WPIA Meeting - OECD. Paris Oct2016

Intangible assets and participation in global value chains:an analysis on a sample of European countries

C. Jona-Lasinio, (ISTAT and LUISS Lab), RomeS. Manzocchi (LUISS), RomeV. Meliciani (LUISS), Rome

WPIA Meeting - OECD10-11 October 2016, Paris, France

This project has received funding from the European Union’s Seventh Framework Programme forresearch, technological development and demonstration under grant agreement No. 612774

C. Jona-Lasinio ISTAT and LUISS Lab of European Economics 1 / 24

Background

• International fragmentation of production, which implies thatcountries specialise in portions of the value chain and trade otherportions of it, has led to widespread processes of globalisation ofvalue chains (GVC) over the past two decades (for recent reviews,see Kaplinsky 2013; De Backer and Miroudot 2013; Timmer et al.2012).

• Baldwin (2011) has defined these as a "second unbundling" ofglobalisation, which has transformed the terms of internationalcompetition and shifted the barycentre of the world’s globalheadquarters and peripheries.

• While the international fragmentation of production has allowedmore countries to be involved in the production of a final good,not all countries have retained the same benefits from suchprocess.


Background

• A growing number of studies have pointed out that gains areunevenly distributed across the value chains (Gereffi, 1994;Kaplinsky, 1998; Schmitz, 2006; Fitter and Kaplinsky, 2001;Kaplinsky and Fitter, 2004; Kaplinsky, 2005; Milberg 2009).

• Some authors have observed that the balance of power oftenfavors nodes with high technology which would imply that firmswhich control technology through mechanisms like patents orlicenses are in extremely powerful positions and are likely toextract maximum rents in GVCs. (Mudambi, 2008; ).

• However, together with technology also better organizational skillsand better marketing capabilities might be crucial. Overall, toextract maximum rents, governance becomes an importantingredient in the value chain.


Background

• Therefore, firms investing in intangible assets (research,marketing, organization capital etc.) should be able to generatehigher returns, ceteris paribus, with respect to other firms(Mudambi, 2008).

• Marcolin, Le Mouel and Squicciarini (2016) provide evidenceabout the linkages between global value chain and investment inintangible assets focusing in particular on organizational capital.Their analysis support the assumption that industry-levelinvestment in KBC is causally linked to GVCs in the form ofbackward linkages with the foreign market.


Aim

• We investigate the role of intangible assets for participation inglobal value chains (GVC) in a sample of European countries.

• The analysis distinguishes between different forms of participationin GVC entailing a different degree of capability to create valueadded domestically

• We examine if and to what extent different intangible assets(R&D, advertising, market research, design, training, organizationcapital) foster countries’ engagement in GVC and the returnsfrom such participation.

What is the role of intangible assets in explaining the degreeand the benefits of countries’ participation in GVC?


Data

We merge intangible data from INTAN - Invest (ed 2014) withEUKLEMS information on value added, and hours worked and withdifferent measures of participation in global value chains fromOECD-WTO Trade in Value Added (TiVA) database:

• Domestic va embodied in export of a foreign country (DVAFEX)

• Domestic va embodied in foreign final demand (DVAFFD)

• Foreign va embodied in export of a domestic country (FVADEX)

• Foreign va embodied in domestic final demand (FVAFFD)

We measure:

• Participation = (DVAFEX+FVADEX) or (DVAFFD+FVADFD)

• Gains = (DVAFEX/FVADEX) or (DVAFFD/FVADFD)


Average VA share of intangible investment 1995-2010

The average VA share of intangible investment is relatively higher in the service sectors(8.2 %) than in manufacturing (7.0%) in most of the sample countries besides SE, DE,FI, IT and FR.


Participation to Global Value Chains

Participation significantly increased over time in manufacturing across the EU countries,but to a lesser extent in Finland, Belgium and in the Mediterranean economies (ES andIT).


Forward Participation to Global Value Chains

Forward participation (here measured in per hours worked) is likely to be higher forcountries (and sectors) involved in upstream production, with output and exports of thatcountry feeding into the production and exports of downstream producers (i.e. forwardintegration).


Backward Participation to Global Value Chains

Backward participation, the extent to which a country’s exports are dependent onimported content, it is likely to be higher for countries (sectors) involved in downstreamproduction.


Gains from Participation to Global Value Chains

Gains from participation refer to the capability of a country to appropriate a large shareof value added. In 2011, UK and NL have relatively higher gains both in manufacturingand services, DK higher in manufacturing and DE in services.


Research hypothesis: 1

HP1: Advanced countries investing relatively more in intangible assetsdisplay higher participation in global value chains.• Advanced countries are expected to organize their productionalong a value chain by keeping at home those activities that havea higher strategic value, are more complex (involve highertransaction costs) and allow them to keep control over the valuechains.

• Assets such as R&D expenditures, training, organizational capitalmay play a strategic role in creating domestic value added in theseactivities.



HP2: Intangible assets provide a different contribution to forward andbackward linkages in GVC. R&D, and design are more relevant toforward linkages while marketing and advertising to backward linkages.

• The role of intangible assets might differ according to the positionof a country in the GVC.

• While assets such as R&D and design may be strategic in theupstream activities stages of the value chains, other assets such asmarketing and advertising may be more important in downstreamactivities.



HP3: Benefits from participation in GVC (in terms of value addedcreation) increase with intangible capital assets in the advancedeconomies.

Overall, there is evidence that a great part of the valued added of a final product is created in the first and last stages of theproduction process (R&D, design, marketing and sales), while firms involved in intermediate stages (such as the production ofcomponents and assembly) reap only a small part of the final value of the good or service produced (Mudambi, 2008).


Participation to Global Value Chains vs Intangible capital


Forward Participation vs Intangible assets


Backward Participation vs Intangible assets


Gains from Participation vs Intangible capital


Econometric analysis

lnYGVCj

c,i,t = α1lnKIntgsc,i,t + α2lnK

ICTc,i,t + α3lnK

Tangc,i,t + α4lnXc,i,t + δt + γi + ε

where:

• c=country, i=industry, t=time

• Y GVCj different indicators for GVC participation (total, forward and backward) andgains from GVC

• K Intgs is intangible capital with s=Tot Intg, R&D, Training, Design, Advertising andmarketing, Organizational capital; K ICT is ICT capital and KTang is tangible capitalstock; X are other controls (corporate income taxes, country size); δt and γi aretime and industry dummies.

• All variables are in per hours term.


Empirical results:factors influencing participation in global value chains

(1) (2) (3) (4) (5) (6) (7) (8)

VARIABLES

lnKIntg 0.169***

(0.050)lnK

Intg(ex-R&D) 0.346*** 0.397***(0.067) (0.120)

lnKR&D 0.247*** 0.143***

(0.035) (0.047)lnK

Train 0.596***(0.049)

lnKAdv-Mkt 0.306***

(0.069)lnK

Design 0.015(0.063)

lnKOrgCap 0.250***

(0.045)lnK

ICT 0.316*** 0.262*** 0.173** 0.025 0.103 0.199** 0.495*** 0.160**(0.080) (0.058) (0.084) (0.091) (0.065) (0.091) (0.077) (0.081)

lnKTang 0.447*** 0.406*** 0.421*** 0.355*** 0.098** 0.412*** 0.359*** 0.539***

(0.065) (0.058) (0.055) (0.062) (0.047) (0.057) (0.061) (0.064)CountrySize -0.148*** -0.124*** -0.223*** -0.230*** -0.283*** -0.197*** -0.103 -0.140***

(0.049) (0.046) (0.052) (0.051) (0.036) (0.058) (0.070) (0.051)Corporateincometaxrate -0.028*** -0.019*** -0.028*** -0.014*** -0.010** -0.031*** -0.028*** -0.030***

(0.006) (0.005) (0.005) (0.005) (0.005) (0.004) (0.005) (0.005)

Observations 92 82 92 82 92 92 92 92Numberofctrysec 18 16 18 16 18 18 18 18

Standarderrorsinparentheses

***p<0.01,**p<0.05,*p<0.1

ParticipationtoGVC


Empirical results:factors influencing forward and backward linkages

(1) (2) (3) (4) (5) (6) (7) (8) (1) (2) (3) (4) (5) (6) (7) (8)

VARIABLES

lnKIntg 0.301*** 0.144***(0.060) (0.046)

lnKIntg(ex-R&D) 0.503*** 0.540*** 0.281*** 0.277*(0.083) (0.116) (0.088) (0.152)

lnKR&D 0.385*** 0.251*** 0.119*** 0.036(0.040) (0.051) (0.041) (0.064)

lnKTrain 0.787*** 0.415***(0.048) (0.063)

lnKAdv-Mkt 0.141* 0.404***(0.076) (0.074)

lnKDesign -0.001 0.158**(0.051) (0.074)

lnKOrgCap 0.124** -0.203***(0.057) (0.050)

lnKICT 0.160** 0.108* 0.005 -0.206** -0.137*** 0.277*** 0.419*** -0.177** 0.385*** 0.399*** 0.299*** 0.254** 0.255*** 0.178* 0.413*** 0.278***(0.074) (0.063) (0.089) (0.093) (0.051) (0.092) (0.051) (0.076) (0.084) (0.070) (0.100) (0.110) (0.088) (0.099) (0.087) (0.077)

lnKTang 0.553*** 0.513*** 0.505*** 0.441*** 0.213*** 0.518*** 0.479*** 0.075 0.325*** 0.319*** 0.256*** 0.264*** 0.084 0.265*** 0.262*** -0.224***(0.069) (0.067) (0.071) (0.071) (0.040) (0.059) (0.050) (0.059) (0.069) (0.066) (0.068) (0.073) (0.063) (0.065) (0.070) (0.059)

CountrySize -0.222*** -0.250*** -0.236*** -0.285*** -0.272*** -0.130* -0.064 -0.248*** -0.158** -0.158** -0.187** -0.234***-0.335***-0.225*** -0.162* -0.293***(0.042) (0.034) (0.049) (0.052) (0.035) (0.067) (0.070) (0.039) (0.075) (0.074) (0.078) (0.078) (0.063) (0.060) (0.084) (0.076)

Corporateincometaxrate -0.032*** -0.015** -0.034*** -0.016** -0.012** -0.029***-0.029*** -0.025*** -0.027*** -0.019*** -0.024***-0.016*** -0.010 -0.026***-0.027*** -0.028***(0.005) (0.006) (0.005) (0.006) (0.005) (0.004) (0.005) (0.006) (0.005) (0.005) (0.006) (0.005) (0.006) (0.005) (0.006) (0.006)

Observations 92 82 92 82 92 92 92 92 92 82 92 82 92 92 92 92Numberofctrysec 18 16 18 16 18 18 18 18 18 16 18 16 18 18 18 18

Standarderrorsinparentheses***p<0.01,**p<0.05,*p<0.1

Forwardparticipation Backwardparticipation


Empirical results:factors influencing gains from participation

(1) (2) (3) (4) (5) (6) (7) (8) (1) (2) (3) (4) (5) (6) (7) (8)

VARIABLES

lnKIntg 0.285*** 0.044*(0.056) (0.026)

lnKIntg(ex-R&D) 0.450*** 0.221* 0.027 0.014(0.071) (0.116) (0.035) (0.063)

lnKR&D 0.152*** 0.095** 0.080*** 0.077***(0.025) (0.043) (0.023) (0.026)

lnKTrain 0.331*** 0.056*(0.049) (0.030)

lnKAdv-Mkt -0.102 0.027(0.080) (0.038)

lnKDesign 0.042 0.054*(0.064) (0.032)

lnKOrgCap 0.297*** 0.033(0.038) (0.021)

Observations 92 82 92 82 92 92 92 92 92 82 92 82 92 92 92 92Numberofctrysec 18 16 18 16 18 18 18 18 18 16 18 16 18 18 18 18

Standarderrorsinparentheses***p<0.01,**p<0.05,*p<0.1

GainsfromGVCparticipation(exportbasedmeasure) GainsfromGVCparticipation(demandbasedmeasure)


Summing up

We find supportive evidence for our 3 main hypothesis.

• HP1: Intangible capital as a whole is positively related to participation in globalvalue chains in advanced countries. Moreover, nonR&D intangibles play a largerrole than R&D with training being the main driver.

• HP2: Intangibles contribute positively but to a different extent to both forward andbackward participation: R&D is more relevant for forward while marketing andadvertising are more important for backward linkages . This evidence supports theassumption that R&D is a factor affecting upstream production while marketingand advertising have a role in downstream production.

• HP3: Intangibles positively affect value appropriation and the results are robust tointroducing separately R&D and other intangible assets. Training andorganizational capital have a large positive effect while marketing and advertisingand architectural design do not.


Next steps

• Further develop the econometrics to control for possibleendogeneity biases affecting our results (IV)

• Extend the database across countries and sectors (INTAN-Invest(2016), SPINTAN (2016) and EUKLEMS (2016))

• Extend the analysis to the total economy evaluating the role ofpublic intangible intensity in affecting country’s participation toGVC.


WPIA Meeting - OECD. Paris Oct2016

Economy & Finance

Transcript of WPIA Meeting - OECD. Paris Oct2016