1. Introduction
According to the Oslo manual conceptualization [
1] (p. 48), product innovation can be defined as “both the introduction of new goods and services and significant improvements in the functional or user characteristics of existing goods and services”. The existing research indicates that, in a world increasingly featured by turbulence, frenetic competition, and temporary advantages [
2], product innovation becomes crucial in sustainable long-term performance and firm’s competitiveness [
3]. Nevertheless, product innovation efforts usually involve significant risks that are due to the irreversible expenses incurred and the uncertain future payoffs [
4,
5].
Family businesses really do matter to the world economy. Family firms represent two-thirds of all firms worldwide, creating approximately 70–90% of annual global Gross Domestic Product (GDP), and providing 50–80% of employment around the world [
6]. In Europe, these types of businesses generate around 50% of GDP and they create more than 60 million jobs in the private sector [
7]. Furthermore, there are solid motives to support that family involvement influences how resources are managed and deployed [
8], as it implies peculiar incentives [
9,
10], authority structures, and accountability norms [
11]. In this regard, families can be involved in a firm through ownership or through management. Family involvement in ownership reflects family participation in the firm property; meanwhile, family involvement in management reflects family participation in strategic decision-making [
12]. In any case, family involvement entails unique strengths, such as tacit knowledge or social capital, and weaknesses, as, for instance, nepotism and altruistic behavior [
13,
14], which may relevantly impact on the features of product innovation in family firms. Thus, family involvement may be essential in differentiating family and non-family firms regarding product innovation [
15].
Nevertheless, family firms are a heterogeneous group [
16], and there is also a need for making a comparison between them [
17]. A major source of these firms’ heterogeneity stands in the extent to which family members occupy the upper echelon positions [
18]. Specifically, high or low family involvement could feature the top management team of the firm [
19]. The level of family management may directly influence the way in which product innovation is conducted. Therefore, family involvement in management affects the way in which the strategy is planned and developed [
20], and it provides a peculiar ability to dynamically orchestrate resources [
21], influencing innovation processes [
4,
22].
The scarce empirical findings regarding the influence of family management on product innovation in public firms are mixed: family management has been shown to have a positive effect on patent counts and the subsequent citation of patents [
23], but it has also been confirmed to exert a negative effect on the patent counts [
24]. For private firms, according to our knowledge, until now prior empirical literature has not explored how product innovation is related to family involvement in management, thus offering us the opportunity to examine this interplay. In short, notwithstanding the importance of this topic, no studies have empirically investigated the mechanisms and antecedents of product innovation in private family managed firms to date. Hence, we answer the call for the study of family firms’ innovative behavior while taking into account their heterogeneity in terms of management [
25]. We also respond to the specific call to clarify the way in which family firms manage product innovations, inasmuch as it continues to be an overlooked topic in already published business research [
15].
By drawing on the socioemotional wealth (SEW) approach [
26,
27], we address this gap in the research by analyzing how the level of family management affects the likelihood of introducing a product innovation through intramural (in-house R&D or internal R&D) and extramural R&D spending (R&D outsourcing or extramural R&D) as the mediating variables. We also examine the mediating effect of the investment strategy that simultaneously combines both intramural and extramural R&D in the family management-product innovation relationship. The function of intramural R&D as a source of innovation is indisputable [
28], extramural R&D has been recognized as crucial for sustained success in product innovation [
29], and the combination of both intramural and extramural R&D has been identified as a successful strategy in obtaining product innovations [
30]. However, research regarding innovation in private family firms has overlooked not only how the presence of family members in top management teams influences product innovation, but also the mediating role of R&D processes that underlie the introduction of product innovations, making it difficult to appreciate why some family firms demonstrate more innovative attitudes and behavior than others.
According to the Oslo Manual [
1] (p. 97), intramural (in-house) R&D is the “creative work undertaken on a systematic basis within the enterprise in order to increase the stock of knowledge and use it to devise new applications. This comprises all R&D conducted by the enterprise, including basic research”. On the other hand, the acquisition of extramural R&D comprises the “same activities as intramural R&D, but purchased from public or private research organizations or from other enterprises (including other enterprises within the group)”.
This paper makes several contributions. Specifically, we first explore the relationship between family involvement in management and the likelihood of introducing product innovations in private firms, advancing the understanding on how, why, and to what extent the level of family management influences product innovation in non-quoted firms. Second, we offer insights into how the degree of family involvement in management produces heterogeneity in the likelihood of introducing product innovations. Third, we focus on R&D investments as a mediating factor that helps to explain the relationship between family management and product innovation in a private firm context. To the best of our knowledge, none of the prior studies investigating the role of R&D investment in family firms had used it as a mediator.
Utilizing a balanced and longitudinal dataset that was made up of 7264 time-series cross-sectional observations, which consisted of 454 Spanish firms over the period 2000–2015, carried out the empirical analysis. We test the theory using data on Spanish manufacturing firms from the database ESEE (Survey on Business Strategies).
The paper is structured, as follows: We start with a brief theoretical discussion of product innovation in family firms and of the relationship between R&D investments and product innovation. We derive hypotheses and empirically test them. We present the empirical results and conclude with a discussion of the implications of our research.
5. Results
Table 3 provides descriptive statistics, frequencies for the categorical variable, and the correlation matrix.
Table 3, Panel B, reflects that the likelihood of obtaining product innovation is reached in 22.21% of the analyzed firms. As reflected in Panel C, multicollinearity should not be a concern in our sample, because the correlations between the various items are well below the 0.80 threshold, above which multicollinearity threats could arise [
146], with the exception of the correlations among intramural&extramural R&D expenses and, intramural R&D on the one hand, and extramural R&D on the other hand, which are not considered in the same regression models. Furthermore, variance inflation factor tests did not exceed 3.01 and the condition index did not exceed 3.271, also suggesting the absence of multicollinearity (results are available from the authors). Therefore, there is sufficient evidence to discard multicollinearity in the data [
147]. We also conducted the Ramsey test to look for model misspecification concerns and concluded that the models were adequately specified [
148].
We provide tests for our hypotheses in
Table 4 and
Table 5, in which the causal step approach is followed [
146]. In
Table 4, Model 1 corresponds to the first condition of mediation. Models 2–4 in
Table 4 correspond to the second condition of mediation and Models 5–7 allow for the evaluation of the third and fourth conditions [
137].
In
Table 4, Model 1 reveals a negative effect of Family management on the likelihood of obtaining product innovation, not supporting Hypothesis 1. Models 2–4 analyze the impact of family management on intramural, extramural, and both intramural&extramural R&D expenses, respectively. In this case, we find support for Hypotheses 2 and 4, to but not for Hypothesis 3. Thus, family management significantly influences intramural R&D and the strategy of jointly developing intramural and extramural R&D investments, but it does not lead to extramural R&D. Model 5 and Model 7 show that the effect of family management on product innovation is mediated by intramural R&D and by intramural&extramural R&D, respectively. Thus, hypotheses 5 and 7 are supported. In both cases, the data show partial mediation [
137].
Table 5 confirms the direct effect of intramural, external, and intramural&extramural R&D on product innovation, which is a necessary condition for mediation.
Hypothesis 6 is not supported to the extent that family management does not lead to extramural R&D. Thus, Model 3 in
Table 4 shows that extramural R&D expenses do not mediate the relationship between family management and product innovation. In this case, conditions 1, 3, and 4 in the mediation model were supported. As stated above, family management negatively influences the likelihood of obtaining product innovation (
Table 4, Model 1). Moreover, the simultaneous effect of family management and extramural R&D on product innovation is significant (
Table 4, Model 6).
We also executed the regression analysis of extramural R&D over product innovation, confirming a positive and significant effect (
Table 5, Model 2), as established in previous literature [
61]. Nevertheless, inasmuch as the relationship between family management and extramural R&D is not accomplished, not all of the conditions are met to find mediation [
137]. Specifically, the second condition is lacking, which is the direct effect of the independent variable on the mediating variable. The Sobel test confirmed non-mediation [
138] (
Table 6).
Table 6 reports the results of Sobel (1982) tests [
138] and Monte Carlo confidence intervals [
139] for all potential mediators. The z-score column contains the Sobel statistic. The mediation relationship exists and there is a significant indirect effect, if the Sobel test
z-value is above 1.9 [
144]. The last columns in this table contain the lower and upper limits of a 95% confidence interval on the indirect effect while using a Monte Carlo simulation with 20,000 repetitions [
139]. When considering the normal sampling errors for the analyzed coefficients, random draws from the coefficients’ distributions are developed to simulate the magnitude of the indirect effect, and then a confidence interval is built form these simulations.
The results in
Table 4 show that intramural R&D and the joint intramural&extramural R&D expenses met the conditions for the mediating relationship between family management and product innovation, while the extramural R&D do not satisfy the requirements for mediation. The tests in
Table 6 reinforce these results: the Sobel test is weakly significant for intramural R&D, insignificant for extramural R&D, and strongly significant for intramural&extramural R&D. Moreover, the Monte Carlo simulation suggests the confidence intervals of the indirect effect associated with intramural and intramural&extramural R&D are statistically significantly different from 0, which allows for us to reject the null hypothesis that these variables have no indirect effect.
Thus, these findings give strong support to Hypotheses 5 and 7, but they are not consistent with Hypothesis 6.
Therefore, intramural R&D, extramural R&D, and the joint development of intramural and extramural R&D investments lead to product innovation, but extramural R&D does not act as a mediator in the family management-product innovation relationship.
Furthermore, although we control for complementarities among intramural and extramural R&D expenses by adding alternative past R&D expenses in both equations, we also checked for multiple mediation by simultaneously introducing both mediators. The findings (not reported) were similar to those in the individual models.
6. Discussion
Product innovation is a major decider factor of firm performance [
149] and it is quite essential in the firm’s survival, competitiveness, and sustainability [
150]. This is also true for family firms, which require product innovation to survive throughout different generations [
151] and to achieve their goals [
152]. However, the study of product innovation in the family business field has been very limited and it mainly focused on public companies. Surveys distinguishing between family and non-family firms [
36,
37,
38] and family ownership [
153] have served as a useful lens for examining product innovation in the scant studies that were conducted in a context of private firms. However, prior literature did not explore how such product innovation is related to the level of family involvement in management in non-quoted firms, thus giving us the opportunity to investigate not only the differences between family and non-family firms, but also the heterogeneity within the pool of family firms [
104].
Our findings reveal that the level of family management negatively influences the likelihood of introducing a product innovation, which is in line with previous research hypothesizing a negative relationship between family management and patents citations [
24]. Our results seem to confirm that the product innovation performance depends on hiring and maintaining greatly competent staff [
154]. However, as family management increases, the level of professional executives will decrease [
155]. Family managers might suffer of a lack of expertise, skills, and resources due to altruism and/or nepotism [
83]. In these situations, untalented family members without the appropriate expertise benefit, provoking an inability of the firm to retain qualified workers [
8]. In short, as family management increases the dark side of SEW [
156], it is more likely to limit merited human capital and intensify conflicts [
157] through nepotism and the generation of jobs for unskilled family members [
158], which seems to result in a reduced likelihood of introducing product innovations.
Second, this research sheds new light on the consequences of the degree of family management on product innovation, exploring the mediating effects of intramural R&D, extramural R&D, and the strategy of simultaneously investing in both intramural and extramural R&D. Our findings show that the level of family involvement in management has a positive and significant effect on in-house R&D and on the investment strategy that combines intramural and extramural R&D and, through them, exerts a positive impact on the introduction of product innovations. Therefore, this manuscript goes beyond input-output models showing that in-house R&D and the investment strategy combining intramural and extramural R&D play a mediating role in the understanding of the connection between the level of family involvement in management and the achieving of product innovations.
This study underlines the importance of R&D investments for family firms and the relevance of explicitly measuring their heterogeneity when addressing different R&D strategies, because the behaviors of family firms are more heterogeneous than those of nonfamily firms [
75]. Specifically, our results support a strong effect of family management on intramural R&D spending. In this way, we extend family business literature illustrating how the level of family management increases intramural R&D investments and disentangle the mixed findings found until now regarding the influence of family management on R&D intensity—without distinguishing between intra and extramural R&D [
44,
77]. Our research shows that when firms decide to specifically carry out intramural R&D investments, the higher the number of family members that are actively involved in firm management, the greater the priority of long-term orientation over risk aversion [
77], and in the end, the higher the spending. As family management augments, family firms seem to lose their fear of jeopardize both their firm’s economic performance and their family goals, when they are making decisions regarding increasing the more controllable in-house R&D. However, there appears to be a trade-off between long-term perspective and risk aversion when making a decision concerning extramural R&D, which results in a non-significant influence of the level of family management on extramural R&D investments. The lower expected pressure of family managers to achieve short-term results and their longer investment horizons [
159] might compensate for the apparent conflict between the pursuance of family-centered goals and extramural R&D investments. This confirms that firms’ governance structures determine their R&D behavior. These results contribute to a better comprehension of family businesses on the whole, family managed firms in particular, and their R&D decisions.
In addition, we refined the product innovation model by introducing and testing the mediating effect of three different R&D strategies: intramural R&D, extramural R&D, and investment strategy combining both intramural and extramural R&D. In support of this thesis, we found different results for intramural and extramural R&D. Our data showed that the level of family management does not led to increasing extramural R&D. Thus, extramural R&D does not mediate the level of family management-product innovation relationship, but it is found to be a direct antecedent of product innovation. Our findings do reveal a very important role of intramural R&D and the investment strategy in both in-house and extramural R&D, in explaining the influence of the level of management on product innovation. Particularly, we argue that, as family management increases, firms invest more in in-house R&D, because they are seen as a source of alignment between economic and family goals. The higher the level of family management, the greater the likelihood of firm awareness that failing to invest in intramural R&D results in a loss of opportunity to ensure and create future socioemotional wealth [
75]. Our empirical results confirm that economic goals, far to collide with, are aligned with family goals [
75,
160,
161], because, as family management intensifies, firms are more likely to invest more in intramural R&D and, in turn, to obtain product innovation. This type of intramural investment may be seen by family managers as a method to defend products from imitation by rivals [
21]. Furthermore, our results suggest that family managers contemplate extramural R&D as a means of obtaining external know-how that can then be applied to internal knowledge of the firm in such a way that this combination will be unique, particular to this business, and thus relevant for obtaining a better likelihood of the achievement of product innovation [
8,
28]. Therefore, this study also confirms the essential role of the intramural R&D investments in balancing the extramural and intramural R&D spending for a greater likelihood of introducing product innovation as the level of family management increases [
61].
The findings of this study have notable implications, both academic and practical. A more nuanced comprehension of the R&D-product innovation relationship in family businesses is essential for investigators and policy planners. Scholars do not only need to consider that family firms are quite heterogeneous, but they may also address distinct R&D strategies to achieve product innovation. The fact of simply considering family firms as homogeneous firms, while only looking at the direct effects on innovation success or neglecting the indirect effects of distinct R&D strategies, may also result in over-simplistic interpretations and biased policy implications. Therefore, heterogeneity in both family firms and R&D strategies should not be disregarded when analyzing the innovative behavior of family firms. Policy makers should abandon the opinion of family business as a type of firm that is deficient in making investments in R&D. Alternatively, they should recognize that different types of family firms might carry out distinct R&D strategies to increase their likelihood of achieving product innovation. Likewise, this study highlights the relevance of R&D investments in introducing product innovations. The significant mediating effect of intramural R&D and the strategy of combining intramural & extramural R&D suggest that practitioners should help top managers to find the right balance between these types of R&D (intramural vs. extramural) to achieve better product innovation outcomes. Furthermore, policy makers, in an attempt to improve the likelihood of introducing product innovations, may be aware of the strengths and weaknesses that are associated to the level of family management, to better align their policies with the particular situation of family businesses. For instance, policy makers might consider new legislation that includes specific policy instruments to incentive the use of intramural R&D and the development of strategies combining intramural and extramural R&D.
In spite of the interesting outcomes that are derived from our study, it has some limitations, which provide opportunities for future research. Given that top managers have immediate power over firm actions and strategies, their decisions are essential in explaining the behavior of family firms regarding R&D investments and product innovation. However, our study does not aim to investigate the impact of the presence of family management, but the influence of the level of family management. Our choice of focusing on the level of family involvement in management is justified, because the degree of family top executives has immediate power over organizational decisions and actions. Nevertheless, differentiating the effects between the mere family presence in management and the level of family management may be helpful in future studies, Furthermore, directly measuring the multiple effects of different types of family involvement (ownership, governance, …) can also be very useful in future studies. It would allow for a more nuanced view of product innovation in family firms by addressing whether different sorts of family involvement conduct distinct R&D strategies and obtain dissimilar results regarding product innovation. We also suggest that future research considers distinct knowledge acquisition modes, while taking into account not only intramural R&D, extramural R&D, and the combination of both types of R&D, but also business acquisitions or the in-licensing of technology. The findings may also be restricted by the use of a particular sample of Spanish manufacturing firms, disregarding public family businesses and family businesses in other countries [
162]. Nevertheless, this database has been already utilized in numerous prior innovation studies among family firms [
44], since it includes firms that tend to particularly rely on innovation because of the great level of obsolescence of their products. Future work may further improve our understanding of the relationship between the family management and firm innovation through the mediating role of R&D, disentangling research activities from development activities, instead of capturing R&D through a single measure [
163].