The Relationship between Foreign Competition and Innovation Activities Based on Quantile Regression

Abstract

This article uses quantile regression to shed light on the complex relationship between foreign competition and innovation activities. Quantile regression is more powerful than classical linear regression since quantile regression can produce estimates for all conditional quantiles of the distribution of the innovation activities variable, whereas classical linear regression only estimates the conditional mean effects. The empirical evidence shows that the effect of foreign competition on make innovation activities is different across the conditional quantiles of the distribution of innovation activities, something classical linear regression would leave unidentified. This finding suggests that estimating the quantile effect of innovation activities variable can well be more insightful than effect. Additionally, this article finds a U-shaped relationship between foreign competition and make innovation activities.

Introduction

Competition has been recognised as a major driving force behind the success of organisations (Alderson, 1965; Li & Vanhaverbeke, 2009; Yuchtman & Seashore, 1967). Academics and managers have become increasingly aware of the importance of competitive advantage (Li & Vanhaverbeke, 2009; Nelson, 1991). The sustainability of a competitive advantage is no longer a matter of better efficiency of existing businesses (Li & Vanhaverbeke, 2009), but rather depends on the innovation capabilities of the organisations (Nelson, 1991). Prior research pays attention on understanding implications and outcomes of innovation capabilities (Bogliacino & Pianta, 2012; Cassiman & Veugelers, 2006; Guan & Chen, 2009; Li & Wu, 2010; McKelvey, Zaring, & Ljungberg, 2014; Pérez-Luño & Cambra, 2013; Prajogo & Ahmed, 2006; Ringler et al., 2014; Roberts & Amit, 2003; Suárez, 2014). Specifically, increasing research interest has been devoted to the relationship between competition and innovation, due to the globalisation of markets and the shortening of technology life cycles (Li & Vanhaverbeke, 2009). Research on innovation (R&D departments) found conflicting results on the influence of competition pressure (Amabile & Gryskiewicz, 1987; Amabile, Conti, Coon, Lazenby, & Herron, 1996). As individuals are more creative in a competitive setting than in a non-competitive setting (Shalley & Oldham, 1997), there is a positive effect on competition with other organisations (Amabile, 1996). A competition can challenge individuals and help them to improve their performance (Shalley & Oldham, 1997). This perception of challenge is expected to lead to a high degree of creativity and innovativeness (Amabile, 1996; Amabile et al., 1996; Tierney & Farmer, 2002). However, Amabile (1996) assumes a negative effect of competition within an organisation (Amabile, 1996).

Economists and strategic management researchers have long been interested in the relationship between competition and innovation (Aghion, Bloom, Blundell, Griffith, & Howitt, 5 Amendola, Gaffard, & Musso, 2000; Greenhalgh & Rogers, 2006; Greis, Dibner, & Bean, 1995; Howitt, Griffith, Aghion, Blundell, & Bloom, 2005; Li & Vanhaverbeke, 2009; Smith & Sharif, 2007; Tang, 2006). Innovation is described as the act or process of innovating; something newly introduced, a new method, custom, device or change in the way of doing thing. (Webster’s New World Dictionary, 1982, Second College Edition.) Despite the progress made in the innovation research domain, the relationship between competition and innovation is still a subject of intense debate (Clement, 2003). Thus, alternative measures of competition are needed to better understand the effect of competition on innovation (Tang, 2006). There are different types of innovation, such as ‘make and buy’ (Liu, Hodgkinson, & Chuang, 2014). The ‘make innovation’ activities refer to enterprises that rely on all their internal resources to develop new products, patents, technology and various other elements. The ‘buy innovation’ activities refer to the enterprises which use money to acquire new technology through outsourcing externally, including domestic and foreign technology transfer (Bin, 2008). There are a few studies that examine how competition affects make and buy innovation activities. For instance, Liu (2014), using panel data from China, examined the relationship between foreign competition and buy and make activities. Building on this research, we focus on the relationship between foreign competition and make innovation activities in the USA. The foreign competition means the competition effect from a foreign firm, which can be measured by the log of the proportion of the import goods to the total production value (Kejžar, 2016; Liu et al., 2014; Wang, Lee, & Hsu, 2014).

The remainder of the article is organised as follows. In the second section, the relationship between competition and innovation is discussed. The third section presents the quantile regression model and data issues. In the fourth section, the results of the quantile regression analysis are presented. Finalsection draws the main conclusion.

The Relationship between Foreign Competition and Make Innovation Activities

Innovation is one of the most important ways for firms to achieve competitive advantages (Barney, 1991). Competition may shape the form of innovation activities (Sakakibara & Porter, 2001; Veugelers, 1997; Veugelers & Cassiman, 1999). A number of studies have highlighted that market structure (product market competition) affects innovation (Jacobides & Winter, 2012; Jung & Lee, 2010; Malerba, 2002, 2005). Technological innovations can be used as either a defensive or an offensive strategy to sustain a firm’s competitive advantage (BurgeSmani & Wheelwright, 2004).

Some researchers predict a monotonic relationship between competition and innovation. On the one hand, some predict that as a negative relationship. Innovation should decline with competition, as more competition reduces the monopoly rents that reward entry by new successful innovators (Caballero & Jaffe, 1993; Dasgupta & Stiglitz, 1980). Interestingly, empirical work has shown a positive correlation between competition and innovation (Blundell, Griffith, & Van Reenen, 1999; Geroski, 1994; Nickell, 1996). The relationship would be negative if firms are value-maximising and positive if they are governed by managers who mainly care about the firm remaining in business (Aghion, Dewatripont, & Rey, 1999).

Other researchers argue that the relationship between competition and innovation is not a monotonic one. Li (2009) find a U-shaped relationship. That means as competition increases from a low to a moderate level, the likelihood of innovation decreases (Li & Vanhaverbeke, 2009). However, the likelihood of innovation increases when foreign competition continues to increase from a moderate to a high level. Conversely, Aghion (2005) found that there is an inverted U-shape relationship between competition and innovation (Scherer & Ross, 1990).

The relationship between competition and innovation is not unambiguous (Li & Vanhaverbeke, 2009). In the order to obtain objective conclusions, it is necessary to approach competition and innovation by specifying the competition context and the type of innovation activities in order to be judged relative and meaningful (Baldwin & Scott, 2013; Tang, 2006). Both competition and innovation have many dimensions, and different innovation activities are associated with different types of competitive pressure. Following Liu (2014), we focus on the relationship between foreign competition and make innovation activities. The main purpose of this study is to investigate how foreign competition may influence the likelihood in generating make innovation activities of a firm.

Innovation is a process of discovery, learning and application of new technologies and techniques from many sources. That is the process of turning knowledge into economic activity (Tang, 2006). According to an anti-competition explanation (Roberts, 1999), under foreign competition in their domestic market (Caves, 1996; Chung, 2001), firms may explore novel technological opportunities and bring about innovations (Cohen & Levinthal, 1990).

In the Schumpeterian model, innovation incentives depend upon post-innovation rents per se (pre-innovation rents were equal to zero), and competition may reduce innovation incentives (Aghion et al., 2002). The large firms with substantial monopoly market power have resources and incentives to innovate (Weinberg, 1992), because they seek profitability that arises from monopoly power (Grossman & Helpman, 1991). The profit from monopoly power will be reinvested in R&D, which leads to more innovations (Greenhalgh & Rogers, 2006). Some studies argue that a less competitive environment was conducive to innovation (Grossman & Helpman, 1991; Ranis & Fei, 1961).

The Schumpeterian model has been extended by allowing incumbent firms to innovate in an alternative approach. In the new model, innovation incentives depend upon the difference between post-innovation and pre-innovation rents (pre-innovation rents were not equal to zero). Therefore, more competition may end up fostering innovations and encourage R&D investments aimed at ‘escaping competition’, as it may reduce a firm’s pre-innovation rents by more than it reduces its post-innovation rents (Aghion, Harris, Howitt, & Vickers, 2001; Aghion, Harris, & Vickers, 1997). A perfectly competitive market is more likely to foster innovation than a monopoly market (Arrow, 1962; Blundell et al., 1999), in the sense that the innovator can license the innovation at full market value (Arrow, 1962). This has been supported by some empirical studies (Arrow, 1962; Blundell et al., 1999), which have found a positive linear effect of competition on innovation (Von Hayek, 2005). However, some results offer stronger support for Schumpeter than for Arrow on the relationship between market structure and the incentives to innovate (Greenhalgh & Rogers, 2006).

The relationship between foreign competition and make innovation activities may be non-monotonic, specifically, a U-shaped relationship. This means as foreign competition increases at a low level, the likelihood of make innovation activities decreases. An increase in foreign competition decreases make innovation activities when the level of innovation activities is relatively low, but increases it when the innovation activities level is relatively high.

Data and Methodology

Quantile Regression Model

To fit regression curves to other parts of the distribution of the response variable is especially problematic for regression models with heterogeneous variances. Therefore, most regression analyses to date provide an incomplete picture of the relationships between variables (Mosteller & Tukey, 1977). In contrast, quantile regression estimates the conditional quantiles of a response variable in a liner model (Cade & Noon, 2003), providing a complete view of the possible relationships between a response variable and explanatory variables (Hallock & Koenker, 2001; Koenker & Bassett, 1978). The regression quantile extends the concepts of quantiles, order statistics and rankings of the linear model (Gutenbrunner, Jurečková, Koenker, & Portnoy, 1993; Koenker & Machado, 1999). Statistical theory and computational routines for estimating and making inferences on regression quantiles are most suited for a linear model (Cade & Noon, 2003; Gutenbrunner et al., 1993; Koenker, 1994; Koenker & Machado, 1999), but are also available for parametric non-linear (Koenker & Park, 1996; Welsh, Carroll, & Ruppert, 1994) and non-parametric, non-linear smoothers (Koenker, 1994; Yu & Jones, 1998).

Quantile regression has been widely applied in different literature streams, such as elasticity of demand work (Goel & Ram, 2004; Manning, Blumberg, & Moulton, 1995; Yoo, Simonit, Kinzig, & Perrings, 2014), educational economics (Arias, Hallock, & Sosa-Escudero, 2001; Eide & Showalter, 1998), economic growth studies (Andrade, Duarte, & Simões, 2014; Barreto & Hughes, 2004), wage analysis (Buchinsky, 1994; Garcia, Hernández, & Lopez-Nicolas, 2001; J.A. Machado & Mata, 2001; Nielsen & Rosholm, 2001), labour economics (Ribeiro, 2001), population economics (Abrevaya, 2002), portfolio investment research (Bassett & Chen, 2001) and ecological science (Cade, Terrell, & Schroeder, 1999; Knight & Ackerly, 2002).

Formally, following Koenker and Bassett (1978), a conditional quantile function can be expressed as follows (Ramdani & van Witteloostuijn, 2010):

Q_{θ} (y_{i} | x_{i}) = α (θ) + {x^{'}}_{i} β (θ) w i t h θ \in (0, 1)

(1)

In this function, y_i is the response variable of observation i, x_i is the vector of covariates representing individual observation i, θ represents the θth quantile, where quantile refers to a point taken along the cumulative distribution, and subscript i = 1,2, . . ., n reflects an index for individual observations. $Q_{θ} (y_{i} | x_{i})$ denotes the θth conditional quantile of y_i given x_i. By way of comparison, recall that the OLS regression function is expressed as $E (y | x) = μ_{y | x} = α + {x^{'}}_{i} β$ , which is the classical linear function that estimates the conditional mean $μ_{y | x}$ , namely, the average value of y for a given value of x (Cade et al., 1999; Koenker & Machado, 1999).

The optimisation problem of the conditional quantile function is as follows:

\frac{\min}{β \in R^{K}} [\sum_{i \in {i : y_{i} \geq {x^{'}}_{i} β^{(θ)}}} θ | y_{i} - {x^{'}}_{i} β^{(θ)} | + \sum_{i \in {i : y_{i} \geq {x^{'}}_{i} β^{(θ)}}} (1 - θ) | y_{i} - {x^{'}}_{i} β^{(θ)} |]

(2)

In this function, R indicates the dimensions of the independent variables (K). The optimisation problem of this function is to search for the θth quantile regression estimators (β ^(θ)) that minimise the absolute value of a weighted sum of the residuals between observed values (y_i) and fitted values $({x^{'}}_{i} β)$ . We assign a weight of θ to the points lying below the quantile regression line (the first term in Equation 2), and a weight of 1–θ to the points located above the quantile regression line (the second term in Equation 2). Furthermore, the estimated covariance matrix $\hat{Σ}$ is estimated through

\sum^{​} = \frac{θ (1 - θ)}{n} \frac{1}{f_{ε^{(θ)}} {(0)}^{2}} {(X X^{'})}^{- 1}

(3)

In this function, $f_{ε^{(θ)}} (0)$ is the probability density of the error term $ε^{(θ)}$ evaluated at the θth quantile of the error distribution (Hao & Naiman, 2007). An estimated standard error for coefficient estimator $({\hat{β}}^{(θ)})$ can be obtained by taking the square root of the corresponding diagonal element of covariance matrix $\hat{Σ}$ .

Note that the estimation of quantile regression coefficients is based on the weighted sum of the residuals for the whole sample, and not just on the portion of the sample at that quantile. Therefore, we never lose degrees of freedom, which is especially important when the number of observations in the sample is not very large. The optimisation problem of the quantile regression function can be solved by linear programming methods (Hao & Naiman, 2007), since the problem is based on order statistics without having an explicit form (Buchinsky, 1994). Additionally, note also that quantile regression classifies the sample into low-up to high-level groups (quantiles) of the dependent variable (y), which is different from simple categorisation. In quantile regression, the grouping of the dependent variable (y) is conditional on the independent variable (x); in sample categorisation, grouping of the dependent variable is just to sort out the value of this dependent variable (Elsayed, 2007).

Although Stata does not provide any specific command to perform a quantile regression-based heteroscedasticity test, a quantile regression-based heteroscedasticity test suggested by Machado and Santos Silva (2000) can be implemented by using qreg2 to solve this problem. Their test statistic can be easily computed as n times the R² of the auxiliary regression of $ρ_{τ} (u_{i} (τ))$ on a constant and on appropriate functions of x. The test can be performed by comparing the test statistic with the critical value from the $χ_{(J - 1)}^{2}$ distribution, where J is the number of parameters in the auxiliary regression (Machado & Santos Silva, 2000).

The Machado–Santos Silva (MSS) test is simple enough to be routinely performed after quantile regression, thereby providing the researcher with information not only about the kind of covariance matrix that is more appropriate but also about the relevance of estimating multiple quantiles (Machado & Parente, 2005).

Data and Measures

Dependent Variables

There is a large body of literature on measuring innovation activities. The most commonly used measures at the firm level are research and development spending, patenting activity, innovation counts and total factor productivity (Klette, 1999; Klette & Griliches, 1996). Following Liu (2014), we divide innovation into make and buy innovations (Liu et al., 2014). The make innovation activities are linked to offering new or significantly improved products (Dasgupta & Stiglitz, 1980), whereas others acquire new technology through outsourcing externally (Bin, 2008). We focus on the make innovation activities, which can be measured by the log of the expenditure on new product and technology development (Li & Wu, 2010; Liu & Buck, 2007).

Explanatory Variables

The prior literature measured competition using different proxies, such as Herfindahl index (Becker & Dietz, 2004), the ratio of product market size against the number of competitors (Ang, 2008) and the ratio of market share against the number of competitors (Wu & Pangarkar, 2010). However, we focus on the foreign competition, which can be measured by the log of the proportion of the import goods to the total production value. Confronted with increasing foreign competition, domestic firms might respond to such competition differently when conducting innovation activities (Liu et al., 2014).

Control Variables

In line with prior studies, we include a number of other control variables, which could have influenced innovation activities. First, we included firm size, because large firms simply have more resources (Ballot, Fakhfakh, Galia, & Salter, 2015), and they are more likely to be more effective in innovation than small firms (Wu, 2012). Small companies may be more likely to explore new technological areas by innovating in less-crowded areas (Almeida & Kogut, 1997). Since firm size has been shown to influence firms’ innovation activities, it is important to control for this variable (Li & Vanhaverbeke, 2009). Following Ballot et al. (2015), we measure firm size using log of the number of employees. Second, we included SIC¹ category of firms with the SIC code. Previous studies found that firms in different industries vary dramatic in their innovation activities (Colombo, Piva, & Rossi-Lamastra, 2014). In neck-and-neck industries, competition is particularly intense and it is also in those industries that the ‘escape competition’ effect pointed out above is strongest (Aghion et al., 2002). Third, following Wu (2012), we controlled for firm age as aged firms would enjoy some advantages of launching new products (Wu, 2012). Fourth, following Li (2009), we controlled for time, considering the timeframe from 1994 to 2004.

Empirical Results

All data used in this study were provided by the World Management Survey². The final sample size, after removing cases with missing values, comprises 211 companies out of the total 290 companies in the USA that met the criteria. Summary statistics and sample sizes are presented in Table 1.

Table 1 provides the relevant descriptive statistics for our variables. Three variables (make innovation activities, foreign competition and firm size) with the exception of dummy variables are transformed into natural logarithms.

Table 1

Descriptive Statistics

Variable	Obs	Mean	Std. Dev	Min.	Max.
Make innovation activities	1993	3.6181	1.5923	–1.7720	7.7200
Foreign competition	1993	–1.5994	0.5749	–3.9094	–0.3844
Firm size	1993	6.7760	1.3822	3.0910	10.5427
SIC category	1993	349.0251	40.4636	203	399
Firm age	1993	45.9252	37.6219	4	204
Time	1993	1999.272	3.0821	1994	2004

Source: World Management Survey (http://worldmanagementsurvey.org).

The main results are presented in Table 2, reporting the OLS and quantile regression outcomes. Both the OLS and robust OLS regression show evidence to support the escape competition effect, which means, on average, the foreign competition is positively associated with make innovation activities (0.5194^***, p < 0.01). In terms of control variables, the variable of firm size is significant indicating that firm size has a positive impact on make innovation activities (0.8387***, p < 0.01). The variable of firm age is significant and negatively influences the make innovation activities (–0.0046***, p < 0.01).

Machado–Santos Silva test results show the following. First, the quantile of 0.25 and 0.50 is significant at the level of 0.1. Second, the quantile of 0.75 and 0.90 is significant at the level of 0.001. This means the error terms are heteroscedastic, and we can reject the null hypothesis ‘the variance of the error term is constant (homoscedastic)’.

The other results in Table 2 are from the quantile regression for θ = 0.1, θ = 0.25, θ = 0.50, θ = 0.75, θ = 0.90. We find that foreign competition has a significantly positive association with make innovation activities, except at θ = 0.1(–0.3593, p = 0.3410 > 0.1). Test shows that the coefficient of foreign competition is not equal at θ = 0.1, θ = 0.25, θ = 0.50, θ = 0.75, θ = 0.90 (F (4,1964) = 2.87**, p < 0.05). The quantile regression results show that the effects of foreign competition differ across the quantiles in the conditional distribution of make innovation activities. To reveal this, the effects for all quantiles are visualized in Figure 1.

Table 2

OLS and Quantile Regression Results

	OLS Regression		Quantile Regression
	OLS	Robust OLS	Q0.10	Q0.25	Q0.5 (median)	Q0.75	Q0.90
Explanatory Variables
Foreign Competition	0.5194^***	0.5194^***	–0.3593	0.6829^***	0.8141^***	0.6787^***	0.6509^***
	(0.0946)	(0.9856)	(0.1399)	(0.1424)	(0.1290)	(0.1092)	(0.0944)
Control Variables
Firm size	0.8387^***	0.8387^***	0.9006^***	0.8428^***	0.8563^***	0.8574^***	0.8446^***
	(0.0211)	(0.0213)	(0.0312)	(0.0318)	(0.0288)	(0.0243)	(0.0211)
Firm age	–0.0046^***	–0.0046^***	–0.0016	–0.0046^***	–0.0045^***	–0.0050^***	–0.0083^***
	(0.0008)	(0.0008)	(0.0012)	(0.0012)	(0.0011)	(0.0010)	(0.0008)
SIC category	Included	Included	Included	Included	Included	Included	Included
Time	Included	Included	Included	Included	Included	Included	Included
Constant	–2.5630^***	–2.5630^***	–5.8723^***	–2.7258^***	–2.4182^**	–2.7181^*	0.1519
	(0.6141)	(0.7304)	(0.9037)	(0.9199)	(0.8328)	(0.7052)	(0.6097)
R ²	0.5128	0.5128	0.3143	0.3064	0.3196	0.3533	0.4191
Machado–Santos Silva Test for Heteroscedasticity
Prob > chi-square			0.295	0.090	0.086	0.000	0.000

Source: Calculated from the data in Table 1 based on the quantile regression model.

Notes: *, ** and *** represent significance at the 10%, 5% and 1% levels, respectively.

We are particularly interested in how the effect of a foreign competition mechanism varies with the quantiles. Note that we only report the findings for the foreign competition variable and not for the control variables. We plot the coefficients of foreign competition variables along the vertical axis and the quantiles along the horizontal axis. The line in the middle of the shaded area reflects the coefficient estimates of the quantile regression in different quantiles. The broken line in each figure provides the standard OLS estimate of the conditional mean effect. The shaded grey area depicts a 90 per cent point-wise confidence band for the quantile regression estimates.

Figure 1

Source: Drawn from quantile regression model.

Figure 1 shows that the foreign competition has the largest positive effect around θ = s0.50, being smaller in all quantiles above and below 0.50. Almost all coefficients of the quantile regression are higher than the estimate from the OLS regression, except at θ = 0.10. This effect is significantly positive, but it is not significant in the quantile lower than 0.20. Here, the quantile regression analysis reveals that the effect of the foreign competition is different across quantiles indeed. In a standard OLS regression, this cannot be revealed as only a single estimate is produced, which is conditional on the mean. There results of quantile regression imply that the foreign competition is an effective factor for firms’ innovation in the quantiles θ = 0.25 and above, but not in the quantiles lower than θ = 0.20(–0.3593, p = 0.3410 > 0.1). In terms of control variables, the variable of firm size is significant and has a positive effect on make innovation activities across quantiles (θ = 0.1, θ = 0.25, θ = 0.50, θ = 0.75, θ = 0.90). The effect of the firm size variable on innovation is negative and significant across quantiles, except the quantile at θ = 0.1(–0.0016, p = 0.2690 > 0.1).

Discussion and Conclusion

Using data of the US firms, we examine the impact of foreign competition on make innovation activities by applying the quantile regression method. This study differs from prior work that analyses the relationship between foreign competition and make innovation activities through mean regression (Aghion et al., 2001; Aghion et al., 2002; Correa, 2012; Greenhalgh & Rogers, 2006; Howitt et al., 2005; Li & Vanhaverbeke, 2009; Liu et al., 2014; Roberts, 1999; Tang, 2006). In contrast, the quantile regression method generates different estimates at conditional quantiles (Ramdani & van Witteloostuijn, 2010). This implies that we can explore the impact of foreign competition on make innovation activities at different levels of innovation activities (0.10, 0.25, 0.50, 0.75 and 0.90). We believe that investigating the effects of foreign competition on make innovation activities at different levels of innovation activities provides a better understanding of the conditionality of this effect, since the empirical literature, to date, does not differentiate these effects across differently innovation activities firms.

Therefore, our study contributes to advance the knowledge on innovation. This study extends prior empirical literature on the effects of foreign competition on make innovation activities. The findings offer valuable empirical evidence on what determines the intensity of make activities. Hence, the findings help to provide new insights into the extent to which foreign competition influences the pattern of innovation activities. Our application of quantile regression revealed that instead of negative (Caballero & Jaffe, 1993; Dasgupta & Stiglitz, 1980), positive (Blundell et al., 1999; Correa, 2012; Geroski, 1994; Nickell, 1996) or inverted-U (Aghion et al., 2002; Howitt et al., 2005), there is a U-shaped relationship between foreign competition and make innovation activities. This means that, as foreign competition increases at a low level, the likelihood of make innovation activities decreases (–0.3593). Our findings are consistent with previous research which found a U-shaped relationship between competition and innovation (Li & Vanhaverbeke, 2009). However, it should be noted that the likelihood of make innovation activities increases when foreign competition continues to increase from a moderate to a high level (0.6829, 0.8141, 0.6787 and 0.6509). This set of findings suggests that the impact of foreign competition is conditioned non-linearly by initial innovation activities.

The reason why foreign competition is negatively associated with make innovation activities at a low innovation activities level is that make innovation activities may be too slow to pre-empt competitive threats (Liu et al., 2014). Firms at a low innovation activities level do not need good development by innovation, but a chance to survive. As firm innovation spending increases, the foreign competition is significantly positive above 0.25 quantiles, reaching its highest point at 0.50 (0.8141^***). As firm moves into a top rank, the impact effect will turn back to 0.6787^*** and 0.6509^***, at θ = 0.75, θ = 0.90, respectively.

The findings in this study have implications for our theoretical understanding. Specifically, the impact of foreign competition depends on the level of firm innovation activities and underscores Contingency Theory (Liu et al., 2014). Contingency theory suggests that all firms’ practices, such as make innovation activities, are contingent on the firm’s environment (Drazin & Van de Ven, 1985). The key concept in a contingent proposition is fit, and the fit is between environmental contingencies and internal configurations, which leads to a greater understanding of this relationship (Donaldson, 2001; Sirmon, Hitt, & Ireland, 2007). Contingency theory has been used in the area of business strategy (Aragon-Correa & Sharma, 2003; Hofer, 1975; Priem & Butler, 2001), quality management (Das, Handfield, Calantone, & Ghosh, 2000), and organisational change (Battilana & Casciaro, 2012). Our findings suggest that the impact of foreign competition is conditional on the level of initial firm innovation activities and the firm’s environment.

This study has also important policy implications. First, the findings from our research help policymakers understand the conditions under which different types of innovation activities occur. Increasing foreign competition will bring new competitive pressures for domestic enterprises but may also represent opportunities. The government can implement a policy on promoting foreign direct investments and imports of foreign technology from other countries (Guo, 2008; Liu et al., 2011). The governments of developed economies can design appropriate policy to create an innovation-enhancing enterprise environment to respond to an intensified foreign competition. Second, our findings imply that it may be crucial for the governments of developed economies to adopt a combined technological development strategy, which encourages indigenous firms to obtain international technology to catch up with technological leaders, under the competitive pressure.

Liu (2014) found that foreign competition is negatively associated with make activities in China. We find that, in the USA, only at 0.1 quantile, foreign competition has a negative association with make innovation activities. Besides, there is a significantly positive association. In China, the foreign competition plays an important role in the technological upgrading of Chinese enterprises (Guo, 2008; Liu and Buck, 2007), and make activities may be too slow to pre-empt foreign competition. It is more rational for enterprises make buy activities to upgrade technology in the face of competitive threats. Hence, make activities are feasible when industries face less foreign competition in China. However, with the continuous improvement of Chinese enterprise innovation ability, the innovation activities of Chinese enterprises may also change. This means that once Chinese enterprise innovation ability comes into the developed area, we should note that the relationship between foreign competition and innovation activities will venture to the opposite side. The government should pay special attention to the guidance and adjustment of macro-policies. At the same time, China’s negative correlation shows that the development of China’s enterprise innovation capability has not reached the developed level. We need to constantly improve the institutional arrangements to strengthen the protection of intellectual property rights.

There are some limitations to this study. The study is limited in the context of a single country. Future research could be extended to other countries, especially in some developing countries, and compare how firms in different countries undertake different innovation activities. This study demonstrates that quantile regression can provide additional insight into the understanding of the relationship between foreign competition and make innovation activities. We believe that there is a research opportunity to apply quantile regression widely in the innovation research.

Footnotes

Acknowledgements

The authors would like to thank Krishna V. V. (Editor in Chief) and two anonymous reviewers for their constructive comments and suggestions to improve the article on an earlier version of the article. This work was supported by the National Natural Science Foundation of China (71403066); the State Key Program of National Social Science of China (14AGL004); the Fund of Ministry of Education of China (20122304120021); China Postdoctoral Science Foundation (2013T60351); the Special Foundation of Central Universities Basic Research Fee (HEUCF170903); Heilongjiang Province Postdoctoral start fund (LBH-Q13050); the Scientific and Technological Projects of Heilongjiang Province (GZ11D203); and Project of Education Department of Heilongjiang Province (12534028).

Notes

References

Abrevaya

(2002). The effects of demographics and maternal behavior on the distribution of birth outcomes. In Economic applications of quantile regression (pp. 247–257). Springer.

Aghion

Bloom

Blundell

Griffith

& Howitt

(2005). Competition and innovation: An inverted U relationship. National Bureau of Economic Research.

Aghion

Dewatripont

& Rey

(1999). Competition, financial discipline and growth. Review of Economic Studies, 66(4), 825–852.

Aghion

Harris

Howitt

& Vickers

(2001). Competition, imitation and growth with step-by-step innovation. The Review of Economic Studies, 68(3), 467–492.

Aghion

Harris

& Vickers

(1997). Competition and growth with step-by-step innovation: An example. European Economic Review, 41(3), 771–782.

Alderson

(1965). Dynamic marketing behavior: A functionalist theory of marketing. IL, Homewood: Irwin Homewood.

Almeida

& Kogut

(1997). The exploration of technological diversity and geographic localization in innovation: Start-up firms in the semiconductor industry. Small Business Economics, 9(1), 21–31.

Amabile

& Gryskiewicz

S. S.

(1987). Creativity in the R&D Laboratory. Technical Report Number 30, May 1987 / Center for Creative Leadership.

Amabile

T. M.

(1996). Creativity in context: update to the psychology of creativity. High Ability Studies,(2), 100–101.

10.

Amabile

T. M.

Conti

Coon

Lazenby

& Herron

(1996). Assessing the work environment for creativity. Academy of Management Journal, 39(5), 1154–1184.

11.

Amendola

Gaffard

J. L.,

& Musso

(2000). Competition, innovation and increasing returns. Economics of Innovation and New Technology, 9(2), 149–181.

12.

Andrade

J. S.

Duarte

& Simões

(2014). A quantile regression analysis of growth and convergence in the EU: Potential implications for Portugal. Notas Económicas, (39), 48–72.

13.

Ang

S. H.

(2008). Competitive intensity and collaboration: Impact on firm growth across technological environments. Strategic Management Journal, 29(10), 1057–1075.

14.

Aragon-Correa

J. A.

& Sharma

(2003). A contingent resource-based view of proactive corporate environmental strategy. Academy of management review, 28(1), 71–88.

15.

Arias

Hallock

K. F.

& Sosa-Escudero

(2001). Individual heterogeneity in the returns to schooling: instrumental variables quantile regression using twins data. Empirical Economics, 26(1), 7–40.

16.

Arrow

(1962). Economic welfare and the allocation of resources for invention. In Richard

(Ed.), The rate and direction of inventive activity: Economic and social factors (pp. 609–626). Calenda: NBER.

17.

Baldwin

& Scott

(2013). Market structure and technological change. London: Taylor & Francis.

18.

Ballot

Fakhfakh

Galia

& Salter

(2015). The fateful triangle: Complementarities in performance between product, process and organizational innovation in France and the UK. Research Policy, 44(1), 217–232.

19.

Barney

(1991). Firm resources and sustained competitive advantage. Journal of Management, 17(1), 99–120.

20.

Barreto

R. A.

& Hughes

A. W.

(2004). Under performers and over achievers: A quantile regression analysis of growth. Economic Record, 80(248), 17–35.

21.

Bassett

Jr.

Gilbert

& Chen

H. L.

(2001). Portfolio style: return-based attribution using quantile regression. Empirical Economics, 26(1), 293–305.

22.

Battilana

& Casciaro

(2012). Change agents, networks, and institutions: A contingency theory of organizational change. Academy of Management Journal, 55(2), 381–398.

23.

Becker

& Dietz

(2004). R&D cooperation and innovation activities of firms—evidence for the German manufacturing industry. Research Policy, 33(2), 209–223.

24.

Bin

(2008). Technology acquisition channels and industry performance: An industry-level analysis of Chinese large-and medium-size manufacturing enterprises. Research Policy, 37(2), 194–209.

25.

Blundell

Griffith

& Van Reenen

(1999). Market share, market value and innovation in a panel of British manufacturing firms. The Review of Economic Studies, 66(3), 529–554.

26.

Bogliacino

& Pianta

(2012). Profits, R&D, and innovation—A model and a test. Industrial and Corporate Change, 22(3), 649–678. doi:10.1093/icc/dts028

27.

Buchinsky

(1994). Changes in the US wage structure 1963–1987: Application of quantile regression. Econometrica: Journal of the Econometric Society, 62(2), 405–458.

28.

BurgeSmani

R. A.

& Wheelwright

S. C.

(2004). Strategic management of technology and innovation. Reading, 1(1), 66–77.

29.

Caballero

R. J.

& Jaffe

A. B.

(1993). How high are the giants’ shoulders: An empirical assessment of knowledge spillovers and creative destruction in a model of economic growth. discussion. NBER Macroeconomics Annual, 8(8), 85–86.

30.

Cade

B. S.

& Noon

B. R.

(2003). A gentle introduction to quantile regression for ecologists. Frontiers in Ecology and the Environment, 1(8), 412–420.

31.

Cade

B. S.

Terrell

J. W.

& Schroeder

R. L.

(1999). Estimating effects of limiting factors with regression quantiles. Ecology, 80(1), 311–323.

32.

Cassiman

& Veugelers

(2006). In search of complementarity in innovation strategy: Internal R&D and external knowledge acquisition. Management science, 52(1), 68–82.

33.

Caves

R. E.

(1996). Multinational enterprise and economic analysis. Cambridge: Cambridge university press.

34.

Chung

(2001). Mode, size, and location of foreign direct investments and industry markups. Journal of Economic Behavior & Organization, 45(2), 185–211.

35.

Clement

(2003). Creation Myths. Reason, 34(10), 30–38.

36.

Cohen

W. M.

& Levinthal

D. A.

(1990). Absorptive capacity: a new perspective on learning and innovation. Administrative Science Quarterly, 35(1), 128–152.

37.

Colombo

M. G.

Piva

& Rossi-Lamastra

(2014). Open innovation and within-industry diversification in small and medium enterprises: The case of open source software firms. Research Policy, 43(5), 891–902.

38.

Correa

J. A.

(2012). Innovation and competition: An unstable relationship. Journal of Applied Econometrics, 27(1), 160–166.

39.

Das

Handfield

R. B.

Calantone

R. J.

& Ghosh

(2000). A contingent view of quality management: The impact of international competition on quality. Decision Sciences, 31(3), 649–690.

40.

Dasgupta

& Stiglitz

(1980). Industrial structure and the nature of innovative activity. Economic Journal, 90(358), 266–293.

41.

Donaldson

(2001). The contingency theory of organizations. Sydney: SAGE Publications.

42.

Drazin

& Van de Ven

A. H.

(1985). Alternative forms of fit in contingency theory. Administrative science quarterly, 30(4), 514–539.

43.

Eide

& Showalter

M. H.

(1998). The effect of school quality on student performance: A quantile regression approach. Economics letters, 58(3), 345–350.

44.

Elsayed

(2007). Does CEO duality really affect corporate performance? Corporate Governance: An International Review, 15(6), 1203–1214.

45.

Garcia

Hernández

P. J.

& Lopez-Nicolas

(2001). How wide is the gap? An investigation of gender wage differences using quantile regression. Empirical economics, 26(1), 149–167.

46.

Geroski

P. A.

(1994). Market structure, corporate performance, and innovative activity. Oxford: Clarendon Press Oxford.

47.

Goel

R. K.

& Ram

(2004). Quantile-regression estimates of cigarette demand elasticities for the United States. Journal of Economics and Finance, 28(3), 413–421.

48.

Greenhalgh

& Rogers

(2006). The value of innovation: The interaction of competition, R&D and IP. Research Policy, 35(4), 562–580.

49.

Greis

N. P.

Dibner

M. D.

& Bean

A. S.

(1995). External partnering as a response to innovation barriers and global competition in biotechnology. Research Policy, 24(4), 609–630.

50.

Grossman

G. M.

& Helpman

(1991). Quality ladders in the theory of growth. The Review of Economic Studies, 58(1), 43–61.

51.

Guan

& Chen

(2009). The technological system of Chinese manufacturing industry: A sectorial approach. China Economic Review, 20(4), 767–776.

52.

Gutenbrunner

Jurečková

Koenker

& Portnoy

(1993). Tests of linear hypotheses based on regression rank scores. Journal of Nonparametric Statistics, 2(4), 307–331.

53.

Hallock

& Koenker

(2001). Quantile regression. Journal of Economic Perspectives, 15(4), 143–156.

54.

Hao

& Naiman

D. Q.

(2007). Quantile regression. Washington, DC: SAGE.

55.

Hofer

C. W.

(1975). Toward a contingency theory of business strategy. Academy of Management Journal, 18(4), 784–810.

56.

Howitt

Griffith

Aghion

Blundell

& Bloom

(2005). Competition And Innovation: An Inverted-U Relationship. Quarterly Journal of Economics, 120(2), 701–728.

57.

Jacobides

M. G.

& Winter

S. G.

(2012). Capabilities: Structure, agency, and evolution. Organization Science, 23(5), 1365–1381.

58.

Jung

& Lee

(2010). Sectoral systems of innovation and productivity catch-up: determinants of the productivity gap between Korean and Japanese firms. Industrial and Corporate Change, 19(4), 1037–1069.

59.

Kejžar

K. Z.

(2016). Shutdown versus M&A: An empirical investigation of Slovenian incumbent firms’ responses to foreign competition. Economic Systems, 40(2), 247–259.

60.

Klette

T. J.

(1999). Market power, scale economies and productivity: Estimates from a panel of establishment data. The Journal of Industrial Economics, 47(4), 451–476.

61.

Klette

T. J.

& Griliches

(1996). The inconsistency of common scale estimators when output prices are unobserved and engogenous. Journal of Applied Econometrics, 11(4),343–361.

62.

Knight

C. A.

& Ackerly

D. D.

(2002). Variation in nuclear DNA content across environmental gradients: A quantile regression analysis. Ecology Letters, 5(1), 66–76.

63.

Koenker

(1994). Confidence intervals for regression quantiles. In Mandl

& Huskova

, (Eds), Asymptotic statistics (pp. 349–359). Heidelberg: Springer.

64.

Koenker

& Bassett

Jr. (1978). Regression quantiles. Econometrica: Iournal of the Econometric Society, 46(1), 33–50.

65.

Koenker

& Machado

J. A. F.

(1999). Goodness of fit and related inference processes for quantile regression. Journal of the American Statistical Association, 94(448), 1296–1310.

66.

Koenker

& Park

B. J.

(1996). An interior point algorithm for nonlinear quantile regression. Journal of Econometrics, 71(1), 265–283.

67.

& Wu

(2010). In-house R&D, technology purchase and innovation: Empirical evidences from Chinese hi-tech industries, 1995–2004. International Journal of Technology Management, 51(2), 217–238.

68.

& Vanhaverbeke

(2009). The relationships between foreign competition, absorptive capacity and pioneering innovation: An empirical investigation in Canada. International Journal of Innovation Management, 13(1), 105–137.

69.

Ying

, & Vanhaverbeke, Wim. (2009). The relationships between foreign competition, absorptive capacity and pioneering innovation: An empirical investigation in Canada. International Journal of Innovation Management, 13(1), 105–137.

70.

Liu

& Buck

(2007). Innovation performance and channels for international technology spillovers: Evidence from Chinese high-tech industries. Research Policy, 36(3), 355–366.

71.

Liu

Hodgkinson

I. R.

& Chuang

F.M.

(2014). Foreign competition, domestic knowledge base and innovation activities: Evidence from Chinese high-tech industries. Research Policy, 43(2), 414–422.

72.

Liu

Xiaohui

Hodgkinson

Ian R

., & Chuang

Fu-Mei

. (2014). Foreign competition, domestic knowledge base and innovation activities: Evidence from Chinese high-tech industries. Research Policy, 43(2), 414–422.

73.

Machado

J. A.

F, & Parente

(2005). Bootstrap estimation of covariance matrices via the percentile method. The Econometrics Journal, 8(1), 70–78.

74.

Machado

J. A. F

& Santos Silva

J. M. C.

(2000). Glejser’s test revisited. Journal of Econometrics, 97(1), 189–202.

75.

Machado

José A. F.

& Mata

(2001). Earning functions in Portugal 1982–1994: Evidence from quantile regressions. Empirical Economics, 26(1), 115–134.

76.

Malerba

(2002). Sectoral systems of innovation and production. Research policy, 31(2), 247–264.

77.

Malerba

(2005). Sectoral systems of innovation: A framework for linking innovation to the knowledge base, structure and dynamics of sectors. Economics of Innovation and New Technology, 14(1–2), 63–82.

78.

Manning

W. G.

Blumberg

& Moulton

L. H.

(1995). The demand for alcohol: The differential response to price. Journal of Health Economics, 14(2), 123–148.

79.

McKelvey

Zaring

& Ljungberg

(2014). Creating innovative opportunities through research collaboration: An evolutionary framework and empirical illustration in engineering. Technovation, 39(11), 26–36.

80.

Mosteller

& Tukey

J. W.

(1977). Data analysis and regression: A second course in statistics (Addison-Wesley Series in Behavioral Science: Quantitative Methods). Mass: Addison-Wesley.

81.

Nelson

R. R.

(1991). Why do firms differ, and how does it matter? Strategic management journal, 12(S2), 61–74.

82.

Nickell

S. J.

(1996). Competition and corporate performance. Journal of Political Economy, 104(4), 724–746.

83.

Nielsen

H. S.

& Rosholm

(2001). The public-private sector wage gap in zambia in the 1990s: a quantile regression approach. Empirical Economics, 26(1), 169–182.

84.

Pérez-Luño

& Cambra

(2013). Listen to the market: Do its complexity and signals make companies more innovative? Technovation, 33(6), 180–192.

85.

Prajogo

D. I.

& Ahmed

P. K.

(2006). Relationships between innovation stimulus, innovation capacity, and innovation performance. R&D Management, 36(5), 499–515.

86.

Priem

R. L.

& Butler

J. E.

(2001). Is the resource-based “view” a useful perspective for strategic management research? Academy of management review, 26(1), 22–40.

87.

Ramdani

& van Witteloostuijn

(2010). The impact of Board Independence and CEO Duality on Firm Performance: A Quantile Regression Analysis for Indonesia, Malaysia, South Korea and Thailand. British Journal of Management, 21(3), 607–626. doi: 10.1111/j.1467-8551.2010.00708.x

88.

Ranis

& Fei

J. C.

H. (1961). A theory of economic development. The American Economic Review, 51(4), 533–565.

89.

Ribeiro

E. P.

(2001). Asymmetric labor supply. Empirical Economics, 26(1), 183–197.

90.

Ringler

Cenacchi

Koo

Robertson

R. D.

Fisher

Cox

C. M.

, et al. (2014). Sustainable agricultural intensification: The promise of innovative farming practices (pp. 43–52). Washington, DC: IFPRI.

91.

Roberts

P. W.

(1999). Product innovation, product-market competition and persistent profitability in the US pharmaceutical industry. Strategic Management Journal, 20(7), 655–670.

92.

Roberts

P. W.

& Amit

(2003). The dynamics of innovative activity and competitive advantage: The case of Australian retail banking, 1981 to 1995. Organization Science, 14(2), 107–122.

93.

Sakakibara

& Porter

M. E.

(2001). Competing at home to win abroad: evidence from Japanese industry. Review of Economics and Statistics, 83(2), 310–322.

94.

Scherer

F. M.

& Ross

(1990). Industrial market structure and economic performance. University of Illinois at Urbana-Champaign’s Academy for Entrepreneurial Leadership Historical Research Reference in Entrepreneurship. Retrieved from SSRN: https://ssrn.com/abstract=1496716

95.

Shalley

C. E.

& Oldham

G. R.

(1997). Competition and creative performance: Effects of competitor presence and visibility. Creativity Research Journal, 10(4), 337–345.

96.

Sirmon

D. G.

Hitt

M. A.

& Ireland

R. D.

(2007). Managing firm resources in dynamic environments to create value: Looking inside the black box. Academy of management review, 32(1), 273–292.

97.

Smith

& Sharif

(2007). Understanding and acquiring technology assets for global competition. Technovation, 27(11), 643–649.

98.

Suárez

(2014). Persistence of innovation in unstable environments: Continuity and change in the firm’s innovative behavior. Research Policy, 43(4), 726–736.

99.

Tang

(2006). Competition and innovation behaviour. Research Policy, 35(1), 68–82.

100.

Tierney

& Farmer

S. M.

(2002). Creative self-efficacy: Its potential antecedents and relationship to creative performance. Academy of Management Journal, 45(6), 1137–1148.

101.

Veugelers

(1997). Internal R&D expenditures and external technology sourcing. Research policy, 26(3), 303–315.

102.

Veugelers

& Cassiman

(1999). Make and buy in innovation strategies: evidence from Belgian manufacturing firms. Research Policy, 28(1), 63–80.

103.

Von Hayek

F. A.

(2005). Die verfassung der freiheit (Vol. 3). Deutscher: Mohr Siebeck.

104.

Wang

L. F. S.

Lee

J. Y.,

& Hsu

C. C.

(2014). Privatization, foreign competition, and social efficiency of free entry. International Review of Economics and Finance, 31(338), 138–147.

105.

Weinberg

J. A.

(1992). Exclusionary practices and technological competition. The Journal of Industrial Economics, 40(2), 135–146.

106.

Welsh

A. H.

Carroll

R. J.

& Ruppert

(1994). Fitting heteroscedastic regression models. Journal of the American statistical Association, 89(425), 100–116.

107.

(2012). Technological collaboration in product innovation: The role of market competition and sectoral technological intensity. Research Policy, 41(2), 489–496.

108.

& Pangarkar

(2010). The bidirectional relationship between competitive intensity and collaboration: Evidence from China. Asia Pacific Journal of Management, 27(3), 503–522.

109.

Yoo

Simonit

Kinzig

A. P.

& Perrings

(2014). Estimating the Price Elasticity of Residential Water Demand: The Case of Phoenix, Arizona. Applied Economic Perspectives and Policy, 36(2), 333–350.

110.

& Jones

M. C.

(1998). Local linear quantile regression. Journal of the American statistical Association, 93(441), 228–237.

111.

Yuchtman

& Seashore

S. E.

(1967). A system resource approach to organizational effectiveness. American sociological review, 32(6), 891–903.