b smes in international trade: stylized facts · 2016-09-27 · smes in international trade:...
TRANSCRIPT
SMEs in international trade: stylized factsEvery firm that contemplates expanding its operations in a foreign country has to choose a specific market entry strategy. As trade is the most common form of internationalization for small and medium-sized enterprises (SMEs), this section surveys available statistical evidence on the participation of SMEs in international trade in both developed and developing economies, and how their activities relate to traditional trade flows and to trade in the context of global value chains. The objective is to provide an accurate and detailed description of the SME trade landscape, but also to identify important gaps in information and data coverage.
B
Contents1. SME involvement in direct trade 31
2. SME involvement in indirect trade and global value chains 39
3. SME participation in international e-commerce 46
4. MSME trade participation over time 51
5. Conclusions 54
Some key facts and findings
•• Trade•participation•of•SMEs•in•developing•countries•is•low,•with•exports•accounting•for•7.6•per•cent•of•manufacturing•sales,•compared•to••14.1•per•cent•for•larger•firms.
•• MSMEs•account•for•34•per•cent•of•exports•on•average•in•developed•countries.•There•is•a•positive•relationship•between•enterprise•size•and•export•participation,•with•lower•rates•of•participation•for•micro•enterprises••(9•per•cent)•and•small•enterprises•(38•per•cent)•than•for•medium-sized••(59•per•cent)•and•large•enterprises•(66•per•cent).
•• In•developing•economies,•indirect•exports•in•the•manufacturing•sector•of•SMEs•were•estimated,•on•average,•at•2.4•per•cent•of•total•sales,•a•level••three•times•lower•than•the•estimated•share•of•direct•exports.•Most•manufacturing•SMEs•in•developing•countries•have•low•levels•of•integration••in•global•value•chains,•with•few•backward•and•forward•linkages•in•production.
•• In•developed•economies,•the•direct•contribution•of•SMEs•to•domestic•value-added•exports•is•predominant•over•indirect•exports.
•• Electronic•commerce•expands•opportunities•for•SMEs•to•participate•in•international•trade.•On•average,•97•per•cent•of•internet-enabled•small•businesses•export.•Meanwhile•export•participation•rates•for•traditional••SMEs•range•between•2•per•cent•and•28•per•cent•in•most•countries.
•• In•developing•countries,•there•is•an•inverse•relationship•between•the•number•of•employees•that•a•firm•has•when•it•begins•operations•and•the•number•of•years•before•it•starts•to•export.•For•large•firms•that•started•as•SMEs,•it•took•17•years•to•export•for•those•that•began•with•five•employees•or•less,•compared•to•five•years•for•those•that•had•60-100•employees.
WORLD TRADE REPORT 2016
30
Internationalization is often defined as the strategyadoptedbyfirmsengagedinoverseasactivities(WelchandLuostarinen,1993).1 Internationalizationmaytakevarious forms, namely: (1) direct exports; (2) indirectexports (i.e. sales of goods through a third domesticparty that exports); (3) non-equity contractualagreements; and (4) foreign direct investment (FDI)andotherformsofequityagreements.
First,SMEscandirectlyserve internationalmarketsbybeginningtoexporttodistributorsortofinalconsumerslocatedinforeignmarkets.Second,SMEsmayoptforanindirect internationalization strategy by providing partsand components or services to other domestic firmsparticipatinginregionalorglobalvaluechains(GVCs)orbysellingproductsorservicestoexportintermediaries,suchaswholesalers,exportbuyingagentsandbrokers,situated in their own countries, who in turn export tointernational markets. Third, SMEs may opt for non-equitycontractualmodes,suchasfranchising,licensingor more structural alliances (e.g. export consortia).Fourth, SMEs can engage in FDI through greenfield investment (i.e. a type of FDI by which a parentcompany foundsanewventure ina foreigncountrybyconstructingnewoperationalfacilitiesfromscratch)andthrough mergers and acquisitions, as well as throughco-investment with other firms, such as joint ventures,withdifferentcontrollevels(e.g.fromminoritysharesto100percentowned).
While SMEs can use one or more of these types ofinternationalization modes, trade, direct or indirect, isoftenconsideredtobethefirststeptowardsengagingin international markets, operating as a platform forgreater future international expansion. Exportingindirectly is typically considered to be the least riskyentry mode to international markets because itenablesSMEstogainaccesstointernationalmarketswithout having to bear the upfront costs (including“sunk”costs, i.e.costs thatcannotberecoveredonceincurred)associatedwithsearchingfornewcustomersand negotiating contracts. Export intermediaries orother firms which undertake transaction sales and/or services in overseas markets on behalf of SMEsbenefit from market knowledge and negotiation skillsthat allow business risks to bepooledand diversifiedand that reduce the searching and matching costsassociatedwithexporttransactions.
Exporting is viewed as less risky than contract- orinvestment-based internationalization strategiesbecause it requires a lesser commitment oforganizational resources, entails fewer financial andcommercial risks,andallowsforgreater flexibilityandmanagerialdiscretion(LagesandMontgomery,2005).In practice, some SMEs export both directly andindirectly, highlighting the potential complementarity
betweenbothforeignmarketentrymodes(Nguyenetal.,2012).
Other forms of internationalization, such as non-equitycontracts and FDI, entail larger fixed costs, which aremore difficult to reverse in particular for SMEs. That iswhySMEs that have chosen in recent years toexpandtheir research and development (R&D), productionand distribution into foreign markets, tend to resort tocontractual arrangements, such as outsourcing, andminority share investment positions, rather than fullownershipofforeignaffiliates(Hollenstein,2005;Nakosand Brouthers, 2002). Since SMEs tend to experiencegreater financial, human and management constraintsthan large companies, and are more adverselyaffected by higher market barriers, it is not surprisingthat exporting continues to be the most commoninternationalization formadoptedby them(Riddleetal.,2007; Westhead, 2008). For instance, less than 3 percent of SMEs located in the European Union have aforeignsubsidiaryoverseas,which is significantly lowerthantheshareofSMEsexportingwithinandoutsidetheEuropeanUnion(EuropeanCommission,2014a).
The availability of data on international trade byenterprise size is limited in many respects. Forthe most part, researchers must rely on a mix ofenterprise surveys and administrative data, with all ofthe compromises that using different data sourcesentail (e.g. incomplete country coverage, inconsistentdefinitions of SMEs across datasets, differences inreportingstandardsacrosscountries,timelinessofdata,etc.).Detailed firm-level datamayalsobe inaccessibledue to confidentiality concerns. The main datasetsused in this sectionof the report are theOrganisationforEconomicCo-operationandDevelopment(OECD)’sTrade by Enterprise Characteristics (TEC) database,whichmostlydealswithdevelopedeconomies,2andtheWorldBank’sEnterpriseSurveys,whichprovidedetailedinformation on a range of developing economies.3These data sources are supplemented with others asnecessary,includingexistingstudiesonSMEs,nationalstatisticsandprivatesectorreports.
A number of findings emerge from this section. WeobservethattheparticipationofSMEsininternationaltrade varies considerably across countries,geographical regions, sectors and enterprise sizeclasses inbothdevelopedanddevelopingeconomies.In developed countries, shares of MSMEs in exportsand importsare relatively small compared to thoseoflargefirms,butthetradeparticipationofmedium-sizedfirmsisgreaterthanthatofmicroorsmallenterprises.A relatively small fraction of SMEs in developingeconomies export, either directly or indirectly,comparedtolargefirms.GVCparticipationofSMEsindevelopingcountriesisespeciallylowinsomeregions,
31
B. S
ME
s IN IN
TER
NA
TION
AL
TRA
DE
: ST
YLIZ
ED
FAC
TSLEVELLING THE TRADING FIELD FOR SMES
andfirmswithfeweremployees indevelopingregionsalso take longer toaccess internationalmarkets thanlargerfirms.
Despite these disadvantages, new technologies areenhancing trade opportunities for smaller firms indeveloped and developing countries alike. Unliketraditional SMEs, a very high percentage of Internet-enabled SMEs engage in international trade. Thissuggests that increasing SMEs’ access to onlineplatforms could potentially raise exports of smallerenterprises, particularly in developing economieswhere Internet access is less widespread than indevelopedcountries.
Finally, available data on SMEs and trade areinsufficient to answer many outstanding questions,in particular questions about the extent of indirectparticipationintradebySMEsandtheirroleinGVCs.
1. SMEinvolvementindirecttrade
“Direct exports” occur whenever an enterprise sellsgoods or services directly to customers in anothercountry.Sincethereisnointermediary,amajorbenefitofexporting inthisway isthattheexportingfirmis indirect contact with its consumers, enabling a betterunderstanding of their needs, thereby creating newbusiness opportunities. In addition, direct exportsprovidefirmswithmoreprotectionoftheirtrademarksorpatentsincaseofinnovativeproducts.
SMEs can export directly if they have the means toreach foreign consumers or GVC partners locatedabroad. However, they may find it difficult to mobilizeall the necessary human and financial resourcesto develop their international trade activities. Thus,exporting can be challenging for SMEs, especially indevelopingeconomies.
This subsection provides details on the directparticipationofSMEsininternationaltradebyfirmsize,sector,and,fordevelopedeconomies,wherepossible,bypartnercountryandregion.
As noted in section A.1, there are no universallyaccepted definitions of enterprise size classes.By default in this report, firms with fewer than 10employeesarereferredtoas“micro”enterprises,firmswith between 10 and 49 employees are classifiedas “small” enterprises, firms with between 50 and249 employees are categorized as “medium-sized”enterprises,andfirmswith250ormoreemployeesareconsidered “large”. These size classes correspond tothose used in the OECD TEC database, but differentcategories will be used in other contexts dependingon the definitions used in particular databases or
studies.Forexample,thecategoriesabovedifferfromthoseemployedby theWorldBank in theirEnterpriseSurveys, in that the latter excludes firms with fewerthan 5 employees and businesses with 100 or moreemployeesfromitsdefinitionofSMEs.Otherdefinitionsarealsoused in researchandstatisticsonSMEs,butnearly all of these encompass businesses with fewerthan500employees.Consequently,thereadershouldbeaware that the termsSMEmay refer todifferentlysized firms in different contexts. The term MSME,referring to “micro, small and medium enterprises”, isalso used in this section and elsewhere in the reportto indicate the inclusionofmicroenterprises in totalswherepossible.
The TEC database provides breakdowns of exportsand imports by economic sector and by partnercountry/region.Tradevalues in theTECdatabasearerecordedincurrentUSdollars,facilitatingaggregation,but country coverage is mostly limited to developedeconomies.Onenotableexception isTurkey,which isusually classified as a developing/emerging economybutissometimestreatedasdevelopedbecauseit isamemberoftheOECD.
The World Bank’s Enterprise Surveys provide detailedinformation by sector and enterprise size for a widerangeofdevelopingcountries,but thedatasuffer fromsomeof thecommonshortcomingsofsurveys,suchasincompleteanswersfromrespondents.AnotherlimitationoftheEnterpriseSurveysisthatthetradevaluesareinnationalcurrencytermsandarelaggedtothefiscalyearprior to that during which the survey was carried out.Convertingtodollarsforaggregationpurposesisanon-trivial exercise, but this has been carried out to arriveat aggregate estimates for least-developed countries(LDCs)andotherdevelopingregions.
Due to differences in coverage and data sources, itis currently not possible to compare the participationof SMEs in developed economies with those in thedevelopinggroup.
(a) DirectparticipationofSMEsandMSMEsintradeofdevelopedcountries
DespitethefactthatMSMEsmakeupthevastmajorityof firms in developed economies (98 per cent ofindustrial firms in OECD countries, according to theTEC database), their direct exports typically accountfor less than half of the value of gross exports. Thisis illustrated by Figure B.1, which shows shares ofSMEs (i.e. excluding micro firms with fewer than 10employees) and MSMEs (i.e. including micro firms)trading with OECD economies. Shares of SMEs intrade were below 50 per cent in every country ontheexportside,andallbutonecountryonthe import
WORLD TRADE REPORT 2016
32
side. Includingmicrofirmswith0-9employeesboostsMSME shares in exports over 50 per cent in a fewcases, but shares of most countries remain below50percent.
(i) Directtradebyenterprisesize
Exportshares forMSMEssignificantlyexceed50percent inasmallnumberofcountries, includingEstonia(69 per cent), Turkey (63 per cent), Cyprus (61 percent) and Ireland (57 per cent). With the exceptionof Turkey, all of the countries with the highest SME
shares inexportvaluesaremembersoftheEuropeanUnion. By comparison, shares for non-EU countriessuch as Canada (29 per cent) and the United States(28percent)areconsiderablylower(seeFigureB.1).
Shares of MSMEs in gross imports tend to besomewhat larger than their shares in exports, withthe largest shares belonging to small countries suchasEstonia (78per cent),Cyprus (75per cent),Malta(74percent)andLatvia(63percent).However,theseenterprisesstillaccountforlessthanhalfofthevalueofimportsinthelargestdevelopedcountries,including
Figure B.1: SME and MSME shares in the dollar value of exports and imports of selected developed countries, 2013 (or latest year)(percentage)
Exports
SMEs (excluding firms with 0-9 employees) Other enterprises MSMEs (including firms with 0-9 employees)
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Estonia Ita
ly
Cypru
s
Turke
y
Portug
al
Croati
aLatv
ia
Nether
lands
Spain
Austria
Slovak
Rep
ublic
Lithua
nia
Denmark
Bulgari
a
Greec
e
Irelan
d
Sloven
ia
Roman
ia
Finlan
d
Belgium
Poland
Malta
France
Sweden
Luxem
bour
g
Hunga
ry
United
King
dom
Canad
a
German
y
Czech
Rep
ublic
United
Stat
es
47 45 44 42 41 37 36 35 34 33 33 33 33 33 31 30 27 27 27 26 25 25 24 24 21 21 20 18 18 17 15
69
51
61 63
5148
53
43 4348
42 4339
43 41
57
36 34 30 31 30
46 44
35
26
46
33 2921 20
28
Imports
SMEs (excluding firms with 0-9 employees) Other enterprises MSMEs (including firms with 0-9 employees)
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Cypru
s
Estonia
Croati
a
Portug
al
Denmark
Austria
Irelan
d
Bulgari
a
Sloven
ia
Finlan
dIta
ly
Lithua
niaLatv
ia
Roman
ia
Nether
lands
Luxem
bour
g
Slovak
Rep
ublic
Greec
e
Turke
y
Sweden
Poland
Malta
France
Spain
Belgium
Czech
Rep
ublic
Hunga
ry
Canad
a
United
King
dom
German
y
United
Stat
es
56 48 45 42 42 41 41 40 38 38 37 37 37 36 36 35 33 32 32 31 31 30 29 29 27 27 26 26 25 23 16
75 78
59 5651
61 6153 51
48 44 49
63
46 46 45 47 4439
50
37
7450
4032 31
4839
41
28 26
Note:Bulgaria,Canada,Ireland,Romania,SloveniaandTurkeyreferto2012,whileLuxembourgrefersto2011.
Source:OECDTECdatabase.
33
B. S
ME
s IN IN
TER
NA
TION
AL
TRA
DE
: ST
YLIZ
ED
FAC
TSLEVELLING THE TRADING FIELD FOR SMES
Germany(28percent)andtheUnitedStates(26percent).
In aggregate, the share of MSME exports in totalexports of developed countries in the TEC databasein 2013 was 34 per cent. The equivalent share onthe import side was 38 per cent. Note that theseshares includeTurkey,which isusuallyclassifiedasadevelopingeconomybutisamemberoftheOECD.
Despite relatively small shares of SMEs in developedcountries’ exports and imports by value, MSMEs (and
micro firms in particular) represent the large majorityof trading firms inmostdevelopedeconomies.This isillustratedbyFigureB.2,whichshowsthepercentageof exporting and importing firms that are MSMEs inselected developed economies by enterprise size in2013 or the latest available year. Shares of MSMEsare lowest in countries with large numbers of firmsof unknown size (e.g. Belgium, Czech Republic andGermany). However, MSMEs account for as much as99 per cent of exporting and importing firms in theNetherlands and more than 95 per cent in Sweden.Shares are considerably smaller if micro firms
Figure B.2: Percentage of exporting and importing firms that are SMEs in selected developed economies by enterprise size, 2013 or latest year(percentage)
Exports Imports
0 10 20 30 40 50 60 70 80 90 100
Romania
Turkey
Canada
Lithuania
Malta
Croatia
Bulgaria
United Kingdom
France
Poland
Austria
Greece
United States
Portugal
Italy
Denmark
Luxembourg
Finland
Sweden
Latvia
Hungary
Slovak Republic
Germany
Ireland
Estonia
Cyprus
Spain
Netherlands
Slovenia
Belgium
Czech Republic
SMEs (10-249) Micro enterprises (0-9) Large enterprises (250+) Number of employees unknown
0 10 20 30 40 50 60 70 80 90 100
Turkey
Canada
Lithuania
France
United States
Romania
United Kingdom
Croatia
Bulgaria
Poland
Italy
Hungary
Greece
Latvia
Malta
Austria
Denmark
Ireland
Luxembourg
Portugal
Estonia
Germany
Slovak Republic
Sweden
Czech Republic
Finland
Netherlands
Cyprus
Slovenia
Spain
Belgium
Note:Bulgaria,Canada,Ireland,Romania,SloveniaandTurkeyreferto2012,whileLuxembourgrefersto2011.
Source:OECDTECdatabase.
WORLD TRADE REPORT 2016
34
(0-9 employees) are excluded, ranging from 8 percentto48percent.Bycomparison,smallenterprises(10-249 employees) account for more than half ofexporting and importing firms in most countries intheTECdatabase.Intotal,theshareofMSMEsinthenumber of exporting and importing firms was 78 percentontheexportsideand76percentontheimportsidein2013(orlatestyear).
Small enterprises in the OECD TEC database maybe more representative of SMEs than either microor medium-sized firms, since the former frequentlyoperate in non-tradable sectors while the lattersometimes resemble larger firms more closely. Thisis especially true when comparing TEC data to theWorld Bank’s Enterprise Surveys, which classifyestablishments with more than 100 employees aslarge enterprises. Focusing on small enterprisesexclusively, we see that their overall share in exports(9percent)issignificantlylessthantheirshareinthenumber of exporting firms (21 per cent). Their sharein imports (11 per cent) is also less than their sharein importing firms (16 per cent), but not dramaticallyso. Meanwhile, medium-sized businesses account foragreaterfractionofinternationaltrade(15percentofboth exports and imports) than their numbers wouldsuggest (7 per cent of enterprises that export and 5percentofthosethatimport).
Ifwerestrictourattentiontoindustrialenterprises,wecan see a positive association between enterprise-size SMEs and participation in international trade.This is shown for developed OECD countries in
FigureB.3.The lowshares formicro firmswith fewerthan10employees(9percentontheexportsideand12 per cent on the import side) have dragged downaverage figures for all size classes due to the largenumber of micro firms in OECD economies. All ofthe other enterprise size classes (small, medium andlarge) have above-average shares of firms engagingin international trade, ranging from 38 per cent to66 per cent on the export side and 40 per cent to70percentontheimportside.Inparticular,exportandimportparticipationratesformedium-sizedenterprisesapproachthoseoflargeenterprises,whileparticipationratesforsmallandmicroenterprisesareconsiderablysmaller.
Insummary,sharesofSMEsandMSME’sintradeflowsof developed OECD countries are generally low, butthere is considerable heterogeneity across enterprisesize classes. In particular, rates of export and importparticipation for medium-sized enterprises are quitehigh,approachingthoseoflargebusinesses.
(ii) DirecttradeofMSMEsbysectorandpartner
Dollar values of trade flows by firm size and sectorare shown in Figure B.4 through 2012, the last yearfor which a complete sectoral breakdown wasavailable in the TEC database for a sufficient numberof countries. Micro enterprises appear to have thelargestshares inexports incertainservicescategoriesincludingaccommodation,arts/entertainment/recreationand other service activities, while large enterprises
Figure B.3: Percentage of industrial firms that are exporting and importing by enterprise size, 2013 or latest year(percentage)
0
10
20
30
40
50
60
70
80
Exports Imports
Total
17 20 9 12 38 40 58 62 66 70
0-9 10-49 50-240 250+
Note:DataforCanadaandIrelandreferto2012.Turkeyisexcludedduetomissingdata.
Source:OECDTECdatabase.
35
B. S
ME
s IN IN
TER
NA
TION
AL
TRA
DE
: ST
YLIZ
ED
FAC
TSLEVELLING THE TRADING FIELD FOR SMES
predominate in sectors such as manufacturing andmining/quarrying. On the import side, micro firms aredominantinservicesectors,includinghealthcare,whilelarge firms account for an outsized share of financialservices imports. There does not appear to be anysystematic relationship between economic sectors andenterprise size other than the fact that more capital-intensivesectors(mining,manufacturing,electricityandgassupply)tendtobedominatedbylargeenterprises.Atahigherlevelofaggregation,itappearsthatmostMSMEexportsandimportsindevelopedeconomiesareinfact
services,with68percentontheexportsideand83percenton the importside (seeFigureB.5,alsowithdatathrough2012).
Two findings regarding the services trade of SMEsare worthy of note. First, those SMEs that begin toexporttendtopersistinthisbehaviour,i.e.theyhaveahighsurvival rateconditionaluponexporting.Second,although a smaller fraction of SMEs engage in tradecompared to large firms, those SMEs that do tradedirect a larger share of their sales toward foreign
Figure B.4: Trade values by sector, exports and imports, 2012(percentage)
Exports
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
59
9 25
14 23 2621 19
5341
5
71
47
118 15
10 15 4 2
53 46 44 41 38 38 35 32 28 24 18 18 16 13 9 8 8 7 3
937970
42
121
79
29
5
5
22
62
13
32322030
3717
29
3 5 3 13 9 11 5 6 2
65 6080 72
17 12 44
Wat
er s
uppl
y; s
ewer
age,
was
te/
rem
edia
tion
Agr
icul
ture
, for
estr
y an
d fis
hing
Tran
spor
tatio
n an
d st
orag
e
Con
stru
ctio
n
Who
lesa
le, r
etai
l tra
dean
d re
pair
Pro
fess
iona
l, sc
ient
ific
and
tech
nica
l act
iviti
es
Info
rmat
ion
and
com
mun
icat
ion
Art
s, e
nter
tain
men
tan
d re
crea
tion
Acc
omod
atio
n an
dfo
od s
ervi
ces
Acc
omod
atio
n an
d fo
odse
rvic
es; n
on-m
arke
t ser
vice
s
Oth
er s
ervi
ce a
ctiv
ities
Fina
ncia
l and
insu
ranc
e ac
tiviti
es
Man
ufac
turin
g
Edu
catio
n
Adm
inis
trat
ive
and
supp
ort
serv
ice
activ
ities
Rea
l est
ate
activ
ities
Hum
an h
ealth
and
soci
al w
ork
activ
ities
Ele
ctric
ity, g
as, s
team
and
air c
ondi
tioni
ng
Min
ing
and
quar
ryin
g
Pub
lic a
dmin
istr
atio
n, d
efen
ce;
soci
al s
ecur
ity
Imports
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Wat
er s
uppl
y; s
ewer
age,
was
te/
rem
edia
tion
Agr
icul
ture
, for
estr
y an
d fis
hing
Tran
spor
tatio
n an
d st
orag
e
Con
stru
ctio
n
Who
lesa
le, r
etai
l tra
dean
d re
pair
Pro
fess
iona
l, sc
ient
ific
and
tech
nica
l act
iviti
es
Info
rmat
ion
and
com
mun
icat
ion
Art
s, e
nter
tain
men
tan
d re
crea
tion
Acc
omod
atio
n an
dfo
od s
ervi
ces
Acc
omod
atio
n an
d fo
odse
rvic
es; n
on-m
arke
t ser
vice
s
Oth
er s
ervi
ce a
ctiv
ities
Fina
ncia
l and
insu
ranc
e ac
tiviti
es
Man
ufac
turin
g
Edu
catio
n
Adm
inis
trat
ive
and
supp
ort
serv
ice
activ
ities
Rea
l est
ate
activ
ities
Hum
an h
ealth
and
soci
al w
ork
activ
ities
Ele
ctric
ity, g
as, s
team
and
air c
ondi
tioni
ng
Min
ing
and
quar
ryin
g
Pub
lic a
dmin
istr
atio
n, d
efen
ce;
soci
al s
ecur
ity
71
929
2519 27
7 11
71
22 17
66
8 6 131
22
8265
268
53 48 45 4431 31 24 23 22 22 20 17 17 14 14 13 10 9 5
8880
6
215
6381
20
72
3
18
50 43
5
52
12
2329231318
2 6 7 6
50
5 6 2
50
1 1 1
59
3 74
SMEs (10-249) Micro enterprises (0-9) Large enterprises (250+) Number of employees unknown
Source:OECDTECdatabase.
WORLD TRADE REPORT 2016
36
markets than large firms.These findingsbyLejárragaet al. (2014) could have important policy implicationsregarding the effectiveness of support for SMEs inaccessinginternationalmarkets.
The vast majority of MSME exports in developedcountriesaredestinedforotherdevelopedeconomies,and most MSME imports also originate in developedeconomies.Chinaisthemainexception,accountingfor2.3 per cent of developed country exports and 7 percentof imports.This isshown inFigureB.6for2012,
thelastyearwithsufficientlydetaileddatabypartner.SharesofdevelopedcountriesaspartnersofMSMEsmaybeexaggeratedduetothefactthatintra-EUtradeis included in the chart. An alternative perspective isprovided by Figure B.7, which shows the same dataexcluding trade between members of the EuropeanUnion. In this case, China’s shares in exports andimports of developed county SMEs rise substantially,to 7 per cent and 22 per cent, respectively, as doshares of other emerging markets such as India, theRussianFederationandTurkey.
Figure B.5: Exports and imports of MSMEs by broad product category, 2012(percentage)
Exports Imports
Agriculture, fuels and mining products (2%)
Manufacturing (30%)Services (68%)
Agriculture, fuels and mining products (1%)
Manufacturing (17%)Services (83%)
Source:OECDTECdatabase.
Figure B.6: Exports and imports of SMEs in developed countries by partner, 2012(percentage)
Exports Imports
Germany (15.3%)
France (10.3%)
Italy (5.6%)
Belgium (4.5%)
United States (4.3%)
Switzerland (3.3%)
Austria (3.2%)
Poland (3.2%)
Russian Federation (2.8%)
Netherlands (5.6%)
United Kingdom(5.9%)
Other (19.1%)
Denmark (1.3%)
Turkey (1.4%)
Portugal (1.5%)
Hungary(1.5%)
Sweden(1.6%)
SlovakRepublic(1.8%)
China (2.3%)
CzechRepublic (2.7%)
Spain (2.7%)
Germany (15.4%)
Netherlands (11.2%)
Italy (5.9%)
Belgium (4.9%)
United Kingdom (4.0%)
United States (3.6%)
Spain (3.1%)
Czech Republic (2.8%)
Poland (2.6%)
France (6.5%)
China (7.0%)
Other (17.5%)
Denmark (1.1%)
Turkey (1.2%)
Hungary (1.5%)
Slovak Republic(1.5%)
Sweden (1.5%)
Japan(1.9%)
Switzerland(1.9%)
Austria (2.4%)
RussianFederation (2.5%)
Source:OECDTECdatabase.
37
B. S
ME
s IN IN
TER
NA
TION
AL
TRA
DE
: ST
YLIZ
ED
FAC
TSLEVELLING THE TRADING FIELD FOR SMES
Oneconclusionthatmightbedrawnfromtheprecedingcharts is that MSMEs in developed countries, andparticularly micro SMEs, have more difficulty inbridging the trade gaps between themselves anddistantordissimilartradingpartners.
(b) DirectparticipationofSMEsintradeofdevelopingcountries
AsnotedinSectionA,SMEsplayanimportantroleineconomicandsocialdevelopment,particularlyinpoorercountries and LDCs. According to WTO calculations,based on data from World Bank Enterprise Surveys,out of more than 15,500 manufacturing and servicesfirms in 41 LDCs, 88 per cent were SMEs, includingsome59percentofsmallfirmsemployingfewerthan20people,and29percentofmedium-sizedfirmswith20-99employees.Ingeneral,theirdirectparticipationin international trade is low. According to WTOestimates,basedondatafromWorldBankEnterpriseSurveys for over 25,000 SMEs in the manufacturingindustry in developing economies, SMEs’ directexportsrepresentonaveragejust7.6percentoftotalmanufacturingsales.4Incontrast,largemanufacturingfirms,withmorethan100employees,directlyexported14.1percentoftheirtotalsales.
The involvement of SMEs in direct exports variessignificantly across developing regions. The highestshares were recorded in Developing Europe, wheretheyaccountedforaround28percentofoverallsalesbySMEs,and in theMiddleEast (16percent).Theseshares are much higher than in SMEs in Developing
Asia (8.7 per cent). SMEs in Africa exported directlyonly3percentoftheirtotalsales(seeFigureB.8).Asindicated above, the World Bank Enterprise surveysexclude micro enterprises (in the class size betweenzero and four employees). However, the World Bankhascollectedmicrofirmsurveysinselecteddevelopingcountries. Using these data, Box B.1 shows that inLDCs, direct involvement in trade of micro firms withlessthanfiveemployeesismarginal.
A sectoral analysis reveals that, in developingeconomies,SMEs’lowerparticipationindirectexportsthanlargerfirmsaffectsallmanufacturingsectors,withthe exception of the wooden furniture manufacturingindustryandthepublishingandprintingindustries(seeFigureB.9).Itshouldbenotedthathighershareswerein both cases predominantly due to SMEs in LDCs(66 and 30 per cent respectively). A considerablenumber of medium-sized firms in several LDCs, suchas Bhutan, Mozambique, Myanmar, Tanzania, Ugandaand Zambia, directly exported wooden sofas, beds,chairs, tables, etc. SMEs did not participate activelyin the direct exports of textiles and garments. Theirshareofdirectexportswasoftenlessthan5percent,well below the high percentages reported by largeenterprises. Another example is manufacturing ofofficeequipmentandelectronics,wherelargefirmsindeveloping economies exported directly, on average,around43percentoftheirtotalsales,comparedwith4percentbySMEs.
Participationbydevelopingeconomies’SMEsindirectservicesexportswasnegligible,atlessthan1percent
Figure B.7: Extra-EU exports and imports of SMEs in developed countries by partner, 2012(percentage)
Exports Imports
United States (13.7%)
Switzerland(10.6%)
RussianFederation
(9.0%)
China (7.4%)
Turkey (4.4%)
Norway (2.4%)
Japan (2.3%)
Other (45.2%)
Canada (1.4%)
India (1.7%)
Mexico (1.9%)
China (21.5%)
United States(10.9%)
RussianFederation
(7.6%)
Switzerland (5.8%)
Japan (5.7%)
Turkey (3.6%)
Other (37.5%)
Norway (3.4%)
India (2.4%)
Canada (1.0%)
Mexico (0.5%)
Source:OECDTECdatabase.
WORLD TRADE REPORT 2016
38
of total services sales, comparedwith32per cent oflarge firms. The difference in performance with largeservices firms is striking, ranging from 16 per centin LDCs to peak to 40 per cent in large enterpriseslocated in other developing economies (seeFigureB.10).
In services, the highest share of direct exports bySMEs in developing economies was in transport (20per cent of total sales). In communications, including
the provision of Internet access, the contribution wasaround 4 per cent. In the accommodation sector, theshare of direct exports by SMEs was below one percent. In LDCs, virtually all SMEs in constructionactivities, often foreign-controlled, supplied thenational market. Finally, SMEs’ participation in directexports in higher-skilled services was marginal.Computer-related activities accounted for less than 1percentoftheirtotalsalescomparedwith23percentoflargefirms.
Box B.1: Participation of micro firms in exports in selected LDCs
Evidence from recentWorldBankMicro firmsurveys inselectedLDCsconfirms themarginalizationofmicrofirms(i.e.,lessthanfiveemployees)ininternationaltrade.Microfirmswereengagedindifferentsectorsoftheeconomyrangingfromfoodmanufacturingtotheretailandwholesaletradeandtheleathergoodsindustry,aswellas restaurantsand ITservices. In2013,outof the412surveyedmicro firms in theDemocraticRepublicof the Congo, only 6 per cent were engaged in exports. The share of exporting micro firms, whether inmanufacturingorservices,inBhutanandEthiopiawasevenlower,at3percentofthetotal.Finally,inMyanmar,lessthanonepercentofthe430surveyedmicrofirmsexportedtheirproductstoforeigncountries.
Microfirmswereyoung,havingstartedoperationsbetween2004and2005,andseveralwererunbyfemalesowners,withatleastsecondaryeducation.InMyanmar,halfoftheownersheldauniversitydegree;inEthiopia,onequarter.
Virtuallyallmicrofirmsweredomestically-ownedandtargetedthe localornationalmarket.Onlyahandful ineachcountryheld internationalcertificatesofproductsand/orprocesses.Whileseveralmicro-firmsusedtheInternettoreachtheirclientsorsuppliers,onlyafewhadtheirownwebsites,rangingfrom2percentinBhutanto20percentinEthiopia.
Figure B.8: SMEs’ shares of direct exports in total sales in the manufacturing sector, by developing region and in the LDCs(percentage of total sales)
0
10
20
30
40
50
60
70
80
90
100
National sales Direct exports
Per
cent
age
Developing economies, of which:
DevelopingEurope
Middle East DevelopingAsia
Latin Americaand the Caribbean
Africa Least-developedcountries
Note:SMEs’sharesof indirectexportarenot included.DevelopingEuropecomprisesAlbania,BosniaandHerzegovina,Montenegro,Serbia,FormerYugoslavRepublicofMacedoniaandTurkey;DevelopingAsia includesallmembersof theWTO’sAsiaregionminusAustralia,JapanandNewZealand;LatinAmericaandtheCaribbeanincludesallmembersoftheWTO’sSouthandCentralAmericaandtheCaribbeanregionplusMexico(seeWTOdocumentWT/COMTD/W/212).DevelopingeconomiesandLDCsaredefinedintheTechnicalnotesandinWTO(2016).
Source:WTOestimatesbasedonWorldBankEnterpriseSurveys.
39
B. S
ME
s IN IN
TER
NA
TION
AL
TRA
DE
: ST
YLIZ
ED
FAC
TSLEVELLING THE TRADING FIELD FOR SMES
Figure B.9: Direct exports by manufacturing sector and firm size in developing economies (percentage of total sales)
05
101520253035404550556065
Man
ufac
turin
g (IS
IC 1
5-3
7)
Pub
lishi
ng, p
rintin
g, re
prod
uctio
n of
reco
rded
med
ia (I
SIC
22
)
Furn
iture
and
oth
erm
anuf
actu
ring
(ISIC
36
)
Toba
cco
(ISIC
16
)
Mot
or v
ehic
les
and
othe
rtr
ansp
ort e
quip
men
t (IS
IC 3
4-3
5)
Pap
er a
nd p
aper
pro
duct
s (IS
IC 2
1)
Cok
e, re
fined
pet
role
um p
rodu
cts
and
nucl
ear f
uel (
ISIC
23
)
Food
(IS
IC 1
5)
Text
iles
and
garm
ents
(ISIC
17
-18
)
Fabr
icat
ed m
etal
pro
duct
s,ex
cept
mac
hine
ry (I
SIC
28
)
Rub
ber a
nd p
last
ics
prod
ucts
(ISIC
25
)
Off
ice
and
elec
tric
al m
achi
nery
, rad
io,
TV a
nd c
omm
unic
atio
n eq
uipm
ent,
med
ical
,pr
ecis
ion
and
optic
al n
.e.s
. (IS
IC 3
0-3
3)
Mac
hine
ry a
nd e
quip
men
t n.e
.s.
(ISIC
29
)
Che
mic
als
and
chem
ical
pro
duct
s(IS
IC 2
4)
Woo
d an
d w
ood
prod
ucts
(IS
IC 2
0)
Leat
her:
lugg
age,
han
dbag
s, s
addl
ery,
foot
wea
r, et
c. (I
SIC
19
)
Oth
er n
on-m
etal
lic m
iner
alpr
oduc
ts (I
SIC
26
)
Bas
ic m
etal
s (IS
IC 2
7)
SMES (<100 employees) Large (+100 employees)
Note:WTOestimatesbasedontheInternationalStandardIndustrialClassificationofAllEconomicActivities(ISIC),Rev.3.1.N.e.s.standsfor“notelsewherespecified”.
Source:WorldBankEnterpriseSurveys.
2. SMEinvolvementinindirecttradeandglobalvaluechains
Rather than exporting directly, SMEs may connectindirectly to global markets by supplying goods andservices to other domestic firms that export. SMEscan use the services of domestic intermediariessuch as agents or distributors to help market theirproducts in foreigncountriesand reachnewmarkets.However, goods and services produced by SMEs canalso be indirectly exported as intermediate inputsincorporatedinproductsexportedthroughlargerfirms.In the manufacturing sector, for example, SMEs maybe contracted to produce certain parts according tospecifications of other companies, often larger ones,andentervaluechains.
Over the last decades, rapid technological changes,coupled with more efficient and less costlytransportation means, have significantly affectedthe ways goods and services are produced and sold.Thanks to lower barriers to international trade, theproduction of goods and services, rather than takingplace in a single economy, is globalized and spreadoverfirmslocatedindifferentcountries,alongachain.TradeinGVCsmainlyreferstotheexchangeofgoods
Figure B.10: Shares of direct services exports by firm size and developing group(percentage of total sales)
SMEs in LDCs
SMEs in otherdevelopingeconomies
Large firmsin LDCs
Large firmsin other
developingeconomies
0 5 10 15 20 25 30 35 40 45
Source:WTOestimatesbasedonWorldBankEnterpriseSurveys.
WORLD TRADE REPORT 2016
40
and services along the production and distributionchains that are fragmented across countries. Theproduction sequence is often supplemented by alogisticsanddistributionnetworkinwhichintermediateandfinalproductscirculatewithinandacrosscountriesuntiltheyreachthefinalconsumptionmarket.AlthoughGVC trade relates essentially to the exchange ofintermediates, exports of final products always takeplacewithinthefinalstageofthechain.Basedontheinputs produced by upstream suppliers, the ultimateenterprise in the production chain, which may or maynot be the lead firm in the chain, completes the finalproductandsendsiteithertointernationaldistributors(wholesalersorretailers)orstraightforconsumptionintheimportingcountry.
Enterprises participate in GVCs in two ways relatedto the linkages with their foreign partners. Backwardlinkages correspond to the import of inputs fromenterprises in order to produce intermediate or finalgoodsandservicesfordomesticconsumptionorfurtherexport.Enterprisesmayalso import finalproducts forfurther distribution through national or internationalnetworks. Backward linkages represent the “buyer’s”perspectiveorsourcingsideinGVCs.Forwardlinkagesrepresent the “seller-related” measure or supply sidein GVC participation, when an enterprise exportsintermediates through the international productionchainorfinalproductstodistributioncircuits.
It isalsonecessary todistinguishbetweendirectandindirect forward linkages. An enterprise contributesdirectly to a GVC when it exports inputs to partnercountries along the production chain for moreprocessing (and subsequent domestic consumption)or further export through international networks.Direct exports of final products through internationaldistributionchainsarealsopartofGVCtrade.
The indirect forward participation in GVCs mainlyconcerns enterprises that provide intermediate or finalgoodsandservicestolargerdomesticfirmsforexportsthroughinternationalnetworks.Inthisway,anenterprisebehaves like an “indirect exporter” by contributing tothe production or distribution of goods and servicesexported by other domestic enterprises. Direct andindirectforwardparticipationinGVCsdealwithexportsofproductsforfurtherexchangeswithintheproductionordistributionchains.FigureB.11 illustrates theabovedefinitions and shows the domestic and internationaltradeflowsrelatedtoGVCs.
(a) IndirectexportsandGVCparticipationofSMEsindevelopedcountries
OnlyafewstudieshaveexaminedtheroleofSMEsinindirectexports. Ina reporton the involvementofUScompanies in international supply chains, Slaughter(2013) stated that US multinational enterprisesin a typical year purchase inputs valued at morethan US$ 3 billion from SMEs in the United States,equal to 25 per cent of total input purchases. Otherestimates from the United States International TradeCommission (USITC) (2010) indicate that in 2007the share of SMEs in gross exports rose from 28per cent to 41 per cent once indirect exports wereconsidered. A similar study on Canadian SMEs fromIndustry Canada (2011) produced estimates showingthat 26 per cent of manufacturing enterprises soldinputs to other Canadian enterprises that were usedin the production of final goods for export. However,Canadian SMEs were actually less likely than largerenterprises to export intermediate goods indirectly.Specifically, 26 per cent of small enterprises and 27percentofmedium-sized firmsexported intermediategoods indirectly, compared to 30 per cent of largeenterprises.
Figure B.11: Schematic presentation of GVC trade flows
Import of intermediate/final goods/services(Backward GVC participation)
Reporting country
Entreprise 1Producer/distributorIndirect GVC exporter
Entreprise 2Producer/distributorDirect GVC exporter
Export of intermediate/final goods/services for:
Further production/distributionwithin GVCs(Forward GVC participation)
Final consumption
Intermediate/final goods/services
Source:WTOSecretariat.
41
B. S
ME
s IN IN
TER
NA
TION
AL
TRA
DE
: ST
YLIZ
ED
FAC
TSLEVELLING THE TRADING FIELD FOR SMES
Official enterprise surveys and business-related datasources such as Trade by Enterprise Characteristics(TEC), Services Trade by Enterprise Characteristic(STEC) or Structural and Demographic BusinessStatistics(SDBS)providerelevantinformationonSMEtradeandotherdomains likeproduction,employment,productivityorconsumptionbuttheydonotnecessarilycontaindetailstodelineatetheactualactivityofSMEsindirectexportsandwithinGVCs.
An alternative is to use the value added approachto trade, which allows the decomposition of grossexports into their domestic and foreign value addedcomponents, and tracking of trade in intermediatestaking place within GVCs. Currently, the OECD-WTO Trade in Value Added (TiVA) database providesestimatesonbackwardandforwardlinkagestoGVCsfor 61 reporters, 34 industries and seven historicalyears. For the time being, the global input-outputtable underlying TiVA and GVC participation datarelies on the hypothesis of the homogeneity of firmsand industries, meaning that all firms within a sameindustry are supposed to have the same productiontechnology and the same share of imported inputs.This does not match the wide variety of enterprisesengaged in GVCs (SMEs, multinational enterprises,processers,multinationalaffiliates).
An expert group on “Extended supply-use tables(E-SUT)” launched in2015by theOECD investigateswaystobetterreflecttheheterogeneityofenterprisesin the national Supply-Use Tables (SUTs) that areused to construct the global input-output table for
the TiVA database. The principle is to combine SUTswith business-related data sources, like TEC, SDBSor Foreign Affiliates Trade Statistics (FATS), to getE-SUTs that will expand the granularity of standardSUTsinseveraldomains(seeOECD,2015b).Basedonsuch developments, TiVA and related GVC indicatorswillbebrokendownby:
• Firm size (micro enterprises, SMEs, largeenterprises,multinationalenterprises).
• Ownership(domesticorforeign,usingFATS).
• Exportorprocessingintensity(companiesinvolvedornotinglobalproduction).
Figure B.12 presents the various data sources andproduction sequence that will be involved to producetrade invalueaddedandGVCstatisticsbyenterprisetype.
TheOECDcarriedoutexploratoryworktofigureoutthetypeof trade-in-value-added indicators thatmay resultfromthefutureextended-SUTs.TheexerciseconsistedinlinkingnationalbusinessstatisticsonSMEswiththeglobal input-output tables developed for the OECD-WTO TiVA initiative. The results were presented in anOECD-WorldBankGroup reportprepared for theG20TradeMinistersMeetingheld in Istanbulon6October2015(seeOECDandWorldBank,2015).
The contribution of SMEs to GVCs is broken downintodirectandindirectdomesticvalueaddedcontents
Figure B.12: Moving towards trade in value added and GVC participation by enterprise characteristics
National Supply-Use table (SUT)
Extended SUT (E-SUT)Bilateral trade matrices(merchandise and services)
TiVA and GVC indicators broken down by:– Entreprise size (SMEs, Large entreprises,...)– Entreprise type of ownership– Other variables
Trade by Entreprise Characteristics (TEC, trade in goods)
Services Trade by Entreprise Characteristics (STEC, trade in services)
Structural Business Statistics(SBS, employment data)
Foreign Affiliates Trade statistics(firm ownership)
Global Input-Output table
Source:WTOSecretariat.
WORLD TRADE REPORT 2016
42
of exports. The direct approach measures thecontributionmadebyanSMEinasectorofactivitytothe production of goods and services for export. Thenotion of indirect value added exports correspondsto the domestic value added originating from SMEsin upstream industries that provide inputs to theexportingindustry.
Formostof theOECDcountriescovered in the report,SMEs accounted for more than 50 per cent of thetotal domestic value added exports in 2009. Generallyspeaking, the direct contribution of SMEs to domesticvalueaddedexportsispredominantoverindirectexports.However, the proportion between direct and indirectexports varies greatly between industries. As shown inFigure B.13, the direct exports made by SMEs in themotor vehicles industryaremarginal,whereasSMEs inother domestic sectors (manufacturing and services)contributemuchmoretotheexportsofthis industrybyprovidingcomponentsorintermediateservicestomotorvehicle exporters. Indeed, the direct contribution ofSMEstoexportsofthebusinessservicesindustryoftenexceeded40percentofthetotaldomesticvalueadded
exportedbytheindustryin2009formostofthereviewedcountries (see Figure B.14). Overall, when cumulatingthedirectexportsofSMEswithupstreamsuppliesfromothersectors,SMEsturnouttobethemainexportersofbusinessservicesinmanyOECDcountries.
(b) IndirectexportsandGVCparticipationofSMEsindevelopingeconomies
TheWorldBankEnterpriseSurveysallow the indirecttrade and potential activity of SMEs within GVCs tobe quantified. This subsection exploits the availableindicators to establish stylized facts for SMEs indevelopingeconomies.
(i) Indirectexports
AccordingtoWTOestimates,indevelopingeconomies,the indirect exports in the manufacturing sector ofSMEs were estimated, on average, at 2.4 per cent oftotalsales,alevelthreetimeslowerthantheestimatedshare of direct exports. Indirect exports account for
Figure B.13: SMEs’ share of total domestic value added contained in exports of motor vehicles, 2009 (percentage)
0 10 20 30 40 50 60 70 80 90 100
Austria
Belgium
Czech Republic
France
Germany
Hungary
Italy
Mexico
Netherlands
Poland
Portugal
Spain
Turkey
United Kingdom
United States
SMEs direct exports(SMEs belonging to motor vehicle industry)
SMEs indirect exports(upstream suppliers from other industries to motor vehicle exporters)
Source:OECDestimates.
43
B. S
ME
s IN IN
TER
NA
TION
AL
TRA
DE
: ST
YLIZ
ED
FAC
TSLEVELLING THE TRADING FIELD FOR SMES
Figure B.14: SMEs’ share of total domestic value added contained in exports of business services, 2009 (percentage)
0 10 20 30 40 50 60 70 80 90 100
Austria
Belgium
Czech Republic
France
Germany
Hungary
Italy
Mexico
Netherlands
Poland
Portugal
Spain
Turkey
United Kingdom
United States
SMEs direct exports(SMEs belonging to business services industry)
SMEs indirect exports(upstream suppliers from other industries to business services exporters)
Source:OECDestimates.
Figure B.15: Shares of direct and indirect manufacturing exports by firm size in developing economies(percentage of total sales)
Indirect manufacturing exports
0
10
20
Per
cent
age
5
15
25
35
30
40
Direct manufacturing exports
SMEs Large firms(>100 employees)
2.4%
7.6%
12.6%
14.1%
Source:WTOestimatesbasedonWorldBankEnterpriseSurveys.
a much larger share of sales in large firms (14.1 percent), suggesting that they can adapt more easilyto product requirements, such as standards andcertification, made by other firms, or have a moreefficient network of intermediaries (see Figure B.15).Overall, in developing economies, SME participationin exports, direct and indirect, was estimated at only10 per cent of total manufacturing sales comparedwithsome27percentinlargerfirms.
SMEs in Developing Europe recorded the highestshare of indirect participation in exports, estimatedat around 9.3 per cent, followed by Developing Asia(3.7percent)andtheMiddleEast(2.4percent),whileAfricanSMEs,excludingLDCs,sawonly1percentoftheirtotalsalesexportedindirectly(seeFigureB.16).
At the product level, SMEs’ highest shares ofindirect exports were found in the manufacturingof various types of machinery, in the publishing andprinting industry and in paper and paper productsmanufacturing, as well as in the automotive industry,where international production is widely organized. Inallthesesectors,theshareofindirectexportsinSMEs’total sales largely outpaced that of large firms (seeFigure B.17). Large firms, by comparison, appeared
WORLD TRADE REPORT 2016
44
tobeheavilyengagedinthemanufacturingoftextilesand garments, office equipment and electronics,tobacco,glassandceramics.And,especially inLDCs,leathergoodsandfootwear.
Services SMEs in developing economies participatedmore in indirect exports than in direct exports.However,theiroverallparticipationinservicesexportsis marginal, at 4 per cent of total services sales. Itis interesting to note that large firms in developingeconomies supply services to foreign consumerspredominantlythroughdirectexports(seeFigureB.18).
(ii) GVCparticipation
The opportunities for SMEs in global value chainsare enormous. Participation in value chains exposesthem to a large customer/buyer base, as well asto opportunities to learn from large firms and fromengagingandsurviving in thehotlycontestedsectorsof the global marketplace. The penetration of globalvalue chains, however, also presents huge and oftendauntingchallengesforSMEs(ADB,2015).
Unfortunately,dataonSMEstradeinGVCsarescarce.Official business data sources, like TEC, STEC or
Figure B.16: SMEs’ shares of indirect exports in total sales in the manufacturing sector, by developing region and in LDCs(percentage of total sales)
0 5 10 15 20 25 30 35 40
Indirect exports (sold domesticallyto a third party that exports)
Direct exports
Percentage
DevelopingEurope
Developing Asia
MiddleEast
Latin Americaand the Caribbean
Africa
LDCs
Note:SMEs’sharesofnationalsalesareexcluded.SeenotesofFigureB.8fordetailsoncountrygroups.
Source:WTOestimatesbasedonWorldBankEnterpriseSurveys.
Figure B.17: Indirect exports by manufacturing sector and firm size in developing economies (percentage of total sales)
0
5
10
15
20
25
30
35
40
45
50
Man
ufac
turin
g (IS
IC 1
5-3
7)
Pub
lishi
ng, p
rintin
g,re
prod
uctio
n of
reco
rded
med
ia (I
SIC
22
)
Mac
hine
ry a
nd e
quip
men
t n.e
.s. (
ISIC
29
)
Pap
er a
nd p
aper
pro
duct
s (IS
IC 2
1)
Mot
or v
ehic
les
and
othe
r tra
nspo
rt e
quip
men
t (IS
IC 3
4-3
5)
Food
(IS
IC 1
5)
Leat
her g
oods
: lug
gage
, han
dbag
s,sa
ddle
ry, f
ootw
ear,
etc.
(IS
IC 1
9)
Text
iles
and
garm
ents
(IS
IC 1
7-1
8)
Rub
ber a
nd p
last
ics
prod
ucts
(IS
IC 2
5)
Toba
cco
(ISIC
16
)
Bas
ic m
etal
s (IS
IC 2
7)
Off
ice
and
elec
tric
al m
achi
nery
, rad
io, T
Van
d co
mm
unic
atio
n eq
uipm
ent,
med
ical
,pr
ecis
ion
and
optic
al n
.e.s
. (IS
IC 3
0-3
3)
Cok
e, re
fined
pet
role
um p
rodu
cts
and
nucl
ear f
uel (
ISIC
23
)
Furn
iture
and
oth
er m
anuf
actu
ring
(ISIC
36
)
Fabr
icat
ed m
etal
pro
duct
s,ex
cept
mac
hine
ry(IS
IC 2
8)
Che
mic
als
and
chem
ical
pro
duct
s (IS
IC 2
4)
Woo
d an
d w
ood
prod
ucts
(IS
IC 2
0)
Oth
er n
on-m
etal
lic m
iner
al p
rodu
cts
(ISIC
26
)
SMES (<100 employees) Large (+100 employees)
Note:BasedontheInternationalStandardIndustrialClassificationofAllEconomicActivities(ISIC),Rev.3.1.N.e.s.standsfor“notelsewherespecified”.
Source:WTOestimatesbasedonWorldBankEnterpriseSurveys.
45
B. S
ME
s IN IN
TER
NA
TION
AL
TRA
DE
: ST
YLIZ
ED
FAC
TSLEVELLING THE TRADING FIELD FOR SMES
Figure B.18: Shares of direct and indirect services exports by firm size in developing economies(percentage of total sales)
0
10
20
Per
cent
age
5
15
25
35
30
40
Indirect manufacturing exportsDirect manufacturing exports
SMEs Large firms(>100 employees)
0.9%
2.6%
31.9%
4.2%
Source:WTOestimatesbasedonWorldBankEnterpriseSurveys.
SDBS,donotalwayscoverGVCactivityandtheyfocuslargelyondevelopedeconomies.
The World Bank Enterprise Surveys shed light onSMEs’ potential activity within GVCs in developingeconomies. The indicator within the EnterpriseSurveys, “Percentage of material inputs and/orsupplies of foreign origin” refers to the upstreamlinkages that SMEs set up with foreign partners toget inputs for their production and related exports.This indicator is used as a proxy for the backwardparticipation inGVCs.Averagebackwardparticipationwascalculatedastheaverageofforeigninputsusedinthemanufacturingprocess foreacheconomy,by firmsizeandbymanufacturingsector.
On the supplier side, two indicators are combined toapproximate SMEs’ forward participation to GVCs,namely the “Sales exported directly as percentageof total sales” and the “Sales exported indirectly aspercentage of total sales”. However, such indicatorspresentsomelimitsinoutliningtheactualroleofSMEsin GVCs, as they do not give information on the end-use category of the exported goods and services.Although no distinction is made between exports ofintermediategoodsandservicesthatarefurtherusedalongtheproductionchain,andproductsdedicatedtofinalconsumption,thesetwoindicatorsareretainedtoestimate the potential of SMEs’ downstream linkagestoGVCs.
Over 33,000 surveyed establishments engaged inthe manufacturing sector in developing economiesreported the values of total sales, largely in localcurrencies, and their percentage breakdown intonational sales, direct exports, and indirect exports.Values of direct exports and of indirect exports werecalculated for each establishment then convertedinto US dollars. Direct exports and indirect exportsdatawere thenaggregatedtoprovideaveragesharesof direct exports and indirect exports for individualeconomies,furtherbrokendownbysizeoffirmandbymanufacturingsector.
According to WTO estimates, SMEs in themanufacturingsectorindevelopingeconomiesarenotactivelyengagedinGVCs.Participationismainlydrivenbyupstreamlinks(backwardparticipation),withSMEsimportinginputsneededinthemanufacturingprocessfrom abroad. However, only a limited part of SMEs’production is exported to foreign countries, whetherdirectlyorindirectly.AsshowninFigureB.19,thevastmajorityofSMEsindevelopingeconomiesarelocatedin the bottom left quadrant, suggesting a low GVCparticipation(lowbackward/forwardparticipation).
The low levels of integration of SMEs into GVCs areevidentespeciallyifcomparedwithlargemanufacturingfirms (Figure B.20). In Developing Asia and in LatinAmericaandtheCaribbean,largefirmsareintegratedinto GVCs, as shown by some economies’ very highvaluesofbackward/forwardparticipation.Bycontrast,SMEs in the region have a low forward participation,with most countries concentrated in the bottom-leftsquare in the chart, suggesting that they are not yetinvolved in GVCs. SMEs in Developing Asia also useonaveragefewerinputsofforeignorigin(FigureB.21).ThiscanbeexplainedbythefactthatAsia’sindustrialnetwork is more advanced than in other developingregions.Asianfirmsarethemselvesthemanufacturersof inputs/intermediate products, for foreign firms inparticular, in developed economies. Necessary inputsare largely availabledomestically and sodonot needtobeimportedfromabroad.
Estimates suggest that in Africa, it is not onlySMEs but also large firms that do not benefit fromparticipation in GVCs. Both SMEs and large firmsin several African economies show high backwardparticipation.Comparedwithotherregions,theyimportalargeshareofinputsfromforeigncountriesinorderto be able to manufacture their products. However,their forwardparticipation is the lowestacrossall thedeveloping regions. A sectoral analysis shows that,in general, SMEs’ poor integration in GVCs affectsall manufacturing industries, with the exception ofthe furniture-making sector, in which SMEs in LDCshave a high share of direct exports (as shown in the
WORLD TRADE REPORT 2016
46
previous subsection). By contrast, large enterprisesare relatively more connected to GVCs in severalsectors, in particular in the textiles and garmentsindustryandinthemanufacturingofofficeequipmentsuch as computers and electronic products. In theseindustries, developing economies have high levels offorwardparticipationinGVCs.Largefirmsalsoshowagoodlevelof integrationinGVCsintheleathergoodsmanufacturingindustry.
Figure B.20 also shows average backward/forwardparticipation in GVCs by firm size, manufacturingsector and ownership. FDI plays an important rolein firms’ integration to GVCs, whether small or large.Estimates show that foreign-owned SMEs have morelinkages to GVCs than domestic SMEs. These firmsimport more inputs to be used in the manufacturingprocess than domestic SMEs, showing higher levelsof backward participation in GVCs. In addition, theycan export a much larger share of their production(forward participation), and this applies to almost allmanufacturingexports.Forexample,intheautomotivesector, direct and indirect exports accounted for over40 per cent of SMEs’ total sales, while in domesticSMEs the share was around10 per cent. Similarly, inthe furniture manufacturing industry, which recordedthe highest share of direct export to total sales, thecontribution was essentially made by foreign-ownedSMEs.
3. SMEparticipationininternationale-commerce
The development of electronic commerce as ameans for firms to reach customers in overseasmarkets promises to dramatically expand exportopportunitiesforSMEsifcertainobstacles–includingthose related to information and communicationstechnology (ICT) infrastructure, and to the legal andregulatory environment, discussed in Section D.4– can be overcome. Retail businesses and serviceproviders such as Amazon, eBay, PayPal and othersnow provide platforms and payment systems thatfacilitate exports by even the smallest firms. Digitaltechnologies reduce trade costs for SMEs and givethem a global presence that was once reserved forlargemultinational firms,allowingsmallbusinesses tocompete directly with larger companies. Some of theservices that the Internet-based technologies havemade more accessible to SMEs include shipping/logistics, international payments, translation services,customerservicesandmarketresearch.
Thissection reviewsavailableevidenceonSMEtradeenabledbyinformationtechnology.Forthepurposesofthis report, e-commerce is defined as the production,advertising,saleanddistributionofgoodsandservicesvia telecommunication networks such as the Internet.
Figure B.19: SMEs in developing economies: backward and forward participation in GVCs(share in total sales and share in total inputs, percentage)
0 10 20 30 40 50 60 70 80 90 1000
10
20
30
40
50
60
70
80
90
100
Dire
ct a
nd in
dire
ct e
xpor
ts (f
orw
ard
part
icip
atio
n)
Use of foreign inputs in manufacturing (backward participation)
Hig
h G
VC
part
icip
atio
nH
igh
back
war
d G
VC
part
icip
atio
n
Hig
h fo
rwar
d G
VC
part
icip
atio
nLo
w G
VC
part
icip
atio
n
Note:EachsquarerepresentstheaverageGVCparticipationofSMEsinagivendevelopingeconomy.
Source:WTOestimatesbasedonWorldBankEnterpriseSurveys.
47
B. S
ME
s IN IN
TER
NA
TION
AL
TRA
DE
: ST
YLIZ
ED
FAC
TSLEVELLING THE TRADING FIELD FOR SMES
Figure B.20: SMEs and large enterprises: backward and forward participation in GVCs by region, ownership and manufacturing sector(share in total sales and share in total inputs, percentage)
Source:WTOestimatesbasedonWorldBankEnterpriseSurveys.
Dire
ct a
nd in
dire
ct e
xpor
ts (f
orw
ard
part
icip
atio
n)
Use of foreign inputs in manufacturing (backward participation)
Large Linear (large)SMEs Linear (SMEs)
00 10 20 30 40 50 60 70 80 90 100
10
20
30
40
50
60
70
80
90
100
Low GVCparticipation
High backward GVCparticipation
High forward GVCparticipation
High GVC participation
DevelopingAsia
Low GVCparticipation
High backward GVCparticipation
High forward GVCparticipation
High GVC participation
Dire
ct a
nd in
dire
ct e
xpor
ts (f
orw
ard
part
icip
atio
n)
Use of foreign inputs in manufacturing (backward participation)
Large Linear (large)SMEs Linear (SMEs)
00 10 20 30 40 50 60 70 80 90 100
10
20
30
40
50
60
70
80
90
100
DevelopingEuropeandtheMiddleEast
Low GVCparticipation
High backward GVCparticipation
High forward GVCparticipation
High GVC participation
Dire
ct a
nd in
dire
ct e
xpor
ts (f
orw
ard
part
icip
atio
n)
Use of foreign inputs in manufacturing (backward participation)
Foreign-owned SMEs Domestic SMEs
00 10 20 30 40 50 60 70 80 90 100
10
20
30
40
50
60
70
80
90
100
Furnitureand other
manufacturing
Automotive
Officeequipment
and electronics
Chemicals,petroleum,
plastic
SMEsbysectorandownership
Low GVCparticipation
High backward GVCparticipation
High forward GVCparticipation
High GVC participation
Dire
ct a
nd in
dire
ct e
xpor
ts (f
orw
ard
part
icip
atio
n)Use of foreign inputs in manufacturing (backward participation)
Large Linear (large)SMEs Linear (SMEs)
00 10 20 30 40 50 60 70 80 90 100
10
20
30
40
50
60
70
80
90
100
LatinAmericaandtheCaribbean
Low GVCparticipation
High backward GVCparticipation
High forward GVCparticipation
High GVC participation
Dire
ct a
nd in
dire
ct e
xpor
ts (f
orw
ard
part
icip
atio
n)
Use of foreign inputs in manufacturing (backward participation)
Large Linear (large)SMEs Linear (SMEs)
00 10 20 30 40 50 60 70 80 90 100
10
20
30
40
50
60
70
80
90
100
Africa
Low GVCparticipation
High backward GVCparticipation
High forward GVCparticipation
High GVC participation
Dire
ct a
nd in
dire
ct e
xpor
ts (f
orw
ard
part
icip
atio
n)
Use of foreign inputs in manufacturing (backward participation)
Foreign-owned SMEs Domestic SMEs
00 10 20 30 40 50 60 70 80 90 100
10
20
30
40
50
60
70
80
90
100Textiles and
garments
Leather
Officeequipment and
electronics
Largefirmsbysectorandownership
WORLD TRADE REPORT 2016
48
E-commercecanbebrokendown into sales (e-sales)and purchases (e-purchases). In its survey on ICTusageinenterprisessurvey,theEuropeanUnionfindsthat purchases by companies are twice as frequentas sales.5 This section discusses cross-border onlinesales, as opposed to domestic online sales. It shouldbeemphasizedfromtheoutsetthatmoste-commercetodayisreportedtobedomesticcommerce,especiallyinlargeeconomies(McKinseyGlobalInstitute,2013a).Cross-border online transactions, as a share of totalonline sales to consumers, are significantly larger insomedevelopingcountries(e.g.morethan50percentin India and Singapore) than in developed countries(e.g.20percentforCanadaand18percentforJapan)(McKinseyGlobalInstitute,2013a).
The Internet has proved to be significantly moreamenabletoSMEsthanprivatebusinessnetworksthatpredated it. The United Kingdom Office for NationalStatistics has estimated that between 2009 and2013, SMEs web-based sales increased five timesfasterthansalesviaEDI(electronicdatainterchange)systems.eBayalsopublishedaseriesofstudies(eBay,2012; 2014; 2016) using data covering transactionson the eBay Marketplace since 2010. To ensure thatthe community of small commercial enterprises isproperly captured, and that small individual sellersare excluded, the data are limited to transactions bysellerswithannualsalesofmorethanUS$10,000(orlocal currency equivalents) on the eBay marketplace.Thesefirmsarereferredtoas“commercialsellers”,orsmallonlinebusinesses.Toallowforcomparisonswith“traditional”, non-Internet-enabled SMEs, eBay has
useddatafrompubliclyavailablesourcesincludingtheWorld Bank, Eurostat, and various national statisticalagencies.
Broadly, these studies find that the vast majority oftechnology-enabled small firms export: 97 per centof them on average, and up to 100 per cent in somecountries. By comparison, only a small percentage oftraditionalSMEsexports (between2percentand28percentforallcountriesexceptItalyandThailand,seeFigureB.22).NotonlydoInternet-enabledcommercialSMEs export at a high rate, they also reach a largenumber of foreign destinations. For example, SMEsin China typically export to 63 countries, and KoreanSMEstypicallyexportto57countries(FigureB.23).6
One difference between exporting SMEs and largeexporters is that shipments from SMEs are often oflow volumes and frequently consist of single itemsshippedthroughtraditionalmailorbyexpressdeliverycompanies. The rapid growth in shipments of parcelsby post offices (Figure B.24) could signify growingshipments by SMEs. Growth has been fastest indeveloped countries (average annual growth of morethan 10 per cent since 2005), but negative in Africa(-3.1percent)andstandsat0percentinAsiaandthePacificandinLatinAmerica.Onepossibleexplanationfor the low rates of postal delivery of packages inAfrica, Asia and Latin America is that shipments inthese regions may be conducted by express deliverycompanies and cost more than traditional mail. The40 per cent rise in the index of international expressdelivery volumes registered by the Global Express
Figure B.21: Use of foreign and domestic inputs in production of SMEs by developing region(percentage)
0
10
20
30
40
50
60
70
80
Domestic origin (% total inputs) Foreign origin (% of total inputs)
Middle East Africa Latin America and the Caribbean Asia
Source:WorldBankEnterpriseSurveys.
49
B. S
ME
s IN IN
TER
NA
TION
AL
TRA
DE
: ST
YLIZ
ED
FAC
TSLEVELLING THE TRADING FIELD FOR SMES
Association (DHL, FedEx, TNT, and UPS) between2008and2013issuggestiveevidenceinthisregard.
Online buying and selling are relevant to trade inboth goods and services. Even when trade in goodsis involved, services also play a role. Online facilities,even those primarily offering merchandise, are alsoa form of retailing service. Moreover, online trade is
naturally relevant for services that can be deliveredelectronically. This encompasses such activities asprofessionalservices,businessprocessing,backofficeservicesanddigitalproductssuchassoftware,music,films, e-books and consultant reports. With the offerofonlinereservations,ticketing,trackingandcustomerservice, tourismwasamong the first services sectorsthatengagedsignificantlyinonlinebusiness.Asshown
Figure B.22: Share of eBay-enabled and traditional SMEs(percentage)
0
10
20
30
40
50
60
70
80
90
100
Italy
United
King
dom
Thail
and
Jord
an
Turke
yChin
aSpa
in
Repub
lic of
Kor
ea
Colombia
South
Africa
Chile
France
German
yBraz
il
Ukraine Ind
iaPer
u
Mexico
United
Stat
es
Indon
esia
Austra
lia
Canad
a
Traditional businesses eBay
Source:DataforallcountriesweresourcedfromeBay(2016)exceptJordan,PeruandUkraine,whichweresourcedfromeBay(2012),andTurkey,whichwassourcedfromeBay(2014).
Figure B.23: Number of export destinations of eBay-enabled SMEs
0
10
20
30
40
50
60
70
China
Thail
and
Repub
lic of
Kor
ea
Indon
esia
Ukraine
Turke
y
South
Africa
Brazil
Chile
Jord
anPer
uSpa
in
Mexico
Canad
aInd
ia
France
United
King
dom
United
Stat
es
Colombia Ita
ly
Austra
lia
German
y
63
57
46
3735 34
29 28 28 2825 24
21 20 20 20 19 18 1816
1412
Source:DataforallcountriesweresourcedfromeBay(2016)exceptJordan,PeruandUkraine,whichweresourcedfromeBay(2012),andTurkey,whichwassourcedfromeBay(2014).
WORLD TRADE REPORT 2016
50
inacasestudyon thesector inEgypt (KamelandElSherif,2001),SMEsparticipatedinthistrend.7
An increasing number of e-commerce platforms areset up or adapted with the specific goal of assistingSMEs or even individual sellers, such as freelancersor designers of arts and crafts.8 For example, Etsy,an online market for artisans and small producers,recorded US$ 2 billion in sales in 2014, with morethan one-third of those representing internationalsales (McKinsey Global Institute, 2015). Large retailplatformsandserviceproviderssuchasAmazon,eBayand PayPal now provide or are developing ancillaryservicesandpaymentsystemsto facilitateexportsbyeven the smallest sellers. Such online marketplacescanofferSMEsameans to scaleupatminimal cost,providing nearly instant solutions that include securepaymentsystems,logisticssupport,andglobalvisibilityofthekindoncereservedforlargefirms.
Another promising development for SMEs engaging inall types of business activities is the growing numberof independent commercial business-to-business(B2B) trade platforms. In the infancy of e-commerce,those that initially emerged were usually corporateprocurementportalsby largemultinationals, permittingsellers to bid. The new models, however, offer sellersthepossibilitytomarkettheirwarestootherbusinesses,and frequently tend to offer a wider range ofgoods and services than consumer-oriented platforms.Indiamart.com and Tradekey.com are two examples,offering on their websites a host not only of businesssupplies and equipment, but also a range of business,professionalandfinancial(e.g.insurance)services.
Online sellers can also benefit from the possibility toanalyse large volumes of data that is available fromweb-basedapplications,often referred toasBigData.
By turning a series of discrete snapshots into a moreholistic view of customer’s behaviour and motivation,Big Data analysis can significantly boost online sales(VanBommeletal.,2014).Suchservicesarebecomingmoreaffordableforstart-upsandSMEs(OECD,2015a).Some e-commerce platforms offer such data to theirsellers, while analytics software, often combined withcloud processing and storage, is also available forcompaniesthatsellviatheirownwebsites.
Eveninsituationswhereformaltradingplatformsarenoteasilyaccessibleoraffordable,socialmediaareplayinganimportantroleinSMEtrade.Somecommercialtradeplatforms, for example, require sellers tobe registeredbusinesses, whereas on social media, informal microenterprises and even individual entrepreneurs canoperate.Suchsitesmayalsobemorereadilyaccessibleby means of mobile technologies for keeping in touchwith customers as well as securing and organizingsales.ResearchbytheMcKinseyGlobalInstitute(2016)showsthatthenumberofSMEswithaFacebookpageis growing, from 25 million in 2013 to 30 million in2014and50million in2015.While local followersarecurrently in themajority,cross-border foreignexposureis significant (at 30 per cent). For example, more than20,000 independent designers and artists showcasetheir work on Pinkoi, an online marketplace basedin Chinese Taipei. The company has connected withcustomers in more than 47 countries, using Facebooktoexpand its reach throughout theAsia-Pacific region(McKinseyGlobalInstitute,2016).
Despite thepromisesofe-commerce,SMEscontinueto be less well represented online than largerenterprises. One reason for this is the requirementsinvolved in establishing a retail-ready website, whichis a very important condition for facilitating onlinesales. In the United Kingdom (where 70 per cent of
Figure B.24: Index for worldwide number of ordinary parcels, domestic and international service, 2000-2014
90
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 20142013
210
270
240
180
150
120
Domestic service International service
Note:2000=100.
Source:UniversalPostalUnion.
51
B. S
ME
s IN IN
TER
NA
TION
AL
TRA
DE
: ST
YLIZ
ED
FAC
TSLEVELLING THE TRADING FIELD FOR SMES
individuals purchase products online, more than anyother country according to Figure II.3 in UNCTAD,2015) in 2013, nearly 77 per cent of firms with 49or fewer employees had a website, whereas nearly99 per cent of the largest firms surveyed had one(seeTableB.1).9FewerSMEs indevelopingcountrieshave a website, as shown in Table B.2. In a dynamicperspective,TableB.3reportstheshareofenterprisesreceivingordersover the Internet,and itsgrowth ratebetween 2010 and 2014. The table confirms thatSMEs persistently rank behind larger firms in termsof online retail, despite moving toward online retail inmosteconomies.
4. MSMEtradeparticipationovertime
Very limited information is available on the evolutionof MSME trade, either direct or indirect, in bothdeveloped and developing economies. No strongtrendeitherupordowninexportparticipationratesofMSMEs(0-250employees)canbediscernedfromthe
OECDTECdatabase,althoughslightlymorethanhalfofcountriesrecordedincreasesoverarelativelyshortperiodof lessthan10years, includinglargecountriessuch as France and the United States (Figure B.24).Meanwhile, among developing countries and LDCscovered in the World Bank Enterprise surveys, onecanobserveamoderategrowthofexportsfromSMEs(5-100 employees) between the first survey and themost recent one (see Figures B.25, B.26 and B.27).However, it isnotpossible to inferasignificant trend,assampledataaretooheterogeneous,havingdifferentbenchmarkyearsdependingoncountries.
Among LDC countries (see Figure B.26), Tanzaniarecorded the highest increase in exports for smallenterprises between the two survey periods, with theshare of exporting small enterprises moving from 2.8percentin2006to11percentin2013.Thisevolutionis particularly due to the rise of indirect exports,most likely through larger enterprises. In general, thesample suggests that medium-sized enterprises inLDCs export more than smaller ones, with a greaterincrease in indirect exports between the two survey
Table B.1: Proportion of businesses in the United Kingdom with a website, by size of business, 2007-2013(percentage)
Year Employment size
10-49 employees 50-249 employees250-999
employees1,000+ employees All size bands
2007 65.8 89.3 94.4 97.6 70.0
2008 70.6 91.3 95.2 97.9 74.5
2009 72.0 91.9 96.9 98.3 75.7
2010 75.3 92.3 96.0 98.7 78.5
2011 78.7 93.6 96.2 98.6 81.4
2012 77.6 92.9 95.7 98.7 80.3
2013 76.6 94.9 95.8 98.6 79.7
Source:UKOfficeforNationalStatistics.
Table B.2: Proportion of businesses in developing economies with a website, by size of business (percentage)
Employment size
0-9 employees 10-50 employees 51-100 employees101-250
employees251+ employees
Developing 22.75 43.94 67.25 75.11 84.79
G20developing 32.33 52.8 72.88 81.37 88.93
Otherdeveloping 23.62 43.79 65.88 73.66 84.88
LDCs 12.33 27.25 53.44 58.08 71.64
Source:WorldBankEnterpriseSurveys(lastavailablesurveypercountry),authors’owncalculations.
WORLD TRADE REPORT 2016
52
Table B.3: Proportion of businesses receiving orders over the Internet(percentage)
10-49 employees 50-249 employees +250 employees
Economy 2010 2014%
change2010 2014
% change
2010 2014%
change
Countries with low proportion of orders received by firms with 10-49 employees in 2010
Bulgaria 3.3 7.9 144% 6.0 11.5 92% 7.3 12.2 67%
Cyprus 5.9 8.7 46% 15.1 26.1 73% 14.8 33.5 126%
Estonia 8.7 12.5 44% 18.6 21.3 15% 29.7 32.0 8%
Greece 8.3 9.3 11% 13.0 18.7 44% 20.0 20.5 3%
Hungary 7.9 11.5 46% 12.9 17.9 39% 23.6 31.1 32%
Italy 4.5 7.3 62% 8.0 13.1 63% 16.6 26.0 57%
Latvia 6.2 7.5 21% 10.0 14.8 48% 15.5 22.8 47%
Poland 7.3 10.1 39% 11.8 16.1 37% 24.5 34.4 40%
Romania 6.4 7.4 15% 6.4 8.9 39% 8.2 15.1 84%
SlovakRepublic 7.1 12.3 72% 10.0 19.5 95% 14.7 29.0 97%
Slovenia 10.0 15.1 50% 18.8 27.6 47% 39.0 50.2 29%
FYROM 3.7 6.8 87% 6.7 7.3 9% 6.0 10.7 77%
Countries with medium proportion of orders received by firms with 10-49 employees in 2010
Austria 15.0 14.9 -1% 28.1 27.1 -3% 46.0 45.7 -1%
Belgium 25.9 22.1 -15% 42.6 32.8 -23% 53.3 49.1 -8%
Croatia 24.4 24.2 -1% 23.2 32.5 40% 29.9 51.9 74%
CzechRepublic 18.8 26.1 39% 25.1 30.4 21% 38.7 45.1 17%
Denmark 27.8 25.8 -7% 33.6 35.4 5% 49.9 53.4 7%
Finland 15.8 15.4 -3% 30.9 32.4 5% 48.7 49.5 2%
France 12.3 12.4 1% 21.5 26.3 22% 34.7 44.2 27%
Germany 21.4 23.6 10% 30.2 30.3 0% 45.1 45.6 1%
Iceland 16.3 29.4 80% 34.7 53.8 55% 51.8 53.1 3%
Ireland 18.1 20.4 13% 33.6 40.4 20% 34.8 45.6 31%
Lithuania 21.3 17.9 -16% 24.0 24.6 2% 27.6 29.8 8%
Malta 15.1 15.9 5% 26.0 29.1 12% 27.8 30.7 10%
Netherlands 21.4 21.8 2% 30.5 31.5 3% 42.0 39.9 -5%
Portugal 18.3 12.6 -31% 24.8 24.3 -2% 36.5 40.6 11%
Spain 11.8 16.8 43% 18.9 26.6 41% 29.0 36.4 26%
Sweden 21.8 23.3 7% 39.6 39.7 0% 54.7 53.1 -3%
UnitedKingdom 14.3 19.3 36% 28.0 29.5 5% 43.8 47.9 9%
Countries with high proportion of orders received by firms with 10-49 employees in 2010
Indonesia 26.1* 35.8 37% 57.8* 46.9 -19% 58.6* 54.6 -7%
Mauritius 35.7 35.4* -1% 36.1 47.4* 31% 46.2 56.2* 22%
Norway 36.3 26.2 -28% 45.7 37.5 -18% 50.4 44.9 -11%
Singapore 42.0 56.3* 34% 61.0 62.9* 3% 57.1 69.2* 21%
Notes:* indicates2013data.Onlyeconomiesforwhich timeseriesdata isavailableare included.FYROMis theformerYugoslavRepublicofMacedonia.
Source:Author’selaborationbasedonUNCTAD(2015)andadditionaldatafromUNCTAD.
53
B. S
ME
s IN IN
TER
NA
TION
AL
TRA
DE
: ST
YLIZ
ED
FAC
TSLEVELLING THE TRADING FIELD FOR SMES
periods. This is an indication of the rising integrationinto domestic value chains and maybe even globalvaluechains,especiallywhenenterprisesactas localsuppliers of foreign-owned corporations. Despite theincreasedcontributiontointernationaltradeovertime,theprogressionnoticedfordirectandindirectexportsvarieswidelyaccordingtocountries.
According to estimates based on the World BankEnterprise Surveys, covering over 3,000 large firms
which started as SMEs in 85 developing economies,there isanegativecorrelationbetween the initialsizeoffirmsandthenumberofyearstheywereinexistencebeforetheystartedtoexport.Inthecaseoflargefirmswhichstartedasmicro firms (one to fouremployees),it took on average 17 years before they exported,slighly less when the number of employees rangedbetweenfiveandten.Thenumberofyearsdrasticallydecreasedforfirmswhichstartedwithaprogressivelyhighernumberofemployees(seeFigureB.27).
Figure B.25: Share of MSMEs in exports of selected developed economies, 2005 and 2013(percentage)
0%
10%
20%
30%
40%
50%
60%
70%
80%
Austria
Canad
a
Czech
Rep
ublic
Denmark
Finlan
d
Estonia
France Ita
ly
Hunga
ry
Poland
Luxem
bour
g
Slovak
Rep
ublic
Portug
al
Sloven
ia
United
Stat
es
Sweden
2005 2013
Source:OECDTradebyEnterpriseCharacteristics(TEC)database.
Figure B.26: Firms in LDCs that export directly and indirectly at least 1 per cent of total sales, by size of firm(percentage of total firms)
Per
cent
age
of to
tal f
irms
Firms exporting directly (at least 1% of total sales) Firms exporting indirectly (at least 1% of total sales)
0
10
5
15
25
20
30
First survey: Small firms(5-19 employees)
Second survey: Small firms(5-19 employees)
First survey: Medium firms(20-99 employees)
Second survey: Medium firms(20-99 employees)
Note:Firstsurveyconductedbetween2006and2010.Secondsurveyconductedbetween2011and2014.
Source:WorldBankEnterpriseSurveys.
WORLD TRADE REPORT 2016
54
OnaverageittooklesstimeforanSMEinDevelopingAsia to start exporting compared with a firm of thesame size in Africa or in Latin America. The longesttime lag to export was found in the food sector, onaveragemorethan14years,twicethetimenecessarytobeginexportinginthetextilesandgarmentsorofficeequipment and electronics manufacturing sectors.This applied to all developing regions, suggestingthat SMEs in the food sector encounter additionaldifficultiestoexportiftheydonotcomplywithsanitaryandphytosanitary(SPS)andtechnicalbarrierstotrade(TBT)standards(seeSectionD).
5. Conclusions
This section has surveyed statistical evidence onthe participation of micro, small and medium-sizedenterprisesininternationaltrade.Ithasfoundthattheshare of exporting SMEs is small when compared tothatoflargefirms,andthatthecontributionofSMEstototalexportsandimportsislow.However,considerableheterogeneityexistsacrossenterprisesizeclasses,aswellasalongotherdimensions.Indevelopedcountriesin particular, the trade participation of medium-sizedenterprisesmayapproachthatoflargefirms,whereassmall and micro enterprises are less active in trade.
Meanwhile, SMEs in developing countries have lowparticipationratesinbothdirectandindirectexporting,andSMEswithfeweremployeestakelongertoaccessinternationalmarketsthanlargerfirms.
Internet-enabled SMEs are an exception to the ruleof low trade participation, with very high rates ofexporting approaching 100 per cent. The spread ofonlineplatformspromisestogivesmallenterprisestheability to reach customers around the world. Reportsfrom eBay find that, while only a small fraction oftraditionalSMEs(between4and28percent)engagein exports, nearly all “Internet-enabled” SMEs do(97percent).
Since data limitations make it difficult to capture thefullextentofindirecttradeandGVCtradeparticipation,this section of the report has also outlined new andbetter approaches to measuring SMEs’ contributionsto GVC trade in value added terms. The integrationof SMEs in developing countries into global valuechainsisstillrelativelylimitedforreasonsdiscussedinSectionsCandDofthisreport,butnewopportunitiesarebecomingavailable.
Figure B.27: Time lag between firms’ start of operations and engagement in exports by selected firm size in developing economies(years and number of employees)
0123456789
1011121314151617181920
Yea
rs fr
om s
tart
of o
pera
tions
unt
il th
e fir
m e
ngag
ed in
exp
orts
1 to 4(micro firms)
6 to 10 11 to 20 21 to 30
Number of employees when firm started operations
31 to 40 41 to 50 51 to 60 61 to 99
Source:WTOcalculationsbasedonWorldBankEnterpriseSurveys.
55
B. S
ME
s IN IN
TER
NA
TION
AL
TRA
DE
: ST
YLIZ
ED
FAC
TSLEVELLING THE TRADING FIELD FOR SMES
Endnotes1 Forotherdefinitionsofinternationalization,seeforexample
Beamish(1999),Karlsenetal.(2003)orZengetal.(2008).
2 TheOECDTradebyEnterpriseCharacteristics(TEC)databaseprovidesinformationonthevalueofexportsandimportsandthenumberoftradingenterprisesin32mostlydevelopedcountries(28EUmembersplusCanada,Norway,TurkeyandtheUnitedStates)brokendownbysector,sizeclassandpartner.Figuresareproducedbynationalstatisticalagenciesbylinkingtransactionsdatainmerchandisetradestatisticstobusinessregistries.Notethatininternationaltradestatistics,firmsizeisgenerallydefinedattheenterpriselevel,althoughtheseenterprisesmaystillbepartofalargerenterprisegroup.
3 TheWorldBank’sEnterpriseSurveyscollectdatafromkeymanufacturingandservicesectorsineveryregionoftheworld.Thesurveysareconductedaccordingtotheglobalsamplingmethodologywhichusesstratifiedrandomsamplingtominimizemeasurementerrorandtoyielddatathatarecomparableacrosseconomies.Thesamplingmethodologygeneratesasamplerepresentativeofthewholenon-agriculturalprivateeconomy,includingservicesindustries,andgenerateslargeenoughsamplesizesforselectedindustriestoconductstatisticallyrobustanalyseswithlevelsofprecisionataminimumof7.5percentfor90percentconfidenceintervals.Thismeansthatthepopulationparameteriswithinthe7.5percentrangeoftheobservedsampleestimate,exceptin10percentofthecases.
4 Around33,800small,medium-sizedandlargefirmssurveyedbytheWorldBankreportedtheamountoftheirtotalsalesandtheirbreakdownintonationalsales,directexportsandindirectexports(soldthroughadomesticpartythatexports).TheWTOSecretariatclassifiedeachestablishmentasamanufacturingorservicesenterpriseonthebasisofthereportedmainproduct/sectorcodeaccordingtotheInternationalStandardIndustrialClassificationofAllEconomicActivities(ISIC),Rev.3.1.Inthesurveys,themainproduct/sectorwastheonethatrepresentedthelargestproportionofannualsales,which,followingcalculations,accountedonaverageformorethan83percentofannualsalesofmanufacturingSMEsandfor81percentofservicesSMEs.Thisinformationwascorroboratedbythedescriptionofthemaintwoproducts/sectorsofactivityasreportedbyeachestablishment.Itshouldbenotedthatonly17percentofallestablishmentssurveyedbytheWorldBankindifferentcountriesandindifferentyearswerepartoflargerfirms;thebulkwerestand-alonefirms.
5 TheEuropeanUnionCommunitysurveyonICTusageande-commerceinenterprisesisanannualsurveyconductedsince2002,collectingdataontheuseofinformationandcommunicationtechnology,theInternet,e-government,e-businessande-commerceinenterprises.
6 RecentresearchfocusingonUSfirms(Lendleetal.,2013)hasalsofoundthatexportsarelessconcentratedinonlineexportersthaninofflineones.Thetop10percentofUSonlineexporterscapturelessthan70percentofexports,whereasofflinethetop10percentofUSofflineexporterscapturemorethan85percentofexports.Thisstudy,however,doesnotfocusonSMEs.
7 KamelandElSherif(2001)arguethate-commerceofferedEgyptianSMEsinthetourismindustry“acompetitivetooltoincreaseprofitabilityusingthewebtechnologyasapromotion,marketing,andsellingtool,withanimmediateeffect”byreducingdependencyoncostlytravelintermediariesandattractingreservationsfromaroundtheworld.
8 GovernmentsandinternationalorganizationshavealsosetupplatformswiththegoaloffacilitatingparticipationbySMEsine-commerce.Analysingtheeffectivenessofgovernment-sponsoredplatformsinAustralia,Gengatharen(2006)notesthat,insomecasesexamined,itwasonlyaftersixyearsofoperationthatsomeSMEsbegantoexperienceeconomicbenefits.Asaresult,thestudystressesthatthefactorscontributingtosuccessfullybenefittingSMEsincludeacommitmenttoalonggestationperiod,andcommensuratefunding,notonlyforthedevelopmentandmaintenanceoftheplatformsbutalsofortheirevaluation,aswellasrecognitionoftheneedtodevotespecificattentiontobuildingthee-competenciesofSMEsthemselves.Foranoverviewofe-commerce-relatedinitiativesestablishedbytheICT,seeSectionDofthisreport.
9 TableB.1,however,alsoshowsthatthesmallestfirmsincreasedownershipofawebsitepresencebytenpercentagepointsbetween2007and2013,andfirmswith50to249employeesbyfivepercentagepoints.