NTMsandEconomicIntegra3on:recentevidence

JosephFrancoisUniversityofBern-WorldTradeIns3tute,

CEPR,CESifo

UNCTADNTMsweek,Geneva,September2017

Overview•  Changingpolicydatalandscape:dataintegra3on

–  Newdata,consistencychecks,andprojectsustainabilitymeansrevisi3ngques3onswithnewevidence.

•  Valuechains,NTMs,andcompe33veness:VCintegra3on–  Weseeevidenceofcostandproduc3vityevidencefromNTMs–  Thereisanopenques3on,givennewdataoftheimpactor

regulatorybarriersonvaluechainintegra3on•  Regulatorydivergence:regulatoryintegra3on

–  Oneareawherewehaveearlierevidencefromservicesratherthangoods

•  Somepolicytakeaways:includingpoli3calsustainability.

ThePolicyDataLandscape

DowntheRabbitHoleLewisCarroll

ThereandBackAgainJ.R.R.Tolkien

DownthePhillipscurvewithgunandcameraRobertSolow

Publicportalopensearly2018–integratedsourcedata,classifica3onconcordances,HSmappings,commonsummaryindexesfrommul3plesourcedata.Alsohats,t-shirts,coffeemugs.

Integra3onofDataSources

appendix A.2 provides the ITC classification; for details see ITC, 2014.

3 NTM data: sources, information contents anddatabases

3.1 Data sources

The information about NTMs comes from di↵erent sources that are used inthe data collection. Table 2 provides an overview of di↵erent sources of NTMinformation that are respectively mapped to the databases for NTMs for goodsand NTMs for services. In the following, we elaborate on the sources by orderingthem according to their completeness and the details they provide about themeasures identified.

Table 2: Overview NTM Data SourcesCollectionFramework

Goods Services

Inventories oflegislation

- NTM TRAINS - World Bank STRI Database- OECD PMR Database - OECD STRI Database- WB Investing Across Borders - OECD PMR Database- World Bank TTBD - WB Investing Across Borders- OECD Export Restrictions - Global Trade Alert- Global Trade Alert

Internationalagreements

- DESTA - DESTA- UNCTAD BIT Database - GATS commitment schedule

- WTO Services RTA Database- UNCTAD BIT Database

Review oflegislation

- WTO Trade Policy Review - WTO Trade Policy Review- WTO DG Monitoring Reports - WTO DG Monitoring Reports- UNCTAD Investment Policy - UNCTAD Investment Policy

Notifications - WTO Notifications - GATS NotificationsSurveys andcomplaintportals

- ITC NTM Surveys- ITC Trade Obstacle Alert- WTO STC- EU Market Access Database- tradebarriers.org

Import refusals - EU RASFF- US FDA OASIS

Other sources - USITC CoRe NTM Database

3.1.1 Inventories of legislation

NTM TRAINS: The most direct source of NTM data are regulatory inventoriesin which national legislation is meticulously reviewed in order to identify whichmeasures are specified in the legal body of a country and thus imposed by thecountry. Regulatory inventories require considerable knowledge on the rulesand regulation that contain NTMs but also the governmental bodies/ministries

9

Source:RauandVogt,2017,“NTMs:Dataconceptsandsources,“PRONTOworkingpaper.

AndMore...-BOWNtraderemediesno3fica3ons-Dataconsolida3onprojects(WTO,ITC,etc)-environmentaltaxes(CEPII)-OECDPSE

Example:WTOI-TIPdatabase•  38,881measuresno3fied1979-2014•  For60%,noHScodeprovided.Wehavereducedthisdrama3callywithcomputa3onal

linguis3cmethods(tryingtosqueezeitfurther).•  SPS&TBTaredominantinthedata.

0

500

1000

1500

2000

2500

3000

3500

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

TBT SPS QR TBTSTC SG SPSSTC ADP CV SSG

Source:WTOI-TIP;Ghodsi,ReiterandStehrer,PRONTOworkingpaper2015.NOTEthesedatahassincebeenupdatedforJan2018release.

NTMcomposi3onpersectoranddatabaseGTA NTM TRAINS WTO Notifications

0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00 0.00 0.25 0.50 0.75 1.00

XXI Works of art

XX Misc manufactured articles

XVIII Arms and ammunition

XVII Optical, medical or surgical

XVI Vehicles, aircraft, vessels

XV Machinery and mechanical appliances

XIX Base metals and articles of base metal

XIV Natural or cultured pearls

XIII Articles of stone, cement, etc.

XII Footwear

XI Textiles

X Pulp of wood, waste and scrap

IX Mineral products

VIII Wood and articles of wood

VII Raw hides and skins

VI Plastics and rubber

V Chemical products

IV Prepared foodstuffs

III Animal or vegetable fats and oils

II Vegetable products

I Live animals; animal products

Percent

HS

Sect

ors

MAST Chapters per HS Sector

A SPSB TBTC Pre Shipment

D Contingent ProtectionE Non AutomaticF Price Control

G FinanceH CompetitionI Investment

J Distribution RestrictionsL SubsidiesM Government Procurement

O Rules Of OriginP Export Related

Composition of Measures per HS Sector

Figure 2: NTM composition per sector and database

[ADJUST TEXT TO FIGURE]

As shown, most incidences of NTMs are found in animal and vegetablesproducts (I and II), plastics and rubber (VI), textiles (XI) as well as vehicles,aircrafts and vessels (XVI). It is interesting to note that a relatively low num-ber of countries are reported for the relatively large number of observations ofWTO notifications in plastics and rubber, and textiles. Overall, the pattern ofthe number of country and the number of observations seems to be rather bal-anced across sector. However, one or two sectors show a disproportionately highnumber of observations. In the GTA, the number of alerts are comparativelyhigh for vehicles, aircrafts and vessels, making up for about 21% of all observa-tions. In NTM TRAINS, about 19% of all regulations are reported for rubberand plastics and most WTO notifications are made in animal and vegetableproducts.

Table 55 presents the number of observations for services by mapping theavailable data to the first level of the Services Sectoral Classification List (SSCL/W/120)covered under the GATS. As mentioned, the information provided in the dif-ferent databases di↵ers and thus the figures presented are hardly comparableacross databases. Furthermore, the number of subsectors that di↵er accordingto classification influences the number of observations per SSCL sector. Forexample, the relatively high number of commitments and reservations as wellas GTA measures in business and transport services could be partly explained

5The World Bank and OECD STRI, and GTA database do not use the SSCL sectoralclassification. The sector mapping has been approximated.

24

Source:RauandVogt,2017,“NTMs:Dataconceptsandsources,“PRONTOworkingpaper.

Example:ProcurementNTMs•  Procurementmeasuresusuallytaketheformoflocaliza3on

requirementsratherthanpreferencesperse•  AllG8countrieshavereliedonthis,butnotallmeasuresareactually

implemented.

Source:EveneiandShingal,PRONTOworkingpaper,(2016).

More than half (51.8%) of US discriminatory procurement measures (n = 328) are not in force. In contrast, 126/151 of non-US discriminatory procurement measures are currently in force.

More than half (51.8%) of US discriminatory procurement measures (n = 328) are not in force. In contrast, 126/151 of non-US discriminatory procurement measures are currently in force.

Valuechainpar3cipa3onandNTMs

2040

6080

1980 1985 1990 1995 2000 2005 2010year

Zambia Spain MaliMozambique SouthAfrica Nigeria

Homophily --Africa's network densities vis-a-vis Spain

Source:dataarefromUNCTADCOMTRADEdatabase(2015).SeeEggeretal(2015)

ComplexityofProduc3onandTrade:valuechainsandtheimpactofNTMs

AVEsofindirecttradecostsoninputs

31

Figure 3 – Country Average indirect AVE for TBT on Inputs - third channel

Figure 4 - Country Average indirect AVE for SPS on Inputs - third channel

31

Figure 3 – Country Average indirect AVE for TBT on Inputs - third channel

Figure 4 - Country Average indirect AVE for SPS on Inputs - third channel

31

Figure 3 – Country Average indirect AVE for TBT on Inputs - third channel

Figure 4 - Country Average indirect AVE for SPS on Inputs - third channel

30

Appendix Note that all the trade policy measures in the third channel presented below are sorted by the AVE averaged across all sectors.

Figure 2 – Country Average Indirect Tariffs on Inputs - third channel

Theyalsoimpactones3matedproduc3vity,and

par3cipa3oningloballysourcedproduc3on.Source:Ghodsi&Stehrer,2016PRONTOWorkingpaper,“NTMsinthePresenceofGlobalValueChainsandtheirImpactonProduc3vity”

SPS,TBTeachhavecomparableimpactsasthoseoftariffsoncosts,whentracedoversupplychains.Addinginothertypeswillonlycompoundthisobserva3on.

RegulatoryDivergenceandIntegra3on

PaiernsofRegulatoryRestric3ons&RegulatoryDivergence

UK Financial Services Exports, 2014

lnX lnX lnX

ln(diverge1) -0.874 -0.810 (0.96) (0.47) lnGDP 0.932** 0.945** 0.945** (7.36) (8.00) (8.00)

EEA 0.180 0.349 (0.51) (64.20)**

lnSTRI 0.109 -0.353 (0.28) (0.60)

ln(diverge2) -0.810 (0.47)

R 2 0.51 0.52 0.52 n 43 43 43

* p<0.05; ** p<0.01

Cross-correlation in STRI indicators, Financial Services (UK partners) 2014

ln(diverge) ln(GDP) EUN EEA ln(STRI)

ln(diverge) 1.0000

ln(GDP) 0.3868 1.0000

EUN -0.6621 -0.3269 1.0000

EEA 0.0966 -0.1537 -0.2676 1.0000

ln(STRI) 0.8067 0.2391 -0.5043 0.2648 1.0000

| lnDIV lnGDP EUN EEA lnSTRI -------------+--------------------------------------------- lnDIV |

Source:Nordås,H.(2016),“ServicesTradeRestric3venessIndex(STRI):TheTradeEffectofRegulatoryDifferences”,OECDTradePolicyPapers,No.189,OECDPublishing,Paris.hip://dx.doi.org/10.1787/5jlz9z022plp-en

UK Financial Services Exports, 2014

lnX lnX lnX

ln(diverge1) -0.874 -0.810 (0.96) (0.47) lnGDP 0.932** 0.945** 0.945** (7.36) (8.00) (8.00)

EEA 0.180 0.349 (0.51) (64.20)**

lnSTRI 0.109 -0.353 (0.28) (0.60)

ln(diverge2) -0.810 (0.47)

R 2 0.51 0.52 0.52 n 43 43 43

* p<0.05; ** p<0.01

PaiernsofRegulatoryRestric3ons&RegulatoryDivergence

Source:Nordås,H.(2016),“ServicesTradeRestric3venessIndex(STRI):TheTradeEffectofRegulatoryDifferences”,OECDTradePolicyPapers,No.189,OECDPublishing,Paris.hip://dx.doi.org/10.1787/5jlz9z022plp-en

TakeawayPoints

Sometakeawaypoints•  RegulatoryTransparency(soyetmorequan3fica3on)makesense:

–  Uncertaintyisaneffec3vecostgenerator(agentsintheeconomyarenotriskneutral)–  Agreedarbitra3onmechanismsandmutualrecogni3onmechanismsmayhelp–  NTMsareataminimumcloselyrelatedtotradefacilita3on.

•  Regulatoryburdens,poorinfrastructurearebothcostgenerators•  NTMrelatedcostsmaymakeithardertointegratetohighervalue

addedpartsoftheglobaleconomy.RISKisthatcountrieswithhighbordercostsmaybeeffec3velylockedoutofemergingregionalproduc3onnetworks.Thiswouldwidengapsweobserveinglobalwagesandproduc3vitypaierns.Itposesadevelopmentchallengeforthetradingsystem.

•  Divergenceinregula3onsmaybearelevanttargetèweneedmuchmoreevidencehere.NTMsonenhaveanunderlyinglogic.Divergencewiththesameobjec3veslessso.Sopermycaveatbelow,thesemaymakemoresense.

•  Genericcaveat:èpoli3cal/distribu3onaladjustmentassistancemaybeneededtoensuretheprocessofopeningbordersispoli3callysustainable.