The structure and evolution of ICT global innovation network

Figures & data

Table 1. Distribution of companies’ activities, by ICB classification.

ICB sector	2011			2015
	% of firms	% of R&D centres	% of international R&D centres	% of firms	% of R&D centres	% of international R&D centres
Aerospace & defence	7	11	38	6	10	45
Automobiles & parts	10	16	73	10	15	74
Electronic & electrical equipment	20	20	50	20	21	60
Financial services	1	1	60	0	1	81
General industrials	3	7	53	4	7	72
Health care equipment & services	3	3	39	3	3	60
Household goods & home construction	2	1	76	3	2	70
Industrial engineering	4	3	70	7	4	71
Leisure goods	8	7	47	7	6	54
Media	1	0	70	1	0	67
Pharmaceuticals & biotechnology	2	2	48	1	2	63
Software & computer Services	5	6	60	3	5	63
Support Services	1	0	67	0	0	67
Technology hardware & equipment	34	24	60	33	24	63
Total	100	100	Average: 58	100	100	Average: 65
Number	172	3154	1771	210	4461	2799

Source: IHS. Own calculations.

Table 2. R&D activity types.

	R&D type	2011		2015
		% of R&D centres with R&D type	% of international R&D centres with R&D type	% of R&D centres with R&D type	% of international R&D centres with R&D type
1	Hardware	62	59	63	61
2	Software	47	46	43	42
3	Components	33	29	33	32
4	Technology	21	19	24	23
5	Research	5	5	8	8
6	Others	1	1	2	2
7	Quality assurance	1	1	0.04	0.07
8	Quality control	1	2	0.40	0.39

Source: IHS. Own calculations.

Figure 1. The ICT global innovation network in 2011.

Figure 2. The ICT global innovation network in 2015.

Table 3. Distribution of companies and ICT R&D centres by country of ownership, location and out- and inward internationalisation in 2011 and 2015.

Nr	Country		2011				2015
			Ownership		Location		Ownership		Location
	Name	Code	% of total	Outward internation-alisation (%)	% of total	Inward internation-alisation (%)	% of total	Outward internation-alisation (%)	% of total	Inward internation-alisation (%)
1	USA	US	39	44	32	33	36	54	29	41
2	Japan	JP	22	45	14	16	22	52	13	15
3	Germany	DE	8	71	7	67	7	72	7	71
4	France	FR	7	73	4	53	6	77	4	61
5	China	CN	4	58	11	83	4	51	11	83
6	Taiwan	TW	3	59	2	40	4	66	2	40
7	Ireland	IE	0	100	1	100	3	99	1	97
8	Netherlands	NL	3	93	1	79	3	94	1	83
9	South Korea	KR	3	69	2	50	2	67	2	56
10	Canada	CA	3	85	2	84	2	88	2	89
11	Switzerland	CH	1	93	1	85	2	92	1	82
12	UK	UK	1	83	4	94	2	87	5	95
13	Italy	IT	2	62	2	57	2	72	2	70
14	Sweden	SE	2	88	1	73	2	88	1	76
15	Finland	FI	2	88	0	50	1	92	0	80
16	Denmark	DK	–	–	0	100	0	100	1	100
17	Turkey	TR	0	33	0	54	0	50	1	71
18	Belgium	BE	0	88	1	96	0	85	1	95
19	Brazil	BR	0	0	1	95	0	33	1	88
20	Russia	RU	–	–	0	100	0	56	1	92
21	India	IN	0	75	4	99	0	60	5	99
22	Singapore	SG	–	–	0	100	0	80	2	99
23	Hong Kong	HK	0	75	0	90	–	–	–	–
24	Australia	AU	–	–	1	100	–	–	1	100
25	Czech Republic	CZ	–	–	1	100	–	–	1	100
26	Spain	ES	–	–	0	100	–	–	1	100
27	Israel	IL	–	–	1	100	–	–	1	100
28	Malaysia	MY	–	–	0	100	–	–	1	100
29	Poland	PL	–	–	0	100	–	–	1	100
30	United Arab Emirates	AE	–	–	0	100	–	–	0	100
31	Argentina	AR	–	–	0	100	–	–	0	100
32	Austria	AT	–	–	0	100	–	–	0	100
33	Bulgaria	BG	–	–	0	100	–	–	0	100
34	Belarus	BY	–	–	1	100	–	–	0	100
35	Chile	CL	–	–	1	100	–	–	0	100
36	Costa Rica	CR	–	–	0	100	–	–	0	100
37	Egypt	EG	–	–	0	100	–	–	0	100
38	Greece	GR	–	–	0	100	–	–	0	100
39	Croatia	HR	–	–	1	100	–	–	0	100
40	Hungary	HU	–	–	0	100	–	–	0	100
41	Indonesia	ID	–	–	1	100	–	–	0	100
42	Luxembourg	LU	–	–	0	100	–	–	0	100
43	Mexico	MX	–	–	0	100	–	–	0	100
44	Norway	NO	–	–	0	100	–	–	0	100
45	New Zealand	NZ	–	–	1	100	–	–	0	100
46	Philippines	PH	–	–	0	100	–	–	0	100
47	Pakistan	PK	–	–	0	100	–	–	0	100
48	Puerto Rico	PR	–	–	1	100	–	–	0	100
49	Portugal	PT	–	–	1	100	–	–	0	100
50	Romania	RO	–	–	0	100	–	–	0	100
51	Slovenia	SI	–	–	0	100	–	–	0	100
52	Slovak Republic	SK	–	–	0	100	–	–	0	100
53	Bangladesh	BD	–	–	–	–	–	–	0	100
54	Estonia	EE	–	–	–	–	–	–	0	100
55	Jordan	JO	–	–	–	–	–	–	0	100
56	Kenya	KE	–	–	–	–	–	–	0	100
57	Morocco	MA	–	–	–	–	–	–	0	100
58	Nigeria	NG	–	–	–	–	–	–	0	100
59	Peru	PE	–	–	–	–	–	–	0	100
60	Qatar	QA	–	–	–	–	–	–	0	100
61	Serbia	RS	–	–	–	–	–	–	0	100
62	Saudi Arabia	SA	–	–	–	–	–	–	0	100
63	Thailand	TH	–	–	–	–	–	–	0	100
64	Ukraine	UA	–	–	–	–	–	–	0	100
65	Vietnam	VN	–	–	–	–	–	–	0	100
66	South Africa	ZA	–	–	–	–	–	–	0	100

Table 4. ICT GIN indices.

	2011	2015
Number of countries	52	65
Number of arcs	347	446
Density	0.117	0.107
Average out-degree	17.4	20.3
Average in-degree	6.7	6.9
Clustering centralisation	46%	53%

Table 5. Countries’ position in the ICT GIN, 2011 and 2015.

Rank	2011								2015
	Out-degree		In-degree		Closeness		Betweenness		Out-degree		In-degree		Closeness		Betweenness
	Country	Value	Country	Value	Country	Value	Country	Value	Country	Value	Country	Value	Country	Value	Country	Value
1	US	43	CN	18	US	0.85	US	0.086	US	51	US	20	US	0.82	US	0.090
2	DE	35	US	17	DE	0.75	DE	0.046	JP	40	CN	20	JP	0.72	CN	0.037
3	JP	31	DE	16	JP	0.70	CN	0.044	FR	38	IN	19	DE	0.70	JP	0.029
4	FR	31	UK	15	FR	0.70	JP	0.021	DE	37	DE	17	FR	0.70	DE	0.023
5	SE	27	FR	14	SE	0.67	FR	0.019	CN	29	UK	17	CN	0.63	FR	0.022
6	CN	25	IN	14	CN	0.65	SE	0.018	NL	26	JP	15	SE	0.61	KR	0.012
7	CA	20	CA	13	CA	0.61	CA	0.015	SE	25	FR	15	CH	0.60	UK	0.012
8	UK	18	JP	12	CH	0.59	CH	0.008	CH	25	IT	15	IE	0.60	SE	0.009
9	NL	18	IT	11	UK	0.59	UK	0.007	IE	23	CA	14	NL	0.59	CH	0.009
10	CH	17	KR	11	NL	0.59	KR	0.005	TW	20	PL	14	TW	0.58	TW	0.008
11	TW	16	BR	11	TW	0.57	NL	0.003	CA	20	RU	13	CA	0.57	FI	0.008
12	FI	15	SE	10	FI	0.57	TW	0.002	UK	19	BR	13	UK	0.56	IT	0.007
13	IT	13	PL	10	IT	0.55	FI	0.002	IT	18	CZ	13	KR	0.56	NL	0.003
14	KR	12	NL	9	KR	0.55	IT	0.001	KR	18	KR	12	FI	0.56	CA	0.003
15	IE	8	TW	9	IE	0.53	BE	0.000	FI	17	SE	11	IT	0.54	IN	0.002

Table 6. Core/periphery model statistics in the 2011 and the 2015 ICT GIN.

Goodness of fit^a
	2011		2015
Starting fitness	0.906		0.827
Final fitness	0.906		0.827
Density matrix
	Core	Periphery	Core	Periphery
Core	0.886	0.308	0.743	0.137
Periphery	0.308	0.007	0.137	0.000

Note: Number of iterations: 50.

Source: Own calculations.

^a Fitness measure: Pearson correlation.

Table 7. Block membership by country in 2011 and 2015 ICT GIN.

Block	Time	Country	Nr of countries and % of total
Core	2011	Canada, China, Finland, France, Germany, India, Italy, Japan, South Korea, Netherlands, Sweden, Switzerland, Taiwan, UK, USA	Nr of countries: 15
			% of total: 28%
	2015	Belgium, Brazil, Canada, Switzerland, China, Czech Republic, Germany, Denmark, Finland, France, United Kingdom, India, Ireland, Italy, Japan, south Korea, Netherlands, Poland, Russia, Singapore, Sweden, Turkey, Taiwan, USA	Nr of countries: 24
			% of total: 37%
Periphery	2011	Argentina, Australia, Austria, Belgium, Bermuda, Brazil, Bulgaria, Cayman Islands, Croatia, Czech Republic, Denmark, Egypt, Greece, Hong Kong, Hungary, Indonesia, Ireland, Israel, Luxembourg, Malaysia, Mexico, New Zealand, Norway, Pakistan, Philippines, Poland, Portugal, Puerto Rico, Romania, Russia, Singapore, Slovakia, Slovenia, South Africa, Spain, Thailand, Turkey, United Arab Emirates, Vietnam	Nr of countries: 39
			% of total: 72%
	2015	Argentina, Australia, Austria, Bangladesh, Belarus, Bulgaria, Chile, Costa Rica, Croatia, Egypt, Estonia, Greece, Hungary, Indonesia, Israel, Jordan, Kenya, Luxembourg, Malaysia, Mexico, Morocco, New Zealand, Nigeria, Norway, Pakistan, Peru, Philippines, Portugal, Puerto Rico, Qatar, Romania, Saudi Arabia, Serbia, Slovak Republic, Slovenia, South Africa, Spain, Thailand, Ukraine, United Arab Emirates, Vietnam	Nr of countries: 41
			% of total: 63%

Table 8. Estimation results of the determinants of the linkages in the GIN.

Variable	Number of R&D centres in 2015
	(1)		(2)		(3)		(4)		(5)		(6)
	Coeff.	SE	Coeff.	SE	Coeff.	SE	Coeff.	SE	Coeff.	SE	Coeff.	SE
Log distance ij	−0.087***	0.021	−0.029	0.021	0.011	0.021	0.008	0.022	−0.019	0.022	−0.021	0.022
Common language ij	0.535***	0.051	0.379***	0.054	0.646***	0.056	0.645***	0.056	0.675***	0.057	0.702***	0.056
Log GDP i, 2015	1.601***	0.038	1.917***	0.056	0.651***	0.082	0.627***	0.101	0.611***	0.101
Log GDP j, 2015	0.799***	0.030	0.142***	0.041	0.063	0.051	−0.024	0.056	−0.098*	0.052
Log Population i, 2014	−0.906***	0.041	−0.934***	0.040	−0.197***	0.052	−0.183**	0.060	−0.173**	0.060	0.123***	0.032
Log Population j, 2014	−0.023	0.020	0.035*	0.020	0.071**	0.024	0.114***	0.026	0.122***	0.025	0.090***	0.018
Log Patent i, 2011	0.089***	0.023	0.072***	0.022	−0.042*	0.025	−0.042*	0.025	−0.027	0.026	−0.025	0.026
Log Patent j, 2011	0.117***	0.012	0.043**	0.014	0.061***	0.013	0.062***	0.013	0.100***	0.015	0.095***	0.014
Out-degree i, 2011			−0.089***	0.013	−0.086***	0.014	−0.104**	0.037	0.000	0.078	−0.051	0.078
Out-degree j, 2011			0.215***	0.010	0.204***	0.011	0.140***	0.016	0.143***	0.016	0.139***	0.016
In-degree i, 2011					0.082***	0.004	0.082***	0.005	0.075***	0.006	0.093***	0.006
In-degree j, 2011					0.007**	0.003	0.011***	0.003	−0.003	0.004	−0.005	0.003
Closeness centrality i, 2011							0.682	1.370	−4.121	3.263	−1.110	3.252
Closeness centrality j, 2011							2.254***	0.461	1.909***	0.425	1.663***	0.402
Betweenness centrality i, 2011									7.190**	3.630	9.508***	3.712
Betweenness centrality j, 2011									8.593***	1.330	8.087***	1.307
Constant	−7.665***	0.247	−6.613***	0.316	−6.245***	0.323	−7.266***	0.625	−4.506**	1.498	−5.240***	1.501
N	1408		1408		1408		1408		1408		1408
Log likelihood	−2244.78		−2002.51		−1809.28		−1794.88		−1772.90		−1793.86
LRχ^2	9185.84		9670.39		10,056.84		10,085.64		10,129.60		10,087.68
Prob > χ^2	0.000		0.000		0.000		0.000		0.000		0.000
Pseudo R^2	0.672		0.707		0.735		0.7375		0.741		0.738

Notes: The dependent variable is the number of R&D centres per link. The list of explanatory variables includes: log of distance between two the capitals of each pair of countries; dummy variable controlling for common language; logs of real GDP, population size, inventive performance, measured by the number of patents of country i and j. Network measures, i.e. out- and in-degree, closeness and betweenness centrality of country i and j, refer to the position of a country in the GIN in 2011. All models report Poisson regression estimates. ***, ** and *Indicate 1, 5 and 10% significance levels, respectively.

Source: Own calculations based on the data from IHS, CEPII and IMF.

Table 9. Estimation results of the determinants of the change of intensity of R&D linkages in the ICT GIN.

Variable	Change in the number of R&D centres between 2011 and 2015
	(1)		(2)		(3)		(4)		(5)		(6)
	Coeff.	SE	Coeff.	SE	Coeff.	SE	Coeff.	SE	Coeff.	SE	Coeff.	SE
Log distance ij	0.000	0.034	0.073**	0.035	0.104**	0.035	0.089*	0.036	0.071*	0.037	0.065	0.037
Common language ij	0.684***	0.080	0.568***	0.085	0.762***	0.088	0.734***	0.088	0.763***	0.089	0.765***	0.087
Log GDP i, 2015	1.559***	0.063	1.908***	0.089	0.714***	0.133	0.334**	0.155	0.347**	0.155
Log GDP j, 2015	0.742***	0.049	0.113**	0.066	0.107	0.083	−0.049	0.096	−0.129	0.092
Log Population i, 2014	−0.848***	0.061	−0.880***	0.059	−0.254***	0.077	−0.070	0.086	−0.065	0.086	0.087*	0.050
Log Population j, 2014	−0.031	0.032	0.018	0.031	0.019	0.039	0.092**	0.042	0.104***	0.041	0.064**	0.029
Log Patent i, 2011	−0.032	0.036	−0.052	0.036	−0.185***	0.042	−0.163***	0.042	−0.165***	0.045	−0.160***	0.045
Log Patent j, 2011	0.137***	0.020	0.067**	0.022	0.084***	0.022	0.089***	0.023	0.125***	0.024	0.116***	0.023
Out-degree i, 2011			−0.099***	0.020	−0.075***	0.021	−0.311***	0.057	−0.365***	0.125	−0.386***	0.125
Out-degree j, 2011			0.207***	0.017	0.202***	0.017	0.061*	0.032	0.084***	0.029	0.080***	0.029
In-degree i, 2011					0.076***	0.007	0.085***	0.007	0.088***	0.010	0.098***	0.009
In-degree j, 2011					0.000	0.005	0.005	0.005	−0.008	0.006	−0.011*	0.006
Closeness centrality i, 2011							9.409***	2.129	11.521**	5.234	13.082*	5.213
Closeness centrality j, 2011							5.307***	1.096	4.222***	0.937	3.905***	0.905
Betweenness centrality i, 2011									−1.802	5.827	−0.429	5.890
Betweenness centrality j, 2011									8.472***	2.271	7.684***	2.220
Constant	−8.236***	0.403	−7.457***	0.520	−6.379***	0.536	−11.742	1.042	−11.945***	2.445	−12.227***	2.452
N	1370		1370		1370		1370		1370		1370
Log likelihood	−1259.392		−1167.381		−1106.424		−1078.190		−1071.133		−1074.748
LRχ^2	3281.400		3465.420		3587.330		3643.800		3657.920		3650.690
Prob > χ^2	0.000		0.000		0.000		0.000		0.000		0.000
Pseudo R^2	0.566		0.598		0.619		0.628		0.631		0.629

Notes: The dependent variable is the change in the number of R&D centres per link between 2011 and 2015. The list of explanatory variables includes: log of distance between two the capitals of each pair of countries; dummy variable controlling for common language; logs of real GDP, population size, inventive performance, measured by the number of patents of country i and j. Network measures, i.e. out- and in-degree, closeness and betweenness centrality of country i and j, refer to the position of a country in the GIN in 2011. All models report Poisson regression estimates. ***, ** and *Indicate 1, 5 and 10% significance levels, respectively.

Source: Own calculations based on the data from IHS, CEPII and IMF.

Table A1. Descriptive statistics.

Variable	Mean	SD	Min	Max
Number of R&D centres per pair of countries	2.0	8.8	0.0	152.0
Number of R&D types per pair of countries	3.4	15.2	0.0	259.0
Distance (km)	6484.5	4497.0	81.0	19,447.4
Official common language	0.1	0.3	0.0	1.0
GDP i (current US$)	2581.1	4091.7	227.5	17,968.2
GDP j (current US$)	1047.8	2589.6	22.9	17,968.2
Population i (Mln)	180,950.6	371,205.6	4612.7	1,364,270.0
Population j (Mln)	88,105.3	227,267.5	556.1	1,364,270.0
Number of patents i	15,145.5	43,218.7	166.1	208,462.1
Number of patents j	5124.4	25,732.1	0.0	208,462.1
Out-degree i	10.4	3.9	5.0	19.0
In-degree i	14.7	12.0	0.0	41.0
Closeness centrality i	0.687	0.101	0.556	0.952
Betweenness centrality i	0.013	0.022	0.000	0.100
Out-degree j	5.1	5.0	0.0	19.0
In-degree j	4.8	9.7	0.0	41.0
Closeness centrality j	0.457	0.272	0.000	0.952
Betweenness centrality j	0.004	0.014	0.000	0.100

Table A2. Pair-wise correlations between variables.

		1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18
1	Number of R&D centres per pair of countries	1.000
2	Number of R&D types per pair of countries	0.995*	1.000
3	Distance (km)	0.011	0.014	1.000
4	Official common language	0.098*	0.097*	0.011	1.000
5	GDP i (current US$)	0.271*	0.271*	0.138*	0.064	1.000
6	GDP j (current US$)	0.454*	0.452*	0.048	0.063	−0.025	1.000
7	Population i (Mln)	0.005	0.006	0.096*	0.043	0.474*	−0.012	1.000
8	Population j (Mln)	0.300*	0.291*	0.059	0.051	−0.012	0.528*	−0.026	1.000
9	Number of patents i	0.010	0.010	0.087*	−0.047	0.522*	−0.013	0.673*	−0.017	1.000
10	Number of patents j	0.288*	0.279*	0.015	0.000	−0.014	0.568*	−0.018	0.683*	−0.026	1.000
11	Out-degree i	0.198*	0.196*	0.164*	0.090*	0.766*	−0.019	0.492*	−0.013	0.438*	−0.012	1.000
12	In-degree i	0.260*	0.253*	0.009	0.001	0.602*	−0.015	0.019	−0.001	0.200*	−0.005	0.614*	1.000
13	Closeness centrality i	0.219*	0.218*	0.148*	0.102*	0.836*	−0.021	0.492*	−0.013	0.458*	−0.012	0.989*	0.639*	1.000
14	Betweenness centrality i	0.315*	0.313*	0.089*	0.069	0.911*	−0.023	0.191*	−0.005	0.260*	−0.007	0.720*	0.783*	0.789*	1.000
15	Out-degree j	0.364*	0.359*	−0.027	0.086*	−0.009	0.646*	−0.006	0.424*	−0.005	0.397*	−0.012	−0.008	−0.012	−0.009	1.000
16	In-degree j	0.337*	0.338*	−0.08*	0.068	−0.012	0.678*	0.000	0.215*	−0.004	0.329*	−0.012	−0.019	−0.012	−0.015	0.725*	1.000
17	Closeness centrality j	0.251*	0.247*	−0.034	0.098*	−0.005	0.435*	−0.003	0.275*	−0.003	0.259*	−0.006	−0.004	−0.006	−0.005	0.823*	0.515*	1.000
18	Betweenness centrality j	0.420*	0.422*	−0.005	0.069*	−0.022	0.923*	−0.005	0.284*	−0.006	0.343*	−0.018	−0.019	−0.020	−0.025	0.612*	0.803*	0.406*	1.000

Note: Pearson pair-wise correlations.

Source: Own calculations.

* Significant at p < 0.001 level.