Which is the most effective innovation protection method for different types of product innovation?Evidence from South Korean manufacturing firms

ABSTRACT

Securing appropriation is crucial for firms to retain their competitive advantage. This article empirically examines the effects of different types of innovation protection methods on different types of product innovation. The study considers six types of innovation protection methods which can be used during product development: patents, a utility model, designs, confidentiality, complex designs, and market preoccupation. Four types of product innovation are considered: product quality improvements, product cost reductions, product diversification, and pioneering a new market. The main findings are as follows: (1) Patents have a positive effect on pioneering a new market compared to other product innovation types. This result means that patents are effective for product innovations with completely novel characteristics. (2) Market preoccupation is most effective for improving quality, reducing costs, and enhancing product diversification compared to other innovation protection methods.

KEYWORDS:

• Legal innovation protection method
• non-legal innovation protection method
• product innovation
• types of product innovation

JEL:

• O32
• O34
• C10

I. Introduction and background

Appropriation is important if a firm wants to gain a competitive advantage (Hurmelinna‐Laukkanen and Puumalainen 2007). If technology developed at the expense of high R&D costs is not protected and easily disclosed, other firms will imitate the new technology at a low cost and low risk in the short term. Therefore, in order to appropriate innovation performances, firms should devise strategies to prevent competitors from imitating their technologies. Representative innovation protection methods include legal protection methods through the registration of intellectual property rights, such as patents, utility models, and designs. There are also non-legal protection methods such as maintaining confidentiality in the firm, adopting a complex design, and occupying the market ahead of competitors.

Previous studies have mainly dealt with (1) the relationship between patents and confidentiality (Levin et al. 1987; Arora 1997) and (2) the relationship between means of cooperation or negotiation and appropriation (Blind et al. 2006: Hertzfeld, Link, and Vonortas 2006). In early research on innovation protection, researchers recognized patents and secrets as mutually exclusive types of protection with alternative relationships (Levin et al. 1987). However, some studies have described the relationship between patents and confidentiality as complementary rather than exclusive (Arora 1997), whereas other studies have found that intellectual property rights are used not only as a means of appropriating technological innovation but also as a strategic bargaining tool with partnerships (Blind et al. 2006: Hertzfeld, Link, and Vonortas 2006).

However, it is difficult to find any comparative study that investigates effective innovation protection methods for different types of product innovation. The main purpose of our study is to examine how different types of innovation protection methods – patents, utility models, designs, confidentiality, complex designs, and market preoccupation – have different effects on different types of product innovation, particularly quality, cost, flexibility, and time based on the theory of competitive priority in manufacturing (Garvin 1993; Kathuria 2000). Based on this model, we measure quality improvements, cost reductions, product diversification, and pioneering new markets in terms of quality, cost, flexibility and time, respectively. The contribution of our study is that it was found that effective innovation protection methods differ across four different product innovation types, which can be an academic framework on the relationship between product innovation and innovation protection methods and a guide for practitioners to establish an innovation protection strategy.

II. Data and model

Our study uses data from the ‘Korean Innovation Survey 2010: Manufacturing Sector (KIS)’ conducted by the Science and Technology Policy Institute (STEPI) of South Korea. The KIS survey measures manufacturing firms’ innovation activities and performance in the three years prior to the survey (2007–2009). This includes the degree of each type of innovation protection method used and the performance of each type of product innovation, all rated on a Likert-type scale. We use 1,539 companies that responded to all of the questions related to our study.

Because we use several dependent variables with the same independent variable, we conduct a multivariate regression analysis. Our econometric model for the multivariate regression analysis is as follows:

$Y_{ij}={\beta }_{0j}+{\beta }_{1j}PA{T}_{i1}+{\beta }_{2j}UT{I}_{i2}+{\beta }_{3j}DE{S}_{i3}+{\beta }_{4j}SE{C}_{i4}+{\beta }_{5j}CO{M}_{i5}+{\beta }_{6j}PREE{M}_{i6}$

${\beta }_{7j\mathrm{\setminus }12j}Control{s}_{i7\mathrm{\setminus }i12}+{\epsilon }_{ij},$

$i=1,2,\dots ,n,j=1,2\dots ,m,here,n=1539,m=4$

Table 1 presents the definitions of the variables. The dependent variables in our study are product innovation performance for major types of product innovation: product quality improvements, product cost reductions, product diversification, and the pioneering of new markets. The subscript j in the dependent variable indicates that the number 1 denotes quality improvement, 2 represents a cost reduction, 3 is product diversification, and 4 is the pioneering of new markets. KIS measures product innovation performance during 2007–2009 using a five-point Likert-type scale. The independent variables are the extent of the use of each type of innovation protection method (patents, utility models, designs, confidentiality, complex designs, and market preoccupation), as measured by KIS for 2007–2009 on a six-point scale (0 to 6): 0 = not using, 1 = very low use, 5 = very high use.

Table 1. Summary of variables

Variables	Definitions
Dependent variables
Yij	Effects of different purpose of product innovations made by each firm during 2007–2009 (from 1 = very low performance to 5 = very high performance) Yi1 is the effect of product quality improvement Yi2 is the effect of product cost reduction Yi3 is the effect of product diversification, Yi3 is the effect of pioneering a new market,
Independent variables	(0 = not using, from 1 = very low using to 5 = very high using)
PATi	Degree of utilizing patents
UTIi	Degree of utilizing utility models
DESi	Degree of utilizing designs
SECi	Degree of utilizing confidentiality
COMi	Degree of utilizing complex designs
PREEMi	Degree of utilizing market preoccupation
Control variables
FIRM_SIZE	Log transformed value of the average number of total employees in 2007–2009
RD_PERS	Percentage value of the number of dedicated research personnel compared to the total number of employees in 2007–2009
RD_EXPN	Log transformed value of the average value of total R&D expenditures in 2007–2009
INTR_RD	Dummy variable for whether internal R&D is used in product innovation (0 = internal R&D not used, 1 = internal R&D used)
COOP	Dummy variable for whether R&D cooperation with external partner is used in product innovation (0 = cooperation is not used, 1 = cooperation is used)

Our study uses five control variables. To control for the effect of firm size, we adopt the log-transformed value of the total number of workers at firms (Wang 2007), averaged over three years. The ratio of R&D personnel is an indicator of a firm’s absorptive capacity (Fritsch and Lukas 2001). R&D expenditure (RD_EXPN) is used to control for the effect of R&D investment in product innovation (Becker and Dietz 2004). The log value of R&D expenditure is used in our study. Internal R&D activities and R&D cooperation are controlled for to measure the net effect of each type of product innovation.

III. Results

Table 2 presents a summary of the descriptive statistics and correlations. Among the types of product innovation, the innovation performance for pioneering a new market is highest, with an average score of 2.559. Regarding the use of each type of innovation protection method, using patents is highest, with an average of 3.463 and using market preoccupation is lowest, with an average of 1.846.

Table 2. Descriptive statistics and correlations

Variables	Mean	SD	PATi	UTIi	DESi	SECi	COMi	PREEMi	FIRM_SIZE	RD_PERS	RD_EXPN	INTR_RD	COOP
PATi	3.463	1.215	1.000
UTIi	2.843	1.430	0.531	1.000
DESi	3.065	1.399	0.388	0.368	1.000
SECi	3.190	1.305	0.363	0.366	0.416	1.000
COMi	2.940	1.468	0.292	0.347	0.357	0.583	1.000
PREEMi	1.846	1.592	0.355	0.411	0.353	0.384	0.434	1.000
FIRM_SIZE	1.958	0.615	0.048	0.200	0.018	0.123	0.128	0.145	1.000
RD_PERS	9.979	11.007	0.065	0.076	0.114	0.104	0.101	0.104	−0.160	1.000
RD_EXPN	2.651	0.915	0.104	0.183	0.093	0.163	0.180	0.148	0.603	0.239	1.000
INTR_RD	0.920	0.271	0.083	0.104	0.043	0.058	0.045	0.086	0.016	0.083	0.049	1.000
COOP	0.338	0.473	0.063	0.100	0.069	0.079	0.107	0.090	0.140	0.043	0.132	−0.413	1.000
Quality	1.923	1.641
Cost	2.519	2.063
Diversification	1.339	1.733
New Market	2.559	1.860

Table 3 shows the results of a multivariate estimation of the effectiveness of innovation protection in different types of product innovation. All of the F values indicate that the models are statistically significant. With regard to effective methods of innovation protection for quality improvements, all innovation protection methods except patents have a positive effect. Regarding cost reductions, a utility model is effective (B = 0.095, P < 0.05) as a legal method and market preoccupation is effective (B = 0.156, P < 0.01) as a non-legal method. For effective methods of innovation protection for product diversification, designs are effective (B = 0.077, P < 0.01) as a legal method, and confidentiality (B = 0.084, P < 0.05) and market pre-emption are effective (B = 0.217, P < 0.01) as non-legal methods. Regarding the pioneering of a new market, all innovation protection methods except market pre-emption have a positive effect.

Table 3. Estimation results of innovation protection methods (2010 KIS)

	Product Quality Improvement		Product Cost Reduction		Product Diversification		Pioneering New Markets
	B	Std. Err.	B.	Std. Err.	B.	Std. Err.	B	Std. Err.
FIRM_SIZE	0.174***	0.062	0.286***	0.108	0.484***	0.093	0.452***	0.097
RD_PERS	0.001	0.003	0.029***	0.005	0.006	0.004	0.020***	0.004
RD_EXPN	0.052	0.042	0.359***	0.073	0.034	0.063	0.068	0.065
INTR_RD	0.119	0.114	0.229	0.198	0.349**	0.171	0.022	0.178
COOP	0.073	0.066	0.295**	0.114	0.191*	0.099	0.090	0.103
PAT	−0.023	0.028	0.065	0.049	0.023	0.042	0.159***	0.044
UTI	0.128***	0.025	0.095**	0.043	0.026	0.037	0.113***	0.038
DES	0.047**	0.023	0.037	0.040	0.077***	0.035	0.065*	0.036
SEC	0.057**	0.028	0.015	0.048	0.084**	0.041	0.155***	0.043
COM	0.081***	0.024	0.077*	0.042	−0.027	0.036	0.111***	0.038
PREEM	0.625***	0.021	0.156***	0.036	0.217***	0.031	0.030	0.032
Constant	−0.704	0.162	−0.759	0.281	−1.121	0.242	−0.701	0.253
R-Squared	0.569		0.180		0.136		0.183
F	83.84		30.46		21.96		31.26

The main findings are as follows. First, Patents have a positive effect on pioneering a new market compared to other product innovation types. Second, Market preoccupation is most effective for improving quality, reducing costs, and enhancing product diversification compared to other innovation protection methods.

IV. Robustness checks

In order to verify the robustness of major findings, additional verification was performed with using 745 firms of 2014 KIS data. As shown in Table 4, the results for the main findings are the same. Patents have strong positive effect on pioneering a new market compared to other product innovation types. Market preoccupation is most effective for improving quality, reducing costs, and enhancing product diversification compared to other innovation protection methods.

Table 4. Estimation results of innovation protection methods (2014 KIS)

	Product Quality Improvement		Product Cost Reduction		Product Diversification		Pioneering New Markets
	B	Std. Err.	B.	Std. Err.	B.	Std. Err.	B	Std. Err.
FIRM_SIZE	0.142	0.062	0.471***	0.091	0.187	0.109	0.002	0.107
RD_PERS	0.001	0.003	0.005*	0.002	0.005	0.003	0.001	0.003
RD_EXPN	0.019	0.042	−0.138**	0.061	0.157**	0.073	0.225***	0.072
INTR_RD	0.127	0.114	0.107	0.144	0.315*	0.174	0.063	0.171
COOP	0.038	0.066	0.081	0.120	0.112	0.144	0.223	0.142
PAT	0.065	0.043	0.065	0.030	0.079	0.048	0.136***	0.047
UTI	0.017	0.039	0.038	0.036	0.028	0.037	0.052	0.043
DES	0.014	0.055	0.039	0.050	0.072	0.035	0.025	0.060
SEC	0.077**	0.032	0.138***	0.033	0.024	0.041	0.050	0.039
COM	0.072	0.067	0.079	0.062	0.011	0.036	0.063	0.073
PREEM	0.100**	0.363	0.222***	0.040	0.067*	0.036	0.088**	0.035
Constant	1.401	0.213	−0.759	0.196	0.705	0.236	0.916	0.253
R-Squared	0.147		0.197		0.065		0.085
F	13.20		16.38		4.69		6.23

V. Discussion

We examined effective methods of innovation protection for different types of product innovation, finding that patents are effective for product innovation with completely novel characteristics, such as the pioneering of new markets, compared to other innovation types. On the other hand, non-legal innovation protection methods through market preoccupation are most effective for improving quality, reducing costs, and for product diversification compared to legal innovation protection methods and other non-legal innovation protection methods.

The contributions of this study can be summarized in two ways. First, from a practical point of view, the results of this study provide important managerial implications for firms who are faced with the task of selecting effective methods of innovation protection for different types of product innovation. Second, theoretically, this study provides a new integrative perspective on innovation protection methods and product innovation performance based on the competitive priorities of manufacturing.

One of the limitations of this study is that it only deals with the manufacturing industry. Therefore, future research should include a comparative study of other industry sectors or should add subsectors.

Disclosure statement

No potential conflict of interest was reported by the author(s).