Optimal path selection of innovation resource allocation in China’s regions with shared inputs

Figures & data

Figure 1. A two-stage process with shared resource inputs.

Source: Adopted from Chen and Guan (2012).

Figure 2. Two-stage process with shared inputs of regional innovative resource allocation.

Source: Author’s own drawing.

Table 1. Descriptive analysis of evaluation indexes.

Sub-systems	Indicator Category	Indicator Name	Mean	SD	Min	Max
Knowledge innovation sub-process	Input indexes (t period)	Full-time equivalent of R&D personnel (man-years)	112170	119889	4008	543438
		Intramural expenditure on R&D (10 000 yuan)	4159641	4764409	70204	20351440
		Fixed asset of S & T (10 000 yuan)	1092424	1488965	9900	10557200
	Output indexes (t + 1 period)	Scientific papers issued (piece)	45051	35371	2521	185601
		Technology market turnover (10 000 yuan)	2796051	6024122	5700	44868900
		Inventions in force (item)	33272	44823	456	208502
Achievements commercialization sub-process	Input indexes (t + 1 period)	Non-R&D personnel (person)	2989981	3162177	86514	14029894
		Expenditure on new products development (10 000 yuan)	3370070	4684388	34745	28286496
	Output indexes (t + 2 period)	Sales revenue of new products (10 000 yuan)	52188750	70118084	85659	393760563

Source: China’s economic and social development statistics database.

Table 2. Efficiency scores of innovation resource allocation.

Region		Two-stage NDEA model with shared inputs				General Two-stage NDEA model
		Overall efficiency$E$	S1 KIS efficiency$E^1$	S2 ACS efficiency$E^2$	Rank	Overall efficiency$E^{\mathrm{*}}$	S1 KIS efficiency$E^{1\mathrm{*}}$	S2 ACS efficiency$E^{2\mathrm{*}}$	Rank
Eastern China	Beijing	1.0000	1.0000	0.9711	1	1.0000	1.0000	1.0000	1
	Tianjin	0.7117	0.4505	0.7117	11	0.6357	0.7893	0.4009	11
	Hebei	0.5937	0.4312	0.6712	18	0.5230	0.6167	0.1984	22
	Liaoning	0.5677	0.7341	0.5677	22	0.4582	0.8684	0.3889	28
	Shanghai	0.9618	0.9680	0.9259	2	0.9094	0.9112	0.8352	2
	Jiangsu	0.8135	0.6256	0.8135	4	0.8150	0.7842	0.3221	4
	Zhejiang	0.9180	0.7673	0.9180	3	0.8515	0.8170	0.4244	3
	Fujian	0.6113	0.4608	0.6113	17	0.5678	0.6402	0.2152	18
	Shandong	0.6159	0.3379	0.6160	16	0.6030	0.8354	0.2387	14
	Guangdong	0.8029	0.8029	0.5634	6	0.7774	0.8801	0.4441	5
	Hainan	0.5047	1.0000	0.5047	26	0.4917	0.8387	0.4304	26
Central China	Shanxi	0.5238	0.4782	0.5238	25	0.6292	0.7383	0.4844	12
	Jilin	0.5635	0.7751	0.5635	23	0.5057	0.6918	0.3461	24
	Heilongjiang	0.4611	0.7476	0.2300	29	0.5382	0.8452	0.4369	20
	Anhui	0.7131	0.6011	0.7131	10	0.6940	0.7149	0.2764	7
	Jiangxi	0.7092	0.5678	0.7092	12	0.6873	0.6546	0.2793	8
	Henan	0.6825	0.4424	0.6825	14	0.6048	0.7032	0.1823	13
	Hubei	0.7488	0.7488	0.7142	7	0.6572	0.7996	0.3192	10
	Hunan	0.8123	0.5734	0.9095	5	0.7183	0.7406	0.3045	6
Western China	Inner Mongolia	0.4690	0.4690	0.5587	28	0.3871	0.6513	0.2527	30
	Guangxi	0.6613	0.7970	0.6477	15	0.5259	0.6925	0.3592	21
	Chongqing	0.7194	0.8255	0.7194	9	0.6831	0.7584	0.3639	9
	Sichuan	0.7387	0.7387	0.5419	8	0.5716	0.7537	0.4257	17
	Guizhou	0.6871	0.9709	0.4403	13	0.5835	0.7098	0.3900	16
	Yunnan	0.5921	0.9529	0.4362	20	0.5945	0.7122	0.4106	15
	Shaanxi	0.5937	0.8428	0.3608	19	0.4413	0.5825	0.2575	29
	Gansu	0.4931	0.9812	0.4931	27	0.4795	0.8427	0.4923	27
	Qinghai	0.5774	0.9381	0.2554	21	0.5018	0.7340	0.3707	25
	Ningxia	0.5322	0.7748	0.5322	24	0.5493	0.7297	0.4020	19
	Xinjiang	0.4389	0.9145	0.4389	30	0.5128	0.7772	0.4613	23
Mean		0.6606	0.7239	0.6115		0.6166	0.7604	0.3904

Source: Author’s own calculations.

Figure 3. Trends in the overall and sub-stages efficiency of regional innovation resources.

Source: Author’s own drawing.

Figure 4. Influencing mechanism of regional innovation resource allocation efficiency.

Source: Author’s own drawing.

Table 3. Qualitative breakpoints for variables fuzzy calibration.

Variable category		Full membership	Cross-over point	Full nonmembership
Outcome variable	E	0.79	0.59	0.41
Causal variable	NS	0.65	0.37	0.23
	NO	0.40	0.21	0.10
	NC	0.45	0.30	0.21
	ED	7.78	4.64	3.53
	BE	0.52	0.16	0.10
	SP	290.84	69.30	25.43
	SC	0.60	0.36	0.20

Source: Author’s own calculations.

Table 4. Necessity analysis of conditional variables.

Conditional variables	Consistency	Coverage	Conditional variables	Consistency	Coverage
fz NS	0.733375	0.789905	∼ fz NS	0.456618	0.470013
fz NO	0.699810	0.805394	∼ fz NO	0.507916	0.492629
fz NC	0.687144	0.685841	∼ fz NC	0.511083	0.569112
fz ED	0.672855	0.728849	∼ fz ED	0.572343	0.573813
fz BE	0.643406	0.696466	∼ fz BE	0.549936	0.551702
fz SP	0.725992	0.807117	∼ fz SP	0.517286	0.506583
fz SC	0.770167	0.800932	∼ fz SC	0.454546	0.473965

Note. the set ‘not A’ is represented as ∼ A, and the symbol ‘∼’ denotes negation.

Source: Author’s own calculations.

Table 5. Truth table of efficient path for regional innovation resource allocation.

NS	NO	NC	ED	BE	SP	SC	number	E	raw consist.
1	1	1	1	1	1	1	2	1	0.932170
0	1	1	1	1	1	1	1	1	0.918239
0	1	1	1	0	0	0	1	1	0.893805
1	0	0	0	0	1	1	1	1	0.886364
1	0	0	1	1	0	1	1	1	0.870861
1	1	0	1	1	1	1	4	1	0.827255
1	1	1	0	1	1	1	1	1	0.818493
0	1	1	1	1	1	0	1	1	0.812102
0	1	0	1	0	0	0	1	1	0.805461

Source: Author’s own calculations.

Table 6. Parsimonious and intermediate for efficient allocation of regional innovation resources.

Intermediate solution
Configuration	Raw coverage	Unique coverage	Consistency
∼NS*NO*NC*ED*BE*SP	0.198227	0.028499	0.841398
NS*NO*NC*BE*SP*SC	0.336922	0.024066	0.895623
NS*NO*ED*BE*SP*SC	0.423686	0.099430	0.868831
∼NS*NO*ED*∼BE*SP*∼SC	0.208993	0.047498	0.841837
NS*∼NO*∼NC*∼ED*∼BE*SP*SC	0.222293	0.075364	0.886364
NS*∼NO*∼NC*ED*BE*∼SP*SC	0.166561	0.028499	0.870861
Solution coverage	0.673844
Solution consistency	0.833203
Parsimonious solution
Configuration	Raw coverage	Unique coverage	Consistency
∼NS*NO	0.699810	0.123496	0.805394
NS*SC	0.706776	0.130462	0.801724
Solution coverage	0.830272
Solution consistency	0.772995

Source: Author’s own calculations.

Table 7. Efficient allocation path of regional innovation resources.

Conditional variables		Efficient configuration
		Ha		Hb
		Ha1	Ha2	Hb1	Hb2	Hb3	Hb4
Network structure dimension	Network scale	$\otimes$	$\otimes$	●	●	●	●
	Network openness	●	●	●	●	$\otimes$	$\otimes$
	Network centrality	●		●		$\otimes$	$\otimes$
Regional Innovation Milieu	Economic development environment	●	●		●	$\otimes$	●
	Basic establishment environment	●	$\otimes$	●	●	$\otimes$	●
	Systems and policies environment	●	●	●	●	●	$\otimes$
	Social-cultural environment		$\otimes$	●	●	●	●
Raw coverage		0.198	0.209	0.337	0.424	0.222	0.167
Unique coverage		0.028	0.047	0.024	0.099	0.075	0.028
Consistency		0.841	0.842	0.896	0.869	0.886	0.871
Frequency cutoff		1
Consistency cutoff		0.805461

Source: Author’s own calculations.

Chen, K., & Guan, J. (2012). Measuring the efficiency of China’s regional innovation systems: Application of network data envelopment analysis (DEA). Regional Studies, 46(3), 355–377. https://doi.org/10.1080/00343404.2010.497479

 Web of Science ®Google Scholar