison shopping service in its search result pages ahead of competingcomparisonshoppingserviceproviders.InJune2017,theEuropeanCommissionannouncedthatithaddeterminedthisconducttobeanabuseofdominance,andthereforefinedGoogle€2.42billion.Googlemay,however,decidetoappeal.Akeyareaofdebateislikelytobehowwidelythemarketshouldbedefined.Nonetheless,theEuropeanCommissionhasstatedthatevenifthemarketweretobemorebroadlydefined,itwouldstillhavecompetitionconcernsrelatingtoGoogle’sconduct.

AnadditionalallegedabuserelatestoGooglerestrictingthirdpartieswebsitesfromdisplayingsearchadvertsfromGoogle’scompetitors, i.e.,advertsonotherwebsitesfacilitatedbyGoogle’sAdSenseplatform.

CASE REVIEW 28 – STREETMAP v GOOGLE

Industry: Online search and mapping services


Court / Competition Authority: UK High Court

Case name and citation: Streetmap EU Limited v Google Inc., Google Ireland Limited and Google UK Limited [2016] EWHC 253 (Ch)

Date of decision: 12th February 2016


3 Case summary

Streetmapwasanonlinemapprovider,launchedin1997.In2005,Googleintroduceditsownonlinemappingservice,‘GoogleMaps’,torivalStreetmap.In2007,Googlelaunchedasmallthumbnailboxfeatureinthetoprightcornerofitsonlinesearchresultspagecalled‘MapsOneBox’,containingamapresultrelatedtotheinitialsearch.Streetmaparguedthatthisformof‘bundling’wasanabuseofGoogle’sdominantpositionintheonlinesearchmarket,anddroveonlinetraffictoGoogleMapsattheexpenseofStreetmap.Streetmap’sargumentwasnotthatthesmallthumbnailmapshouldnotbepresent,butthat it shouldfeature resultsfromotheronlinemapproviders.However, theHighCourtintheUKrejectedStreetmap’sclaims.ThejudgeconcludedthatGooglewasobjectivelyjustifiedtoincludeMapsOneBoxwithinthesearchresultsasitimprovedthegeneralsearchenginetothebenefitofusers.InFebruary2017,StreetmapwasdeniedtheoptiontochallengethedecisionbytheUKCourtofAppeal.

95

267 40099 Google Android (2017).

9.2.7. TheEuropeanCommission is currently investigatingGoogle for an alleged abuseof dominanceofits position in the mobile phone operating systemmarket (see Case review 29),267 regarding theallegedbundlingofGoogle’sAndroidoperatingsystemwithGoogleapps,inadditiontootherpotentialinfringements.

CASE REVIEW 29 – GOOGLE ANDROID

Industry: Mobile operating systems

Country / Union of countries: Europe, US, Korea, Russia


Case name and citation: 40099 Google Android



3 Case summary

InApril2015theEuropeanCommissionopenedaformalinvestigationintoGoogle’sAndroidmobileoperatingsystem.SimilarcasesarealsobeinginvestigatedintheUS,Korea,andRussia.InparticularitisbeinginvestigatedwhetherGooglehaseitherenteredintoanti-competitiveagreementsand/orhasabusedapositionofdominance.

TheEuropeanCommissionisinvestigatingthreeallegations:

1.Whether rivalmobileapplicationswerehinderedasa resultofGooglerequiringor incentivisingdevicemanufacturerstoexclusivelypre-installGoogle’sownappsandservices;

2.Whethersimilarharmwascausedbythetying/bundlingofGoogleappsandservicesonAndroiddevices; and

3.WhetherpreventingdevicemanufacturersfromdevelopingmodifiedandcompetingversionsofAndroidonotherdevicesinhibitedcompetitionintheoperatingsystemsmarket.

InitsStatementofObjectionsinApril2016,theEuropeanCommissionoutlinedthatithadreachedapreliminaryviewthatGooglehasabused itspositionofdominanceby imposingrestrictionsonAndroid devicemanufacturers andmobile network operators. Specifically, Google Search is pre-installed and set as the default search engine on most Android phones. Additionally, financialincentivesareoftenofferedtomanufacturersandmobilenetworkoperatorsthatexclusivelypre-installGoogleSearch.Finally,manufacturersarepreventedfromsellingsmartmobiledevicesrunningoncompetingoperatingsystemsbasedontheAndroidopen-sourcecode.

TheEuropeanCommissionisconcernedthatthiswillstrengthenGoogle’spositionintheinternetsearchmarket, and inhibit competitors to Google Chrome in themobile browsers market. TheEuropeanCommission isalsoconcernedthatthedevelopmentofnewoperatingsystemsbasedonAndroidsourcecodeisbeinginhibited,therebyharmingconsumersthroughlimitingchoiceandstiflinginnovation.

96

CASE REVIEW 30 – MyEG

Industry: Online Foreign Workers Permit Renewal applications / Insurance

Country / Union of countries: Malaysia

Court / Competition Authority: MyCC

Case name and citation: My E.G. Services Berhad

Date of decision: 24th June 2016


3 Case summary

InJune2016,MyE.G.ServicesBerhad(MyEG)wasfoundtohaveabuseditspositionofdominanceinthemarketforonlineForeignWorkersPermitRenewalapplicationsfollowingcomplaintsbyotherparties.

Specifically,MyEGwasfoundtohave inhibitedcompetition inthesellingofmandatory insurancepolicies – products that are also sold by a number of competitors. Complainants argued thatemployersofforeignworkerswere“forced”topurchase insurancesthroughMyEG,andwhentheemployerswereallowedtopurchasetheinsurancesfromotherinsurancecompaniesorinsuranceagents,itwasallegedthatMyEGhadimposedunfairandunreasonableconditionsonsuchparties.

AfinancialpenaltyofRM307,200wasimposedonMyEG,whowasalsorequiredtoremoveexistingagency agreements with regard to mandatory insurances, and provide entry for all insurancecompaniestosellmandatoryinsurances,allowingthemtocompeteatthesamelevel.

9.3. Predatory pricing

Overview of conduct

9.3.1. Predatorypricingoccurswhen:

“adominantfirmdeliberatelyreducespricestoaloss-makinglevelwhenfacedwithcompetition from an existing competitor or a new entrant to the market; the existing competitors having been disciplined, or the new entrant having been foreclosed, the dominantfirmthenraisesitspricesagain,therebycausingconsumerharm”.269

9.3.2. Insingle-sidedmarkets,pricingbelowaveragevariablecostmaythereforebeconsideredindicativeofapredatorystrategy.Inonlineone-sidedmarkets,thesameapproachtodeterminingwhenbelow-costpricingispredatorybehaviourbyadominantfirmcanbeusedasintraditionalbrick-and-mortarone-sided markets.

9.3.3. Inmulti-sidedonlinemarkets,however,below-costpricingononesideofamarket isacommonstrategyemployedbyfirmstoattractusersonanothersideofaplatform,duetotheexternalitiesbetweenthedifferentsidesofamarket.Thismaybetrueeveninthelongrun,beyondaninitialphaseof ‘penetrationpricing’thatafirmmay implementwhenenteringamarket.270Forexample,onlinesearchandsocialmediaservicesaretypicallyfreeforindividualsinordertoattractadvertiserswhoarechargedforusageoftheplatform.Socialmediausersarethereforechargeda“price”(equaltozero)belowthecostoftheservicetotheplatform.Thisisnotconsideredanti-competitivebehaviour.

9.2.8. AlocalexampleofanabuseofdominancecaseinASEANisoutlinedinCasereview30.268Specifically,theMalaysiaCompetitionCommissionfoundthatMyEGhadabuseditsdominantpositionintheonlineprovisionofForeignWorkersPermitRenewalapplicationsbyrequiringsomecustomerstoalsopurchaseinsurancethroughitssite,therebypreventingcompetitioninthemarketfortheseinsuranceproducts.

268 Press release: MyCC issues final decision against MY E.G. Services Berhad (MyCC, 2016).

269 Whish, R. and Bailey, D. (2015).

270 Evans, D. (2004).

97

271 Recoupment of profits is not often a necessary part of the assessment. The ability to recoup profits in the future is part of the assessment in the United States, but is not in the European Union.

272 12/02931 Google/Evermaps (2015).

9.3.4. Inassessingallegedpredationinmulti-sidedmarkets,competitionauthoritiesmaythereforeneedtolookatthepricechargedtoallsidesofamarket,andcostsincurredinservingallcustomergroups.AsEvans(2004)discussed,thiscanbedonebycomparingthetotalpricechargedtoallsidesofamarketpertransactionwiththeincrementalcostpertransactiontoallsides.Or,ifpriceorcostpertransactioncannotbedetermined,for instanceasusersarechargedanaccessfeeasopposedtoatransactionfee,thetotalrevenuecanbecomparedwiththetotalvariablecosts.Asinone-sidedmarkets,competitionauthoritiesmaythenwishtoexplorewhetherthedominantfirmhasareasonableprospectofrecoupingprofitsbychargingahighertotalpriceinthefuture,271oncecompetitorshaveleftthemarket,againconsideringallsidesofthemarket.

Insights from cases

9.3.5. The approach discussed abovewas utilised by the Paris Court of Appeal in its review of allegedpredationbyGoogle(seeCasereview31below).272Inparticular,itwasallegedthatGooglehadforeclosedEvermapsfromthemarketbyofferingmappingservicestoretailersforfree.However,theParisCourtofAppealfoundthatGooglewasinfactcoveringitscostswhenitalsoconsideredrevenuesobtainedfromadvertisingontheothersideofthemarket.Itwasthereforeconcludedthatthepricingpracticewasnotpredatory.

CASE REVIEW 31 - GOOGLE MAPS

Industry: Online mapping services

Country / Union of countries: France

Court / Competition Authority: Paris Commercial Court / Paris Court of Appeal

Case name and citation: 12/02931 Google/Evermaps

Date of decision: 25th November 2015

Type of alleged infringement: Abuse of dominance through predation

3 Case summary

InJanuary2012,theParisCommercialCourtfoundthatGooglehadabuseditspositionofdominanceinonlinemappingservices(allowingretailerstoprovidedirectionsandlocationinformationontheirwebsite)bypricingitsservicebelowcost(orrather,forfree),therebyforeclosingEvermaps(formerlyBottinCartographes)fromthemarket.Googlewascharged€500,000indamages.However,Googleappealed,and,inNovember2015,theParisCourtofAppeal,havingsoughtadvicefromtheFrenchCompetitionAuthority,ruledthatthepricingstructurewasnotpredatoryasincomefromadvertisingontheothersideofthemulti-sidedmarketmeantthatGooglewasinfactcoveringitscosts.

Initsdecision,theAppealCourtexplainedthat:

“The Authority has rightly observed that for operators on multisided markets it may be rational…to provide free products or services in a market not to foreclose competitors but to increase the number of users on the other market [and that] the free business model is quite widespread in electronic markets”

98

273 OECD (2016), para. 1.

274 OECD (2016), para. 17.

275 Ibid, para. 18

276 Ibid. para. 32.

277 Ibid. para. 123.

9.4. Price discrimination

Overview of conduct

9.4.1. Price discrimination occurs when “two similar products, which have the samemarginal cost toproduce, are sold by a firm at different prices”,273 and exists in both online and offlinemarkets.Pricediscriminationisnotnecessarilyaconcernforcompetitionauthoritiesandit isnotgenerallyregardedasaviolationofcompetitionlawasitcangiverisetoefficienciesbyincreasingtradeanddrivingcompetition.However,pricediscriminationcanraisecompetitionissuesifithasexploitative,distortionaryorexclusionaryeffects.

9.4.2. Pricediscriminationiscategorisedunderthreedifferentgroups:

a. First-degree,orperfectpricediscrimination,involvesafirmsettingpriceequaltoeachcustomer's willingnesstopayforthatgood/service.

b. Second-degreepricediscriminationisindirectasitinvolvessettingamenuofpricesfordifferent versionsoftheproduct.Thedecisionofwhattopaythereforerestswiththecustomer.274Business classandeconomyairfaresmaybeconsideredanexampleofsecond-degreepricediscrimination wherethecustomers“self-select”andchoosetheclassoffaresthemselves.

c. Third-degreepricediscrimination involvesafirmsettingdifferentprices fordifferent groupsof consumers(e.g.lowerpricesforpensionersorstudents).275

When competition authorities may investigate price discrimination

9.4.3. Fundamentally,pricediscriminationisnotperseanti-competitive.Itcanincreaseallocativeefficiencythroughmoreconsumersbeingserved.Someconsumerswhowouldnotpurchaseaproductundersinglepricing arenowable to afford theproduct. Pricediscrimination isvisible inmanydifferentmarkets,andfirmsusethesestrategiesregardlessoftheirlevelofmarketpower.Becauseofthis,theOECDsuggestthatcompetitionauthoritiesshouldhavearebuttablepresumptionthatanyobservedpricediscriminationschemehasabenignorbeneficialimpactonconsumers.276However,therearetimeswhencompetitionauthoritiesmightwanttoinvestigatepricediscrimination,ontheprincipleofprotectingtheinterestsofconsumers.Inotherwords,byandlargetheissueofconcernwithpricediscriminationmightberelevantforconsumerlawandpolicyratherthanforcompetitionlaw.

9.4.4. Authoritiesmayalsowishtolaunchaninvestigationwhendistortionarypricediscriminationoccursupstream,as itcan result inhigherpricesbeingchargedtofinalconsumers. Inthissituation,theactionsofadominantupstreamfirmcanleadtoadownstreamfirmpayinghigherpricesfortheirinputs,whicharethenpassedontoconsumers.

9.4.5. TheOECD(2016)listedseveralotherreasonsastowhypricediscriminationmaybescrutinised,inparticularfor:

“concepts of fairness, or other policy goals, such as the desire to operate a single market, or to protect domestic producers and consumers from excess production by organisations in non-market economies.” 277

99

278 Where the marginal cost of a good is close to zero, the scope for price discrimination is greater, as a supplier or retailer is incentivised to sell the greatest quantity possible in order to cover their fixed costs and return a profit. As this is often the case for digital goods or services in online markets, price discrimination is common where retailers attempt to maximise output. Looking at mobile applications, an app developer may implement such a strategy by offering a free basic version of an app, whereby revenue is generated from advertisements, in addition to offering a superior advert-free version of the app sold for a small fee. The superior version targets users who place a higher value on the app and therefore have a greater willingness to pay.

279 Mikians et al (2012).

280 OECD (2016), para. 144.

281 OECD (2016), para 152.

282 OECD (2016), section 4.

283 Ibid. para 147.

284 OECD (2016).

285 European Commission (2017c).

Price discrimination in E-commerce markets

9.4.6. ItisgenerallyeasierforfirmstoimplementpricediscriminationstrategiesinE-commercemarketsasconsumerscanbeofferedatailoredpricebasedondatathatafirmholdsonthatconsumer.278 This data is most likely to be technological/system based, geographic, or personal/behaviouralinformation.279 Firmsusethisinformationtoassessaconsumer’swillingnesstopayforaproductorservicebasedontheirbehavioursand/orcharacteristics.280

9.4.7. Intraditionalbrick-and-mortarmarkets,tailoredpricingtothisextentisrarelypossible,asitwouldtakesignificanttimeandrequireanad-hocdatacollectionexercisefortheretailertomakeareasonableestimateofaconsumer’swillingnesstopay.Suchconstraintsareno longerpresent intheonlinespace,withonline retailersbeingabletogathervastamountsofdataand resort topersonalisedpricing.TheOECD (2016)281,however,highlightedthatpricediscriminationofthisform isnot likelytobeaconcernifallcompetitorshaveaccesstosuchdata;and,asdiscussedinSection4.3ofthishandbook–ifthedatathatfirmsholdisnon-rivalrous,non-excludable,andcanbepurchasedfrommultiplesources,thisconditionislikelytobemetinmostmarkets.

9.4.8. Whilstpricediscriminationisnotanewphenomenonarisinginonlinemarkets,firmsnowhavethetoolsanddatatotargetspecificconsumersbasedoncertainattributes.Thisshouldbeofconcerntoauthoritieswhenfirmswithfewcompetitorsextractconsumersurpluswithoutexpandingoutput,usingpersonalisedprices.Thesepartitioningstrategiesmayfacilitateexploitativepricediscrimination,increasingmark-upsandmarketpowerattheexpenseofconsumers.282

9.4.9. Personalised price discriminating strategies deployed by E-commercefirms include price testing.Pricetestingoccurswhenafirmoffersdifferentpricesdependingonthetimeofday, geographiclocationofthecustomer,orothercharacteristicsthatallowthefirmtodeveloppredictivemodelsonagivenindividual’swillingnesstopay,andtheirelasticityofdemand.283Itispossibleforonlinefirmstochangetheirpriceseveryminute,especiallyusingautomatedpricingsoftware,apracticethatisnotconvenient,orevenpossible,forbrick-and-mortarstores.

9.4.10. Itisimportanttonote,however,asmentionedabove,thatpricediscriminationsuchaspersonalisedpricingisnotacompetitionprobleminitself,butmaygiverisetoconcernsaroundfairness.Issuesoffairnessarebetteraddressedviamoresuitedpolicyinstrumentssuchasconsumerlaw,ratherthancompetitionlaw.

9.4.11. Oneofthekey issuesforcompetitionauthoritieswithrespecttopricediscrimination issimilar inonlineandofflinemarkets;that istopreventpricediscriminationthatstrategicallyexcludesrivals.Exclusionarypricediscriminationofthisnaturecancreate,buildandprotectmarketpowerattheexpenseofconsumers.Competitionauthoritiesshouldfocusoninstanceswherepricediscriminationisusedasameanstoexcludearivalwhichdoesnotrequirethefirmtosacrificeprofits(i.e.marginsqueeze,fidelityrebatesandbundleddiscounts).284

Insights from cases

9.4.12. The European Commission is currently investigating price discrimination in the online hotelaccommodationmarket, followingcomplaintsfromconsumers.Theagreements inquestionarebetweenlargetouroperators(suchasKuoni,REWE,ThomasCookandTUI)andhotels(MeliaHotels),whichmaydiscriminatebetweencustomersbasedontheirnationalityorcountryofresidence.285 Whilst the competition authoritywelcomes innovativepricingmechanisms, they cannot lead topricediscriminationbasedonacustomer’slocation.

100

286 OECD (2016), para. 107.

287 ASEAN Today (2016).

288 See for example OECD (2002).


290 Whish, R. and Bailey, D. (2015), page. 770.


9.5. Fidelity rebates or loyalty discount schemes

Overview of conduct

9.5.1. Anextensionofpricediscriminationstrategiesareloyaltydiscountschemes,alsoknownasfidelityrebatesorexclusivityrebateschemes,presentinbothbrick-and-mortarandE-commercemarkets.Theseoccurwhenadominantselleroffersamorefavourableprice, rebateorfinancialadvantageto the buyer, conditional on their loyalty in the purchases theymake.286 Rebates are common in E-commercemarketswherefirmsprovidecustomerswithfinancialincentivesinreturnforfeedback,orreviewsonarecentpurchase.287

9.5.2. Under certain circumstances, such a practice can foreclose competitors and reinforce a firm’sdominantpositioninthemarket.Whilstofferingrebatestocustomersisnotinitselfanti-competitive,assuchpricingstructurescanintensifycompetitionamongstsuppliers,caselawfromaroundtheworldsuggeststhatrebatesandloyaltypricepracticescanhaveadetrimentaleffectoncompetition.288

EUcompetition lawhastraditionallyfound loyalty rebateschemestoconstituteanabuseofthatdominant position.289Suchcase law,however,hastypicallypursuedastrictapproachfocusingonthe structure of the rebate (a ‘form-based approach’)which can be loyalty-inducing, rather thanexaminingtheactualimpactonthemarketvis-à-vistheabilityofcompetitorstomatchthoserebatesandcounterapotentialforeclosure(an‘effects-based’approach).

Insights from cases

9.5.3. Simplequantity rebatesthatareonly linkedtothevolumeofsalestoacustomerarecommonlypresumedtobe lawful.On theotherhand, exclusivity rebates, inwhichdiscounts areoffered toconsumerswhopurchasefromadominantfirm,aretypicallyconsideredunlawfulunlessobjectivelyjustified.290

9.5.4. Caselawonrebatesreliesoninsightsfrombrick-and-mortarmarkets.Thereisnoreasontobelievethat online rebates,with the exception of the considerations presented in Section 9.4 in relationto price discrimination, would require any specific approach which would differ from a properassessmentoftheabilityofcompetitorstocompeteeffectivelyonthemarketandtherebyavoidanypotentialforeclosure.Therearecurrentlynorelevantcasesconsideredbyanycompetitionauthoritywhichexamineloyaltyrebatesinonlinemarkets.

9.6. Imposing vertical conditions (e.g. quantity forcing)

Overview of conduct

9.6.1. Dominantfirmsthat imposevertical restraintsonotherpartiesatdifferentstages inthechainofproductionmaybedeemedtohaveabusedtheirpositioninthemarket(adominantfirmmaylackthe incentivestogenerateand/orpassonefficiencygainstoconsumers).Consequently,averticalrestraintwhichmaintains,createsorstrengthensadominantpositioninthemarketcannotnormallybejustifiedonthegroundsthatitcreatesefficiencygains.291

101

292 Abuse of a Dominant Position by SISTIC.com Pte Ltd CCS/600/008/07 (2010).

293 Lexis (2017).

CASE REVIEW 32 – SISTIC

Industry: Online event ticketing



Case name and citation: Abuse of a Dominant Position by SISTIC.com Pte Ltd (CCS/600/008/07)

Date of decision: 4th June 2010

Type of alleged infringement: Abuse of dominance through exclusive agreements

3 Case summary

In2010CCSfoundthatSISTIC.com(SISTIC)hadabuseditspositionofdominanceintheticketserviceprovidersmarket.Ticketserviceprovidersactasaplatformconnectingeventpromotersandticketbuyers.ItwascalculatedthatSISTIChadapersistentmarketshareof85-95%inthismarket.

SISTICwasfoundtohaveabuseditspositionofdominancethroughitsexclusivityagreementswithcertainvenues.Forexample,keyvenuessuchasTheEsplanadeandSingaporeIndoorStadiumwererequiredtouseSISTICasthesoleticketproviderforallevents.

Itwasfoundthattheseagreementsrestrictedthechoiceofvenueoperators,eventpromotersandticketbuyers.EvidenceofthiswastheabilityofSISTICtoincreaseitsbookingfeeforticketbuyersby50%toS$3in2008.

Initsdecision,CCSinstructedSISTICtochangeitsagreements;inparticularremovingclausesthatrequiredSISTIC'scontractualpartnerstouseSISTICexclusively.SISTICwasalsofinedS$989,000forinfringingsection47oftheSingaporeCompetitionAct(abuseofadominantposition).

On 3rdAugust2010,SISTICappealedthisdecision.TheCompetitionAppealBoardoftheRepublicofSingaporeupheldCCS’sdecisionbutreducedthefinancialpenaltytoS$769,000.

Insights from cases

9.6.2. In2010,CCSfoundSISTICtohaveabuseditspositionofdominanceinthemarketforonlineticketsalesasaresultofimposingexclusivityagreementsoneventvenues(seeCasereview32).292


9.7.1. Thecasesoutlinedabove illustratethat instancesofallegedabuseofdominance inE-commercemarketstendtodiffergreatly innature.However, inall instances,existingcompetitionpolicyandlawhasbeensufficient todealwiththe issuesarising. Inapplyingthis law, there isnoone-size-fits-allapproachtodeterminingwhencertaintypesofconductsuchastyingandbundlingareanti-competitive.Thereforean‘effectsbasedapproach’isrecommended,aimedatexploringwhethertheconductinquestionconstitutesanti-competitivebehaviour. It is importanttodisentangleconductwhichharmscompetitors(allcompetitionharmscompetitorsbydefinition)fromconductthatharmscompetition,andtherebyconsumers.

9.7.2. ThegrowthofIPrightsrelatingtoE-commercemarketsisanewfactorforauthoritiestobemindfulofintheassessmentofdominance.TheownershipofIPrights,asdiscussedfurtherinSection14,maynotonlycreateamonopoly,butalsoconstituteabarriertoentryforcompetitorsincircumstanceswherethepatentedtechnologyiscrucialforentry.293

102

9.7.3. When conducting an assessment of alleged abuse of dominance in E-commerce markets, it isimportantthatcompetitionauthoritiesrecognisewhenadominantfirmisabusing itsposition;forexampleforeclosingafirmfromthemarket,asintheMyEGcase(Casereview30).However,authoritiesshouldalsoconsiderwhetheradominantfirmisinsteadsimplyinnovatingfasterthanitsrivals,tothebenefitofconsumers;forinstanceasdeterminedintheUKcasebetweenGoogleandStreetmap(Casereview28).Toevaluatethis,andotherimportantfactorsinallegedabuseofdominancecasesinE-commercemarkets,competitionauthoritiesshouldconsiderthefollowingquestions:

Define the relevant market or markets:

• What is/are the relevant product market/s?

Thisusuallyrequiresan identificationofthepotentialeconomicsubstitutesfromtheconsumers’pointofview(demandsidesubstitution),andtheabilityofsupplierstouseexisting capacity tobeginproducing theproduct inquestion (supply sidesubstitution).(InternationalCompetitionNetwork,2011).

• What is/are the relevant geographic market/s?

Thekeyquestioniswhetherconsumerswouldsubstitutetherelevantproductofsuppliers inothergeographicareas insufficientvolumetoconstraintheexerciseofmarketpowerbyahypotheticalmonopolist.Thegeographicmarketcanbethelocationofsuppliersoftherelevantproduct,oritcanalsobedefinedasthelocationofcustomersinthegivenmarket.(InternationalCompetitionNetwork,2011).

1.

Next, the competition authority should determine whether the firm is in a position of dominance by considering the following questions:

• Is the firm in a position of dominance? I.e. is it able to profitably increase prices above the competitive level for a significant period of time?

Insomejurisdictionsaroundtheworld,dominance isassumedifafirm’smarketshareisaboveacertainthreshold.Inotherjurisdictions,suchasMalaysia,thefactthatafirm’smarketshareisaboveorbelowaparticularlevelisnotdeemedtobeconclusiveastowhethertheyoccupy,ordonotoccupy, adominantposition inthemarket.Dominanceshouldbedeterminedbyanumberoffactorssuchasthepositionofactualandpotentialcompetitors,barrierstoentry,andthecountervailingbuyer power of customers. Moreover, dominance in itself is not a competitionproblem,rather,particularformsofconductmaygiverisetoanabuseofsuchapositionandwouldthereforebeanti-competitive.AsdiscussedinSection6.3,afirmmayalsobeinapositionofcollectivedominancealongsideothercompetitors.

• Is the market multi-sided in nature?

Ifyes,indefiningtherelevantmarkets,themarketforfacilitatingtransactionsand/ormatchingdistinctsidesshouldalsobeconsidered.Additionally,ifaninvestigationisrequired,allsidesofthemarketshouldbeconsideredtogether,andinisolation,takingintoaccountthepresence,directionandmagnitudeofanynetworkeffectsbetweenorwithinthedistinctsidesofthemarket,aswellasanyfeedbackeffects.

2.

103

If a position of dominance is determined, a more in-depth analysis of the alleged anti-competitive unilateral conduct should be undertaken by asking the following questions:

• Absent the alleged anti-competitive conduct, would prices be lower, investment/innovations greater, consumer choice more diverse and/or quality of goods/services enhanced?

Tosupportsuchananalysisthefollowingfactorscanbeconsidered:

• How competitive is the market?

Thestronger isthepositionofthedominantfirm,theweakeractualorpotentialcompetitorsare,andthehighertheimpactofbarrierstoentry,themorelikelyanti-competitiveforeclosureis,ceterisparibus.

• What proportion of the dominant firms sales are affected by the conduct?

Thehighertheshare,themorelikelyanti-competitiveforeclosureis,ceterisparibus.

• What is the duration of the alleged anti-competitive conduct?

The longer the duration, the more likely anti-competitive foreclosure is, ceterisparibus.

• Is there evidence that the conduct has caused competitors’ market shares to fall and/or firms to leave the market?

Ifyes,thismaybedirectevidenceofanti-competitiveforeclosure.

• Can an equally efficient firm (i.e. a firm as efficient in production as the dominant firm) compete with the pricing set by the dominant firm? E.g. in cases of predation or fidelity rebates.

Ifyes,anti-competitiveeffectsfromadominantfirms’pricingstrategiesare lesslikely.Inmulti-sidedmarkets,thisevaluationshouldconsiderthecostsinserving,andpricecharged,toallsidesofthemarket,inadditiontoconsideringeachsideofthemarketinisolationifrelevant,asdiscussedinSection9.3above.

• Has the dominant firm imposed vertical restraints on other firms at different stages of production? E.g. exclusivity clauses.

Ifyes,firmsmaybeanti-competitivelyforeclosedfromthemarket.

• Are tied or bundled goods distinct products? I.e. absent the tying/bundling, a large proportion of the customers would not buy the tied or bundled good.

Ifyes,anti-competitiveforeclosureismorelikely.

3.

104

Following an analysis of the extent of any anti-competitive effects from the alleged abusive conduct, competition authorities should consider if there are any efficiency benefits from the behaviour by asking the following questions:

• Does the tying/bundling generate any efficiencies to the benefit of consumers, such as a reduction in transaction costs for consumers and/or a reduction in production costs for the firm, for example through economies of scope?

If yes, such efficiency benefits should be compared with any anti-competitiveeffects.

• Do consumers receive a fair share of any efficiency gains? I.e. at least compensating for the anti-competitive effects resulting from the conduct.

Ifno,anypro-competitiveeffectsarelessimportantinanassessmentoftheallegedanti-competitiveconduct.

• Are there alternative, less restrictive options available to the firm which are more beneficial/less costly to consumers? I.e. have greater pro-competitive effects and/or fewer anti-competitive effects.

Ifyes,explorethepossibilitythatthesecouldbeimplementedbyfirmsinsteadoftheexistingconduct.

4.

105

10.1. Introduction

Competition authorities’ role

10.1.1. Inadditiontoassessingcasesofallegedanti-competitiveconduct,ifthereisamergerregimeinplaceitistheresponsibilityofcompetitionauthoritiestoassessproposedmergersthatmaysignificantlylessencompetitioninaparticularmarket.294

10.1.2. Mergerscanbeeithervertical(betweenfirmsatdifferentstagesofproduction),horizontal(betweencompetingfirms)orconglomerate(betweenfirmswithnohorizontalorverticalconnection).Generally,verticalmergersandconglomeratemergersdonotposecompetitionissuesunlessaspecificmergergives rise to the incentive and theability to foreclosecompetitors.Authorities shouldbealert tohorizontalmergersthatgiverisetomoreimmediateanddirectcompetitiveconcerns,suchastheacquisitionofadirectcompetitororofastrongfringeplayerinthesamerelevantmarket.295

Structure of section

10.1.3. ThissectionfirstconsiderswhetherexistingcompetitionrulesareeffectivelyabletodeterminewhenandhowtoassessaproposedmergerinE-commercemarkets,beforelookingattheimplicationsofnetworkeffectsonmergerassessments,andfinallyconsidersstructuralandbehaviouralremedieswhennetworkeffectsarepresent.Keythemesemergingfromrelevantcasesinjurisdictionsaroundtheworldarehighlightedaswell.

10.2. Ability of existing competition rules to capture relevant transactions

Standard approach to merger investigations

10.2.1. A merger investigation is normally concerned with the horizontal, vertical and/or conglomerateeffectsofamerger(oracombinationofthesethree).Specifically,authoritiesaregenerallyconcernedwithprotectingcompetitionintherelevantmarketinordertomaximiseconsumerwelfare,ortotalwelfareinjurisdictions,suchasSingapore,whichadoptatotalwelfareratherthanaconsumerwelfarestandard.Determiningwhetheramergergivesrisetoanti-competitiveeffectsistypicallybasedonastaticframeworkofanalysisfocusingonthedegreeofoverlapintheproductsorservicessoldintherelevantmarket(generallymeasuredbymeansofthemergingparties’combinedmarketshare).296

Applyingthistestmeansthatininstanceswherethereisnooverlapintheproductsorservicessoldintherelevantmarket(s),somemergersinvolvingonlineplayersmaynotbeinvestigated.

10.2.2. Thesubstantiveassessmentofamergernormallydevelopsaroundfourkeyareas,namely:marketdefinition,assessmentofmarketstructureandconcentration,unilateralandcoordinatedeffects,andmarketentryandexpansion.297Althoughthisassessmentisprimarilyfocusedontheexistingfeaturesofcompetitioninthemarket,aforward-lookingviewofthemarket,capturingthedynamicnatureofcompetition,isparticularlyimportant,andevenmoresoinonlinemarketscharacterisedbyquicklydevelopingtechnologies.Theassessmentofamergerthereforerequiresaproperunderstandingofhowcompetitionworksinthemarketandacleartheoryofcompetitiveharmastowhyconsumerswillbeimpacted,aswellasevidencetosupportthetheoryofharm.Todothis,authoritieswillneedtoconsiderboththelikelyfuturedevelopmentofthemarketpost-merger,andthecounterfactualscenarioifthemergerwasnottooccur.298

294 In ASEAN, all AMS except Malaysia have merger controls in place.

295 The pooling of data between two merging firms may pose competition concerns, though this is likely to be mitigated if the data is not unique and can be replicated or purchased by competitors. As discussed in Section 4, and Section 6, most data that firms collect is easily replicable, or can be purchased from other sources, therefore it does not necessarily lead to an increase in a firm’s market power.

296 See for example, European Commission (2013).


298 Ibid. page 862.

Mergersandacquisitions

10

106

Merger assessments in E-commerce markets

10.2.3. Given the relatively low barriers to entry in onlinemarkets, as discussed in Section 4, the likelydevelopmentofaparticularmarketcan includeentryofplayersfromneighbouringmarkets.Asaresult,althoughamergermayinvolvefirmsoperatinginseeminglyunrelatedonlinemarkets,eitherfirm, or indeed both firms, may be potential future entrants into each other’s market. Dynamiccompetitionmaythereforebeharmedasaresultofamergerduetotheremovalofapotentialfutureentrantinagivenmarket.Insomejurisdictions,suchmergersmaynotbecaughtbycurrentmergercontrolpolicyasthemergertestsfornotificationmaynotcapturemergerswherethereisanabsenceofcurrentoverlapsinproducts/servicesbetweenthemergingparties,orwhererevenuesofoneofthefirmsarelow,and,therefore,fallbelownotificationthresholds.Additionally,ifsuchmergersarecapturedbytherelevanttest,thecurrentanalyticalframeworkforreviewislargelybasedonastaticapproachwhichmaynot consider dynamic aspects of competition.Consequently, potential anti-competitivemergersmaynotbecapturedbycurrentmergercontrolrules.

Implication of dynamic competition on merger controls

10.2.4. Competition authorities may therefore wish to consider whether their notification thresholdsto determinewhen a proposedmerger is reviewed arefit for purpose. In caseswhere amergerregulationregimeisnotpresent,theremaybeaneedtointroduceone.Insomejurisdictions,suchastheUS,competitionauthoritiesareabletocapturecasesofpotentialcompetitionthankstoamergertestwhichincludesthevalueofthetransaction.Somecompetitionauthoritieswhodonothaveanequivalenttestareeitherconsultingontheadoptionofone(asisthecasefortheEU)orarealreadyadaptingtheirmerger regimestoensuretheyareableto investigatesuchmergersbyaddingatestbasedonthevalueofthetransaction(asisthecaseforGermany).299 Atransactionvaluethresholdwouldenablecompetitionauthoritiesto investigatemergersandacquisitionswherethepurchasepriceisoveracertainthreshold,thuscapturingmergersinE-commercemarketswherethetransactionvalueishighbasedonthemarketvalueofnewtechnology,orIP,yettherevenueoftheacquiredfirm,oritsmarketshare,islowandtherearenoclearcurrentoverlapsbetweenthemergingparties.

10.2.5. Establishingthebestsuitedmergernotificationthresholdsisextremelyimportantforauthorities.Ifthresholdsaresettoohigh,thereisariskthatsomeanti-competitivemergerswillnotbecaptured.Ifthresholdsaretoolow,thecostsandadministrativeburdenoncompetitionauthoritiesandbusinessesmightexceedthebenefitsfromhavinganexantemergercontrol.300TheOECD(2016c)notesthatifthresholdsaresettoolow,unnecessarycostswillbefacedbothbythemergingpartiesandbytheauthorities.301Therefore,ifauthoritiesaretoimplementtransactionvaluethresholds,itisimportantthatthesearesetatanoptimallevelthatdoesnotdiscouragestart-upsandsmallbusinessesfrommerging,therebyforgoinganypotentialefficiencybenefitstoconsumersfromeconomiesofscopeorinnovationinthelongrun.

Implication of dynamic competition on merger controls

10.2.6. TheGermancompetitionauthority,theBundeskartellamt,iscurrentlyupdatingitsMergerRegulationtoenableittoreviewmergersbasedonthetransactionvalue.302Anewlawwillstatethatamergeristobesubjecttonotificationwhenthevalueoftransactionisaboveacertainthreshold,inadditiontorelyingonthesizeofrevenueofthetwofirms,asiscurrentlythecase.Forexample,theFacebook/WhatsAppmergerof2014wasnotsubjecttonotificationinGermanyasrevenueswerebelowthethresholdforreview,despitethedealbeingworth$19billion.Thismergerwouldnowbecapturedundertheproposedtransactionvaluethreshold.TheBundeskartellamtarguesthatrelyingonrevenuethresholdsdoesnottakeintoaccountfuturevaluesthatcouldberealisedthroughnewtechnologieswhicharenotyetdrivingsignificant revenues.This isparticularlytrue inmulti-sidedE-commercemarketswherefirmsmaytaketimetobuilda largecustomerbase, forexampleduetonetworkeffects.

299 See, for example, European Commission (2016b).

300 Buccirossi, Cervone and Riviera (2014), chapter 6.

301 OECD (2016c), para. 16.

302 See, for example, http://www.bmwi.de/Redaktion/DE/Dossier/schwerpunkte-wirtschaftspolitik.html.

107

10.2.7. Austria is another example of a country updating its merger controls to account for the digitaleconomy, enabling it to prevent large mergers from being completedwithout a suitable review.BeginningNovember 2017,Austriawill introduce a transactionvaluemerger notification thresholdof€200million(approx.S$300million)providedthatthetargetcompanyhassignificantactivitiesinAustria.TheEuropeanCommissionisalsocurrentlyundertakingaconsultationontheimplementationofatestonthetransactionvalue.303ThedeadlineforsubmissionswasJanuary2017,thoughfindingsfromthisconsultationareyettobepublished,asofJune2017.

Merger regimes in ASEAN

10.2.8. Table11providesabreakdownofthecurrentmergercontrolrulesinplaceacrossASEAN,inadditiontotheUKandUSAforcomparison.AMSarecurrentlyatdifferentstagesindevelopingtheirmergercontrols.OtherthanCambodia,whohasrecentlydiscusseditsdraftlawwithAustralianexpertstoincorporatemergercontrols in its lawby2017,Malaysia istheonlyAMSthatstilldoesnothaveamergercontrolregimeinplace.DespitethelackofmergerregulationsinMalaysia,mergingpartiesmust ensure the post-merger outcome does not breach any prohibition under the MalaysianCompetitionAct2010,i.e.prohibitionofanti-competitiveagreementsandtheabuseofadominantposition.Nevertheless,theMalaysianAviationCommission(MAVCOM),anindependentstatutorybody,prohibitsanymergerwhichwillsubstantiallylessencompetitioninanyaviationservicemarket.Thus,mergerprovisionsrelatedtocompetitionintheaviationsectorareenshrinedintheMalaysianAviationCommissionAct2015.Presently,noAMShasatransactionvaluethresholdfornotificationwithintheirmergerregimes.

303 European Commission (2016b).

Table 11: Merger controls across ASEAN and other jurisdictions

ASEAN Member State Brunei Darussalam

Cambodia

Indonesia

LaoPDR

Type of regime (voluntary / mandatory) Voluntarynotification

Mandatorynotificationbasedondraftlaw

Voluntarypremergernotification.

Mandatorypost-mergernotificationifthresholdsare met.

Mandatorypre-mergernotificationunlessbusinessisclassedasanSME.SMEsarerequiredtonotifypost-closing(basedonlegislationthatcameintoeffectinDecember2015)

Competition Law CompetitionOrder,2015,Chapter4.

DraftlawonCompetitionofCambodia,Version5.6(May2017).

Article28,LawNo.5of1999ontheprohibitionofMonopolyandUnfairBusinessCompetition Practices.

LawonBusinessCompetition (No.60/NA)

Merger control thresholds N/A

TBC

Domestic assets and turnover.Notification is mandatory if a mergermeets one or more of the followingthresholds:

(1)TheassetvalueofthemergedentityexceedsIDR2.5trillion;or

(2) The turnover of the merged entityexceedsIDR5trillion.

These thresholds are not applicableto transactions involving banks. Formergers involving two or more banks,the threshold for notification is IDR 20trillion.

N/A

108

304 Whilst Malaysia does not have a general merger regime in place, as noted above there are sector-specific rules set by MAVCOM, which establish a merger control regime for within the aviation sector.

ASEAN Member State Malaysia

Myanmar

ThePhilippines

Singapore

Thailand

Vietnam

Type of regime (voluntary / mandatory) N/A

Mandatorynotificationsubjecttothresholds

Mandatorynotificationsubjecttothresholds

Voluntarynotificationisencouragedformergersthatarelikelytosubstantiallylessencompetition.

Partiesarerequiredtodoaself-assessmentonwhetheramergernotificationshouldbemadetotheCCS.

MandatoryfilingifthemergermayresultinamonopolyorunfaircompetitionassetoutbytheTradeCompetitionCommission.

Mandatorymergernotificationifthresholdsare met.

Competition Law Nogeneralmergercontrolregimeatpresent.304

TheCompetitionLaw,ThePyidaungsuHluttawLawNo.9,2015),ChapterX

ThePhilippineCompetition Act No. 10667

Section54,CompetitionAct,Chapter50B.

TheTradeCompetition Act BE2542,1999,Section 26.

Section16to24,TheCompetitionLawNo:27/2004/QH11.

Merger control thresholds N/A

Market shares.Mergers will not be permitted if theintention is to create excessive market dominancewithinacertainperiod.

Domestic turnover and asset sizeMandatorynotificationif:

(1) The annual gross revenues in, into,or fromthePhilippines, orvalueof theassetsinthePhilippinesoftheultimateparent entity of at least one of theacquiringoracquiredentitiesexceeds1billionpesos;or

(2)Theaggregatevalueoftheassets,orrevenues generated from the assets isgreaterthan1billionpesos.

Market shares.The CCS is generally of the view thatcompetition concerns are unlikely toariseunless:

(1)Themergedentitywillhaveamarketshareof40%ormore;or

(2)Themergedentitywillhaveamarketshareofbetween20%to40%andthepost-mergercombinedmarketshareofthethreelargestfirmsis70%ormore.

TheCCSisunlikelytoinvestigatemergersinvolving small companies where theirturnover in Singapore in the financialyear preceding the transaction of eachoftheparties isbelowS$5millionandthecombinedworldwideturnoverinthefinancialyearprecedingthetransactionisbelowS$50million.

Jurisdictionalthresholdsaretobesetbynotification,butnonotificationshaveyetbeenissued.

TheTradeCompetitionCommissionhasnotissuedanyminimumthresholdsfornotification of mergers, therefore pre-mergerfilingisnotrequired.

Market shares.Merging parties that have a combinedmarketshareofbetween30%and50%arerequiredtonotify.

109

305 The term ‘person’ refers to the ultimate parent entity of the acquiring and acquired firm.

306 These figures are adjusted each year based on changes in the US gross national product.


308 European Commission (2004), Article 2.

ASEAN Member State UnitedKingdom

USA

Type of regime (voluntary / mandatory) Voluntary,howeverifthetransactionmeetsthejurisdictionalthresholdsandthepartiesdonotnotify,theCMAcanopenaninvestigation.

Wheretheapplicablethresholdsaremetandthetransactionisnototherwiseexempt,notificationismandatory.

Competition Law Enterprise Act 2002,EnterpriseandRegulatoryReformAct2013.

Section7oftheClaytonAct,enactedin1914,amended in 1950.

Merger control thresholds Domestic turnover and market share. Ananti-competitivesituationmayariseifthejurisdictionalthresholdsaremet:

(1)Thetarget’sUKturnoverexceeds£70million

(2)Thetransactionresults inashareofsupplyover25%.

Commerce test, size of person305, and size of transaction thresholds: (1)Thecommercetest–thistestismetifeitherpartyisengaged,oraffectedbycommerce.

(2)Thesizeoftransactiontest–mergersor acquisitions in excess of US$80.8million306maybesubjecttotheHSRAct.

(3)Thesizeofpersontest–thepartiestothe transactionmustmeetcertain sizerequirements if the transaction test ismet.Thesizeofpersontestisgenerallymet where a person with annual netsalesortotalassetsofUS$161.5millionormoreacquiresapersonwithannualnet sales or total assets of US$16.2millionormore,orviceversa.

Sources:InternationalFinancialLawReview(2016),WongPartnership(2016),Rodyk(2013)andCompetitionActsfromrespectivecountriesavailablefromtheASEAN-competitionwebsite.

Substantive test in merger controls

10.2.9. Anecessaryelementofanymergercontrolisasubstantivetestthatcanbeusedtodeterminewhetheramergershouldbeapproved,modifiedbymeansofremedies,orprohibited.307Mergercontrols inthe US and UK rely on analysis aimed at investigatingwhether themerger substantially lessenscompetition,oftenreferredtoasanSLCtest.Otherjurisdictionsrequireinterventionwhereamergerwouldcreateorstrengthenadominantpositionthatsignificantlyimpedeseffectivecompetition,308 aformofdominancetest.Somenations,suchasFranceandGreece,mayuseboththeSLCanddominancetestswhenassessingthepotentialimpactofamerger.

110

309 Sidak and Teece (2009), page 581.


311 Financial Times (2017).

10.3. Innovation and dynamic competition in merger assessments

Assessment of dynamic competition in merger reviews

10.3.1. Whendeterminingwhetheramergerwillresultinasubstantiallesseningofcompetition,competitionauthoritiesmay need to look beyond the static change in the distribution ofmarket shares andconsider how the long-run incentives to compete, for example through innovation, are affected.Furthermore,astaticassessmentofmarketsharesmaybeirrelevantifthereisnocurrentoverlapingoods/serviceofferingsfromthemergingfirms.

10.3.2. The concept of dynamic competition and the removal of a potential entrant or innovator fromamarketisadifficultareaofcompetitionpolicyduetotheinherentspeculative,andcomplexnatureofassessingpotentialfuturecompetitivescenarios.Competitionauthoritiesaswellastheirinternationalnetworksshouldmonitordevelopmentsintheliteratureandpolicydebatestogaininsightsfromnewresearch intheareaandensurethattheyareabletoreflectadvances inthegeneralcompetitionpolicydebateintheirownpractice.SidakandTeece(2009),intheirdiscussionofhowinnovationanddynamicmarketsimpactcompetitionlaw,explainedthat:

“[a] neo-Schumpeterian framework for antitrust analysis that favors dynamic competition over static competition would put less weight on market share and concentration in the assessment of market power and more weight on assessing potential competition and enterprise-level capabilities.” 309

10.3.3. Whilstcompetitionauthoritiesshouldevaluatethepotentiallossofdynamiccompetitionresultingfromamerger,thereshouldalsobeaconsiderationofadditionalpotentialfuturedynamiccompetitionfacedbythemergedentityarisingfrommaverickfirmsorotherpotentialentrantsinthemarket.

Insights from merger reviews

10.3.4. TheacquisitionofLinkedInbyMicrosoft isagoodexampleofamergerbetweentwo largeonlineplayerswheretheoverlapinproductsandserviceswasverylimited.Initsapprovaloftheacquisition,310

theEuropeanCommissionconsideredonlyminoroverlapsinonlineadvertisingwithnoreferencetotheremovalofapotentialentrant intosocialmediamarkets,andtheresultantthreattodynamiccompetition inthe longrun.TheconcernsexploredbytheEuropeanCommissionfocusedonthepotentialforbundlingortyingbetweenMicrosoft’sproductsandLinkedIn’sservices,andtheriskoflessfavourabletreatmentofLinkedIn’scompetitorsbyMicrosoft.

10.3.5. Incentivestoinnovateanddynamiccompetitionhavenotyetbeenconsideredinhorizontalmergercases in E-commercemarkets, however such factors have been evaluated in othermarkets. Forexample,theEuropeanCommission (2016d)providesa reviewof relevantcases inothermarkets,such as pharmaceuticals,where these issueswere of critical importance for the assessment ofthemerger. IntheapprovedUS$130billionmergerbetweenagrichemicalfirmsDowChemicalandDuPont (expected to closeAugust 2017), oneof theEuropeanCommission’smajor concernswasthatinnovationwouldbeadverselyaffectedinthecropprotectionmarket,worthanestimated€60billionannually.311 Evidence suggested incentives to innovatewouldhavebeen lowerpost-merger,andthereforethelevelsofinnovation,hadthetwocompaniesremainedseparate,wouldhavebeenhigher.Therefore,theagreementforthemergertoproceed requiredDuPonttodivestmostof itsglobalresearchanddevelopmentoperationswithinthecropprotectionmarket.

111



314 For example, a computer game that cannot be played on a rival console is a good example of a case where products are not interoperable. Conversely, a DVD can be played in any DVD player.

315 European Commission (2015b).

316 Case No COMP/M.4731 – Google/ DoubleClick (2008).

317 Federal Trade Commission (2007).

318 Case M.8124 – Microsoft / LinkedIn (2016).

10.4. Network effects in merger assessments

Assessing whether a merger will result in a tipping point being reached

10.4.1. Asdiscussed inSection6,thepresenceofnetworkeffects inmulti-sidedonlinemarketshas ledsometoconsiderwhetheratippingpointmayexist incertainmarkets,which,oncereachedbyaplatform,wouldprovideitwithacriticalsizemakingcompetitorsnolongerabletocompete.312Inthisinstance,allcustomersinaparticularmarketoptfortheproductorservicesuppliedbythedominantfirm,creatinganentrybarrierthatistoogreatforpotentialentrantsandsmallerfirmstoovercome.Inthecontextofmergercontrol,theremaybeconcernsthatamergerinmarketscharacterisedbythepresenceofstrongnetworkeffects,andinwhichconsumerstendtosingle-home,mayleadtothemergedfirmreachingatippingpointinthemarketandtherebysubstantiallylesseningcompetition.313

10.4.2. A number of factors affect whether a tipping point is likely to occur. In particular, a lack ofinteroperabilitybetweenproductsinamarket,314andhighswitchingcostsforusersmayincreasethelikelihoodofatippingpoint.However, ifbarrierstoentryarelow,networkeffectsarelesslikelytobeproblematic.Atippingpointisalsolesslikelytooccurifusersmulti-home.Forexample,ifthereisdifferentiationbetweenplatforms,consumersmayusecompetingplatformsdependingonwhichsuitstheirparticularneedbestatthatpointintime.Thelikelihoodofatippingpointoccurringalsodependsonarangeofotherfactors,suchasthenatureoftheproduct/service,thesizeanddirectionofnetworkexternalities,andthedegreeofdynamiccompetition.Therefore,athoroughmarketreviewisrecommendedformergerassessmentswhennetworkeffectsarepresent.


10.4.3. AgoodexampleofacasewherenetworkeffectswereconsideredistherecentreviewoftheFacebook/WhatsAppmerger.315TheEuropeanCommission concluded that post-merger, network effects “donotconstitutean insurmountablebarrier”,citingthepresenceofmulti-homing,theabilityofnewentrantstorecreateauser’snetworkthroughaccesstothatuser’sphonebook,lowswitchingcostsforconsumers,lowbarrierstoentry,andrapidinnovationinthemarket.

10.4.4. A similar conclusion was reached in the Google/DoubleClick merger decision of 2008.316 GooglepurchasedtheonlineadvertisingfirmDoubleClickforUS$3.1billion.TheUSFTCreviewedthemerger,andagreedthatitcouldproceedasthetwocompanieswerenotdirectcompetitorsinanymarket.TheFTC’sstatementaddressedtheissue:

“The markets within the online advertising space continue to quickly evolve, and predicting their future course is not a simple task. Accounting for the dynamic nature of an industry requires solid grounding in facts and the careful application of tested antitrust analysis. Because the evidence did not support the theories of potential competitive harm, there

was no basis on which to seek to impose conditions on this merger.” 317

10.4.5. Thisdecisionwasmadeonthebasisthatcustomerscouldeasilyswitchtoalternativeprovidersinthepost-mergerscenario,andmulti-homingwascommonplaceinthemarket.

10.4.6. Incontrast,inthedecisiontorequirecommitmentsintheMicrosoft/LinkedInmergerin2016,318theEuropeanCommissioncitedtheexistenceofnetworkeffectsasacontributingfactor.Consideringthemarketforprofessionalsocialnetworks,itwasdeemedthattheabsenceofmulti-homing,theexistenceofswitchingcostsassociatedwithcreatingandmaintaininganewprofileonrivalplatformsforusers,andalowlikelihoodofentryfromneighbouringplatformsallmeantthatatippingeffectwasmorelikelytooccur.

10.4.7. Giventheheterogeneityofmarketswherenetworkeffectsarepresent, forexampletheextentofmulti-homing,barrierstoentry,andconsumerswitching, inadditiontothelevelof interoperabilitybetweencompetingplatforms,anin-depthassessmentofthenatureofnetworkeffectsandmarketfeaturesisrecommended.

112

10.5. Structural and behavioural remedies where network effects are present

Overview of remedy design

10.5.1. TheapproachtodeterminingifstructuralorbehaviouralremediesarerequiredtopreventamergerfromgivingrisetoasubstantiallesseningincompetitionisequivalentinE-commercemarketsandofflinemarkets.Thesameistrueforthemethodsfordesigninganysuchcommitmentsorstructuraldivestments,bothwhennetworkeffectsare,andarenot,present.Someauthoritiesmayconsiderstructural remediestobepreferable, asmonitoring isnot requiredandany issuesareaddressedat theoutset; thoughbehavioural remediescanstill beusedeffectively, as longasacomplianceandmonitoringprocessisinplace.Thissectiondiscussesspecificwaysinwhichremediescanbedesignedformergerswherethepresenceofnetworkeffectsislikelytoleadtoasubstantiallesseningofcompetition,inordertoreducethelikelihoodofatippingpointbeingreached,andensureamulti-sided market remains competitive.


10.5.2. Firstly,thepresenceofnetworkeffects,asdiscussedinSection10.4,shouldbeevaluatedinconjunctionwiththewiderassessmentofthemerger inordertoconsider ifthere isaneedforremedies.Forexample,inboththeGoogle/DoubleClickandFacebook/WhatsAppmergers,despitethepresenceofnetworkeffects,itwasdeterminedthatremedieswerenotrequired.

10.5.3. IntheMicrosoft/LinkedInmergerof2016,319theEuropeanCommissiondidhaveconcerns.Despitetherebeinglittleoverlapbetweenthetwomergingfirms,theEuropeanCommissionwasworriedthat,post-merger,thenewentitymaybeabletofurtherenhanceLinkedIn’spositionintheProfessionalSocialNetwork(PSN)marketthroughbundlingandtyingstrategies.ToovercometheseconcernstheEuropeanCommissionacceptedcommitmentsthatreducedthemergedentity’sabilitytobundleortieMicrosoft’sproductswithLinkedIn’sPSN.Specifically,itwasagreedthatpersonalcomputer(PC)manufacturerswouldbefreetonotinstallLinkedInonWindows,anduserswouldbeabletouninstallLinkedInifPCmanufacturerschosetopre-installtheserviceoncomputers.Additionally,alternativeprofessionalsocialnetworkswouldremaininteroperableonMicrosoft’sOfficesoftwarepackage,sothatLinkedInwasnotfavouredaboveotherPSNs.

10.5.4. The JobStreet/SeekAsiamerger320 in Singapore provides another good example of amerger in amulti-sidedmarketwhereremedieswererequired;specifically,inthemarketforonlinerecruitmentservices.Takingintoaccountthedynamicnatureofthemarket,thecommitmentswereputinplaceforaperiodofthreeyears.Uponreview,theCCSconcludedthatthetwofirmswereeachother’sclosestcompetitor,andthemergermay reducecompetition,giving risetoprice increasesand/orexclusive contracts thatwould ultimately harm consumers. In order to address these concerns,SeekAsiaofferedcommitmentsincludingapricecapandnon-exclusiveagreements,aswellastodivestthecompleteassetsofjobs.com.sg.TheCCSclearedthemergerfollowingacceptanceofthebehaviouralcommitmentsanddivestitureofferedbySeekAsiaasitwasfoundthatthelikelyadverseeffectsofthemergerwouldbemitigated.

10.5.5. Thesecases illustrate that if it isdeemedthat remedies are required inmergersbetweenonlineoperatorswhere network effects are present, intervention should focus on encouraging the pro-competitivefactorsdiscussedinSection10.4,forexample:

a.enhancingormaintaininginteroperabilitybetweencompetingplatforms;

b.reducingswitchingcoststousers;

c.encouragingmulti-homing;and/or

d.reducingbarrierstoentry.

10.5.6. Asshownbythemergersreferredtoabove,remediesshouldbeselectedonacase-by-casebasistakingintoaccountthespecificmarketcharacteristicsandthedynamicnatureofcompetition.

10.5.7. Networkeffectsareinevitablecharacteristicsofmulti-sidedmarkets,andhavesignificantbenefitstousersofplatforms.Inordertopreservethesebenefits,networkeffectsshouldnotbeprevented,butratherharnessedinsuchawaytopreventatippingpointfrombeingachieved,forexamplebyloweringthebarriersfacedbycompetingplatformsthroughreducingswitchingcostsforusersorencouragingmulti-homing.

319 Case M.8124 – Microsoft / LinkedIn (2016). 320 CCS (2014).

113

11.1. Introduction


11.1.1. ThissectionlooksatthevariouscompetitionpoliciesandlawsinAMS,andtheextenttowhichtheyareabletodealwiththechallengesposedbytheemergenceandgrowthofE-commerce,asoutlinedinSections6-10ofthishandbook.ThisassessmentisbasedonexistinglawandpracticeasofJune2017.ReassessmentmayberequiredgoingforwardascompetitionauthoritiesacrossASEANupdateandimprovethedesignofcompetitionpolicyandtheenforcementofcompetitionlawaspartoftheASEANCompetitionActionPlan2016-2025.

Basis of assessment

11.1.2. Therecommendationsincludedwithinthissectionarebasedoninsightsderivedfrominternationalbestpracticeandthelatestdebateinthefieldofcompetitionpolicyandlaw,informedbyeconomicanalysisofE-commercedynamics.Thissectionalsodrawsonfindingsfromaquestionnairewhichwasdesignedspecificallytoinformthishandbook,andcompletedbycompetitionauthoritiesinfiveAMS.321Eachoftheseauthoritiesprovideddetailsofthedesignandenforcementofcompetitionpolicyandlawintheirjurisdiction,inadditiontotheirviewsonthechallengesarisingfromtheemergenceandgrowthofE-commerceinASEAN.

11.1.3. To date, two out of the five AMS that responded to the questionnaire have already consideredE-commercewhenformulatingtheir jurisdiction’scompetitionpolicyandlaw,andonlyoneoftherespondentssaidtheywouldnotbeconsideringE-commerceifandwhentheyrevisetheircompetitionpolicyandlaw.

11.2. Design of competition policy and law

Ability of existing legal framework to deal with challenges in E-commerce markets

11.2.1. Bydrawingoncaseexamplesfromvariousjurisdictionsaroundtheworld(seesections6–10),thishandbookhasconsideredtheimpactofE-commercedevelopmentsoncompetitionpolicyandlaw.Ithasfoundthattheexistinglegalframeworkisbroadlysufficienttodealwithcasesinbothonlineandofflinemarkets.

11.2.2. ThisobservationisconsistentwiththeconclusionreachedattheOECD’s2012“HearingsonTheDigitalEconomy”,whichexplainsthat“existingcompetitionlawsaresufficientlyflexibleandnuancedtobeappliedinthedigitaleconomy.”322ItisthereforeapparentthatawidescaleoverhaulofcompetitionpolicyandlawmaynotbeneededtodealwiththechallengescurrentlybeingposedbyE-commerce.AsimilarconclusionappearstohavebeenreachedbytheEUinitsFinalReportontheE-commerceSectorInquiry.Forexample,itisoutlinedhowtheEuropeanCommissionseesnoneedtoacceleratetheexistingreviewprocessofitsverticalblockexemptionregulation:“TheVBERexpiresinMay2022,andthe resultsof thee-commercesector inquiry confirmthat there isnoneedtoanticipate itsreview.” 323

321 Competition Commission of Singapore, the Philippines Competition Commission, Vietnam Competition Authority, Malaysia Competition Commission, Indonesia Competition Authority.

322 OECD (2012), page 7.


Recommendationsonimprovingthedesignofcompetitionpolicyandenforcementofcompetitionlawtoproscribeanti-competitiveconductrelatingtoE-commerceforAMS

11

114

11.2.3. ToaidtheeffectivedesignofcompetitionpolicyandlawinASEAN,thefollowingsectionsprovideaseriesofrecommendationsthatcompetitionauthoritiesmaywishtoconsiderwhenformulatingorrevisingtheircompetitionregimestoensurethattheyareabletoeffectivelydealwiththechallengesarisingfromE-commerce.

Anti-competitive conduct

11.2.4. Competitionauthoritiesoftenfacechallengesintheassessmentoftheimpactofacertainconductof competition. Unless a per se illegal breach of competition law has occurred,324 theremay bebothpro-andanti-competitiveeffectsresultingfromverticalandhorizontalagreementsandcertaintypesofunilateralconductbydominantfirms.ThisishighlightedinSections7-9ofthishandbook.Consequently,acase-by-caseapproachtoassessingsuchconduct isrecommended,applyingtheprinciplesandguidelinesdiscussedinthosesections; inparticularweighingupanypro-andanti-competitiveeffects,andevaluatingtheextenttowhichconsumersbenefitfromanysuchefficiencies.Considerationshouldalsobegiventowhetherfirmscouldadoptalternativeagreements,orconduct,insteadofthosebeinginvestigated,thatmayachievethesame(orgreater)efficiencybenefits,and/orincurfeweranti-competitiveeffects.

11.2.5. Inorderforauthoritiestobeabletoevaluateallegedanti-competitiveconductitisimportantthatcompetitionpolicyandlawhasanumberoffeatures;inparticular,allowingfor:

a. anefficiencydefenceofhorizontalagreementsbetweenfirms;325

b. anefficiencydefenceofverticalagreementsbetweenfirms;and

c. aneffectsbasedapproachintheassessmentofallegedunilateralanti-competitiveconduct(i.e. abuseofdominance)byfirms.

11.2.6. Table 12 below summarises the presence of these features in the competition policy and law ineachoftheAMS,withtheexceptionofLaoPDRasanEnglishtranslationofthecompetitionlawisunavailable.

324 In some jurisdictions hardcore restrictions may also be defensible in some situations (see for example Case review 10)

325 For example, efficiencies resulting from combining complementary skills (Whish, R. and Bailey, D., 2012, page 591).

Table 12: Competition policy and law relating to efficiency arguments in AMS

ASEAN Member State Brunei Darussalam

An effects-based approach in the assessment of alleged unilateral anti-competitive conduct by firms Nolawinplaceallowingforthis.

An efficiency defence of agreements between firms Section 9 of the Third Schedule of theBrunei Darussalam Competition Order2015statesthat:“Thesection11prohibitionshallnotapplytoanyagreementwhichcontributesto-

a.improvingproductionordistribution;or

b. promoting technical or economicprogress,

butwhichdoesnot-

i. imposeontheundertakingsconcernedrestrictionswhicharenotindispensabletotheattainmentofthoseobjectives;or

ii. affordtheundertakingsconcernedthepossibility of eliminating competitionin respect of a substantial part of thegoodsorservicesinquestion.”

Feature of competition policy and law that allows for:

115

ASEAN Member State Cambodia (Draft–Version5.6)

Indonesia

An effects-based approach in the assessment of alleged unilateral anti-competitive conduct by firms Article14ofthedraftlawstatesthat:

“This Article 14 does not prohibit a personor personswith a dominant position fromconductinganyactionwhichhasalegitimatecommercial reason for particular actions,andthatactionswerenotintendtoprevent,restrict and distort competition.”

Nolawinplaceallowingforthis.

An efficiency defence of agreements between firms Article17ofthedraftlawstatesthat:

“Articles 11, 12, 13 and 15 will not apply if aperson who is party to the agreement canprovethat:

a. there are significant identifiabletechnological, efficiency or social benefitsdirectlyarisingfromtheagreement;

b. the benefitswould not arisewithout theagreementhavingtheeffectofpreventing,restrictingordistortingcompetition;

c.thebenefitsoftheagreementoutweighitsanticompetitiveeffect;and

d.theagreementdoesnotallowtheenterpriseconcerned to eliminate competitioncompletelyinrespectofasubstantialpartofthegoodsorservices.”

Article 50 of the Law Number 5 Year 1999Concerning The Prohibition Of MonopolisticPractices And Unfair Business Competitionstatesthat:

“ Excluded from the provisions ofthis law shall be the following: …

c)agreements for the stipulation oftechnical standards of goods and orservices which do not restrain, andor do not impede competition; or …

e)cooperation agreements in the field ofresearchforraisingor improvingthe livingstandardofsocietyatlarge;or

…”


116

ASEAN Member State Malaysia

Myanmar

An effects-based approach in the assessment of alleged unilateral anti-competitive conduct by firms

Section 10 (3) of the CompetitionAct 2010statesthat:

“Thissectiondoesnotprohibitanenterprisein a dominant position from taking anystep which has reasonable commercialjustification or represents a reasonablecommercialresponsetothemarketentryormarketconductofacompetitor.”


An efficiency defence of agreements between firms Section5oftheCompetitionAct2010statesthat:

“Notwithstanding section 4, an enterprisewhich is a party to an agreement mayrelieveitsliabilityfortheinfringementoftheprohibition under section 4 based on thefollowingreasons:

a)there are significant identifiabletechnological, efficiency or social benefitsdirectlyarisingfromtheagreement;

b)the benefits could not reasonably havebeen provided by the parties to theagreement without the agreement havingthe effect of preventing, restricting ordistortingcompetition;

c)the detrimental effect of the agreementon competition is proportionate to thebenefitsprovided;and

theagreementdoesnotallowtheenterpriseconcerned to eliminate competitioncompletelyinrespectofasubstantialpartofthegoodsorservices.” Section14oftheMyanmarCompetitionLawstatesthat:

“TheCommissionmay,byspecifyingacertainperiod, exempt in respect of agreementon restraint on competition which intendsto lessen the expense of consumers if it isinclusiveinanyofthefollowingmatters;

a.reforming formation and type of anybusiness to improve the capability ofbusiness;

b.upgrading of technology and technologylevel in order to improve the quality ofgoodsandservices;

c.ensuring to be uniform development oftechnologicalstandardsandqualitylevelofdifferentproducts;

d.ensuring to be uniform in thematters ofcarryingoutbusiness,distributionofgoodsandpaymentnotconcernedwithpriceorfactsrelatedtoprice;

e.ensuringtoraisecompetitivenessofsmallandmediumenterprises;

f. ensuring to raise competitiveness ofMyanmar businesses in the internationalmarket.”


117

ASEAN Member State ThePhilippines

Singapore

An effects-based approach in the assessment of alleged unilateral anti-competitive conduct by firms

Section15ofthePhilippineCompetitionActstatesthat:

“Itshallbeprohibitedforoneormoreentitiestoabusetheirdominantpositionbyengaginginconductthatwouldsubstantiallyprevent,restrictorlessencompetition.”


However,initsGuidelinesonTheSection47Prohibition,CCSdoesstatethat:

“In considering whether there has beenan abuse of dominance, CCSwill conducta detailed examination of the relevantmarkets concerned and the effects ofthe undertaking’s conduct.” (para. 2.1)

“Inconductinganassessmentofanallegedabuse of dominance, CCS will undertakean economic effects-based assessment inorder to determine whether the conducthas, or is likely to have, an adverse effectontheprocessofcompetition.Theprocessofcompetitionmaybeadversely impacted,for instance, by conduct which wouldbe likely to foreclose, or has foreclosed,competitors in the market. CCS considersthat factors which would generally berelevant to its assessment include: thepositionoftheallegedlydominantpartyanditscompetitors;thestructureof,andactualcompetitive conditions on, the relevantmarket;andthepositionofcustomersand/orinputsuppliers.”(para.4.4)

An efficiency defence of agreements between firms Section14(c)ofthePhilippineCompetitionActstatesthat:

“Agreements other than those specified in(a) and (b) of this section which have theobject or effect of substantially preventing,restricting or lessening competition shallalso be prohibited: provided, those whichcontribute to improving the productionor distribution of goods and servicesor to promoting technical or economicprogress, while allowing consumers a fairshare of the resulting benefits, may notnecessarilybedeemedaviolationofthisAct.

Section 9 of theThird Schedule [ExclusionsFromSection 34 ProhibitionAndSection 47Prohibition]oftheSingaporeCompetitionActstatesthat:

“Agreements with net economic benefit 9.Thesection34prohibitionshallnotapplytoanyagreementwhichcontributesto—

a.improvingproductionordistribution;or

b.promotingtechnicaloreconomicprogress,

butwhichdoesnot—

i. impose on the undertakings concernedrestrictionswhicharenotindispensabletotheattainmentofthoseobjectives;or

afford the undertakings concerned thepossibility of eliminating competition inrespectofasubstantialpartofthegoodsor


118

ASEAN Member State Thailand

Vietnam

An effects-based approach in the assessment of alleged unilateral anti-competitive conduct by firms Nolawinplaceallowingforthis.


An efficiency defence of agreements between firms

Section 27 of the Thailand Competition Actstatesthat:

“Inthecasewhereitiscommerciallynecessarythattheactsunder[section27](5),(6),(7),(8),(9) or (10) be undertakenwithin a particularperiod of time, the business operator shallsubmit an application for permission to theCommissionundersection35.”

Article 10 of the Vietnam Competition Law2004statesthat:

“An agreement in restraint of competitionstipulatedinclause2ofarticle9ofthisLawshall beentitled toexemption for adefiniteperiod if it satisfies one of the followingcriteria aimed at reducing prime costs andbenefitingconsumers:

a)It rationalizes an organizational structureor abusiness scaleor increasesbusinessefficiency;

b)It promotes technical or technologicalprogressor improvesthequalityofgoodsand services;

c)Itpromotesuniformapplicabilityofqualitystandardsandtechnicalratingsofproducttypes;

d)Itunifiesconditionsontrading,deliveryofgoodsandpayment,butdoesnotrelatetopriceoranypricingfactors;

dd) It increases the competitiveness ofmediumandsmallsizedenterprises;

e)It increases the competitiveness ofVietnameseenterprisesintheinternationalmarket.”


Source:Asean-competition.org.

11.2.7. AscanbeseenfromtheanalysisinTable12above,therearesomedifferencesinthepresenceofthesefeaturesincompetitionpolicyandlawacrossAMS.Forexample,onlythePhilippines,MalaysiaandCambodia (draft)havea law inplacethatallowsforaneffectsbasedapproachtoassessingallegedanti-competitiveunilateralconduct.However, it isacknowledgedthataspecific law isnotnecessarily required for suchanapproachtobeapplied, as is thecase inSingapore,whereCCSguidelinesontheapplicationofthelawoutlinethisinstead,326aswellasthroughdecisionalpracticeandcourtdecisions.AllAMSdohavelawsinplace(ordraftlaws)permittinganefficiencydefenceofagreementsbetweenfirms.Therearehoweversomedifferencesinthebreadthandscopeoftheselaws.Forexample,inIndonesia,therelevantlawdoesnotcoverasmanyareasasintheotherAMS.

326 CCS (2016b); Guidelines on The Section 47 Prohibition.

119

327 See, for example, http://www.bmwi.de/Redaktion/DE/Dossier/schwerpunkte-wirtschaftspolitik.html.


Merger control regimes

11.2.8. Asdiscussedabove,thresholdsfornotificationareimportantinordertoavoidawasteofresourcestoassessmergersthatareunabletocauseasubstantiallesseningofcompetition.AsdiscussedindetailinSection10,thereis,however,ariskinE-commercemarketsthatcertainmergersthatmayhaveanti-competitiveeffects inthe long runmaynotbecapturedbyexistingmergerthresholds.For instance, if there isnocurrentoverlap inproducts/servicesoffered,butthemerger results intheremovalofalikelypotentialfuturecompetitor.Asaresult,existingthresholdsmaynotcapturemergerswheredynamiccompetitionmightbeadverselyaffected,therebyharmingconsumersinthelongrun.

11.2.9. ThiswashighlightedwhentheGermanauthorities’revenuethresholdsdidnotcapturetheFacebook/

WhatsAppmerger, despite the deal beingworth US$19 billion. The Bundeskartellamt is currentlyadapting its merger control rules to include a threshold based on the value of transaction.327 Furthermore,theGermanauthoritiesarenotactinginisolation,asotherauthoritiesareconsideringtakingthisstep.328

11.2.10. Table11inSection10presentsareviewofthecurrentmergercontrolrulesinAMS,fromwhichitisevidentthatnoAMScurrentlyhasatransactionvaluethresholdinplace.Inordertoensurethatallmergersthatmayleadtoanti-competitiveeffectsinthelongrunarecapturedandassessedinfull,AMSmaywishtoconsiderimplementingatestonthetransactionvalueintheirmergercontrolrules.

11.3. Enforcement of competition law

11.3.1. In assessing alleged anti-competitive conducts or reviewing proposed mergers, competitionauthoritiesshoulddrawontheinsightsfrompreviouscases,andthelessonsthatcanbelearntfromjurisdictionsacrosstheworld.

11.3.2. InallofthecasesreviewedinSections7-10ofthishandbook,existinglawandenforcementpracticehasshowntobelargelysufficientinidentifyinganddealingwithallegedinstancesofanti-competitiveconduct in E-commerce markets, and in assessing proposed mergers. In multi-sided markets,however,adjustmentstotheapproachfollowedmayberequired.Recommendationsonhowbesttodealwithmulti-sidedmarketshavebeenpresentedthroughoutthishandbook,howevernoultimateapproach has yet been developed, and debate on thematter is ongoing. For instance, an OECDHearingentitled‘Rethinkingtheuseoftraditionalantitrustenforcementtoolsinmulti-sidedmarkets’wasrecentlyheldinJune2017.

11.3.3. Thissub-sectionpresentsfurtherrecommendationsonhowcompetitionauthoritiesacrossASEANcanadapttoensuretheyarebetterplacedtodealwiththechallengesarisingfromthegrowthofE-commerce.

Capability building and technical assistance

11.3.4. Asdiscussedintheprecedingsections,theenforcementofcompetitionlawmaybemorecomplexinE-commercemarketsincomparisontoanalogousinvestigationsinbrick-and-mortarmarkets.Thisisparticularlytruewhenmarketsaremulti-sidedinnatureasanumberofadaptationstoexistingapproachesmay be required, such aswhen defining the relevantmarket (ormarkets) andwhenassessingmarketpower,asoutlined inSection6.Therefore, ifAMScompetitionauthoritiesaretoupskilltheirstaff,forinstancethroughtrainingandlearningfrommoreestablishedauthorities,theanalysisofmulti-sidedmarketsshouldbehighontheagenda.

11.3.5. AMScompetitionauthoritiesmayalsowanttoconsiderintheirrecruitment,andupskillingofexistingstaff, the analysis of price-fixing algorithms and pricemonitoring tools. As E-commercemarketscontinuetogrow intheregion,onewouldexpecttheuseofsuchtoolsto increase inprevalenceamongfirms,thereforemakingcasesinvolvingthesecomplexitemsofsoftwaremorelikelytoarise.

120

11.3.6. TheassessmentofmarketpoweranddominanceinE-commercemarketsshouldalsobeincludedonthelearningagendaforAMScompetitionauthorities.Inparticular,thechangestobarrierstoentryinE-commerce retailers incomparisontobrick-and-mortarmarkets,asdiscussed inSection4.3,shouldbecovered.

11.3.7. CompetitionauthoritiesinASEANmayalsobenefitfromparticipatingininternationalroundtablesontheemergingchallengeswhendealingwithissuesrelatingtohorizontalandverticalcoordinationinE-commercemarkets,consideringissuessuchastheadoptionofMFNclauses.Onasimilarnote,thepossibilityofsecondingstafftocompetitionauthoritiesinotherjurisdictions,whereE-commercemarketsaremoredeveloped,couldbeexplored.

Data gathering and analysis

11.3.8. Whenenforcingcompetitionlaw,competitionauthoritiesmayneedtomakepracticaladjustmentstothewayinwhichtheycollectandanalysedata.QuestionnaireresponsesindicatethatissuesofdatacollectionandreliabilityconcernsareposingchallengestocompetitionauthoritiesinASEANwhendealingwithcasesinE-commercemarkets.Casesofthisnaturerequireauthoritiestounderstandwhattypeofdataisrequired,andthereforethenecessaryskillsneededtoanalysethisinformation,if,forexample,anauthorityneedstoexaminethemechanicsbehindapricingalgorithm.

11.3.9. In investigating a platform business in amulti-sided onlinemarket, a competition authoritymaywishtocollatedataonthenumberofusersoneachdistinctsideofthemarket,andthenumberoftransactionsfacilitated,inadditiontoinformationonpricingandsales.Whentherequireddataisnotavailable,authoritiesmaywishtoconsiderconductingspecificsurveywhichwouldenablethemtoobtainthenecessarydata.

Monitoring of ongoing developments in E-commerce markets internationally

11.3.10. Asexplainedthroughoutthishandbook,thedebateoncompetitionlaw,anditsenforcement,inlightofthechallengesbroughtaboutbyE-commerce, isstillatarelativelyearlystageofdevelopmentacrosstheworld.Forsomeofthechallengeswhichhaveemergedtodate,internationalconsensushasnotyetbeenreached(as,forexample,ontheuseofwideversusnarrowMFNsinEurope,seeCasereview17).Progresstowardsinternationalcoordinationishoweverbeingmade,forexamplewiththeOECDamongotherinternationalgroups,promotingwide-rangingdebateontherelevantissues.Competition authorities should therefore follow the international debate and thedevelopment ofcaselawaroundtheworldinordertokeepabreastofdevelopments.

11.3.11. The importance of keeping up-to-datewith developments in the understanding of E-commercepractices and adopting the correct competition policy approach is all the more important forpractices,suchaspricesettingalgorithmsfosteringcollusion,whereaclearresponsehasnotyetbeenidentified.

11.3.12. Newchallengesarealsolikelytoemerge.AsdiscussedbyDavidCurrie(2017),thechairmanoftheUKCMA,automatedpricingalgorithmsmayprogresstosuchapointthattheyareabletoindependentlyestablishcollusivebehaviourbetweencompetingfirmsinordertomaximiseindustryprofits,reachingthis conclusion without the need for human programming due to machine learning capabilitiesbuilt intothetools. Ifsuchaneventualitydoesoccur, it isunclearwhetherthiswouldrepresentabreachofcompetitionlaw.ItisthereforeimportantthatcompetitionauthoritiesmonitoranyofsuchdevelopmentsinE-commercemarkets.

121

Harmonisation of competition policy and law across ASEAN

11.3.13. InordertosupportthedevelopmentofanintegratedASEANmarket,competitionauthoritiesshouldcontinuetoworktowardscoordinationintheinterpretationandenforcementofcompetitionpolicyandlaw.

11.3.14. Currently,therearedifferencesinthedesignandenforcementofcompetitionpolicyandlawinAMS.AshighlightedintheSection11.2,onlysomeAMShavealawinplacethatpermitsaneffects-basedassessmentofallegedanti-competitiveunilateralconduct.Asa resultofdifferencessuchasthis,businessesmayneed to adapt their practices according towhichAMSthey areoperatingwithin,inordertocomplywithcompetition law.Consequently,cross-bordertrademaybe inhibited,andbreachesofcompetitionlawsmaybemorecommon,asfirmsoperatinginternationallymayfailtomodifytheiroperationstosuittherelevantjurisdiction.

11.3.15. Intheshortterm,competitionauthoritiescouldlooktoreleaseshorthandbookstohelpbusinessesoperatingintheirjurisdictionunderstandthespecificlawsandapproachinthatjurisdiction.Inthelongrun,however,harmonisationwouldbeadvisable.

11.3.16. The challenges faced by businesseswhen competition authorities adopt different approaches isdemonstratedby theBooking.comcase in Europe (Case review 17),where authorities have takencontrastingstancesontheuseofwideandnarrowMFNclausesinthehotelbookingindustry,despiteattemptstofacilitatecoordination.329

11.3.17. AMSshouldthereforeworktogethertocometoacoordinatedviewonthevarioustypesofconductdiscussedinSections7-9ofthishandbook.ProducingguidelinesforbusinessestooutlineandclarifythesepositionswouldhelpbuildconfidenceamongfirmsoperatinginternationallyinASEAN.

329 An international working group including ten competition authorities (Belgium, Czech Republic, France, Germany, Hungary, Ireland, Italy, Netherlands, Sweden and the UK) was set up to coordinate actions for a possible harmonisation of approach on wide and narrow MFN clauses across jurisdictions.

122

CompetitionpolicyandlawcompliancechecklistforbusinessesengagedinE-commerce in ASEAN

12

12.1. Introduction

12.1.1. The followingchecklist aimstoprovide guidance thatbusinessesengaged inE-commercewithinASEANcanfollowtominimisetheriskofbreachingcompetitionlaw.

12.1.2. Therearethreecoreareasofconductthatbusinessesshouldbemindfulof:

a. Coordinationwithcompetitors;

b. Otheranti-competitiveagreements;and

c. Individualanti-competitiveconducts(i.e.abuseofdominance).

12.1.3. Examplesofactionsthatcouldbedeemedasanti-competitiveconductsacrosseachoftheseareasinE-commercemarketsincludethefollowing:

a. Coordinationwithcompetitors–e.g.agreementstofixpricesonanonlineplatform,limitsupply, orsharecustomers;

b. Otheranti-competitiveagreements–e.g.fixingorrestrictingthepricethatretailerscansellaton onlinemarketplaces,longexclusivitycontracts;and

c. Anti-competitiveconductsby individualfirms(i.e.abuseofadominance)–sellingasignificant shareoftheproducts/servicesinanindustryandexploitingthisposition,forexampleby:tying/ bundlingtocreateorraisebarriers,refusingtosupply,orincreasingswitchingcostsforconsumers.

12.1.4. Therearesomeoverarchingprinciplesthatbusinessesshouldadhereto,anditistheresponsibilityofbusinessesandtheiremployeestoensurecompliancewithcompetitionlaw.Inparticular:

a.Companiesshouldnotenterintoanyagreementorpracticethatinfringescompetitionlaw;

b. Ifbusinessesareawareofanti-competitivebehaviourbyanemployeewithinthefirm,acompetitor, supplier,orotherbusiness,itmustbereportedimmediatelytotherelevantcompetitionauthority; and

c.Tacitparticipationmaystillinfringecompetitionrules.

12.1.5. Underpinningthisisacommitmentthatcompliancewithcompetitionlawsisdrivenfromthetopoftheorganisation.

12.1.6. Questionnaire responses highlight that competition authorities in AMS receive a relatively smallnumberofcomplaintsfromconsumersorfirmsregardinganti-competitivebehaviour.Consumersandbusinessesshouldfeelempoweredtospeakupiftheyareawareofbehaviourwhichtheybelieveinfringescompetitionlaw.Inordertomakeacomplaint,theCCSsuggestscollectingthefollowinginformation:330

a. Informationaboutyourself,andtheorganisationyourepresent(ifapplicable);

b. Informationaboutthepartyorpartiesinvolved;

c. Abriefdescriptionoftheagreement,conductormergerthatyouarecomplainingabout;and

d. Anyotherrelevantinformationandsupportingdocuments.

330 CCS (2017).

123

12.2. Stages of risk management to avoid competition law infringement

12.2.1. Toreducetheriskofbehavinganti-competitivelyinE-commercemarkets,firmsshouldfollowafour-stepprocesswhichisinaccordancewithinternationalbestpractice.331

a. Identifyrisks;

b. Assess risks;

c. Takeactiontoreducerisks;and

d. Review processes.

12.2.2. Eachofthesestagesisconsideredingreaterdetailbelow.

12.3. Identify risks

12.3.1. Firmsshouldbeawareofgeneralguidanceoncompetitionlawinordertoensurecompliance.Someimportantquestionsfirmsshouldconsider forE-commercemarketsarehighlightedbelow.Firmsshouldseeklegaladviceifaconductgivesrisetoariskofinfringement.

1. Coordination with competitors:

a. Do your employees have contact with competitors via online communication channels?Ifso,theseemployeesshouldbethoroughlytrained,andcloselymonitored.Contactbyitselfisnotwrong,butthedetailsthatarecommunicatedmaycauseconcerne.g.regardingbiddingorpricingbehaviour.

b. Do you communicate with competitors, for example at industry events, or trade association meetings?

Asin(a).

c.Inyourmarket,doemployeesmovefrequentlybetweencompetingfirms?If yes, the likelihoodof sharing confidential information is increased.Any employeewhohasrecentlyworkedforacompetitor,orisleavingtojoinacompetitor,shouldbetrainedonwhatisandisnotappropriatetosharewiththeirnewemployer.

d. Do you ever work alongside competitors?Ifso,competitionlawrelatingtohorizontalagreementsshouldbethoroughlyreviewed,andstaffworking closelywith competitors shouldbe trainedonwhat is and is not allowed, includingsensitiveinformationthatshouldn’tbeshared.

e. Do you use algorithms to adjust your prices subject to movements in your competitors’ prices?Ifso,itshouldbedeterminedwhetherornotthesupplierofthealgorithmalsosuppliescompetingfirms.Themechanicsofthealgorithmshouldalsobefullyunderstoodi.e.docompetitorswhousethesamesupplieralsohavesimilarresponsivepricing.

f. Do you explicitly or implicitly agree with your competitors a limit to supply? Ifso,theagreementshouldbereviewedthoroughlyandassessedastowhetheritconstitutesillegalcartelbehaviour.

g. Do you sign up to terms with an online platform that are common across competitors, and do these terms restrict in any way the price that you are able to sell your goods/services at? Or are you a platform that sets contractual terms for business users, and do these terms include clauses that restrict the price that users can sell goods or services at?

Ifso,thetermsshouldbethoroughlyreviewedbyanexperttrainedincompetitionlaw,astheymayamounttohorizontalcoordinationamongcompetitors.

331 CMA (2014).

124

2. Other anti-competitive agreements: (Note: These questions are framed on the basis of a manufacturerforminganagreementwitharetailer.Retailersshouldalsoreviewthisguidanceto identifythetypesofclausesinagreementsthatmaybedeemedanti-competitive)

a. Do you require your retailers to enter into exclusive contracts for long periods of time?Insomejurisdictionsthismaybetreatedasananti-competitiveagreement.Itshouldthereforebedeterminedwhetherthisisthecaseinthejurisdiction(s)wherethecontractshaveeffect.

b. Does your business impose restrictions on retailers and online marketplaces that sell your products? E.g. the retail price they can sell at, who/where they can sell to, conditions that must be met for them to be able to sell the product; the quantity of the product they must buy/sell.

Ifso,theguidelinesonuseofverticalagreementsintherelevantjurisdiction(s)shouldbereviewedtodeterminewhethertherestrictionsarepermitted.Therulesandtheirinterpretationsmayvaryacrossdifferentjurisdictions.

c. Do you set a recommended retail price to your retailers or a minimum advertised price, and do you monitor compliance to this?

If so, such behaviourmay be deemed to be equivalent to Resale Price Maintenance (RPM),conduct that isprohibited in some jurisdictions. If the recommended retail price is enforced(i.e.withpunishmentfordeviators–e.g.withdrawalofsupplyorreductioninsales),itislikelytoconstituteRPM.

d. Do you exclude or restrict retailers from selling online?Suchconductmaybedeemedanti-competitiveinsomejurisdictions.Guidanceoninternetsalesbansshouldbereviewedintherelevantjurisdiction(s),and/orthecompetitionauthorityshouldbecontactedtoseekclarityonwhetherthebanispermitted.

e. Do you treat retailers operating online differently from retailers with physical stores?Suchconductmaybedeemedanti-competitiveinsomejurisdictions.Guidanceondiscriminationbetween channels should be reviewed in the relevant jurisdiction(s), and/or the competitionauthorityshouldbecontactedtoseekclarityonwhethertheconductispermitted.

f. Do you charge different wholesale prices (or offer different incentives) to retailers depending on whetherproductsaresoldonlineoroffline?

Asin(e).

g. Do you exclude or restrict retailers from selling on online platforms such as marketplaces? Suchconductmaybedeemedanti-competitive in some jurisdictions.Guidanceonplatformbansshouldbereviewedintherelevantjurisdiction(s),and/orthecompetitionauthorityshouldbecontactedtoseekclarityonwhethertheconductispermitted.

h. Do you enter into ‘best price’ guarantees with retailers?Ifso,itshouldbedeterminedwhethertheclausebreachescompetitionlawinthejurisdictioninwhichitisbeingimplementedi.e.insomejurisdictionsdifferentforms(orallforms)ofpriceparityorMFNclausesmaybeprohibited.

i. Do you restrict retailers from using price comparison websites (PCWs)?Such conduct may be deemed anti-competitive in some jurisdictions. Guidance on PCWrestrictionsshouldbereviewedintherelevantjurisdiction(s),and/orthecompetitionauthorityshouldbecontactedtoseekclarityonwhethertheconductispermitted.

125

3.Individualanti-competitiveconductbyfirms:

a. Are you a business with a large share of any of the markets in which you operate (i.e. over 40%)332ordoyousellasignificantshareoftheproducts/servicestradedinthemarket?

Ifso,youmaybeconsidereddominantinsomejurisdictions,thereforecertaintypesofconductmaybedeemedanti-competitivewhentheyotherwisewouldnot.Note,theexactdefinitionofdominancewillvaryamongdifferentjurisdictions.

b.Doyouoperateanonlineplatform/websitethroughwhichyoucoverasignificantshareof the activity in the market (on any side of the platform)? E.g. transactions made, or platform users.

If so,youmaybe considereddominant even if themarket share of sales is not beyond thethresholdfordominance.

c. Do you impose any restrictions on advertisers/retailers that sell through your platform? E.g. the price they can sell at, or restrictions on which other websites they can sell on.

Suchconductmaybedeemedanti-competitiveinsomejurisdictions.Guidanceonconductthatmaybedeemedanti-competitiveshouldbereviewedintherelevantjurisdiction(s),and/orthecompetitionauthorityshouldbecontactedtoseekclarityonwhethertheconductispermitted.

d.Doyourefusetosupplycustomerswithnoobjectivejustification?Asin(c).

e.Doyouofferdifferentpricestosimilarcustomerswithoutobjectivejustification?Asin(c).

f. Do you use bundling or tying strategies, whereby you sell/package products that you have market powerinalongsideotherproductswherecompetitionwithotherfirmsisfiercer?

Asin(c).

g.Doyouimposetermsondownstreamfirms?E.g.aminimumpurchasequantityoranexclusivity clause.

Asin(c).

h. Do you charge a price below average variable cost and how do you recoup these costs?Someformsofbelowcostpricingmaybeconsideredpredation,and,assuch,infringecompetitionlaw.Ifso,seekguidancefromin-houseorexternallawyersastowhetherthisconductconstitutespredatorypricing.

332 Market definition may vary on a case-by-case basis. The general rule is that above 40% dominance is assumed. In your risk assessment in order to be risk averse assume the narrowest market definition in terms of goods/services included and geographic area covered.

126

12.4. Assess risks

12.4.1. Having identifiedany areaswhere theremightbe a riskofbreaching competitionpolicy and law,companiesshouldassessthelikelihoodofanybreachesoccurring,andtakeactionstopreventthesefromtakingplace.

12.4.2. Companiesshouldunderstandinmoredetailthecompetitionlaw(s)theyareatriskofbreaching.Thismayvaryineachofthejurisdictionsinwhichtheyoperate.FormoreinformationoncompetitionpolicyandlawacrossASEAN,andtherestoftheworld,businessesshouldconsulttheresourcesavailableat:

www.asean-competition.org/ www.internationalcompetitionnetwork.org/working-groups/current/cartel/awareness/business.aspx

12.4.3. Companiesshouldalsoidentifywhichoftheiremployeesaremostatriskofbreachingcompetitionlaw,forexamplethosewhoagreecontractualtermswithcustomers,suppliersorusersofanonlineplatform,orthosewhohavecontactwithcompetitors,orareinsalesroles.

12.5. Take action to reduce risks

12.5.1. Businesses involved in E-commerce should set up processes to reduce the risk of breachingcompetitionlaws.Forinstance,firmsmaywantto:

a. Implementaproceduretoregisterwhenanemployeeisattendingeventswherecompetitorswill bepresent,andprovideguidancetothesestaffinsuchcircumstances;

b. Createalogthatcapturesallcorrespondencewithcompetitors(whetherthisisfacetoface,or via online communication channels) and have someone review this against what is and isn’t allowedundercompetitionpolicyandlaw;

c. Have a trained employee in competition policy and law review any contracts before they are enteredintoe.g.whensigninguptoanonlineplatform;

d. Train employees on relevant competition policy and law, and how these laws are enforced, highlightingthepotentialconsequencesofanybreaches;

e. Establish a whistleblowing telephone hotline so that employees can confidentially raise any competitionpolicyandlawconcernsthattheymighthave,oraskforadvicewheninanuncertain situation;and/or

f. Consultcompetitionlawyers.

12.6. Review processes

12.6.1. Firms should establish a periodic review process of their competition law compliancemeasuresbasedontherisksidentified.

127

Section for Competition Advocacy

PartC:

128

RegulatoryandlegalbarriersinASEAN to E-commerce and as impedimentstoasingledigitalmarket

13

13.1. Introduction

13.1.1. ThissectionlooksfirstataccesstoE-commerce inASEANbeforeconsideringcybersecurity intheregion.

13.2. Access to E-commerce

Customs and tax regimes

13.2.1. DifferingcustomsandtaxregimeswithinASEANaffectaccesstoE-commerceintheregion.Thishasoften ledtouncertaintyaroundcostsforfirms,whichhashadanegative impactonthepotentialforeconomiesofscaleforcompanieswithinthesector.333However,theestablishmentoftheASEANEconomicCommunityin2015shouldactasanenablertoovercomethisbarrier,asoneofitsobjectivesistostreamlinecustomsandtaxrulesacrossAMS.334

Online connectivity

13.2.2. AverageconnectionspeedsintheASEAN6arefasterthanintheAmericas,MiddleEastandAfrica,andaresimilartoworldwideaverages,asshowninFigure4.However,asevidencedbyTable7inSection4.2,thereisasignificantspreadintermsofinternetspeedswithinASEAN,withconnectivityspeedsinsomeAMSsignificantlyslowerthanglobalaverages.

13.2.3. Questionnaire respondents identified slow internet speeds as one of the key barriers to thedevelopmentofE-commerceintheregion.Oneoftherootcausesofthisisthelimitedoverallnetworkcoverage,drivenbythehighrisksforprivatefirmstoinvestininfrastructureduetotheuncertaintyabouttheirabilitytogenerateanadequatereturnontheirinvestmentinremoteandruralareas.Thecostofconnectionwithintheregionisalsohigh,withonlySingaporeandMalaysiaconsideredtohaveaffordablebroadband,335asdiscussedinSection4.


334 ASEAN Economic Community (2015).


129


337 Vela Asia online shopper (2013).

338 AT Kearney (2015).




Figure 4: Average internet speeds across the worldAverageConnectionSpeed(Mbps)

ASEAN 60

2

4

6

8

10

12

14

16

Worldwide AmericasAPAC Europe MiddleEastandAfrica

Source:Akamaistateoftheinternet,Q12017report.Note:EachbarconsistsofallcountrieslistedwithintheAkamaireportforthatregion.Forexample,'Americas'consistsofdatafrom15differentcountriesacrossNorth,SouthandCentralAmerica,withinternetspeedsrangingfrom1.4MbpsinParaguay,to18.7MbpsintheUnitedStates.

13.2.4. Stateaidcouldbeusedtoincreasebroadbandcoverage.ThishasbeensuccessfulintheEuropeanUnion,wherefundingwasusedtoencouragebroadbanddevelopmentinruralareaswhichwouldhaveotherwisebeenunattractivetoprivateinvestors.Thisprogrammehasbeensuccessfulinprovidingconsumerswithequalaccesstobroadband.336

13.3. Cybersecurity

The current challenge

13.3.1. Asof2013,only2%-11%ofdigitalbuyersuseonlinepaymentsintheregion.337Oneofthedriversofthislowtakeuprateisconcernsassociatedwithdatasecurityandcybercrime.

13.3.2. One of the five key actions identified by the ASEAN Business Club Forum in 2014 to promoteE-commerce was to reinforce cybersecurity.338 This involves “increasing information sharing andbilateral assistance, harmonising existing legislative frameworks, and creating a regional onlinedispute-resolutionfacility”.339

13.3.3. Currently, there isno regionalentity setuptofightcybersecurity issues.Thishascreatedanxietyamongstconsumers inthe region.340EstablishingE-payment-specific regulationsandharmonisingE-paymentregulationsregionallywouldhelptoaddresscybersecurity issues.341Thiswouldhelptoimprovethe leveloftrustamongconsumerswhowouldthenbemore likelytoactivelypurchaseitemsonline.Figure5,presentsthecurrentproblemacrossASEAN.

130

Source:ATKearney(2015).

E-payment solutions

13.3.4. E-paymentsolutionsarehelpingtoovercometheseissues,suchasAmazonPaymentsandGHLSystems,anITservicemanagementcompany.ThegovernmentsofSingaporeandMalaysiahavealsoproposedpotentialsolutionstoimprovepaymentregulations.TheMonetaryAuthorityofSingapore(MAS)hasrecentlyproposedanewregulatoryframeworkandgovernancemodelforpayments,whichaimstobringpaymentregulationsunderasingleframeworktostrengthenstandardsofconsumerprotection, anti-moneylaunderingandcybersecurity.MAShasalsoproposedaNationalPaymentsCouncil,tocoordinateinitiativessuchaspromotinginteroperabilityandadoptingcommonstandardsbetweenpaymentsolutions.342InMalaysia,thegovernmenthassetthreegoalsforanewintegratedpaymentsystemcalledtheEntryPointProject(EPP),whichisduetobeimplementedin2020andisaimingto:reducecashtransactionsfrommorethan90%to63%;increaseE-paymentsto200percapitaperyear;andincreasethenumberofpoint-of-saleterminalsto25per1,000inhabitants.

13.3.5. ForE-payments, regulation is required to ensure that current legal uncertainties canbe reduced.Forexample, inASEAN,cross-bordertransactionsoftenrequiregoingthroughaheavy‘know-your-customer’processinordertocomplywithlocalanti-moneylaunderingregulations.This increasescompliancecostsgreatlyandhasanegativeimpactontheexperienceofthecustomer.Thisshouldalso strengthen regulatory symmetry within the region between financial institutions and otherpayment agents to foster a fair and competitive environment, as long as this occurs throughoutASEANasawhole.343

342 Monetary Authority of Singapore (2016).


Figure 5: Digital Buyers who say they do not trust giving their credit card information online

Singapore

Malaysia

Vietnam

Indonesia

Thailand

Philippines

0% 10% 20% 30% 40% 50% 60% 70% 80%

131

14 Theimpactofintellectualpropertyrights(includingitsterritorialnature)as a barrier to E-commerce in ASEAN andasanimpedimenttoasingledigitalmarketinASEAN

14.1. Introduction

Overview of intellectual property rights

14.1.1. Intellectualproperty(IP)isdefinedas“creationsofthemind,suchasinventions;literaryandartisticworks;designs;andsymbols,namesandimagesusedincommerce.”344Moreover,IPrightsaredefinedas“theassignmentofpropertyrightsthroughpatents,copyrightsandtrademarks.Thesepropertyrightsallowtheholdertoexerciseamonopolyontheuseoftheitemforaspecifiedperiod”.345

14.1.2. Thereareeconomiccostsassociatedwithgrantingsuchmonopolypowerasaresultofcreatingabarriertoentryforfirmsthatdonothaveaccesstotheprotectedproperty,asdiscussedinSection4.3.However,thebenefitstosocietyfromincentivisinginnovationbygrantingIPrightsaregenerallyregardedtooutweighthesecosts.346Specifically,IPrightsfosterinnovation,creativity,entrepreneurship,investmentinknowledge-basedassetsandgrowth,347bothinofflineandonlinemarkets.

14.1.3. Competitionauthorities’objectivesofpreservingcompetition,andtheobjectivesofIPgrantingtheirowners exclusive rights, both have the ultimate objective of promoting consumerwelfare via anefficientallocationof resources.Theeffectivecoverage,operationandenforcementof IP rights isthereforevitalforthepromotionofcompetitionbycreatingtheright incentivesfor investments ininnovation.

The importance of an effective IP rights system

14.1.4. IfthereareinefficienciesinIPrightssystems,thedevelopmentofmarketsmaybeinhibitedasfirms’incentivestoinvestarediminished.Thisistrueinbothbrick-and-mortarandonlinemarkets.However,issues relatingto IP rightsareparticularly important inE-commercemarkets,forexample innewdigitalcontentmarketswhichhaveemergedwhereIPrightsarerequired,suchasforE-books.Inthisregard,twooutofthefivequestionnairerespondentscitedIPrightsasabarriertothedevelopmentofE-commercemarketsintheirjurisdiction.

14.1.5. InfringementofIPrightsisnotonlyaformofeconomicinefficiency.ThegrowingriskofcounterfeitgoodsposesathreatfortheinnovativebusinessesthatholdIPrights,andalsoconsumers,relatingtothesafety,healthandsecurityimplicationsofgoodsthataresoldthroughE-commerce.348

Structure of this section

14.1.6. This sectionfirst outlines how IP rights can create a barrier to the development of E-commercemarkets,beforediscussingtheterritorialnatureof IP rightsandthe importanceofaneffective IPrightssystemforthedevelopmentofasingledigitalmarketinASEAN.Finally,theroleofcompetitionauthoritiesinpromotinganeffectivesystemofIPrightsallocationandenforcementisalsoconsidered.

344 WIPO (2017).

345 OECD (1993).

346 OECD (1993).

347 OECD (2015b), page 12.

348 OECD (2007b).

132

349 OECD (1997).

350 US Department of Justice (2004).


14.2. Intellectual property rights as a barrier to E-commerce

14.2.1. IP rights are crucial for thedevelopmentof E-commercemarkets as theyprovidefirmswith theconfidencethattheoutputsfromtheir investmentswillbeprotected,therefore incentivisingsuchinvestmentbyensuringsufficientreturns.349

14.2.2. Bycontrast,intheabsenceofaneffectivesystemofIPrightsinE-commercemarkets,firmswillnothavetheincentivetoinvest,thusharmingthedevelopmentofsuchmarkets.Forexample,ifpiracyiscommonindigitalcontentmarkets,thereislessincentiveforfirmstodevelopsuchcontent.Onequestionnairerespondenthighlightedthatthisiscurrentlythecaseinitsjurisdiction.Inaddition,IPrightsmustbeeffectivelyenforcedtogivefirmstheconfidenceandsecuritytomakeinvestments.

14.2.3. TheriseofE-commercehasfuelledglobalisationasithasallowedfirmstomanufactureandmarkettheirgoodsandservicesonaglobalscale.Therefore,forfirms intheglobalmarketwhichrelyonlicensesforIP,theserightsnowneedtoapplyglobally,ratherthannationally.Asaresult,differencesbetween licensing rules around theworld can potentially prevent firms from engaging in cross-bordertradeinbothonlineandofflinemarkets.350LackofharmonisationofIPrightsamongAMSmaythereforecreateabarrierandpreventfirmsfromoperatingandtradingeffectivelyonaglobalscale.

14.3. Intellectual property rights as a barrier to a single digital market in ASEAN

14.3.1. IPrightsaretypicallygrantedonaterritorialbasis;i.e.givingafirmprotectioninacertainlocationforaspecifiedperiodoftime.CurrentlyinASEAN,IPrightsaregrantedandenforcedonanationalbasis.AsaresultoftheterritorialnatureofIPrights,firmsmaynotbeabletoofferconsumersinanotherAMSaparticulargoodorservice.Thismayinhibitcross-bordertradeandformabarrierasASEANcontinuestomovetowardsanintegratedmarket.

IP rights in digital content markets

14.3.2. Thisisparticularlyrelevantindigitalcontentmarkets.InitsFinalReportontheE-commerceSectorInquiry,theEuropeanCommission(2017b)considersIPrightsindigitalcontentmarketsindetailastheseissuesarehighlyrelevantforasingledigitalmarketsuchastheEuropeanUnion.Itishighlightedhowtheemergenceofdigitalmarketshasledtothedevelopmentofarangeofcomplexlicensingarrangements,withrightstypicallylicensedonanexclusiveornon-exclusivebasisforcertainterritoriesoveraspecifiedlengthoftime.Itisconcludedthatexclusivelicensingonaterritorialbasisisnotinitselfproblematic,butcompetitionconcernsmayarise ifcertainothercontractualrestrictionsarepresent,suchasrestrictionsoncross-borderpassivesales.However,theEuropeanCommissiondoesnotcommittoafirmstanceeitherway,butrathersuggeststhatitwillassesslicensingarrangementsonacase-by-casebasis,takingintoaccount“thecharacteristicsofthecontent industry,thelegalandeconomiccontextofthelicensingpracticeand/orthecharacteristicsoftherelevantproductandgeographicmarkets.”351Additionally,theEuropeanCommissionexplainshowbundlingofdigitalcontent (for instance alongside offline content) may raise concerns if such conduct leads to arestrictionofoutput,forinstanceifalicenseedoesnotfullyexploittheonlinerightsithasacquired.Thedurationofcontracts,orthetermsofrenewalmayalsoconstitutebarriersfornewentrants.

Allocation and enforcement of IP rights in AMS

14.3.3. ForafirmseekingtoobtainIPrightsacrossASEAN,itmustdososeparatelyineachAMS,asIPrightsandpatentsareonlyvalid in theterritory inwhichtheyaregranted, thereby raisinganadditionaladministrativeburdenandcostforfirms.AlthoughfirmsmaybeabletoregisterIPacrosstheentireregion, lengthy processes for obtaining IP rights in some countriesmay deterfirms frommakingtheinvestment,ashighlightedinTable13.Furthermore,ifregion-widesalesarerequiredinordertomakeasufficientreturnonaninvestment,investmentacrosstheregionasawholemaybeinhibited,thusrestrictingthegrowthanddevelopmentofE-commercemarkets.ThedifferencesbetweenIPofficesinAMSarehighlightedinTable13below.Anumberofperformancemetricsarepresented,andcomparedwiththeequivalentfiguresintheUS.

133

Table 13: AMS IP Office performance

ASEAN Member State

Brunei Darussalam

Cambodia Indonesia

LaoPDR

Malaysia

Myanmar

ThePhilippines

Singapore

Thailand

Vietnam

USA

Source:Averagetimetoregisterapatent/trademarkinAMS-BernardandWedel(2011);Averagetimetoregisterapatent/trademarkinUS-IPSpotlight(2016);IPProtectionRank–WorldEconomicForum(2017);TimelinessofReceivingOfficetotransmitcopiesofPCTfilingstotheInternationalBureau–WIPO(2016);Notes:IPProtectionRankbasedonWorldEconomicForumExpertOpinionSurvey“Inyourcountry,towhatextentisintellectualpropertyprotected?”;ThePatentCooperationTreaty(PCT)supportsapplicantsgetquickerinternationalpatentprotection.ThefirststageofthisprocessisfortheReceivingOffice(wherethepatentisfirstsubmitted)totransmitthefilingtotheInternationalBureau.

14.3.4. TherearealsocurrentlydifferencesinthelevelandspeedofIPrightsenforcementacrossAMS,ashighlightedbythevaryingIPProtectionRanksamongtheAMSinTable13.SomecountriessuchasSingaporeandMalaysiaperformrelativelystrongly,andarerankedinthetopquartileofcountries,howeveritisevidentthatimprovementsinIPenforcementarerequiredinotherAMS.SpecificissuesoutlinedbytheEuropeanCommission(2015c)352includehighlevelsofpiracyandcounterfeitgoods,aswellasalackofregulatorydataprotection.Itisalsohighlightedthattheprocessofundertakingjudicialprocessescanbe lengthy,thereforefirmsare less inclinedtopursuesuchformsofenforcement.Consequently,cross-bordertradeisinhibitedasfirms’incentivestoinvestarereducedduetofearsthattheirinvestmentswillnotbeprotectedeveniftheyaregrantedIPrights.

Average time to register a patent (2011)

-

3years

5.6yearswithnormalprocess,4.9yearsviaPCT

4years

1.6yearsviafast-trackbasiswithoutobjections;5.4yearsviaParisConvention;2.2yearsviaPCT

-

4-5years

3-4years

3years

2-3years

2.3years

Average time to register a trademark (2011) -

3months

14months

6months

17-24months

1month

10months

6-8months

12-18months

15-18months

10months

IP Protection Rank (out of 138) (2016)

58

130

50

96

27

-

74

4

121

92

16

Timeliness of Receiving Office to transmit copies of PCT filings to the International Bureau (proportion where the application transmittal delay was 2 weeks or less) (2016)

-

-

14.3%

54.8%

-

-

30.0%

99.5%

20.6%

16.7%

58.0%

352 Available at: https://euipo.europa.eu/ohimportal/documents/11370/0/ Report+on+the+protection+and+enforcement+of+intellectual+property+rights+in+third+countries

134

353 US commerce department, (2017).


355 Ibid.

356 WIPO (2017).

357 At time of writing there are 152 PCT contracting states.

Harmonisation of IP rights allocation and enforcement

14.3.5. Improvementsin,andconsistencyofIPrightsenforcementamongAMSarethereforehighlyimportant.AllAMSaremembersoftheWorldTradeOrganization(WTO)andarerequiredtocomplywiththeAgreementonTrade-RelatedAspectsof IntellectualPropertyRights (TRIPS),whichsetsminimumstandardsonIPrightsprotectionandenforcement,353thereforebasic legislativeframeworksare inplace.Furthermore,progressisbeingmadeacrosstheregiontoimproveIPrightsenforcement.Forexample,thePhilippineshasrecentlygrantednewIPrightsenforcementand inspectionfunctionstothePhilippinesIPOffice,processeswhichwerepreviouslyonlygrantedtothePhilippineNationalPolice,theNationalBureauofInvestigation,theBureauofCustomsandtheOpticalMediaBoard.354

Indonesia has also followed a similar trajectory of improvement, for example by introducing newlegislationtotackleonlineinfringementinitsCopyrightLaw(No.28/2014).355

14.3.6. As ASEAN continues to move towards a single integrated market following the creation of theASEANEconomicCommunityandimplementationofthesingledigitalmarket,cooperationbetweencompetitionandIPauthoritiesisvital,bothnationallyandwithinASEAN.Coordinationbetweenthesegroups is important to help to bridge the gap between parliaments that set the IP rules, and IPauthoritiesandcompetitionauthoritieswhoenforcethese rules.Theworkof theASEANWorkingGroupon IntellectualPropertyCooperation (AWGIPC)willhelptodrivethisacrossASEANthroughinitiativessuchastheASEANIntellectualPropertyRightsActionPlan2016-2025whichoutlines19newinitiativesforthecoming10years.

14.3.7. Cooperationofthiskindshouldcoverthefollowingissues:356

a. Exchangeoftechnicalinformation;

b. Sharingofprocedural routines, guidelinesandstandardsforthetreatmentofmergersoranti- competitivepracticesinvolvingIP;

c. SharingofstudiesconcerningtherelationshipbetweenIPrightsandantitrust;and

d. MutualtrainingofpersonnelfromIPandcompetitionauthorities.

14.3.8. Theseprinciplesarereflectedinthe19initiativesincludedintheASEANIPRightsActionPlan(2016-2025).

14.3.9. An important stepforward in theharmonisationofpatents regimesacrossASEAN is throughthePatentCooperationTreaty(PCT),ofwhichallAMSexceptMyanmar(whichisamemberofTRIPS)aremembers.ThePCT,originatedin1970,providesaunifiedprocedureforfilingpatentsineachof itsMemberStates,andprovidesassistancetobusinessesandnationalpatentoffices.Therefore,afirmthatfilesaninternationalpatentapplicationunderthePCTcanbenefitfromprotectionforitsinventionacrossalargenumberofcountries.357Thisisimportantforbreakingdownthebarrierstocross-bordertradethatarisefromtheterritoriallylimitednatureofIPrights,andensuringtheeffectiveoperationofasingledigitalmarket.Table13does,however,highlightdifferencesintheperformanceofAMSIPofficesinpassingonpatentapplicationstotheInternationalBureauaspartofthePCTprocess.

135

358 Thomson Reuters (2007).




362 Thomson Reuters (2007).

363 DLA Piper (2015).

364 Slaughter and May (2016).

365 Case reference: T-167/08

14.4. The role of competition authorities

Interaction between IP law and competition law

14.4.1. IP rights promote innovation and lead to economic growth, competitiveness and job creation,therefore competition law should complement IP law. Through their advocacy role, competitionauthorities arewell positioned topromote the effective enforcement of IP rights, in particular byencouragingcoordinationbetweendifferentcountries.However,thecurrentinteractionbetweenIPlawandcompetitionlawislimited.ThemainreasonforthisisthatIPrightstypicallyhavecheckstolimitthepossibilityofabusesthatwouldviolatecompetitionlaw,althoughissuesaremorelikelytooccurinrelationtonewsubjectmatterswhichwerepreviouslyunprotectedbyIPlaw.358

14.4.2. ThereisariskthatIPlawbyitselfdoesnotpromoteconsumerwelfareasinnovatorsmayattempttostiflefuturecompetition,orconsumersmaynotbenefitfromfairaccesstotheseinnovations.359

Competition authorities can therefore help to find the right balance between the interests ofdistributors,artists,inventorsandcreators,andtheinterestsofconsumers.360 Effectivecompetitionrulesenforcedbycompetitionauthoritiesarepartoftheanswer,thoughifthereisafundamentalflawinIPrules,thiscanonlybesolvedbyIPlegislation.Nevertheless,competitionauthoritiescansupportIPlegislatorstoeffectivelydesigntheserules.

Supporting the harmonisation of IP rights allocation and enforcement

14.4.3. CompetitionauthoritiescansupporttheimprovementandharmonisationofIPrightsallocationandenforcement acrossASEAN by providing guidance and support to IP offices in ensuring that theoptimal level of IP rights is granted throughout the region. Competition authoritieswithinASEANshouldlistentolocalbusinessesandconsumerscommunitiesinordertounderstandwhenIPrightsmaybeunfairly impactingcompetition,andthereforeharmingconsumers.AshighlightedbyWIPO(2017b),toomuchIPmayinhibitcompetitionwhenfirmsaregrantedexclusivityfornon-differentiatingfeatures.However,WIPO(2017b)alsooutlineshowtoolittleIPissub-optimalduetounder-investmentresultingfromalackofprotectionforthereturnsfromfirms’innovation.

Competition authority intervention

14.4.4. Manyfirmsholdpatentsfortechnologythatisregardedas‘standard’forthatindustrytofunction,suchasthetechnologyrequiredtosendapicturemessageviaamobilephone.Regardlessofthemanufacturerofthephone,orthemobilenetworkthatisused,itisassuredthatthepicturewillbedelivered.361 ThisistheresultofStandardEssentialPatents,orSEPs.SEPsarecommoninE-commercemarkets,inparticularintheelectronics,computingandcommunicationssectors.362OwnersofSEPsshould licencetheirtechnologythroughfair, reasonableandnon-discriminatory (FRAND)terms,toencouragestakeholderstouseandimplementanindustrystandard,whilststillensuringthatownersofSEPsareappropriatelyrewarded.363AkeypracticethatcompetitionauthoritiesshouldbewaryofiswhenanSEPholder imposesunreasonabletermsand/orexcessivelyhighpricesonthosewhorequiretheuseofthistechnology,thereforerestrictingaccesstothetechnologyandharmingbothconsumersandinnovationwithintheindustry.

14.4.5. Competitionauthoritiesshouldalso intervenewhen IP rightsareabusedmoregenerallyby rightsholders,thoughthisisonlylikelytobeinexceptionalcircumstances.Forexample,interventionmayberequirediftechnologytransferagreementsincludeprice-fixingrestrictions,limitationsofoutput,orallocationofcustomersormarkets.364Refusalstolicensecanbedeemedasanabuseofdominance,asseeninthecaseagainstMicrosoftforrefusingtodiscloseinteroperabilityinformation.365Furtherconcernsmayariseifalicensingfirmforeclosescompetitorsthroughtyingorbundlingstrategies.

136

15 Recommendationsonthestrategies,toolsorapproachesAMScanadopttohelpgovernmentbodieswithintheirrespectivecountriestounderstandtheimpactoftheirpoliciesandinitiativesoncompetitionintheE-commerce sector

15.1. Introduction

Importance of effective policy implementation

15.1.1. It is important that governments maintain and promote competitive markets in order to fosterproductivity growth in their countries. Over the past thirty years the competition policy, law, andeconomicsdebatehashighlightedhowoverlyregulatedproductmarketscaninhibitthedevelopmentofcompetitionandtherebyhamperproductivitygrowth.Itisthereforeimportantthatgovernmentsattempt tominimise thedegreeof regulation inmarketswhenthe sameobjective (ofpreventingconsumerdetriment)canbeachievedbytheeffectiveapplicationofcompetitionlaw.Atthesametime,thiscouldminimisetheregulatoryburdenonbusinesses,hencepromotingentryandfosteringfurthercompetition.ThisisparticularlytrueinE-commercemarkets,wherethereissignificantgrowthpotentialandthepaceofinnovationisparticularlyfast.


15.1.2. This section presents a roadmap to aid government bodies in conducting an assessment of thelikely impactofnewpoliciesoncompetition inE-commercemarkets,and inevaluatingtheeffectofanewpolicyfollowingimplementation.Theseguidelinescanalsobeusedtodeterminewhetheritisbeneficialtoremoveaparticularpolicyorregulationfromamarket.Thissectionfirstconsidersthe role that competition authorities canplay in supporting government bodies to conduct suchassessments,andthenprovidesguidanceonconductingexanteandexpostassessmentsofpoliciesin E-commerce markets.

137

15.2. Role of competition authorities and regional bodies

Support that competition authorities can provide government agencies

15.2.1. Inordertofosterabroadapproachtopolicy-makingwhichconsiderstheimplicationsoncompetitionandmarketsofnewproposedpolicies,centralgovernmentsaswellascompetitionauthoritiescanplayasignificantrole.Thisroleisessentiallytwofold.First,theycanpromoteawidedebateacrossagencieswhichfosterstheexchangeofexperienceandexpertise,thusprovidingavoicetocompetitionauthoritiesinaconstructivedialoguewithsectorregulatorsandotherrelevantgovernmentbodies.Relevantexamples in theUKare theUKCompetitionNetwork,366 andtheUKRegulatorsNetwork(UKRN),367whoseaimsaretosupportandenablecompetitioninvarioussectorsacrosstheeconomy.In ASEAN, there is the CCS’s Community of Practice for Competition and Economic Regulations(COPCOMER)whichprovidesaninter-agencyplatformforCCS,sectorregulatorsandothergovernmentbodies to share best practices and experiences on competition and regulatorymatters. Second,a requirement to conduct a competition impact assessment for newly proposed policies can bean extremely effective tool in reducing barriers andmitigating potential regulatory challenges forbusinesses.

Support for government agencies assessing proposed policies

15.2.2. Competitionauthorities should support governmentbodies in applying the guidelinesprovided insections15.3and15.4whendecidingwhetherornottointerveneinaparticularE-commercemarket,and,ifso,howbesttodosowithoutharmingcompetition.Forinstance,competitionauthoritiescanprovideadviceonthebestdata touse, andthequantitative techniquesthat shouldbeadopted,drawingonexperiencesfrompreviousinvestigationsormarketstudiesinrelatedindustries.

15.2.3. Thereareanumberofresourcesthatcompetitionauthoritieshaveproducedwhichshouldbesharedwithgovernmentbodiestoassisttheminassessingtheimpactsofpolicies.FurtherinformationontheseresourcesarehighlightedinSection15.3.

15.2.4. Competitionauthoritiescanalsoworkalongsidegovernmentbodiestoconductjointmarketstudies,especiallywhentherearebothregulatory issuesandcompetitionconcernsinaparticularmarket.Sharing of knowledge between the two bodies can ensure that all relevant information is beingconsidered,andthecorrectconclusionsarereached.

15.2.5. Toensurethatgovernmentbodiesconductassessmentsbeforeimplementinganewpolicy,competitionauthoritiescanalsoplayastrongeradvocacyrolebypro-activelyreachingouttogovernmentbodiestoexplaintheimportanceofconductingcompetitionassessments,andbyhighlightingtheadverseeffects that can occur in the absence of a thorough assessment. In such dialogues, competitionauthoritiesshouldalsoexplainthesupportthattheycanprovideinconductingtheseassessments,andhighlightpreviousexperiencethattheauthorityhas inthatmarket,orother relatedmarkets.CompetitionauthoritiesshouldalsodrawupontheirnetworkofothercompetitionauthoritiesacrossASEAN,andinotherjurisdictions,whomayalsohaverelevantexperienceinaparticularmarket.

Role of regional bodies

15.2.6. Regionalbodies,suchastheASEANExpertsGrouponCompetition(AEGC)alsoplayavitalrole inassistinggovernmentbodiestounderstandthe impacttheirpolicieshaveoncompetition.AbodysuchastheAEGChelpstostrengthentheregulatoryenvironmentacrossASEANbyhostingtraining,workshopsandseminarstostrengthenthecapabilitiesofcompetition-relatedagencies,andoperatesasaforumtodiscussandcoordinatecompetitionpoliciesintheregion.

366 https://www.gov.uk/government/groups/uk-competition-network

367 http://www.ukrn.org.uk/

138

368 CMA (2015b), page 7.

15.3. Ex ante evaluations of policies

Overall approach

15.3.1. Asoundassessmentofthe implicationstocompetitionwhichaproposedpolicymaybringaboutwouldstartfromconsideringthefollowingsetofquestions:368

a. Willthemeasuredirectlyorindirectlylimitthenumberorrangeofsuppliers?

b. Willthemeasurelimittheabilityofsupplierstocompete?

c. Willthemeasurelimitsuppliers’incentivestocompetevigorously?

d. Willthemeasurelimitthechoicesandinformationavailabletoconsumers?

15.3.2. Iftheanswertoanyofthesequestionsisyes,policymakersshouldconductamorethoroughreviewofwhetherornottoimplementthepolicy,andevaluatetheeffectoncompetitionfromalternativeformsofthepolicyintervention.

15.3.3. Firstly,governmentbodiesshouldidentifyalloftheaffectedmarkets.Importantly,thismayextendbeyond the products or services immediately affected (both horizontally and vertically – up anddownstream)andgeographicareasthatthepolicydirectlytargets.Additionally,asdiscussedthroughoutthishandbook,manyE-commercemarketsaremulti-sidedinnature.Intheseinstances,allsidesofthemarketshouldbeevaluated,andrelatedmarketsconsidered;forexample,intheonlinesearchmarketbothwebbrowsersandadvertiserswouldbecovered.Next,ineachoftheaffectedmarkets,governmentbodiesshouldassesstheextenttowhichcompetitionwillbeadverselyaffected,andcompare this to a counterfactual scenario of no intervention. Comparisonswith alternativewaysinwhichthepolicyobjectivemaybeachieved(e.g.throughadifferentpolicyoralternativeformofintervention)shouldalsobeundertaken.

15.3.4. Inadditiontothefourquestionsabove,furtherquestionsforacompetitionimpactassessmentinanE-commercemarketincludethefollowing:(forallquestions,iftheanswerisyes,competitionismorelikelytobeharmed)

a. Isexclusivitygrantedtoasinglefirm,orlicensesgiventoarestrictednumberofcompanies?

b. Arefirms’costsincreasedasaresultofthepolicy,andwillthisincreasethelikelihoodoffirms findingitdifficulttooperateorleavingthemarketasaresult?

c. Willitbeharderfornewfirmstoenterthemarketasaresultofthepolicy?

d. Willsomefirms(e.g.smallfirms)bemoreadverselyaffectedthanothersasaresultofthepolicy?

e. Willfirmshavelessflexibilitytosetprices?

f. Willfirmsbelessabletocompeteonthequalityofgoodsand/orservicesoffered?

g. Willthepolicyfavoureitherbrick-and-mortaroronlineretailersmorethantheother?

h. Doestheregulationmakeiteasierforcompetingfirmstoworktogetherasopposedtoincompetition?

i. Willcustomershavelessdegreeofchoiceastowhichfirmtheypurchaseagoodorservicefrom?

j. Willconsumershaveaccesstothelessinformationfollowingimplementationofthepolicy,oris informationbehardertounderstand?

k. Willitbeharderforcustomerstoswitchfromonefirmtoanother?

l. Willitbeharderforconsumerstomulti-homefollowingimplementationofthepolicy?

m.Willthepolicyresultinatippingpointinthemarketasaresultofnetworkeffects?

n. Willthemarketgrowataslowerrateasaresultoftheregulation?

139

Types of assessment

15.3.5. Assessmentsofproposedpoliciesmaybequalitativeorquantitativeinnature.Qualitativeassessmentsmay combine economic argumentswith insights from research, and studies on similar previouspolicies. For example, conclusions may be reached following a comparison between themeritsandweaknessesofdifferent interventions.Although relativelyeasytounderstandand implement,qualitativeassessmentsareunabletoputvaluesoncertaincostsandbenefits.Itisthereforedifficulttoweighttherespectiveadvantagesanddisadvantagesofvariouspolicies,andcometoaconclusionastowhichisbest.Formorerobustanalysisenablingsuchweighting,quantitativeapproachescanbeused(e.g.cost-benefitanalysis),thoughthesemethodsaretypicallyhardertoimplement,andaretoalargeextentreliantontheavailabilityofdata.

15.3.6. InconductingquantitativeassessmentsinE-commercemarkets,governmentsmaylookattheeffectofsimilarpoliciesinrelatedproductorgeographicmarkets(controllingformarket-orplace-specificfactorsrespectively),orpreviouspoliciesinthetargetedmarket(controllingfortime-variantfactors).However,giventherapidgrowthandchangingnatureofE-commercemarkets,itmaybedifficulttocontrolfortime-variantfactors.

15.3.7. SpecificdatathatgovernmentbodiesmayfindhelpfultoconsiderwhenconductingassessmentsinE-commercemarketsinclude:369

a. Levelsofmarketconcentration(i.e.thedistributionofmarketsharesoffirms);

b. Levelsofentryintothemarketbyfirms(consideringboththeoveralllevelandthelevelofentry amongonlineandbrick-andmortarretailersseparately);

c. Levelsof informationavailabletoconsumersandthedegreetowhichthis informationcanbe understoodandeasilyaccessedbyconsumers;

d. Costsofentrytothemarketforfirms;

e. Costsofexitingthemarketforfirms;

f. LevelsofinnovationorR&Dspendintheindustrybyfirms(e.g.newfeaturesavailableonwebsite);

g. Levelsofconsumerswitchingbetweenfirms;

h. Switchingcostsforconsumers;

i. Extenttowhichconsumersmulti-home;

j. Coststoconsumersinmulti-homing;

k. Interoperabilitybetweenonlineplatforms;

l. Pricelevelsinthemarket;

m.Efficiencyoffirmsinthemarket(e.g.costsofproduction);

n. Quality of services provided to customers (e.g. delivery success rates, delivery times, returns policies,featuresofwebsites);

o. Qualityofgoodsprovidedtocustomers;

p. Degreeofdiversity offeredbyfirms inproducts/services (e.g. diversity of features availableon differentwebsites);and

q. Quantityofgoods/servicesprovided(and/orrateofgrowthinthis).

369 CMA (2015b), page 9.

140

15.3.8. Foramoredetailedoutlineoftheprinciplesandapproachescompetitionauthoritiesshouldfollowin conductingimpactassessmentssuchasthese,theOECD(2015),CMA(2015c)andCCS(2016)provideusefuldiscussions,includingexamplesofpreviousassessments.

15.4. Ex post evaluations of policies

15.4.1. Ifagovernmentbodydecidestoimplementapolicyinstages,itcanassessthepolicy’sinitialimpacton competition before decidingwhether to continue to implement the policy, or expand itmorewidely.Agovernmentbodymayalsowanttoevaluatehowaccurateitwasinitsexanteevaluationinordertolearnlessonsforfutureassessments.Forbothofthesepurposesagovernmentbodymaywanttoconductanexpostevaluationofapolicy,andconsidertheeffectthatitsinterventionhadoncompetitioninthemarket.

15.4.2. Whenconductingexpostevaluationssuchasthese,themarketbeingevaluatedshouldbecomparedtoabaselinemarketunaffectedbythepolicy.Examplesofsuchbaselinesmaybe:thesamemarketintheperiodbeforethepolicywasimplemented(controllingfortime-variantfactors);ageographicareawherethepolicywasnotimplemented(controllingforplace-specificfactors);orasimilarrelatedmarket that does not have an equivalent policy in place (controlling formarket-specific factors).Expostevaluationscanalsobeconducted relativetoabaselineofalternativepoliciesthatwereconsideredintheexanteassessment.Asdiscussedintheprevioussub-section,controllingfortime-variantfactorsmaybechallenginginE-commercemarketswheremarketcharacteristicsareoftenquick tochange, therefore thefirstof thethreeapproachesabovemaybehardto implement inpractice.

15.4.3. Datatobeconsideredintheseexpostevaluationsarethesameasthosediscussedintheprevioussub-sectionforquantitativeexanteevaluations.

15.4.4. TheOECD(2015)advisethatthetimeperiodtowaitbeforeconductinganexpostevaluationshouldbe carefully considered, ensuring sufficient time for the policy to have an effect, but notwaitingtoo longsuchthat itbecomesdifficult to separate theeffectof thepolicy fromgeneral shifts inthe market. A case-by-case approach should be adopted, though the OECD (2015) recommendthatgovernmentbodiesshouldtypicallywait2-3yearsbeforeconductinganexpostassessment.Additionally,adifferentteamshouldconducttheexpostevaluationtothatwhichconductedtheexanteassessment,inordertoensurethattheapproachtakenisnotbiasedinanyway,andthatmistakescanbeidentifiedandlessonslearned.

141

Conclusions

142

16 Conclusions

16.1.1. E-commerce markets have rapidly emerged and grown across ASEAN over the past decade.Currently E-commercemarkets have reached a totalmarket size of US$7 billion across the sixlargesteconomies inASEAN;370andmarketsarepredictedtocontinuetogrowacrosstheregion.Overthecoming3years,B2CE-commercesalesinSoutheastAsiaalonearepredictedtogrowatanannualrateof17.1%.371

16.1.2. Thereare,however,anumberofbarrierswhichmayraisehurdlesforthisgrowthtobeachieved.Thelevelofdevelopmentoftechnologicalinfrastructureintheregionisonesuchbarrier.Legalandregulatory frameworks can also inhibit cross-border trade by failing to provide full and adequateprotectiontoconsumersfromonlinethreatstopersonaldataandfinancialinformation.Piracyandthesaleofcounterfeitgoodsarealsocommonthreatstoconsumersandbusinessesalike.

16.1.3. ConsumershavehoweverlargelybenefittedfromtheemergenceandgrowthofE-commerceintheregion,inparticulardueto:

a. Areductioninsearchcosts;

b. Greaterpricetransparency;and

c. Widerdiversityofgoodsavailable.


371 Frost & Sullivan (2016b).

E-commerce markets in the six largest economies in ASEAN have reached a total market size of

US$7billion

B2C E-Commerce sales in Southeast Asia alone are predicted to grow

17.1%annually

143

16.1.4. Inensuringthesebenefitsarefullyrealised,competitionauthoritiesaroundtheworldhaveencounteredaseriesofchallengesinapplyingtheircompetitionlawtoensurethatE-commercemarketsremaincompetitive,notably:

a. Manynewmulti-sidedmarketshaveemerged,suchasonlinemarketplacesandPCWs.Multi-sided marketsarenotanewphenomenon.However,theincreaseintheirprevalenceindigitalmarkets hasmadetheneedtorethinktraditionaltoolsdesignedfortheanalysisofcompetitioninsingle- sidedmarketsallthemoreapparent,frommarketdefinitiontotheassessmentofmarketpower;

b. Markets are more dynamic in nature. The importance of innovation for the growth and competitivenessofonlinemarketshasemphasisedtheneedtoexaminepotentialcompetition andmovebeyondthestaticframeworkofanalysisadoptedforcompetitionassessments;

c. Newverticalrestraintshaveemergedorexistingrestraintshaveincreasedinprevalence(e.g.MFN clauses, platform bans, geo-blocking strategies, and dual-pricing systems). These restraints havebeentheobjectofin-depthscrutinybothviaaSectorInquiryconductedbytheEuropean Commission and through a number of cases investigated in several jurisdictions around the world. Broad consensus indicates that most of these restraints have been adopted to address potential problems such as free-riding, and incentivising investments. Nevertheless, somespecific instanceshave raisedquestionsastotheircompatibilitywithcompetition rules and/orwidersingledigitalmarketobjectives;and

d.Horizontalcoordinationbetweencompetingfirmshasbecomeeasierduetotheemergenceof pricemonitoringtoolsandprice-settingalgorithms.

16.1.5. Byreviewingrelevantcasesfromjurisdictionsaroundtheworld,thishandbookhasfoundthattheexistinglegalframeworkhasbeenbroadlysufficienttodealwiththeemergingchallengesresultingfromthegrowthofE-commerce.Thereare,however,asmallnumberofinstanceswhichrequireabroaderapproachininvestigationsinE-commercemarketsthathavebeenidentifiedfromcasesandeconomicliteratureinthefield.Specifically:

a. In investigating multi-sided markets, a holistic approach is required which goes beyond the applicationoftraditionalantitrustanalyticaltools.Allsidesofthemarketshouldbeconsideredin anyassessment,takingintoaccountthepresenceanddirectionofnetworkeffectsandfeedback effects.Thisapplieswhendefiningrelevantmarkets,assessingmarketpower,evaluatingalleged harm,andreviewingproposedmergers;

b. Inassessingactualorpotentialmarketpower,andwhenreviewingproposedmergers,dynamic competition shouldbeconsidered i.e.will amerger result in the removalof apotential future entranttoamarket,orarethereothercompetitiveconstraints,includingotherpotentialentrants, whichcanmitigatethisconcern;and

c. Toenablecompetitionauthoritiestoreviewmergersthatmayleadtoalesseningofcompetition inthelongrun,atransactionvaluethresholdmaybeneededinmergercontrolrules.

144

16.1.6. There are, however, a numberof areaswhere international consensushasnotyetbeen reached.Competition authorities in ASEAN should therefore closely monitor emerging case law in jurisdictionsaroundtheworld,inadditiontotheongoingdebateintheantitrustcommunityinthefollowingareas:

a. Theanalyticalframeworkstousewhenassessingmulti-sidedmarkets;

b. TheuseofwideandnarrowMFNclausesbyfirms.AlthoughwideMFNshavebeenbroadlyregarded asgivingrisetoanti-competitiveeffects,differentjurisdictionshaveprovideddifferentresponses totheadoptionofnarrowMFNs,withsomebanningthemaltogetherinthehotelbookingmarket (e.g.Germany,Italy,France,andAustria),andothers,includingtheUSandtheUK,allowingsuch clauses;

c. TheuseofplatformbansandrestrictionsonPCWsbyfirms.Thoughmarketplacebanstypically donotconstituteatotalbanoninternetsales,thereisongoingdebateonwhethersuchrestrictions maybe justified.A landmark judgement in Europe is duewithin the nextyear. Similar debate regardingrestrictionsonPCWsisalsotakingplacesimultaneously;and

d. Thepotentialforprice-fixingalgorithmstoself-learnthatcoordination isoptimal.Althoughthe effectofsuchtoolswouldundeniablybeanti-competitive,thelackofdirectobjecttocoordinate inthefirms’adoptionofsuchtoolsraisesanimportantquestionontheapplicabilityofexisting competitionlaw.

16.1.7. Finally,tosupportthegrowthofE-commercemarketsacrossASEAN,itisrecommendedthatAMScompetitionauthoritiesshouldconsider:

a. Workingtowardsharmonisationonsomekeyareaswhichareattheheartofthedevelopmentof onlinemarkets.Harmonisationontheinterpretationofexistingcompetitionpolicyandlawinthe regioninE-commercemarkets(e.g.ontheuseofMFNs,geo-blockingstrategies,platformbans, andrestrictionsonPCWs)andclearcommunicationoftheseinterpretationstobusinesseswould alsofosterthedevelopmentofE-commercemarkets;

b. Workingalongsideregulatorybodiesandcross-ASEANgroupstosupporttheharmonisationand improvement of the regulatory regime that firms face, for instance with regards to IP rights enforcementandcybersecurity;and

c. Fosteringdialogueandsupportgovernmentbodiesindesigningpoliciessuchthatcompetitionin E-commercemarketsisnotadverselyaffected.

145

Bibliography

146

BibliographyResources cited

ADBI.(2016).TheDevelopmentDimensionofE-commerceinAsia:OpportunitiesandChallenges.Policybrief.No.2016-2(June).

Affuso, L. and Hall, G.(2016).Doescompetitionpolicyneedtoadapttothedigitalage?PwCEconomicsinBusiness,November2016.

Akamai. (2016).Akamai’sstateoftheinternetQ22016report.Volume9/Number2.

Akamai. (2016b).Akamai’sstateoftheinternetQ32016report.Volume9/Number3.

Akamai. (2016c).Akamai’sstateoftheinternetQ42016report.Volume9/Number4.

Akamai. (2017).Akamai’sstateoftheinternetQ12017report.Volume10/Number1.

Allen & Overy. (2017).TheNewCompetitionLawinThailandApproved.

Analysis Group.(2016).IsBigDataaTrueSourceofMarketPower?Availableat:http://www.analysisgroup.com/is-big-data-a-true-source-of-market-power/

ASEAN.(2011).ASEANIntellectualPropertyRightsActionPlan2011-2015.

ASEAN.(2013).HandbookonCompetitionPolicyandLawinASEANforBusiness.

ASEAN. (2015).TheASEANICTMasterplan2020.

ASEAN.(2016).ASEANCompetitionActionPlan(ACAP)2016-2025.

ASEAN. (2017).ASEANStrengthensCapacityforCross-BorderEnforcement.

ASEAN Economic Community (AEC).(2015).Blueprint2025.

ASEAN Today. (2016).Theroleofrebatesone-commerce.

ASEANUP.(2017).TopE-commercesitesinMalaysia.

Asiatoday.com.(2016).ThailandDraftsRoadmapFor"DigitalEconomy".Availableat:http://www.asiatoday.com/pressrelease/thailand-drafts-roadmap-digital-economy-0

Associated Chinese Chambers of Commerce and Industry of Malaysia (ACCCIM).(2012).Reportof2012SMEsSurvey.

AT Kearney. (2014).Beauty&theE-commerceBeast.

AT Kearney. (2015).LiftingtheBarrierstoE-CommerceinASEAN.

AT Kearney. (2015b).TheASEANDigitalRevolution.

AT Kearney.(2016).MalaysianE-commerceroadmapoverview.

147

B2bsource.co.uk. (2016).EffectofEcommerceonFashionSupplyChains.Availableat:http://www.b2bsource.co.uk/consultancy/ecommerce-supplychains.html

Bernard, R. and Wedel, P. (2011). Comparative Assessment Study of Patent and Trademark Offices inSoutheastAsia.KenanInstitute.

Bishop, S. and Walker, M. (2010). The Economics of EC Competition Law: Concepts, Application andMeasurement.

Bloomberg. (2011).TheRiseandIngloriousFallofMyspace.BloombergBusinessWeek,June2011.

Brunei Darussalam, Department of Economic Planning and Development. (2015). Competition Order.OrdermadeunderArticle83(3).

Buccirossi, P., Cerbone, R., and Riviera, C. (2014).Optimalpre-mergernotificationthresholds:AcontributiontotheItaliandebate.ItalianAntitrustReview.

BusinessDictionary.com. (2017). Supply chain.Available at: http://www.businessdictionary.com/definition/supply-chain.html.

Cambodia, Ministry of Commerce Department of Legal Affairs. (2016). Draft Law on Competition ofCambodia. Version 5.5.

Cambodia, Ministry of Commerce. (2016).Cambodiaisspeedingtheadoptionofe-commercelaw.Availableat:http://www.moc.gov.kh/en-us/press-center/details/content/211601

Center of the International Cooperation of Computerization. (2016). E-Government Status of Laos.Availableat:http://www.cicc.or.jp/japanese/kouenkai/pdf_ppt/pastfile/h27/151209-5MPT_LAO.pdf

Center on Law and Information Policy, Fordham Law School. (2011).AnAnalysisoftheEconomics/LegalLiterature on the Effects of IP Rights as a Barrier to Entry. Prepared for theWorld Intellectual PropertyOrganization.

Chen, M.K. and Sheldon, M. (2015).DynamicPricinginaLaborMarket:SurgePricingandFlexibleWorkontheUberPlatform.UCLASchoolofManagement.

Competition and Markets Authority (CMA).(2014).QuickGuidetoComplyingwithCompetitionLaw.CMA19.

Competition and Markets Authority (CMA). (2015). Productivity and Competition: a summary of theevidence. CMA45.

Competition and Markets Authority (CMA). (2015b).CompetitionImpactAssessment–Part1:Overview.CMA50.

Competition and Markets Authority (CMA). (2015c).CompetitionImpactAssessment–Part2:Guidelines.CMA50.

Competition and Markets Authority (CMA). (2017).Digitalcomparisontoolsmarketstudy.Updatepaper.

148

Competition Commission of Singapore (CCS). (2014)GroundsofDecision,NotificationforDecisionfortheproposedacquisitionbySeekAsiaInvestmentsPte.Ltd.oftheJobstreetBusinessinSingaporepursuanttosection57oftheCompetitionAct,13November2014,CCS400/004/14

Competition Commission of Singapore (CCS). (2015).Anythingwrongwithaskingforthebestprice?

Competition Commission of Singapore (CCS). (2016). Government and Competition – A toolkit forgovernmentagencies.

Competition Commission of Singapore (CCS). (2016b).CCSGuidelinesontheSection47Prohibition.

Competition Commission of Singapore (CCS).(2015).E-commerceinSingapore–Howitaffectsthenatureofcompetitionandwhatitmeansforcompetitionpolicy.OccasionalPaperSeries.

Competition Commission of Singapore (CCS).(2017).MakingComplaints.Availableat:https://www.ccs.gov.sg/approach-ccs/making-complaints

Coppock and Maclay. (2002).RegionalElectronicCommerceInitiatives:Findingsfromthreecasestudiesonthedevelopmentofregionalelectroniccommerceinitiatives.TheInformationTechnologiesGroup,CenterforInternationalDevelopmentatHarvardUniversity.

Credit Suisse. (2016).TheEffectofE-CommerceontheFashionSupplyChain.Availableat:https://www.credit-suisse.com/au/en/about-us/responsibility/news-stories/articles/news-and-expertise/2016/06/en/the-effect-of-e-commerce-on-the-fashion-supply-chain.html

Currie, D. (2017).DavidCurrieontheroleofcompetitioninstimulatinginnovation.SpeechgivenbytheCMAChairmanattheConcurrencesInnovationEconomicsConference,King’sCollegeLondon.

Cyber Security Agency of Singapore (CSA). (2016).ASEANMemberStatesCall forTighterCybersecurityCoordination in ASEAN. Available at: https://www.csa.gov.sg/news/press-releases/asean-member-states-call-for-tighter-cybersecurity-coordination-in-asean

DBS.(2015).E-commerceinAsia:BracingforDigitalDisruption.SectorBriefing,No.13.

Deloitte. (2015).Digitalbankingforsmallandmedium-sizedenterprises:Improvingaccesstofinancefortheunderserved.

Department of Trade and Industry Philippines. (2017).PhilippineE-commerceroadmap.Diaz,P.S.(2017).PriceParityClauses:HastheCommissionLetSliptheWatchdogsofWar?EuropeanJournalofLegalStudies.

Digital News Asia. (2016).Indonesiaannouncese-commerceroadmapaspartofJokowi’snewesteconomicreform package. Available at: https://www.digitalnewsasia.com/digital-economy/indonesia-announces-e-commerce-roadmap-part-jokowi%E2%80%99s-newest-economic-reform-package#sthash.LlAV64PW.dpuf

DLA Piper. (2015).Stardard-essentialpatentsandtheRANDrequirement:recentdecisionsonreasonablyandnon-discriminatoryroyalties.

DotEcon.(2015).E-commerceanditsimpactoncompetitionpolicyandlawinSingapore.

Ecommerce IQ Asia. (2017).eIQInsights:CanthePhilippinesReachItsEcommercePotential?Marketwatch,Philippines,research&insights.

eMarketer. (2015).WorldwideRetailE-commerceSales:eMarketer’supdatedestimatesandforecastthrough2019.WorldwideE-CommerceReport.

149

Entrepreneur.com. (2016). Small Business Encyclopedia. Available at: https://www.entrepreneur.com/encyclopedia/fulfillment

Euromonitor. (2013).GlobalApparel(Part1):ShiftingDistributionLandscapeandMarketPerformance.

Euromonitor. (2016).InternetRetailing.Availableat:http://www.euromonitor.com/internet-retailing

Evans, D.S. and Schmalensee, R. (2007). Industrial Organization of Marketswith Two-Sided Platforms.CompetitionPolicyInternational,Vol.3,No.1,Spring2007.

Evans, D.S. and Schmalensee, R. (2016).WhyWinner-Takes-All Thinking Doesn’t Apply to the PlatformEconomy.HarvardBusinessReview.

Evans, D. S. (2003).TheAntitrustEconomicsofMulti-SidedPlatformMarkets.YaleJournalonRegulation,Vol20,Iss2,Article4.

Euromonitor. (2017).InternetRetailinginthePhilippines.CountryReport.

European Commission.(1997).EuropeanCommissionNoticeonthedefinitionoftherelevantmarketforthepurposesofCommunitycompetitionlaw.OfficialJournaloftheEuropeanCommunities,97/C372/03.

European Commission.(2004).Councilregulationonthecontrolofconcentrationsbetweenundertakings(theECMergerRegulation).No139/2004.

European Commission. (2009).GuidanceontheCommission’senforcementprioritiesinapplyingArticle82oftheECTreatytoabusiveexclusionaryconductbydominantundertakings.Ref:2009/C45/02.

European Commission.(2010).GuidelinesonVerticalRestraints.CommissionNotice,SEC(2010)411.

European Commission. (2011).GuidelinesontheapplicabilityofArticle101oftheTreatyontheFunctioningoftheEuropeanUniontohorizontalco-operationagreements.

European Commission.(2017)FinalreportontheE-commerceSectorInquiry,SWD(2017)154final.

European Commission. (2015). A Digital Single Market Strategy for Europe. Communication from theCommissiontotheEuropeanParliament,theCouncil,theEuropeanEconomicandSocialCommitteeandtheCommitteeoftheRegions,COM/2015/0192final.

European Commission. (2015b).What'sUpwithMerger Control in theDigital Sector? Lessons from theFacebook/WhatsAppEUmergercase.Competitionmergerbrief,issue1,2015.

European Commission.(2015c).Reportontheprotectionandenforcementofintellectualpropertyrightsinthirdcountries.Commissionstaffworkingdocument,SWD(2015)132final.

European Commission. Intellectualpropertyandcompetition,19thIBACompetitionConference,Florence11 September 2015 - version 01

European Commission. (2016). PreliminaryReportontheE-commerceSector Inquiry.CommissionStaffWorkingDocument,SWD(2016)312final

European Commission. (2016b).EvaluationRoadmap:EvaluationofproceduralandjurisdictionalaspectsofEUmergercontrol.

European Commission.(2016c).Mergers:CommissionapprovesacquisitionofLinkedInbyMicrosoft,subjecttoconditions.EuropeanCommission-Pressrelease.

150

European Commission.(2016d).EUmergercontrolandinnovation.CompetitionPolicyBrief,Issue1,2016.

European Commission.(2017).DigitalSingleMarket.Availableat:https://ec.europa.eu/digital-single-market/en/digital-single-market

European Commission. (2017).FinalreportontheE-commerceSectorInquiry,SWD(2017)154final.

European Commission. (2017c). Commission opens three investigations into suspected anticompetitivepractices in e-commerce.

European Commission. (2017d).FinalreportontheE-commerceSectorInquiry,CommissionStaffWorkingDocumentAccompanyingTheDocument,SWD(2017)154final.

European Commission.(2017e).ReportonthemonitoringexercisecarriedoutintheonlinehotelbookingsectorbyEUcompetitionauthoritiesin2016.

Export.gov. (2016a). Brunei – E-commerce. Available at: https://www.export.gov/article?id=Brunei-eCommerce

Export.gov. (2016b).Lao–E-commerce.Availableat:https://www.export.gov/article?id=Lao-eCommerce

Export.gov. (2016c). Cambodia – E-commerce.Available at: https://www.export.gov/article?id=Cambodia-eCommerce

Export.gov. (2016d). Burma – E-commerce. Available at: https://www.export.gov/article?id=Burma-e-commerce

Ezrachi, A. and Stucke, M.(2016).VirtualCompetition.HarvardUniversityPress,Cambridge.FederalTradeCommission.(2007).StatementofFederalTradeCommissionConcerningGoogle/DoubleClick.FTCFileNo.071-0170.

Ferguson, C.W. and Yen, D.C. (2007).UsingtheCATEmodeltohelpSMEsexpandtoglobale-commercemarkets.InternationalJournalofManagementandEnterpriseDevelopment.Volume4,No.1.

Filistrucchi, L. (2017).MarketDefinition inMulti-SidedMarkets.PaperpreparedfortheOECDHearingon“RethinkingtheUseofTraditionalAntitrustEnforcementToolsinMulti-SidedMarkets”heldinParison22ndJune2017.

Financial Times.(2017).DowandDuPontseektoaddressBrussels’antitrustconcerns.

Foley.(2013).USandEUConvergingOnDominant-Firm-AbuseTheory.

Foreign Trade University. (2017).VietnamE-commercemarket:Overviewandtrends.Availableat:https://www.ccs.gov.sg/~/media/custom/ccs/files/education%20and%20compliance/events/ccs%20iseas%20symposium/vietnam%20case%20study_nguyen%20van%20thoan%20and%20nguyen%20thi%20hong%20van.ashx

Freshfields Bruckhaus Deringer.(2017).BigData.Availableat:http://antitrust.freshfields.com/big-data

Frost & Sullivan. (2016).SoutheastAsia’sE-CommercemarkettosurpassUS$25billionby2020despitemarketchallenges.Availableat:https://ww2.frost.com/news/press-releases/southeast-asias-e-commerce-market-surpass-us25-billion-2020-despite-market-challenges-finds-frost-sullivan/

Frost & Sullivan. (2016b).AnalysisofSoutheastAsianeCommerceMarket.

151

Galexia.(2013).Reviewofe-commercelegislationharmonizationintheAssociationofSoutheastAsianNations (ASEAN). Available at: http://www.galexia.com/public/research/assets/unctad_asean_ecommerce_review_2013/unctad_asean_ecommerce_review_2013-THE-4.html#Heading95

Getting The Deal Through. (2017).Dominance–Singapore.AuthoredbyKin,L.andChew,C.,Drew&NapierLLC.Availableat:https://gettingthedealthrough.com/area/10/jurisdiction/58/dominance-singapore/

GFK Singapore.(2014).April-May2014,TravelBookingsSurvey.Availableat:https://www.gfk.com/sg/news-and-events/press-room/press-releases/Documents/SG%20Travel%20Infographic.pdfGlobalCompetitionReview(GCR).(2017).ASEAN:Againsttheprotectionistwave.GCRliveinSingapore.

Hagiu, A. and Wright, J.(2015).Multi-sidedplatforms.HarvardBusinessSchoolWorkingPaper,15-037.

Hesen, R. and Soven, J. (2006).Definingrelevantproductmarketsinelectronicnetworkantitrustcases.AntitrustLawJournal,Vol.73,No.3(2006),pp.709-738.

Hsu, P.F., Kraemer, K.L. and Dunkle, D. (2006).Determinantsofe-businessuseinUScompanies.InternationalJournalofElectronicCommerce.Volume10.No.4.

Indonesia LawonMonopolisticPracticesandUnfairBusinessCompetition,LawNo.5of1999.

Infocomm Development Authority of Singapore (IDA). (2014). Annual Survey on Infocomm usage inhouseholdsandbyindividualsfor2014.

International Air Transport Association (IATA). (2008). E-ticketing. Available at: http://www.iata.org/whatwedo/stb/Pages/e-ticketing.aspx

International Competition Network. (2011). Unilateral Conduct Workbook. Chapter 3: Assessment ofDominance.

International Financial Law Review. (2016).ThemarchofASEANcompetition.

International Monetary Fund (IMF). (2016).WorldEconomicOutlook–October2016.

International Telecommunication Union. (2016).ICTDevelopmentIndex.InternetWorldStats.(2017).InternetUsageStatistics.Availableat:http://www.internetworldstats.com/stats.htm.

Investopedia.com. (2017). BusinessToBusiness.Available at: http://www.investopedia.com/terms/b/btob.asp

IP Spotlight.(2016).HowlongdoesittakefortheUSPTOtoissueapatentorregisteratrademark?(2016edition). Available at: https://ipspotlight.com/2016/12/31/how-long-does-it-take-for-the-uspto-to-issue-a-patent-or-register-a-trademark-2016-edition/

Javalgi, R. and Ramsey, R. (2000). Strategic issuesof e-commerce as an alternative global distributionsystem.InternationalMarketingReview.Volume18,No.4.

Katz, M.L. and Carl Shapiro, C.(1985).NetworkExternalities,Competition,andCompatibility.TheAmericanEconomicReview,Vol.75,No.3,pp.424-440.

152

Kraemer et al. (2002). Environment andPolicyFactorsShapingE-commerceDiffusion:ACross-CountryComparison.AssociationforInformationSystems.

Lee, I. (2016).EncyclopediaofE-CommerceDevelopment,Implementation,andManagement.Volume1.

Lexis PSL.(2017).Abusingadominantposition–overview.

Lindsay, M.(2017).OverviewofstateRPM.TheAntitrustSource.

Malaysia CompetitionAct2010(Act712)

Malaysian Digital Economy Corporation. (2016).Malaysia’sNationaleCommerceStrategicRoadmap.

Mikians, J., Gyarmati, L., Erramilli, V., Laoutaris, N. (2012).DetectingpriceandsearchdiscriminationontheInternet.

Ministry Of International Trade & Industry, Malaysia. (2017).E-commerceinMalaysia.Availableat:https://www.ccs.gov.sg/~/media/custom/ccs/files/education%20and%20compliance/events/ccs%20iseas%20symposium/malaysia%20country%20study.ashx

Mondaq.com. (2015). Common issues of trademark infringement in e-commerce and enforcement. Available at: http://www.mondaq.com/china/x/365690/Trademark/Common+issues+of+trademark+in+fringement +in+ecommerce+and+enforcement

Monetary Authority of Singapore. (2016). MAS Proposes New Regulatory Framework and GovernanceModelforPayments.Availableat:http://www.mas.gov.sg/News-and-Publications/Media-Releases/2016/MAS-Proposes-New-Regulatory-Framework-and-Governance-Model-for-Payments.aspx

Myanmar, Ministry of Commerce, Department of Trade. (2015). TheCompetition Law.The PyidaungsuHluttawLawNo.9.

NATO Cooperative Cyber Defence Centre Of Excellence (CCDCOE). (2017).ASEANtoFocusonCybersecurityCapacity – and Confidence-Building in 2017. Available at: https://ccdcoe.org/asean-focus-cybersecurity-capacity-and-confidence-building-2017.html

nchannel.com.(2016).Multichannelinsightsblog.Availableat:https://www.nchannel.com/blog/efulfillment-services/

News.com.au. (2016).FlightCentretoexpandonlinepresencewhile rollingoutflagshipstores.Availableat: http://www.news.com.au/finance/business/travel/flight-centre-posts-245-million-loss-amid-price-war/news-story/3f5efc69f85a0b5e1ae7a573aea8d640

Nielsen. (2014).ASEAN2015:SeeingAroundTheCornerInANewAsianLandscape.

Nomura. (2016).Aseanlogistics:Deliveringthelastmile.GlobalMarketsResearch.

Numbeo.com. (2017). PricesbyCountryof Internet.Availableat:https://www.numbeo.com/cost-of-living/prices_by_country.jsp?displayCurrency=USD&itemId=33

OECD. (1993).GlossaryofIndustrialOrganisationEconomicsandCompetitionLaw.

OECD. (1997).Dismantlingthebarrierstoglobalelectroniccommerce.

OECD. (2002).LoyaltyandFidelityDiscountsandRebates.

OECD. (2007).CompetitionandBarrierstoEntry.PolicyBrief,January2007.

153

OECD.(2007b).CompendiumofOECDworkonintellectualproperty(IP).

OECD.(2011).GlossaryofStatisticalTerms.Availableat:https://stats.oecd.org/glossary/detail.asp?ID=4721

OECD. (2012).TheDigitalEconomy.OECDHearings,DAF/COMP(2012)22.

OECD. (2013).DefinitionofTransactionforthePurposeofMergerControlReview.PolicyRoundtables,DAF/COMP(2013)25.

OECD.(2015).CompetitionAssessmentToolkit-VolumeIII:OperationalManual.Version3.0.

OECD.(2015b).Enquiriesintointellectualproperty’seconomicimpact.Chapter1.SynthesisReport.

OECD.(2016).PriceDiscrimination.BackgroundnotebytheSecretariat.

OECD.(2016b).BigData:BringingCompetitionPolicytotheDigitalEra.BackgroundnotebytheSecretariat.

Open Signal.(2017).GlobalStateofMobileNetworks(February2017).Availableat:https://opensignal.com/reports/2017/02/global-state-of-the-mobile-network

Oxera and Accent. (2016). Vertical restraints: new evidence from a business survey. Prepared for theCompetitionandMarketsAuthority.

Philippine PhilippineCompetitionAct[RepublicActNo.10667]

Philippine Competition Commission. (2017). Competition and E-commerce in the Philippines. Available at: https://www.ccs.gov.sg/~/media/custom/ccs/files/education%20and%20compliance/events/ccs%20iseas%20symposium/philippines%20country%20study_shanti%20aubren%20prado%20and%20meg%20reganon.ashx

Philstar Global.(2016).PhilippineInternetisfast:Wejustcannotaffordit.Availableat:http://www.philstar.com/technology/2016/09/26/1627308/philippine-internet-fast-we-just-cannot-afford-it

Qandme.net. (2016). Vietnam E-Commerce Market Survey. Available at: https://www.slideshare.net/CVEN2016/vietnam-ecommerce-report-2016

Rajah & Tann. (2017).ASEANCountry Studies II: Singapore.Available at: https://www.ccs.gov.sg/~/media/custom/ccs/files/education%20and%20compliance/events/ccs%20iseas%20symposium/singapore%20country%20study_tanya%20tang.ashx.

Rodyk & Davidson LLP. (2013).CompetitionLawsInASEAN–OverviewoftheMainProhibitions.

Rosch, J.T.(2012).Developmentsinthelawofverticalrestraints:2012.PractisingLawInstitute.NewYork.

Sengupta, A. and Wiggins, S. (2006).AirlinePricing,PriceDispersion,andTicketCharacteristicsOnandOfftheInternet.

Sidak, J.G. and Teece, D. J. (2009).DynamicCompetitioninAntitrustLaw.JournalofCompetitionLaw&Economics,5(4),581–631.

154

Singapore CompetitionAct(Cap.50B),RevisedEdition2006

Singapore.(2016).DPMTharmanShanmugaratnamattheopeningofLFLogisticsRegionalDistribution Centre on 26 April 2016. Available at: http://www.pmo.gov.sg/newsroom/dpm-tharman-shanmugaratnam-opening-lf-logistics-regional-distribution-centre-26-april

Singapore Post.(2014).TheChangingFaceofEcommerceandLogisticsinAsiaPacific.

Singapore Post. (2015). SingPost to develop Singapore’s first shopping mall with eCommerce logisticsservices.Availableat:http://www.singpost.com/about-us/news-releases/singpost-develop-singapores-first-shopping-mall-ecommerce-logistics-services

Singapore Post. (2016). SingPost introduces Singapore’s first island wide open parcel locker service. Availableat:http://www.singpost.com/about-us/news-releases/singpost-introduces-singapore%E2%80%99s-first-islandwide-open-parcel-locker-service

Singh, S., Kumar, A., and Agarwal, G.(2011).NextGenerationBusinessModelsinECommerce.InternationalJournalofAdvancedResearchinComputerScience.Volume2,No.2.

Slaughter and May. (2016).TheEUcompetitionrulesonverticalagreements.Aguidetotheassessmentof vertical agreements (including the European Commission’s block exemption regulations on verticalagreementsandmotorvehicledistribution).

Sophos. (2013).SecurityThreatReport2013:NewPlatformsandChangingThreats.

SPRING Singapore. (2016). Retail Industry Transformation Map to Drive E-Commerce & Omni-ChannelFormatstoEnhanceGrowthandCompetitiveness.Availableat:https://www.spring.gov.sg/NewsEvents/PR/Pages/Retail-Industry-Transformation-Map-to-Drive-E-Commerce--Omni-Channel-Formats-to-Enhance-Growth-and-Competitiveness-20160915.aspx

SPRING Singapore. (2016b). Supporting Retailers in their Adoption of technology to boost SMEs’ Productivity and Growth. Available at: https://www.spring.gov.sg/NewsEvents/PR/Documents/2016Sep15_Annex_Initiatives%20under%20the%20Retail%20ITM.pdf

Statcounter. (2016). Mobile and tablet internet usage exceeds desktop for first time worldwide. Available at: http://gs.statcounter.com/press/mobile-and-tablet-internet-usage-exceeds-desktop-for-first-time-worldwide

Statista.com. (2017a).E-commerce inThailand.Availableat:https://www.statista.com/outlook/243/126/e-commerce/thailand#market-revenue

Statista.com.(2017b).E-commerceinSingapore.Availableat:https://www.statista.com/outlook/243/124/e-commerce/singapore#

Statista.com.(2017c).E-commerceinIndonesia.Availableat:https://www.statista.com/outlook/243/120/e-commerce/indonesia#market-revenue

Statista.com.(2017d).E-commerceinthePhilippines.Availableat:https://www.statista.com/outlook/243/123/e-commerce/philippines

Statista.com. (2017e). E-commerce inMalaysia.Available at: https://www.statista.com/outlook/243/122/e-commerce/malaysia

Statista.com. (2017f). E-commerce inVietnam. Available at: https://www.statista.com/outlook/243/127/e-commerce/vietnam

Statista.com.(2017g).ShareofpopulationinIndonesiathatuseasmartphonefrom2015to2021.Availableat:https://www.statista.com/statistics/321485/smartphone-user-penetration-in-indonesia/

155

Statista.com.(2017h).ShareofpopulationinMalaysiathatuseasmartphonefrom2015to2021.Availableat:https://www.statista.com/statistics/321485/smartphone-user-penetration-in-Malaysia/

Statista.com. (2017i). Share of population inThe Philippines that use a smartphone from 2015 to 2021.Availableat:https://www.statista.com/statistics/321485/smartphone-user-penetration-in-philippines/

Statista.com.(2017j).ShareofpopulationinSingaporethatuseasmartphonefrom2015to2021.Availableat:https://www.statista.com/statistics/321485/smartphone-user-penetration-in-singapore/

Statista.com.(2017k).ShareofpopulationinThailandthatuseasmartphonefrom2015to2021.Availableat:https://www.statista.com/statistics/321485/smartphone-user-penetration-in-thailand/

Statista.com.(2017l).ShareofpopulationinVietnamthatuseasmartphonefrom2015to2021.Availableat:https://www.statista.com/statistics/321485/smartphone-user-penetration-in-vietnam/

Statista.com.(2017m).NumberofinternetusersinIndonesiafrom2015to2021.Availableat:https://www.statista.com/statistics/254456/number-of-internet-users-in-indonesia/

Statista.com.(2017n).TotalpopulationofIndonesia.Availableat:https://www.statista.com/statistics/378558/total-population-of-indonesia/

Statista.com.(2017o).TotalpopulationofMalaysia.Availableat:https://www.statista.com/statistics/378558/total-population-of-malaysia/

Statista.com.(2017p).TotalpopulationofPhilippines.Availableat:https://www.statista.com/statistics/378558/total-population-of-philippines/

Statista.com.(2017q).TotalpopulationofSingapore.Availableat:https://www.statista.com/statistics/378558/total-population-of-singapore/

Statista.com.(2017r).TotalpopulationofThailand.Availableat:https://www.statista.com/statistics/378558/total-population-of-thailand/

Statista.com.(2017s).TotalpopulationofVietnam.Availableat:https://www.statista.com/statistics/378558/total-population-of-vietnam/

Statista.com.(2017t).Retaile-commercesalesinIndonesiafrom2015to2021(inbillionU.S.dollars.Availableat:https://www.statista.com/statistics/280925/b2c-e-commerce-sales-in-indonesia

Statista (2017u).AsiaPacificregion:Totalpopulation in2015,bycountry(inmillions).Availableat:https://www.statista.com/statistics/632565/asia-pacific-total-population-by-country/

Statista(2017v).AsiaPacific:Numberofinternetusersin2015,bycountry(perhundredpeople).Availableat:https://www.statista.com/statistics/632510/asia-pacific-number-of-internet-users-by-country/

Thailand TradeCompetitionAct,B.E.2542 (1999).TheTradeCompetitionActB.E.2560 (2017),whichwillreplacetheTradeCompetitionActB.E.2542(1999),willcomeintoeffecton5October2017.

Thatcher, S.M.B., Foster, W. and Zhu, L. (2006). B2B e-commerce adoption decisions in Taiwan: theinteraction of cultural and other institutional factors. Electronic Commerce Research and Applications.Volume5,No.2.

The Business Times. (2017).SATSunveilsS$21me-commerceairhubatChangi.Availableat:http://www.businesstimes.com.sg/companies-markets/sats-unveils-s21m-e-commerce-airhub-at-changi

The Economist.(2006).China’spiedpiper.21/09/2006.

Thepaypers.com.(2017).EcommercefraudandonlinefraudpreventioninThailand.Availableat:http://www.thepaypers.com/online-fraud-prevention/thailand/27

156

Thomson Reuters. (2007).PracticalLaw.Theinterfacebetweenanti-trustandintellectualpropertylawinEurope.

Translate Media. (2017). How the Cosmetics Industry Embraced Technology. Available at: https://www.translatemedia.com/translation-blog/cosmetics-industry-embraced-technology/

United Nations Economic and Social Commission for Asia and the Pacific (UNESCAP). (2013).AnIn-DepthStudyofBroadbandInfrastructureintheASEANRegion.

United Nations Conference on Trade and Development (UNCTAD). (2013).Reviewofe-commercelegislationharmonizationintheAssociationofSoutheastAsianNations.

United Nations Conference on Trade and Development (UNCTAD). (2015). Unlocking the Potential ofE-commerceforDevelopingCountries.InformationEconomyReport2015.

United Nations Conference on Trade and Development (UNCTAD).(2016).Non-tariffmeasuresinASEAN.

United Nations Conference on Trade and Development (UNCTAD).(2016b).E-commerceandDevelopment:KeyTrendsandIssues.WorkshoponE-commerce,DevelopmentandSMEs.

United Nations Conference on Trade and Development (UNCTAD). (2016c).Newinitiativetohelpdevelopingcountriesgrasp$22trillione-commerceopportunity.Availableat:http://unctad.org/en/pages/newsdetails.aspx?OriginalVersionID=1281

United Nations Conference on Trade and Development (UNCTAD). (2016d). In Search of Cross-borderE-commerceTradeData.Technicalnote.No.6.

United States Department of Justice.(2004).Internationalantitrustandintellectualproperty:challengesontheroadtoconvergence.

Vela Asia Online Shopper. (2013).Onlinesurveyofdataprotection.

Verlinda, J. and Lane, L.(2004).TheEffectoftheInternetonPricingintheAirlineIndustry.

Vietnam CompetitionLawNo.27/2004/QH11

Vietnam E-Commerce Association. (2017).Availableat:http://www.vecom.vn/en/

Vietnam Net. (2016).VNGovernmentsetstargetsfore-commercein2016-2020.Availableat:http://english.vietnamnet.vn/fms/business/162619/vn-government-sets-targets-for-e-commerce-in-2016-2020.html

Whish, R. and Bailey, D.(2012).CompetitionLaw.Seventhedition.

Whish, R. and Bailey, D.(2015).CompetitionLaw.Eighthedition.

Wired.(2016).AllhailGrab-thetaxicompanythatfaceddownUberinAsia.Availableat:http://www.wired.co.uk/article/grab-taxi-company-asia

Wong Partnership.(2016).CompetitionLawinASEAN:Wherearewenow,andwhereareweheaded?

World Applied Programming. (2011). Taking a look at different types of E-commerce. WAP journal. Volume1,No.2.

157

World Bank.(2016).HowtoscaleupfinancialinclusioninASEANcountries.Availableat:http://blogs.worldbank.org/eastasiapacific/how-to-scale-up-financial-inclusion-in-asean-countries

World Bank. (2017).InternetUsersasapercentageofpopulation.Availableat:https://www.google.com.au/publicdata/explore?ds=d5bncppjof8f9_&met_y=it_net_user_p2&idim=country:MMR:LAO:KHM&hl=en&dl=en

Worlddata.info.(2015).Averageincomearoundtheworld.Availableat:https://www.worlddata.info/average-income.php

World Economic Forum. (2017).TheGlobalCompetitivenessReport2016/17.InsightReport.

World Intellectual Property Organization (WIPO). (2016).TimelinessofROtotransmitcopiesofPCTfilingstoIB.WIPOIPStatisticsDataCenter.Availableat:https://www3.wipo.int/ipstats/

World Intellectual Property Organization (WIPO). (2017).WhatisIntellectualProperty?Availableat:http://www.wipo.int/about-ip/en/

World Intellectual Property Organisation (WIPO).(2017b).IPandCompetitionPolicy.Availableat:http://www.wipo.int/ip-competition/en/

Yusof Ishak Institute. (2017).Indonesia’sE-commerce.

Competition cases reviewed

E-book MFNs and related matters (Amazon) –EuropeanCommission-40153-2017

Amazon Japan –JFTC-2017

American Express – UnitedStatesofAppealsfortheSecondCircuit–No.15-1672-2016

Apple/book publishers –EC–COMP/39.847–E-BOOKS–2012

Asics –Bundeskartellamt,Germany-ASICS(B2-98/11)–2015

BMW –CMA(https://www.gov.uk/government/news/bmw-changes-policy-on-car-comparison-sites-following-cma-action)–2017

Bosch –Bundeskartellamt,Germany-BoschSiemensHausgeräte;B7-11/13–2013

Catering equipment and bathroom fittings –CMA–CE/9856-14;CE/9857-14–2016

Coty –ECJ-CotyGermanyGmbHvParfümerieAkzenteGmbH,CaseC-230/16–Ongoing

Dow/Dupont – EC - M.7932 - 2017

Eturas –CourtofJustice(CJEU)-Eturas(CaseC-74/14)–2016

Facebook/WhatsApp –EC-COMP/M.7217–2014

Flight Center –ACCC-CaseB15/2016FlightCenter-2016

Gardena –FederalCartelOffice,Germany-B5-144/13–2013

Garuda/Abacus –KPPU-SupremeCourtDecisionNo.01K/KPPU/2004-2005

Google Adsense – EC - 40411

158

Google Android – EuropeanCommission-40099-ongoing

Google/DoubleClick – EC-COMP/M.4731-2008

Google Maps –ParisCommercialCourt/ParisCourtofAppeal-12/02931-2015

Google comparison shopping –EC-39740GoogleComparisonShopping–Ongoing

Google/Streetmap –HighCourt(UK)-Streetmap.EULimitedvGoogleInc.,GoogleIrelandLimitedandGoogleUKLimited[2016]EWHC253(Ch)–2016

Hotel accommodation – EC - AT.40308 - 2017

iFAST –CCS-CCS500/003/13-2016

Lego –Bundeskartellamt,Germany(https://www.bundeskartellamt.de/SharedDocs/Meldung/EN/Pressemitteilungen/2016/18_07_2016_Lego.html)–2016

Life insurance –CCS-CCS500/003/13–2016

Mastercard –EC-C-382/12P-2014

Microsoft –EC–T-167/08-2012

Microsoft/LinkedIn –EC-M.8124Microsoft/LinkedIn–2016

Mobility scooters –OFT/CMA-Mobilityscooters,CE/9578-12–2014

MyEG –MyCC-MyE.G.ServicesBerhad–2016

Online food delivery –CCS-https://www.ccs.gov.sg/media-and-publications/media-releases/investigation-of-online-food-delivery-industry-2016

Online hotel booking –CMA,Bundeskartellamt-CE/9320-10(CMA),B9-121/13(Bundeskartellamt)–2015

Online retail sector –OFT-CE/9692/12

Peugeot –EC-37275SEPetautres/AutomobilesPeugeotSA–2005

Pierre Fabre –ECJ–PierreFabreDermo-CosmétiqueSASvPrésidentdel’AutoritédelaconcurrenceandMinistredel’Économie,del’Industrieetdel’Emploi.(C-439/09)-2011

Ping Europe Limited (Ping) –CMA(https://www.gov.uk/government/news/cma-alleges-breach-of-competition-law-by-ping)–ongoing

SISTIC –CCS-CCS/600/008/07–2010

Tooltechnic - ACCC - A91433 - 2014

Trod/GB Eye –CMA-Onlinesalesofpostersandframes(50223)–2016

Yamaha –EC-37975PO/YAMAHA–2003

Visa –CCS-CCS400/001/06–2013

159

Annexes

160

Annex1:TechnicalInformationA1.1. Multi-sided markets

Definitionofamulti-sidedmarket

A1.1.1. Atwo-ormulti-sidedmarket isoneinwhichdistinctbutrelatedcustomergroupsareconnectedbyacommonplatform.EvansandSchmalensee(2007)372explainedhow“thecorebusinessofthetwo-sidedplatformistoprovideacommon(realorvirtual)meetingplaceandtofacilitateinteractionsbetweenmembersofthetwodistinctcustomergroups.”Forexample,anewspaperconnectsreadersandadvertisers;ahotelbookingwebsiteconnectshotelswithtravellers;andasatellitetelevisioncompanyconnectsviewerswithadvertisersandTVchannels.

Externalities between sides

A1.1.2. Eachsideofamulti-sidedmarkettypicallygivesrisetoexternalitieswhich impacttheother,andthiscanaffectthewayinwhichfirmssettheirpricingstructures.373Considering,fordemonstrativepurposes,anofflinemarket,suchasnewspapers:inordertoattractadvertisers,manynewspapersaresoldbelowcosttoreaders,orevengivenawayforfree.Thisisasaresultofanexternalityimposedbyonesideofamarket(readers)onanotherside(advertisers) i.e.themorereadersthereare,thehigherthevaluetoadvertisers,and,therefore,themoretheyarewillingtopaytoadvertiseinthatnewspaper.Similardynamicsareinplayinonlinemarketssuchassocialmediaandonlinesearchorshopping.

A1.1.3. Externalitiescanbepositiveornegative.Intheexampleabove,ahighernumberofreadersgeneratesapositiveexternalityforadvertisers.Conversely,asthenumberofadvertsinthenewspaperincreases,anegativeexternalityforthereaderemerges,asthevaluethereaderderivesfromthenewspaperfalls.Therefore,newspapercompanieswillneedtofindtherightbalancebetweentheneedtoincreasedemandbyadvertisers,aswellasbyreaders,insettingtheirprices.Iftheyweretochargeahigherpricefornewspapers,thenumberofreaderswoulddecrease,thusreducingthevaluetoadvertisersandtheirdemandforadverts.Conversely, iftheysetaloworzeropricefornewspapers,theycanincreasedemandbyreaders,therebyincreasingdemandbyadvertisers.Newspapersmaybeabletorecoupthecostsfacedonbothsidesofthemarketbychargingadvertisersahigherprice.

A1.1.4. In assessing competition cases inmulti-sidedmarkets, it is therefore important that authoritiesconsidertherelationshipbetweeneachsideofthemarket;forexample,whenassessingthemarketpowerheldbyaplatform.Asdemonstratedbytheexampleabove,ifthenewspaperissoldatapricebelowcosttoreaders,thisdoesnotnecessarilymeanthatthenewspaperinquestionispursuingapredatorypricingstrategy.Thepricechargedforadverts,whichcoversthecostsincurredinservingbothsidesofthemarket,wouldcompensatethenewspaper.

A1.1.5. Theexternalitiesdiscussedherearealsodefinedasnetworkeffects.Networkeffectsarisewhen“theutilitythatagivenuserderivesfromthegooddependsuponthenumberofotheruserswhoareinthesame"network"as isheorshe”.374Forexample,networkeffectsexist insocialmediamarketswherethevalueone individualplacesonaplatform increasesasmoreofthat individual’sfriendsalsousethesameplatform.Theself-reinforcingnatureofnetworkeffectscanenableaplatformtogrowrapidly,andalsoimposeabarriertoentryandexpansiononnewentrantsandsmallerplayers(asdiscussedinSection4).Inassessmentsofmulti-sidedmarkets,competitionauthoritiesshouldthereforecarefullyexamineifnetworkeffectsarepresent.Forexample,asdiscussedinSection10,theremaybeconcernsthatfollowingamergeroftwocompetingplatforms(andthecombiningoftheirnetworks)a‘tippingpoint’maybereached,and,asaresult,smallerfirmsarenolongerabletocompete.

372 Evans, D. and Schmalensee, R. (2007); page 151.

373 “Externalities refers to situations when the effect of production or consumption of goods and services imposes costs or benefits on others which are not reflected in the prices charged for the goods and services being provided.” (OECD, 1993).

374 Katz, M. and Shapiro, C. (1985); page 424.

161

Importance of single- and multi-homing

A1.1.6. Insuchassessments,animportantfactorcompetitionauthoritiesshouldconsideristheextenttowhichusersmulti-orsingle-home.Auserwhosingle-homeswillonlyuseoneplatforminagivenmarket;forinstance,someonewhousesonlythesocialmediawebsiteFacebook.Contrastingly,anindividualwhomulti-homes uses a number of competing platforms; for example, someonewhousesbothFacebookandTwitterwouldbesaidtomulti-homeinthesocialmediamarket.Arelevantexamplewouldbe intheassessmentofaproposedmergerbetweentwoplatforms inamarket.Ifuserstypicallymulti-home,themergerwouldbelessofaconcerntoauthoritiesthanamergerbetweentwoplatformsinamarketwhereuserstypicallysingle-home,ceterisparibus.Foratippingpointtooccur, itmaybesufficientforonlyonesideofamulti-sidedmarkettosingle-home.Asaresult,whenassessingthelikelihoodofatippingpointoccurring,allsidesofthemarketshouldbeconsidered.

A1.2. Block exemptions and hardcore restrictions

Block exemptions

A1.2.1. Ablockexemptionallowscertainpracticestobeexemptfromtheapplicationofcompetitionlaw,wherepro-competitivebenefitsaredeemedtosignificantlyoutweighanyanti-competitiveeffects,orwhereacompanyholdsaverysmallshareofthemarket,makinganti-competitiveeffectsunlikelytoarise.

A1.2.2. Blockexemptionsmayapplytocertainverticalandhorizontalagreements,suchastechnologytransferagreements,orresearchanddevelopmentagreements,inordertopromotesustainablecompetitionwithincertain industries. Insome jurisdictions, suchastheEU,manyvertical restraintsfallunderblockexemptionregulation,aslongastheparties’marketsharesarebelowacertainthreshold.

Hardcore restrictions

A1.2.3. However,competitionauthoritiesmayestablishalistofhardcorerestrictionsthatfalloutsideofanexemption. For example, the European Commission regardsminimum andfixed resale prices ashardcorerestrictions(seeSection7.4onResalePriceMaintenance),meaningtheyareexcludedfromthescopeofblockexemptionregulations.

Example of block exemption on vertical agreements in Europe

A1.2.4. Economictheorysuggeststhat“unlessfirmspossessandexercisemarketpower,theyareunableto affect competition adversely”.375 Consequently, in Europe, firms that have entered into verticalagreementsaregrantedanautomaticexemptionfromtheapplicationofArticle101oftheTFEUforcertainclausesiftheyhaveamarketshareoflessthan30%,asitisunlikelytheseagreementswillgiverisetoanti-competitiveoutcomesgiventhepositionofthefirminthemarket.If,however,theverticalagreementinquestioncontainsahardcorerestriction,thentheinfringingfirmwillbesubjectedtotheapplicationofcompetitionlaw.


162

Annex2:Government initiatives on E-commerce

A2.1. Brunei Darussalam

A2.1.1. In2015,theGovernmentofBruneiDarussalamannouncedtheDigitalGovernmentStrategy,withthemission“toleadthedigitaltransformationandmakegovernmentservicesimpler,fasterandmoreaccessible”.Sixfocusareashavebeenidentifiedtoachievethismission.Theseare:

a. Service innovation:Withanincreasinglysophisticatedanddynamicsociety,governmentagencies mustdevelopnewandinnovativewaystodeliverservicestocitizensandbusinesseswithgreater transparencyandaccountability;

b. Security:Followingonfromthepreviousstrategicplan2009-2014,securitywillremainakeyfocus area.Thegovernmentneedstomaintainsituationalawarenessofitsdigitalassetsandenvironment at all times. Adequate measures will be taken to minimise risks and increase capabilities to respondtocyber-incidentseffectively;

c. Capability & Mind-set: People will always remain the key that will lead to the successful implementation of any technology. It is essential to foster a forward-thinking mindset and collaborativeculture.Thiswillhelptoincreasethespeedofadoptingnewsystems,rateofutilising systemsandproficiencyofgovernmentofficials;

d. Enterprise Information Management: Withtoday’sknowledgedriveneconomy,informationisa fundamentalbuildingblockthatcanadvanceanation.Itiscriticalthatthegovernmentmanage theexplosivegrowthofdatabystructuring,describingandgoverninginformationassetsthatcan thenbeusedtogenerateinsightsthataiddecision-making;

e. Optimisation: To keep pace with the rapid development of technology, the government has beenimplementingvariousITsystemsandplatforms.Movingforward,thegovernmentneedsto optimise the use of these digital assets to ensure effectiveness, minimise redundancy and maximisevalueformoney;and

f. Collaboration & Integration: Government agencies are required to work together to face an increasinglycomplexenvironment.This requiresaWhole-Of-Governmentapproachtoenhance thecollaborationandintegrationofgovernmentbusinessprocesses.

A2.1.2. SixprogrammeshavebeenidentifiedtorealisethevisionandtoachievetheBruneiDigitalGovernment Strategy2015-2020:

a. Advancing digital services:Ensuringkeyservicesareaccessibleanytimeanywhere,andmanaging theGovernmentRevenueCollectiondigitally;

b. Implementing Universal Access for Government Systems:OneIDforcitizens,businesses,and servicesthatsupportsoneID;

c. Strengthening Securities:Anintegratedapproachbyallsectorstowardnationalcybersecurity;

d. Enhancing Stakeholder Engagement:Creatinganewplatformforstakeholderengagement,and agovernanceframeworkformanagingstakeholderengagement;

e. Optimising Digital Assets:Maximisingthevalueofexistingdigitalassets;and

f. Developing Enterprise Information Management Capability:Processes,toolsandcapabilities forEnterpriseInformationManagement.

163

A2.2. Cambodia

A2.2.1. TheGovernment of Cambodia is close to approving an E-commerce law. InNovember 2016, theCambodianMinistryofCommerceannouncedthat the90articles longAct isbeingfinalised.TheobjectiveoftheE-governmentpolicyoftheRoyalGovernmentis“toconnectthepublicadministrationinordertoprovideefficientpublicservicestothecitizens”,andthelawwillbeconsistentwiththis.TheMinistryofCommercewebsitestates:

“E-CommerceLawwillcreateanewbusinessenvironment,calledCyberspaceandallowyouthsdoingtradewithoutbordersatanytimewithmillionconsumersaroundtheworldtobringmorerevenuesforthecompanyandthecountry.Moreover,theE-Commercewillhelppromotethecountry'sreputationontheinternationalstagebecausethisbusinesswillfacilitatetheintegrationofCambodia’sgoodsintotheregionalandglobalproductionnetwork."376

A2.3. Indonesia

A2.3.1. In November 2016, Indonesia announced its 14th economic reform package, which includes theE-commerceroadmap.TheroadmapinvolveseightfocusareasaimingtosupportthedevelopmentofE-commerceinIndonesia.Theseare:377

a. Funding: Includingmicrocreditprogrammestocoverplatformandappdevelopers, grantsfor businessincubatorsandstart-upmentorshipprogrammes;

b. Taxation: Includingloweringthetaxrateforlocalinvestorsinvestinginstart-upsandaneasein thetaxationproceduresforE-commerceventureswithatotalturnoverofRp4.8billion(US$357,191) andbelowperyear,therebylevellingtheplayingfieldintaxationforallE-commerceplayers;

c. Consumer protection: Involvingregulatingelectronictransactionstoallowfortransactionsand governmentspendingthroughE-commerce;

d. Education and human resources:ThegovernmentwillstartanationalE-commerceawareness campaignalongwithanationalincubationprogramme,andE-commerceeducationprogramme forallstakeholders;

e. Logistics:IncludingallowingE-commerceplayerstoleverageontheNationalLogisticsSystem. In2001,theblueprintwassetupforthistobecreatedfromscratch(Sislognas),however,despite this,developmenthasseenextremelyslowprogress.Ofthe30firstprogrammes listed inthe annex to the Sislognas Perpres, which are intended to improve Indonesia’s logistics network, onlyahandfulareinoperation;378

f. Strengthen communications infrastructure:Throughnationalbroadbanddevelopment;

g. Cyber security:IncludingsettingupanationalsurveillanceandE-commercemonitoringsystem; and

h. Form an operating management structure:tomanage,monitor,andevaluatetheimplementation oftheE-commerceroadmap.

376 Cambodia Ministry of Commerce (2016).

377 Digital News Asia (2016).

378 Indonesia Infrastructure Initiative (2015).

164

379 Center of the International Cooperation for Computerization.

380 Malaysian Digital Economy Corporation, (2016).

A2.4. Lao PDR

A2.4.1. LaoPDRhasanE-governmentDevelopPlan(2013-2020).Therearethreekeystagestotheplan.Theseare:

a. Presence Stage (2013-2015):FocusonG2Gapplications(maintainandrebuildtheseapplications, which was established under the E-government project phase I: 2006-2012, mainly in some governmentofficesinVientianeandprovincialgovernors.Somedistrictandvillageadministration officeswillbesettingupanddistributingITequipmentinphaseII);

b. Interaction Stage (2016-2018): This involves integrating the government data into one single service,andinitiatingG2Bserviceapplications;and

c. Transaction Stage (2019-2020): Thisincludesfullycomputerisingtheadministrationsystemand e-Service,especiallyE-commercebygovernmentofficers.ItalsoincludesinitiatingG2Cservice applications.379

A2.5. Malaysia

A2.5.1. An important initiative istheNationalE-commerceStrategicRoadmap,380whichwassetupbytheMalaysianGovernmentandlaunchedinOctober2016.Thisroadmaphassixkeyareas,listedbelow:

a. Accelerate seller adoption of E-commerce;

b. Increase adoption of eProcurement by businesses;

c. Lift non-tariff barriers.ThisincludesincreasingthelevelofmaturityinthedomesticE-fulfilment sector (which will be done by providing economic incentives and preferential schemes for the online environment, offering companies to convert warehouses into fulfilment centres and increasing ICT spending, with accelerated capital allowances), an increase in theadoptionofE-payments(whichwillbedonebyofferingmoreinnovativepaymentproducts, improving service levels, and encouraging adoption and use. Initiatives such as enhancing the infrastructuretokeeppacewithinnovationandmeetinguser’sneeds,andputtingcapsonfeesfor using credit cards), and augmenting and increasingmass awareness of consumer protections (which involves rolling out advocacy programs to protect consumers’welfare on E-commerce platformsandtoincreaseawarenessofconsumers’rightsandredresschannels);

d. Realign existing economic incentives;

e. Make strategic investments in select E-commerce player(s), byprovidingeconomicincentives andpreferentialschemeswhichwillbealignedfortheonlineenvironment.Thiswillincludeoffering companiesincentivestoconvertwarehousesintofulfilmentcentresandtoincreaseICTspending; and

f. Promote national brand to boost cross-border E-commerce. In Malaysia, MATRADE has launchedanationwideadvocacyprogramviaeTRADE,agovernmentinitiativetoaccelerateexports byencouragingSMEstoparticipateinleadinginternationalonlinemarketplaces.Theobjectivesof the program are to widen market access, establish cooperative relationships with already- established onlinemarketplaces, and reduce the cost of exporting products. Ongoing eTRADE initiatives includeassessingpotentialonlinemarketplaces,establishingstrategiccollaborations, compilingalistofE-fulfilmentproviders,promotingMalaysianproductstopotentialonlinebuyers, creating an international sourcing program for buyers, raising awareness about E-commerce among Malaysian exporters (SMEs and non-SMEs), matching Malaysian companies with e-marketplaces,andmonitoringtheimpactofE-commerceadoptionforstakeholders’reporting.

165

A2.5.2.CriticalsuccessfactorswhichwillinfluencethelikelytrajectoryofE-commercewithinMalaysiainclude:

a. Favourable demographic and economic trends: HealthyGDP growth, a high level of internet usageandatechnologicallysavvypopulation;

b. The current level of E-commerce infrastructure:Twothirdsofthepopulationusetheinternet, four-fifths ofwhomhave purchased online. Credit card usage is 12%,381 the second highest in ASEAN,andtherearealreadylargeE-commerceplatformsinexistence;and

c. Government interventions which are required to boost E-commerce development: The ElectronicCommerceActandthePersonalDataProtectionActshavebeenpassedrecentlywith theaimofsupportingthedevelopmentoftheE-commercemarket.

A2.5.3.TheMalaysiaDigital EconomyCorporation roadmap identifies thepotential for these initiatives toalmostdoublethegrowthrateofE-commerceintheregion,fromaCAGRofjustunder11%inthebusinessasusualscenario,tojustunder21%withtheseinterventions.382

A2.5.4.AlthoughtheNationalE-commerceStrategicRoadmapistheprinciplepolicytopromoteE-commerce,there are more than 40 E-commerce related initiatives or programmes involving more than 20ministriesoragencies.AnotherexampleofakeyinitiativeistheBusinessAccelerationProgramme2.0,whichaimstoprovidecapacity-buildinginitiativestoassistSMEstogrowtheirbusinesslocallyandabroad.AMemorandumofUnderstandinghasbeensignedbetweentheMalaysianGovernmentandPayPal,eBayandGoogletoencourageSMEstogodigitalandsellonline.383

A2.6. Myanmar

A2.6.1. TheMyanmarE-governmentICTMasterPlan,thedraftofwhichwasissuedinJanuary2017,hasthefollowingobjectives:

a. To form specific organizations involved in the implementation of E-government in Myanmar, andtodefinetheirresponsibilities;

b. To be aware of the existing implementation progress of E-government and the benefits of E-governmentinMyanmar;

c. To shape the requirements of the implementation of E-government basedon the information collectedfromdiscussionmeetingswithimplementingagenciesofE-governmentinMyanmarand feasibilitystudies;

d. BasedonanalyticalstudiesofbestpracticesofcountrieswithsuccessfulE-governmentsystems, toputaprojectmanagementframeworkinplacetobetterprepareforE-governmentprojects;

e. To create a better andmore comprehensive integrated computer system for government by reviewingexistingICTinfrastructure,andtheapplicationofE-governmentsystemsinMyanmar;

f. Evaluating the skills and the gaps in skill development, and to set necessary measures for narrowingthesegaps;

g. To provide feedback on the required organizational structures and administration, and definingresponsibilitiesinformingtheimplementingagencyfortheeffectiveimplementationof governmentsystems;

381 Malaysian Digital Economy Corporation (MDEC), (2016).

382 Malaysian Digital Economy Corporation (MDEC), (2016).

383 Ministry Of International Trade & Industry, Malaysia, (2017).

166

384 Ministry of Communications and Information Technology, (2017).

385 Philippine E-commerce roadmap, (2017).

h. Toconstitutepoliciesandstandardsrequiredfortheeffectiveandefficientimplementationof theE-governmentsysteminMyanmar;

i. Toensuretheaccessibilityofthesystemfortheusers(government,businesses,citizensand otherstakeholdersorganization);and

j. Todeveloparoadmaptospecifybudgetallocationsrequiredforimplementingtheproject.384

A2.7. The Philippines

A2.7.1. ThePhilippinesGovernmentDepartmentofTradeandIndustryE-commerceRoadmap2016-2020,385

hassixkeyrecommendationstofacilitategrowthinE-commerceinthecountry:

a. Infrastructure: The need for an appropriate supply chain, communications, and applications infrastructure.ThisistobeaddressedbyrollingoutinternetinfrastructureviaaNationalBroadband Masterplan, settingupan ‘E-government’whichentailsmandatingelementsof government to haveelectronicfilingandelectronicpaymentfacilities.GuidelinesonE-commerceimplementation willalsobeissued,andworkwillbedonetodigitisebanking,taxandlogistics;

b. Investment: The ability to promote and support a range of investment opportunities fromForeignDirectInvestmentstocapitalflows.Thiswillbeaddressedbyprovidinganincentive packagefordigitalstart-upsandamendingtheCorporationCodetoallowonepersoncorporations;

c. Innovation: Theabilitytofosterandsupportinnovation,includingtheabilitytoprotectinnovation andinvestmentinresearchanddevelopment.ThiswillbeaddressedbyamendingtheRetailTrade LiberalizationActandassessingotherlegislation;

d. Intellectual Capital: The ability to foster the appropriate skills and training, ranging from technological to linguistic and entrepreneurial.Thiswill be addressedbyofferingE-commerce training in colleges, government trainingprograms inE-commerce, and includingE-commerce subjectsthroughoutallschoollevels;

e. Information Flows:Theabilitytouse,transfer,andprocess information–thecurrencyofthe digital economy – while promoting privacy and a trusted internet environment. This will be addressedbycreatingtheDataPrivacyCommission,DataPrivacyGuidelinesforthegovernment and updating Data Privacy Guidelines for the Information and Communications Systems in the Private Sector, aswell as promotingCybercrimeOnline Reporting and LegalAssistance Network,andsettingupaCybercrimeInvestigationandCoordinationCentreandNationalComputer EmergencyResponseCentre;and

f. Integration: The ability to connect domestic industrieswith the global economy. Thiswill be addressedbyidentifyingandpromotingE-commerceplatforms,implementingcapacitybuilding topromote internationalnetworking,encouragingtheavailabilityofnextgenerationhigh-speed broadband, and identifying and promoting policies and regulatory frameworks for creating a conduciveenvironmentforE-commerce.

167

A2.7.2. Therearealsoanumberofotherinitiativesalongsidethisroadmap,suchastheNationalBroadband PlanandtheNationalRetailPaymentSystemproject.386Criticalsuccessfactorsoftheprojectare:387

a. 100,000SMEsparticipatinginE-commerce;

b. 40-50%ofinternetusersengaginginE-commerce;

c. Cybercrimeenforcementandprotection;and

d. Onlineandconnectedgovernment.

A2.7.3. Keyrequirementstofacilitatetheseoutcomesbeingmetinclude:388

a. Increasinginternetspeeds(whichshouldbemetbysuccessful implementationoftheNational BroadbandPlan).Currentlytheaverageconnectionwithinthecountryis4.2Mbps,whichcompares toanaverageof11.4MbpsinAsia-Pacificasawhole;389

b. InvestingineducationtobetterexplainhowE-commerceworks;

c. OrganisingtrainingforSMEstoassisttheminexploringotherpotentialsaleschannels;

d. Improvingsecurityofwebsites(includingaddingsecurepaymentmethods);and

e.Diversifyingthetypesofproductsandserviceswhicharesoldonline.

A2.8. Singapore

A2.8.1. TherearemanygovernmentpoliciesaimedatthedevelopmentofE-commerce.TheSMEsGoDigitalProgrammehasoverSGD$80millionsetasidebytheGovernmenttoencourageSMEstomakeuseoftechnology.SPRINGSingapore,anagencyundertheMinistryofTradeandIndustry,partneredwithSingPosttolaunchanintegratedend-to-endE-commercesolutiontosupportandenableSMEsinSingaporetoexpandtheirE-commercebusiness.Outputsfromthis includetheMarketReadinessAssistance programme,which provides a grant of up to 70% of eligible third-party costs, whichcoveractivitiessuchassettingupinoverseasmarkets,identifyingbusinesspartners,andoverseasmarketpromotion.Another istheGlobalCompanyPartnershipprogramme,whichgroomsgloballycompetitive companies through building internal capabilities, developing manpower, accessingmarkets,andprovidingaccesstofinancingthroughgrants.BothofwhichaimtohelpbusinessesinSingaporeentermarketsoverseas.

A2.8.2.In2016,SPRINGSingaporelaunchedtheRetailIndustryTransformationMap(RITM),withtheaimofcreatingavibrantretailindustryandincreasingproductivity.Aspartofthistransformation,SPRING,andtheInfo-communicationsMediaDevelopmentAuthorityofSingapore(IMDA)areworkingtogetheron initiativesaimedatboostingthe roleofE-commerce, andattemptingtoencouragetraditionalbrick-and-mortarcompaniestoadoptastrategywiththeuseofdesktopormobileE-commerce.

A2.8.3.Oneof theways inwhichthis isbeingundertaken isbyusingEorM-commerceto teachdigitalmarketingmasterclasses,whichareaimedatretailexecutives,andfocusonwebanalyticsandsearchengineoptimisation.BydoingthisSPRINGandIMDAhopeto improvetheproductivityoftheretailworkforcewithinSingapore.

A2.8.4.Singapore plans to drive E-commerce and other areas of the digital economywithin the region,when it assumeschairmanshipofASEANnextyear.This could include streamliningE-commercerules.SingaporewilluseitschairmanshiptostreamlineregionaltraderulesgoverningE-commerce,improvedigitalconnectivityintheregionandloweroperationalbarrierstoentry.TheGovernmentalsointendstomaketrademoreefficientbyworkingcloselywithotherASEANstatestosetupaself-certificationregime.Thiswillallowauthorisedexporterstoself-certifythattheirgoodsmeetASEANrequirementsforpreferentialtreatment.Anotherinitiativeistospeedupcustomsclearanceviatheelectronicexchangeofinformationacrossborders,facilitatingthemovementofgoodsandloweringcostsforbusinesses.

386 Philippine E-commerce roadmap, (2017).

387 Ibid.

388 Ecommerce IQ Asia (2017).

389 Philippine Competition Commission (2017a).

168

390 Electronic Transaction Development Agency, (2017).



A2.9. Thailand

A2.9.1. Thailandhasrecentlylauncheditslatesteconomicgrowthplan,entitledThailand4.0,whichaimstomakeThailandavalue-creatingdigitaleconomy.Itplanstoachievethisbyfacilitatingkeysectorsofthedigitaleconomy,suchasE-commerce.Initiativesincludeprovidingaffordablebroadbandservicesnationwide,aswellasimprovingITservicesacrossthegovernment.ThailandalsohasanE-commerceplanoverthenextfouryears(2017-21),thevisionofwhichis“increasingvolumeandvalue”.ThefourelementstothisvisionareE-commercesystemdevelopment,standardsdevelopment,buildingandecosystemandpublic-privatecollaboration.Therearefivestrategiesunderthis:

a. ImprovingE-commercecapabilitiesofentrepreneursandenterprises

b. Tradefacilitationanddevelopment

c. EcosystemDevelopmenttosupportE-commerce

d. CreateopportunitiesandexperienceforanyonetobuyandsellthroughE-commerce

e. Buildtrustandconfidenceforconsumer.390

A2.10. Vietnam

A2.10.1.ThegovernmenthaveapprovedaplanfordevelopingE-commerceovertheperiod2016-20.391Keytargetsofthisplaninclude:

a. Bolsteringtheefficiencyofgovernmentadministrativeservices;

b. Ensuring30percentofthepopulationbuygoodsandservicesonline;

c. EnsuringanaveragespendofUS$350perpersononline;

d. IncreasingrevenuefromonlineB2CtoUS$10billion;

e. EnsuringB2Brevenueaccountsfor5%oftotalretailspend;and

f. EnsuringonlineB2Bturnoverisworth30%oftotalturnoverin2020.

A2.10.2.Aswellasthis,thegovernmentisaimingfor:392

a. 50%ofenterprisestoupdatetheirwebsitesonafrequentbasis;

b. 80%oforderscomingfromE-commerceapplications;

c. AllsupermarketstoacceptPOSandnon-cashpayments;

d. 70%ofelectricity,water,telecommunicationsandTVproviderstoacceptnon-cashbill payments;and

e. 50%ofindividualsandhouseholdsinmajorcitiestousenon-cashpaymentswhenspending.

www.ccs.gov.sg