Family ties and involvement impact on innovation process in high-tech firms

Размещено на http://www.allbest.ru//

Размещено на http://www.allbest.ru//

NATIONAL RESEARCH UNIVERSITY HIGHER SCHOOL OF ECONOMICS

FACULTY OF ECONOMIC SCIENCES

DEPARTMENT OF FINANCE

BACHELOR GRADUATION WORK

Family ties and involvement impact on innovation process in high-tech firms

Polina Khmeleva, BEC 141

Research Adviser: Stepanova A.N.

MOSCOW 2018



Abstract

This paper investigates the impact of family participation in business and family connections between top management and governanceon innovation process in high-tech companies from S&P500 index over the period 1999-2017. High-tech companies appear to be a good sample choice, considering well-known paradox of family firms: they tend to invest less in innovation, while producing them more effectively. In this paper we demonstrate that this paradox is not so obvious for technological companies. We conclude that founder`s involvement and CEO`s ownership of an equity share leads to higher R&D expenditures, income margins and greater number of patents in pharmaceutical and IT sectors. However, consistent with the previous studies of family participation in business, family ownership accompanied by holding offices of CEO and Chairman, has negative impact on amounts spent on innovations.

Аннотация

Данная работа посвящена изучению влияния различных форм семейного участия в работе современных высокотехнологичных компаний, работающих в сфере фармацевтики и информационных технологий, на их исследовательскую деятельность. Высокотехнологичные компании для данного вопроса вызывают особенный интерес в связи с часто изучаемым исследователями парадоксом семейных компаний: они склонны инвестировать в исследования меньше, чем другие фирмы, однако на выходе получают большие результаты. По результатам анализа выявлено противоречие с существующими утверждениями: работа основателя компании в качестве главного исполнительного директора и материальная заинтересованность главного исполнительного директора, выражающаяся в его владении частью акций компании (что часто используется авторами, как определение «семейности» компании), оказывают положительное влияние на инвестиции в разработку, что может быть объяснено спецификой данных индустрий (высокая конкуренция, постоянная необходимость инноваций и осознание необходимости вложений как первичной меры для сохранения бизнеса). Выводы о положительном влиянии семейного участия на инновационный процесс, выражаемый в доходной марже, и о негативных последствиях двойственной роли исполнительного директора, сделанные в данной работе, соответствуют превалирующей на данный момент теории и дополняют её с точки зрения выборки, включающей в себя именно инновационные компании.



1.Introduction

innovation family participation business

Family involvement in business is not an uncommon thing - large amount of today`s technological companies were founded by families and still some of them are governed by their members. In most of the countries from 50% to 80% of GDP is produced by family companies, moreover, it turns outthat they frequently are able to provide higher returns than local stock market indexes (Cruz, 2013). However, stereotypes that characterize family businesses as inefficient and inert to innovation are persistent.Family participation in businessis often associated with risk-averse and biased decision-making, exploitation of minority shareholders and nepotism, that results in unprofessional management and governance(Jiang, Jiang, Kim, & Zhang, 2015).

The role and the effect of family ownership and involvement in businesshave been assessed through different perspectives: from psychological to economical and lots of confronting results have been achieved. Shaping family ties between management and governance by including family members on board is believed to be a way of enhancing interaction between top management team and the board(Chahine & Goergen, 2014a). At the same time, existence of such ties or one large shareholder (family) will lead to lack of monitoring and ability to expropriate minority investors, especially in countries with low legal protection (Amit, Ding, Villalonga, & Zhang, 2015). Regarding innovation process, family involvement results in smaller amounts invested in innovations, but at the same time their research and development process is at least as effective as in companies without family participation(Chen & Hsu, 2009; Classen, Carree, van Gils, & Peters, 2014; Matzler, Veider, Hautz, & Stadler, 2015). At this point current researcher`s opinions about such a paradox differ: on the one hand, it can be partly explained by non-economic goals and willingness to preserve their socioemotional endowment, on the other hand, it may be due to realizing their advantages in interaction process and specific social capital inherent to firms with family partaking(Chen & Hsu, 2009; De Massis, Frattini, & Lichtenthaler, 2013; Isakov & Weisskopf, 2015; Minin, 2015). Starting from this point, we assume that family involvement in business which is dependent on innovations (technological companies) will result in more efficient research and development process: from innovation input - money invested, to innovation output, measured in patents and patent applications, and profit margins.

This paper`s central proposition is motivated by the fact that high-tech companies are todays leaders in capitalization, at the same time around 30% of the companies from S&P 500 index are family controlled, and it seems that in technological firms in United States family ownership is becoming popular once again(Kokoreva, Stepanova, & Karnoukhova, 2016; ValueWalk, 2017). In this case information on how technological companies, that need to survive in highly competitive environment, deal with problems and advantages arising from family involvement is of immense importance. Nevertheless, previous researches in this sphere were limited to overall assessment of possible impact of family ownership or governance and management on distinct aspects of company`s activities in public firms from all industries or limited to privately held companies in one industry. In this work companies from S&P 500 index operating in pharmaceutical and informational technology industries, which are classified as high-technology industries by technology intensity definition of OECD Directorate for science, will be considered.

To test the proposition that family involvement in technological companies influence the R&D process in positive way data on 108 companies (including 6 companies, that were excluded from analysis due to lack of information about R&D activities) regarding family ownership, family ties existence, and characteristics of firm`s management and governance structure was manually collected from filings published on U.S. Securities and Exchange Commission site and analyzed usingOLS or Prais-Winsten regressionapproach. The results show that founder involvement in management and governance, as well as CEO`s ownership of more than 5 % of company`s equityhave positive influence on investments in R&D activities, while Chairman ownership (>5% of voting rights), accompanied with family member role of CEO and Chairman, or family ties within the firm at the same time decrease the investments. Impact on income margins was tested on a smaller period due to the lack of required information. However, we find significant positive effects of founder`s involvement, family ties, and ownership by CEO or Chairman, proving the idea of special social capital, smooth interconnection and leadership benefits for innovation process. Unfortunately, information about patent applications was provided only by 19 pharmaceutical companies, and only 25 had information about patents. However, positive effect of family ties and family power on number of patent applications and founder involvement and ownership on income margins was observed reaffirming the proposition of more effective innovation process within the firms with family involvement (results are not included in the paper but provided in appendixes). Overall, this paper makes several important contributions to the literature on innovation companies and managerial literature concentrating on family participation influence. Using this specific dataset, we were able to reaffirm the results from latter literature connected with innovation output and provide new evidence on specific investment behavior for technological companies.

In the next section, we review the literature on family business and innovation processes and develop hypothesis. In chapter 3 we discussthe data, main variables and methods. Chapter 4 presents the results of empirical research followedbe conclusions.



2.Literature review and hypotheses development

2.1 Family ownership and involvement influence on a firm performance

Agency theory, one of the perspectives most frequently used in family business analysis, says that separation of ownership and control leads to misuse of free cash flows and financing projects that may not be economically profitable in a long-run. Family ownership and control may solve this problem as it can provide powerful monitoring of management, albeit, conflict between majority and minority stakeholders may arise (Delbufalo, Poggesi, & Borra, 2016; Isakov & Weisskopf, 2015).

One of the propositions of this theory is that diversification in regions and products should be used by family companies to mitigate the financial risks. Notwithstanding, this hypothesis was not confirmed on a sample of Italian firms, confirming the other theory that concentrates on socio-emotional wealth. SEW theory consider diversification strategy hazardous to socio-emotional wealth and family control (Delbufalo et al., 2016, p.663).

Main idea of SEW theory is that in family companies non-economic utilities may be more important for decision-maker, resulting, for example, in taking excessive or economically needless risks to preserve socio-emotional endowment or not taking risky but profitable potentialities which can appear to be a threat to it (Gуmez-Mejнa, Haynes, Nъсez-Nickel, Jacobson, & Moyano-Fuentes, 2007). Nepotism which is a common case in family business recruitment policies is an superb example of how socioemotional gains may overweight future losses from lack of needed experience and expertise (Firfiray, Cruz, Neacsu, & Gomez-Mejia, 2018). In the case of nepotism SEW gains are clearly observed, while risks are not so evident. However, when we talk about investments in research and development process which always takes a lot of time until it brings some profit if will, the risks for family capital is obvious, while future gains are unclear.

SEW theory may also explain why family firms are reluctant to use patent system as protection tool for their intellectual property. They consider patents as a threat to privacy of their information and tend to apply other ways of gaining advantages from inventions: first-mover advantages, commercial secret, complementary products, etc, probably gaining less profits but keeping their SEW protected (Bannт, 2016).

Following the SEW theory family CEO should gain less satisfaction from economic performance comparing to non-family as his socioemotional wealth is dependent on different personal goals (extension of business, acquiring more control, ability to pass business to future generations). However, recent study of companies from high- and medium-high technology sectors found that there`s no such a difference, contradicting the ideas of socio-emotional wealth theory (Garcйs-Galdeano, Larraza-Kintana, Cruz, & Contнn-Pilart, 2017).

Other studies mention that family involvement in business leads to longer investment horizons (Stein, 1989), corrects informational asymmetry (Anderson, Mansi, & Reeb, 2003; Chahine & Goergen, 2013) and provides better understanding of strategic goals, also using the specific social capital (Cabrera-Suбrez & Martнn-Santana, 2015)

2.2 Innovation process and family involvement

Innovation process in scientific literature is often divided into following parts: innovation inputs or money invested in R&D activities, innovational activities or how process in the company is organized and on what it mainly focuses (this part is often omitted in empirical literature) and innovation outputs or results of these activities (often measures in patents, patent applications, new products issued or profit margins) (Minin, 2015). The main paradox widely analyzed by authors of family business studies is that family involvement has negative impact on the first part of innovation process, but often provide better results on the last one, overperforming their non-family competitors (Chrisman, Chua, De Massis, Frattini, & Wright, 2015; De Massis, Frattini, Pizzurno, & Cassia, 2015).

For technological companies suffering from high competition producing new products is a question of immense importance, thus, family members will tend to invest in innovations to protect and maintain their socio-emotional wealth. However, they also try not to get into risky and costly projects which may be a cause of financial distress, lessening the control (when attracting investor`s capital), and potential losses in the case of failing. This may be a good explanation to the fact that companies with family involvement tend to invest smaller but still positive amounts in R&D activities (Classen et al., 2014). Unfortunately, studies concentrating on R&D activities in family high-tech companies could hardly been found. Nonetheless, there is an opinion that family involvement may result in much more active investing in innovations, when it is considered a vital for surviving in competition advantage (De Massis et al., 2013).

Other explanation of smaller amounts invested be family-controlled firms lies in their conservatism and realization of their advantages in communications and process-building. However, such conservatism may be a problem when all the power is concentrated in one hands (especially in case of CEO duality) and there comes the lack of expertise and independent opinions. In such situation having an independent director with adequate experience and autonomous vision of industry can help to lessen the negative effects (Chen & Hsu, 2009). Many companies separate the roles of CEO and Chairman to ensure that chief executive officer will be focused on day-to-day business, while allowing the chairman and the board to provide advice and independent oversight of management.https://www.sec.gov/Archives/edgar/data/1396814/000104746913004862/a2214633zdef14a.htm Moreover, separating these roles is believed to encourage free and open dialogue.https://www.sec.gov/Archives/edgar/data/1596783/000119312516709810/d253874ddef14a.htm

To the contrary, CEO duality may unite the management and governance function, resulting in better understanding of global strategic goals (Cabrera-Suбrez & Martнn-Santana, 2015). Here comes the knowledge- and resource-based views that emphasize specific capital relating to human resources characteristics and organizational structure, which contribute to creativity and trust among the employees due to interconnection of management and governance functions (Chrisman et al., 2015; Matzler et al., 2015). Among other things companies` governance mention that having one person serve as both Chairman and Chief Executive Officer provides clear leadership and ensures accountability for success and failures, as well as creates the link between management and Board with regular flow of information, enabling it to perform its monitoring function with the benefits of management`s perspective on business, and involving independent directors provides strong oversight of the management team. Moreover, separating positions may result in divided leadership, interfering with good-decision-making and weakening the ability of developing and implementing strategy, when combining them will provide clear chain of command to execute strategic initiatives. In addition, in case of a founder serving as CEO it may be better to use his knowledge and experience rather than of less informed but independent Chairman.https://www.sec.gov/Archives/edgar/data/1596783/000119312516709810/d253874ddef14a.htm 

Crucial factor usually skipped by researchers is family involvement heterogeneity. The absence of unified definition of family firm and family involvement is widely discussed (Harms, 2014; Kraiczy & Hack, 2013). Family involvement may be provided by ownership, management and governance. Resource-based view explains why active family participation in managing and supervising helps in enhancing the efficiency of R&D efforts. On the dataset consisting of largeGerman publicly traded companies it has been shown that family engagement in management and governance has positive effect on innovation output (results of the R&D activities), and negatively on innovation input (money spent on them) (Matzler et al., 2015). This result reaffirms the prevalent opinion about family firms` paradox, however, they have also shown that separate ownership has no significant effect on any of parameters.



2.3 Family involvement and innovation process, testable hypothesis

Involvement of a founder in the management and governance processes is associated with clear understanding of strategic goals and better market performance (Gill & Kaur, 2015). It implies that having founder as chief executive officer results in leadership and better understanding of a business by top management team. Therefore, first two hypotheses are as follows:

H1a. Having founder on the role of CEO leads to more active investing in research and development activities.

H1b. Having founder on the role of CEO results in more effective investing: more patents, patent applications, higher income margin.

CEO duality is also associated with performance alignment with strategic goals and more effective decision-making process(Cabrera-Suбrez & Martнn-Santana, 2015). So next hypotheses are:

H2a. CEO duality provides higher R&D expenses.

H2b. CEO duality provides better innovation outcomes (more patent applications, higher margins).

In this work we classify firm as family one, if a family owns at least 5% of voting rights, following Anderson et al., 2003 and Cronqvist & Nilsson, 2003, also adding the condition, that member of this family also serves as chairman or chief executive officer (making our definition consistent with classical European Union definition 2009 https://ec.europa.eu/growth/smes/promoting-entrepreneurship/we-work-for/family-business_en). This circumstances will lead to high interest of management in high performance, thus they`ll make appropriate amount of investments in R&D and will control and manage innovation process to get the highest gains from money invested (Chahine & Goergen, 2014b).

H3a Family firmsinvest more in innovations.

H3b Family firms have more patent applications and higher margins.

Family ties between top management and the board lead to smoother interaction between them (Chahine & Goergen, 2014b). Therefore, it would be easier to come to agreement on R&D strategy and organize the processes more effectively. Moreover, family ties within the company`s management and governance may make CEO regard the company as a family one, so that protecting socio-emotional wealth will be more important than rational economic estimates. In the case of high-tech firms, where producing innovations is vital to surviving, we expect higher R&D expenditures and results when family ties exist, following (De Massis et al., 2013).

H4a. Family ties between CEO and board lead to higher R&D investment.

H4b. Family ties between CEO and board result in greater number of patent applications and higher income margins.



3. Data and methodology

3.1 Data sources

First, data on companies with highest capitalization was obtained using information about S&P 500 companies from Capital IQ database. Then companies from pharmaceutical and informational technology industries were selected, as the most numerous representatives of high-technology companies, following OECD definition of high-technology industries. After that all financial data, as well as data on ownership and board composition from 1999 to 2017 was obtained from Capital IQ. The data on family ties and family involvement, including the total sums of ownership of voting power was collected manually from filings published on U.S. Securities and Exchange Commission site (form SCHEDULE 14A proxy statement) which uncover family involvement in ownership, management and control. Moreover, companies` histories were reviewed to resolve descendent issues. It should be mentioned that we follow U.S. Securities and Exchange Commission definition of a family member, that includes children, stepchildren, grandchildren, parents, stepparents, grandparents, spouses, former spouses, siblings, nieces, nephews, mothers-in-low and fathers-in-low, daughters-in low, sons-in-low, brothers-in-low and sisters-in-low. In the end we`ve got the data on 108 companies, however, 6 of them were excluded, due to changes in the organizational form within selected time period (e.g. PayPal was acquired by eBay, but then got separate again) and lack of information about R&D activities. This yields 1757 firm year observations for the period 1999 through 1998.

Surprisingly, only one company from our sample is consistent with family firm definition requiring the firm management role to be passed to next generations. In QUALCOMM Incorporated (NasdaqGS:QCOM) founder and CEO son served as Executive Vice President, then became CEO, with his father playing the role of a chairman and his brother working a senior director. In other cases, family ties mostly included spouses, brothers and sisters and were not a common thing. Furthermore, most of the companies have CEO`s or director`s family members employed by the company but not on top management or director`s roles.

3.2 Measuring family involvement and innovation input and output

As it was already mentioned, U.S. Securities and Exchange Commission definition of a family member was used to reveal family ties and involvement. Dummy variables were created to indicate founder involvement, CEO duality, CEO-board family tie and family involvement characteristics (CEO owns > 5% and serves as a chairman).

To measure innovation inputs we used logarithm of R&D expenditures to previous year revenues ratio. As R&D expenditure cannot be an estimation of innovation process effectiveness, we had to find other measures for that. Past studies of innovation processes have developed diverse ways of its productivity measurement. Patent applications seem to be a good estimation of CEO incentives, but at the same time they are often unequal (some of them are completely useless, while others are extremely valuable (Berrone, Makri, & Gomez-Mejia, 2008). Moreover, as it was already mentioned, family firms prefer not to use patents to preserve their privacy (Bannт, 2016). Data on new product issuance was not available, so we decided to use data on income margin,that can show us the quantitative estimation of advantages of the firm due to creating new innovative products, patent and patent applications. We should also mention that innovations are sometimes divided into science-based and technological-based once, but this distinction won`t be made in this paper.

Controls for firm performance, size, age, leverage, country and industry are also included.



3.3 Methodology

To examine the influence of family involvement on innovation input and output, we use the following variablesand equations:

Variables	Description
Dependent variables
RDRev
incmar	Net income margins, %
Pat	Number of patents company has
patapp	Number of patent applications made this year
Independent variables
Found	Dummy variable, 1 if founder performs CEO role, 0 - otherwise
Duality	Dummy variable, 1 - if CEO serves as Chairman, 0 - otherwise
Ties	Dummy variable, 1 - if CEO has family ties within the board
faminv	Categorical, 0 - neither CEO, nor chairman owns >5% of voting power; 1 - if CEO owns >5%; 2 - if Chairman owns>5%; 3 - if CEO serves as Chairman and owns > 5%.
Control variables
ind	Categorical, industry: 1 - pharmaceutical, 2 - IT
coun	Categorical, country of incorporation, 1 - United States, 2 - Netherlands, 3 - Ireland, 4 - Cayman Islands, 5 - Singapore, 6 - Switzerland
age	Age of the company
cap	Logged market Capitalization, $USDmm
rev	Logged revenue of the previous period, $USDmm
DEq	Total debt to equity, logged %
RD	Logged R&D expenditures in $USDmm

To estimate founder`s involvement impact:

(1)

(2)

(3)

To estimate the influence of CEO duality on innovation input and output, we proposethe following equations:

(4)

(5)

(6)

To evaluate the impact of family ties between CEO and board of directors on innovation input and output, we proposethe following equations:

(7)

(8)

(9)

To evaluate the family involvement impact on innovation input and output, we proposethe following equations:

(10)

(11)

(12)

We also looked at joint influence of founder`s involvement and family ties, and CEO`s and/or Chairman ownership with family ties:

(13)

(14)

Descriptive statistics

From the following table it is clearly seen that there`s a downward trend in number of companies where the roles of Chief executive officer and Chairman of the board are not separated. However, the number of companies with founder involvement rises between 2008 and 2017, what can be explained by emerging of new companies which are able to get high capitalization in a very small period of time. The number of companies with family ties decrease slightly from 4 to 2%, as well as the number of companies matching with our definition of family firm and the CEO`s share of voting rights. However, number of companies with woman on the post of CEO increases sharply.

Table 1. Main independent variables characteristics and trends

Variable	1999	2008	2017
	Mean	std dev	Mean	std dev	mean	std dev
CEO duality	0.591	0.495	0.359	0.482	0.313	0.466
Founder inv.	0.197	0.400	0.157	0.366	0.186	0.391
FAMILY TIES	0.041	0.201	0.034	0.183	0.029	0.170
FAMILY FIRM	0.1	0.301	0.032	0.177	0.029	0.169
CEO owned	5.474	14.949	2.129	6.197	2.671	10.71
CEO woman	0 .028	0 .166	0.033	0.181	0.069	0 .255

Empirical results.

To test our hypothesis linear regressions with panel-corrected standard errors were used.

Table 2 presents descriptive statistics for technological companies (102 companies, 40 operating in pharmaceutical industry and 62 in informational technology) in our sample. The average market capitalization is $52 732,63 million, revenue - $12 981 million, age - 40 years. However, there were firms which age was just 1 year in 2016 (it was due to company reorganization).Table also shows some characteristics of dependent variables: R&D expenditures and income margins, as it can be seen not all companies provided information about the last parameter, so it was decided to use pairwise method of counting covariances to deal with unbalanced panel not losing observations when running the regressions.

Table 2. Main sample characteristics in 2017

	(1)	(2)	(3)	(4)	(5)
VARIABLES	Observations	mean	sd	min	max
Age, years	102	40.08	35.81	1	167
Market Capitalization, $USD mm	102	52,733	115,966	262.8	718,409
R&D expense, $USD mm	101	1,539	2,864	0	13,948
Revenuest-1, $USD mm	102	12,729	28,742	0	233,715
Netincomemargin, %	93	18.2	275.1	-279	6,702
Numberofpatents	22	512.3	1,086	1	4,800

VARIABLE RDRev =
mean	sd	min	max	p1	p10	p25	p50	p75	p90	p95	p99
1.545	6.78	0	86	0	0.023	0.081	0.16	0.24	0.93	6.94	37.9
VARIABLE Income margin, %
mean	sd	min	max	p1	p5	p10	p25	p50	p75	p90	p95	p99
18.2	275.1	-279	6,702	-212	-32.8	-5.7	6.51	13.05	20.55	26.7	31.9	63.7

In Table 3 main results considering investments in research and development are presented. Founder`s involvement in business as well as CEO ownership of a share over 5% show the positive impacton innovation input, proving hypothesis H1a and H3a. Coefficientsequal to 0.413 for founder`s involvement and around 0.299 for CEO owning more than 5% of company`s equity, is rather high, considering that mean value of the dependent variable is 1,545 and median - 0,157. Such a result seems to be of a special interest, as it can be interpreted as double or threefold increase for more than half of the companies of our sample in R&D spending when the CEO is a founder, and double increase when CEO owns a significant part of company`s shares, proving the idea of importance of founder`s leadershipand motivation when taking decision on risky and long-term investment projects suggested by(Gill & Kaur, 2015) and (Chahine & Goergen, 2014b).

However, CEO duality accompanied by shares ownership, family ties between CEO and corporate board and Chairman ownership decreases the investments, contradicting to (Cabrera-Suбrez & Martнn-Santana, 2015) idea of positive influence of uniting management and governance in understanding strategic goals. Most substantial impact is providedby situation when Chairman is owning more than 5% of shares, coefficient -0.988 means that for the most companies of our sample (90% percentile shows RDrev parameter value of 0,933) such a situation leads to incredibly low money spending on research and development activities (less than a 1% of revenues). -0.657 for family ties and -0.679for combining CEO duality with equity ownership also show large negative impact on R&D expenditures even though they are not so dramatic.Such a result contradicts hypothesis H2a and H4a, and can be explained by lack of independent vision, as founder playing the CEO`s role will use his power to build the company`s strategy in accordance with his own vision, limiting opportunities for independent experts` ideas, the same explains the situation when Chairman is not independent, having a share of the company.Moreover, not independent Chairman may block useful initiatives and interrupt company`s work with too active and severe monitoring. So we find explanation given by Chen & Hsu, 2009 rather suitable for this case: in such situation decision-making will be characterized by conservatism and lack of independent oversight. That is why powerful family may not be a good decision for governance structure in technological companies. Last but not the least, revenue of a previous period has a strongly negative effect on the R&D spending, however, it can be explained by the nature of this parameter - as we use a logged quotient of R&D expenditures in revenues of a previous year, while R&D spending is always connected with long-term projects that need a stable amount of financing if company is not taking up new projects.

Next table (Table 4) is dedicated to analysis of income margin. Most coefficients used turned up to be insignificant in prediction of income margins. That is explainable, as this parameter is not a direct reflection of effectiveness of R&D process. However, family ties showed positive effects on income margins, proving hypothesis H4band idea of specific social capital and usefulness of such interconnection between board and management provided by (Chrisman et al., 2015; Matzler et al., 2015). Coefficient 31.25 for family ties means more than a twofold rise in income margins if such interconnection between management and governance exist (based on a mean and median value of parameter income margin - 18 and 13 respectively).Contrarywise, CEO duality accompanied with owning of rather high share of company`s equity has negative effect (-31.16), again contradicting to (Cabrera-Suбrez & Martнn-Santana, 2015) idea of uniting management and governance teams. It contradicts ourH3bhypothesis, meaning that it will have completely opposite effect to that, provided by family ties, and can be explained by negative effect of such governance and management scheme on R&D investments, as even with smooth communication within the firm, special social capital cannot replace essential research financing and independent oversight, following Chen & Hsu, 2009.

Surprisingly, in case of income margin Chairman ownership has positive effect, which is a controversy with our results connected with R&D spending. We find an explanation of such phenomena in ambiguous nature of dependent parameter, as income margin depends on many factors including overall monitoring function and all processes of organization.

Unfortunately, the data on patents and patent applications was not available for most of the companies in our dataset (only 25 companies from pharmaceutical industry have data on patents, and 19 - on patent application). For this reason, results are not presented in a paper, nevertheless, they are available in Appendix 1,2.

Considering robustness check, we can see that our estimations of coefficients are robust in different model specifications. Moreover, we should say that we used linear regressions with panel-corrected standard errors that account for panel-specific autocorrelation (AR1) and heteroskedasticity of disturbances, and we have controlled VIFs for all our variables in our model specifications. For example, market capitalization and revenues were not used at the same specification due to high correlation (0,82). For most specifications of models for R&D expenditures estimation we`ve got R squared value of more than 0,74, that shows a good explanatory ability of proposed models.

Despite rather low R squared value (around 0,25) for the models with income margin, we believe that our results are significant, however more parameters should be added in the specification model considering specific nature of this parameter which we mention above. However, looking at different model specifications used for estimation, we can say that results are robust, and we are able to estimate the character of our parameter`s influence.

Table 3.This table presents the results of linear regressions with panel-corrected standard errors (based on OLS approach) of family involvement parameters on R&D expenditures. Our sample is 97 high-tech firms from S&P index. Main dependent variable RDRev is a logged R&D expenditures in $USD mm scaled by previous year revenues, $USD mm, main independent variables: dummy Duality - 1 - if CEO plays dual role, 0 - otherwise, Founder CEO - 1 if founder is a CEO, 0 - otherwise, TIECEOboard - dummy, 1 if CEO has family member working at the board, faminv - categorical, 1 - if CEO owns > 5%, 2 - if chairman owns >5 %, 3 - if CEO owns >5% and serves as CEO. More information about variables (including controls is provided in chapter



3.3 Methodology

Standard errors in parentheses,*** p<0.01, ** p<0.05, * p<0.1

	(1)	(4)	(7)	(10)	(13)	(14)
VARIABLES	RDRev	RDRev	RDRev	RDRev	RDRev	RDRev
Duality		-0.0431
		(0.0322)
FounderCEO	0.413***	0.431***			0.375***
	(0.126)	(0.126)			(0.127)
TieCEOboard			-0.657***		-0.696***	-0.823***
			(0.202)		(0.248)	(0.219)
CEOowns> 5%				0.299**		0.259**
				(0.130)		(0.131)
Chairman owns> 5%				-0.988***		-1.154***
				(0.167)		(0.172)
CEO-Chair owns >5%				-0.679***		-0.569***
				(0.206)		(0.179)
Total Debt to Equity	-0.00218	-0.00243	-0.00550	-0.128***	-0.00424	-0.125***
	(0.00981)	(0.00981)	(0.00970)	(0.0244)	(0.00993)	(0.0241)
Age	0.00140	0.00141	-0.000352	0.000505	-0.000729	0.000642
	(0.00185)	(0.00189)	(0.000887)	(0.00087)	(0.00187)	(0.000869)
Dummy Netherlands	-0.0528	0.0310	-0.208	-0.938***	-0.218	-0.888***
	(0.237)	(0.257)	(0.258)	(0.154)	(0.223)	(0.162)
Dummy Ireland	-0.157	-0.0682	-0.230	-0.833***	-0.222	-0.850***
	(0.181)	(0.201)	(0.214)	(0.0713)	(0.196)	(0.0787)
Dummy Cayman Islands	2.340***	2.426***	2.221***	2.246***	2.334***	2.108***
	(0.234)	(0.241)	(0.259)	(0.453)	(0.261)	(0.444)
Dummy Singapore				0.735***		0.738***
				(0.237)		(0.229)
Dummy Switzerland	-0.126	-0.127	-0.520***	-0.427***	-0.505***	-0.401***
	(0.0986)	(0.0956)	(0.132)	(0.0544)	(0.114)	(0.0500)
Revenue(t-1)	-0.467***	-0.474***	-0.445***	-0.306***	-0.420***	-0.321***
	(0.0239)	(0.0242)	(0.0289)	(0.0188)	(0.0280)	(0.0177)
Dummy Industry (IT)	0.270	0.379*	0.412**	-0.571***	0.338**	-0.636***
	(0.170)	(0.197)	(0.207)	(0.0906)	(0.147)	(0.101)
Constant	1.180***	1.156***	1.261***	1.288***	1.080***	1.453***
	(0.142)	(0.149)	(0.162)	(0.0770)	(0.147)	(0.0765)
Observations	1,247	1,247	1,247	1,266	1,242	1,247
R-squared	0.747	0.739	0.735	0.485	0.731	0.498
Numberofcompanies	97	97	97	97	97	97

Table 4.This table presents the results of linear regressions with panel-corrected standard errors (based on OLS approach) of family involvement parameters income margin. Our sample is 95 high-tech firms from S&P index. Main dependent variable -Net income margin, measured in percent, main independent variables: Founder CEO - 1 if founder is a CEO, 0 - otherwise, dummy Duality - 1 - if CEO plays dual role, 0 - otherwise, faminv - categorical, 1 - if CEO owns > 5%, 2 - if chairman owns >5 %, 3 - if CEO owns >5% and serves as CEO, 0 - otherwise, TIECEOboard - dummy, 1 if CEO has family member working at the board, 0 - otherwise,. More information about variables (including controls is provided in chapter 3.3 Methodology). Standard errors in parentheses,*** p<0.01, ** p<0.05, * p<0.1

	(2)	(5)	(8)	(11)
VARIABLES	Incomemargin	Incomemargin	Incomemargin	Incomemargin
FounderCEO	9.259
	(7.207)
Duality		-1.490
		(2.011)
CEOowns> 5%	1.871			5.429
	(8.575)			(9.560)
Chairman owns> 5%	27.89***			18.59**
	(6.053)			(8.411)
CEO-Chair owns >5%	-16.71*			-31.16**
	(8.755)			(13.43)
TieCEOboard			31.25**
			(14.19)
Log R&D	-1.171	-1.392*	-1.438*	-3.135
	(0.766)	(0.804)	(0.799)	(2.254)
Total Debt to Equity	0.134	-0.198	-0.0818	-1.189
	(1.221)	(1.083)	(1.083)	(0.901)
Age	0.195***	0.183***	0.178***	0.203***
	(0.0449)	(0.0391)	(0.0427)	(0.0616)
Revenue(t-1)	-0.000304***	-0.000262***	-0.000280***	-0.000424***
	(5.90e-05)	(5.15e-05)	(5.87e-05)	(8.81e-05)
Dummy industry(IT)	22.94***	20.03***	21.08***	26.26***
	(3.686)	(3.325)	(3.516)	(6.826)
Constant	-134.4***	-118.5***	-126.7***	-173.1***
	(15.68)	(14.01)	(15.62)	(25.99)
Observations	773	773	768	774
R-squared	0.253	0.225	0.232	0.275
Numberofcompanies	95	95	95	95



Conclusion

Family involvement has a controversial influence on firm`s decision-making and performance. Despite the vast amount of research done on this topic, no clear resolution of this paradox has been found and it is still unclear whether family involvement is good or bad for the business. Furthermore, no clear explanation of family firm`s successive features that are helping them to outperform their competitors is given.

This work aims to deliver new evidence on family firms, more precisely, technological ones in their struggle to remain competitive on rapidly changing markets by analyzing their R&D processes. We`ve shown that founder`s leadership and CEO`s interest formed be equity ownership, as well as shared vision of family members may be an effective way of realizing family involvement advantages in building competitive strategies and processes within the firm.

Our results show that founder`s involvement as well as CEO`s equity ownership has positive effect on R&D spending (provides twofold increase of R&D spending quotient in previous year revenue), contradicting to existing paradox (family firms tend to invest less, while doing it more effectively).At the same time Chairman`s and CEO`s ownership of company` equity, as well as CEO duality accompanied by share ownership and family ties between CEO and board of directors have significant negative effect on this parameter, consistent with the prevalent idea of family involvement impact on innovation input.As for innovation output family ties and Chairman`s ownership have opposite effect - income margin is larger when family ties or Chairman`s ownership exist. That can be explained by special social capital, smoother interconnection and more intensive monitoring of research processes, as well as by ambiguous nature of the dependent parameter, nevertheless it reaffirms the main thesis of family involvement paradox. However, CEO duality with ownership has negative effect on innovation output, too, that is explained by conservative decision-making, lack of oversight and independent expertise.

This information may be rather useful for business management and governance as it revealsthe weaknesses and opportunities arising from family participation in ownership, management and governance. Considering our results, company`s shareholders should be aware of negative impact of CEO duality and Chairman affiliation and try to avoid arising risks or minimize negative impact by adding powerful independent Lead Director in case of affiliated Chairman and improving corporate rules restricting duality.

We contribute to the literature by concentrating on high-technological firms and family ties in companies that were able to become leaders in capitalization and which are believed to be pacemakers in modern economies. However, our study is not free from limitations - jut a small part of our sample got information about patents and patent applications. Moreover, not much of them had family ties, so more close examination of these relationships should be done.



References

Amit, R., Ding, Y., Villalonga, B., & Zhang, H. (2015). The role of institutional development in the prevalence and performance of entrepreneur and family-controlled firms. Journal of Corporate Finance, 31, 284-305. https://doi.org/10.1016/j.jcorpfin.2015.01.001

Anderson, R. C., Mansi, S. A., & Reeb, D. M. (2003). Founding family ownership and the agency cost of debt. Journal of Financial Economics, 68(2), 263-285. https://doi.org/10.1016/S0304-405X(03)00067-9

Bannт, M. (2016). Journal of Family Business Strategy Propensity to patent by family fi rms, 7, 238-248. https://doi.org/10.1016/j.jfbs.2016.07.002

Berrone, P., Makri, M., & Gomez-Mejia, L. R. (2008). Executive compensation in North American high-technology firms: A contextual approach. International Journal of Human Resource Management, 19(8), 1534-1552. https://doi.org/10.1080/09585190802200314

Cabrera-Suбrez, M. K., & Martнn-Santana, J. D. (2015). Board composition and performance in Spanish non-listed family firms: The influence of type of directors and CEO duality. BRQ Business Research Quarterly, 18(4), 213-229. https://doi.org/10.1016/j.brq.2014.08.001

Chahine, S., & Goergen, M. (2013). The effects of management-board ties on IPO performance. Journal of Corporate Finance, 21(1), 153-179. https://doi.org/10.1016/j.jcorpfin.2013.02.001

Chahine, S., & Goergen, M. (2014a). Top management ties with board members: How they affect pay-performance sensitivity and IPO performance. Journal of Corporate Finance, 27, 99-115. https://doi.org/10.1016/j.jcorpfin.2014.04.007

Chahine, S., & Goergen, M. (2014b). Top management ties with board members: How they affect pay-performance sensitivity and IPO performance. Journal of Corporate Finance, 27, 99-115. https://doi.org/10.1016/j.jcorpfin.2014.04.007

Chen, H.-L., & Hsu, W.-T. (2009). Family Ownership, Board Independence, and R&amp;D Investment. Family Business Review, 22(4), 347-362. https://doi.org/10.1177/0894486509341062

Chrisman, J. J., Chua, J. H., De Massis, A., Frattini, F., & Wright, M. (2015). The ability and willingness paradox in family firm innovation. Journal of Product Innovation Management, 32(3), 310-318. https://doi.org/10.1111/jpim.12207

Classen, N., Carree, M., van Gils, A., & Peters, B. (2014). Innovation in family and non-family SMEs: An exploratory analysis. Small Business Economics, 42(3), 595-609. https://doi.org/10.1007/s11187-013-9490-z

Cronqvist, H., & Nilsson, M. (2003). Agency Costs of Controlling Minority Shareholders. The Journal of Financial and Quantitative Analysis, 38(4), 695. https://doi.org/10.2307/4126740

De Massis, A., Frattini, F., & Lichtenthaler, U. (2013). Research on Technological Innovation in Family Firms. Family Business Review, 26(1), 10-31. https://doi.org/10.1177/0894486512466258

De Massis, A., Frattini, F., Pizzurno, E., & Cassia, L. (2015). Product innovation in family versus nonfamily firms: An exploratory analysis. Journal of Small Business Management, 53(1), 1-36. https://doi.org/10.1111/jsbm.12068

Delbufalo, E., Poggesi, S., & Borra, S. (2016). Diversification, family involvement and firm performance. Journal of Management Development, 35(5), 663-680. https://doi.org/10.1108/JMD-09-2015-0129

Firfiray, S., Cruz, C., Neacsu, I., & Gomez-Mejia, L. R. (2018). Is nepotism so bad for family firms? A socioemotional wealth approach. Human Resource Management Review, 28(1), 83-97. https://doi.org/10.1016/j.hrmr.2017.05.008

Garcйs-Galdeano, L., Larraza-Kintana, M., Cruz, C., & Contнn-Pilart, I. (2017). Just about money? CEO satisfaction and firm performance in small family firms. Small Business Economics, 49(4), 825-839. https://doi.org/10.1007/s11187-017-9863-9

Gill, S., & Kaur, P. (2015). Family Involvement in Business and Financial Performance: A Panel Data Analysis. Vikalpa, 40(4), 395-420. https://doi.org/10.1177/0256090915605756

Gуmez-Mejнa, L. R., Haynes, K. T., Nъсez-Nickel, M., Jacobson, K. J. L., & Moyano-Fuentes, J. (2007). Socioemotional Wealth and Business Risks in Family-controlled Firms: Evidence from Spanish Olive Oil Mills. Administrative Science Quarterly, 52(1), 106-137. https://doi.org/10.2189/asqu.52.1.106

Harms, H. (2014). Review of Family Business Definitions: Cluster Approach and Implications of Heterogeneous Application for Family Business Research. International Journal of Financial Studies, 2(3), 280-314. https://doi.org/10.3390/ijfs2030280

Isakov, D., & Weisskopf, J. P. (2015). Pay-out policies in founding family firms. Journal of Corporate Finance, 33, 330-344. https://doi.org/10.1016/j.jcorpfin.2015.01.003

Jiang, F., Jiang, Z., Kim, K. A., & Zhang, M. (2015). Family-firm risk-taking: Does religion matter? Journal of Corporate Finance, 33, 260-278. https://doi.org/10.1016/j.jcorpfin.2015.01.007

Kokoreva, M. S., Stepanova, A. N., & Karnoukhova, E. V. (2016). What We Do not Know about the Ownership Structure of the Largest U. S. Companies? Economic Policy (in Russian), 11(6), 36-59. https://doi.org/10.18288/1994-5124-2016-6-02

Kraiczy, N., & Hack, A. (2013). Innovations in small and medium-sized family firms: An analysis of innovation related top management team behaviors and family firm-specific characteristics. Innovations in Small and Medium-Sized Family Firms: An Analysis of Innovation Related Top Management Team Behaviors and Family Firm-Specific Characteristics, (1999), 1-173. https://doi.org/10.1007/978-3-658-00063-9

Matzler, K., Veider, V., Hautz, J., & Stadler, C. (2015). The impact of family ownership, management, and governance on innovation. Journal of Product Innovation Management, 32(3), 319-333. https://doi.org/10.1111/jpim.12202

Minin, A. Di. (2015). Family-Driven Innovation:, 58(1), 5-19.

Poza, E. (2013). Family Business, (June), 480. Retrieved from https://books.google.com/books?hl=en&lr=&id=7dQWAAAAQBAJ&pgis=1

Stein, J. C. (1989). Efficient Capital Markets, Inefficient Firms: A Model of Myopic Corporate Behavior. The Quarterly Journal of Economics, 104(4), 655. https://doi.org/10.2307/2937861



Appendix 1

VARIABLE patents

mean	sd	min	max	p1	p5	p10	p25	p50	p75	p90	p95	p99
487.2	1,084	1	4,800	1	2	14	24	73.50	285	1,209	4,800	4,800

As it can be seen, founder`s involvement, CEO`s and/or Chairman ownership has robust significant negative effect on number of patents company has. For more than 75% of our sample that have information about patents, it means completely refusing of such intellectual property protection tool. While it contradicts our hypothesis H1a, H3a, it can be explained by the results of the already mentioned paper of (Bannт, 2016), who proved unwillingness of family firms to disclose information through patent system.

	(3)	(6)	(12)	(9)
VARIABLES	Patents	Patents	Patents	Patents
FounderCEO	-453.2***
	(131.6)
1.CEOCHAIROWN5	143.2			-137.9**
	(95.33)			(59.65)
2.CEOCHAIROWN5	-744.7***			-778.3***
	(127.5)			(108.7)
3.CEOCHAIROWN5	-507.4**			-620.3***
	(201.3)			(136.7)
lrd	283.7***	39.73	281.1***	215.1***
	(92.11)	(36.94)	(67.87)	(73.73)
lcap	295.8***	182.7***	354.9***	340.1***
	(94.88)	(55.37)	(82.68)	(91.02)
ldeq	21.99	31.91***	32.23**	38.90**
	(24.29)	(10.82)	(14.29)	(15.13)
age	4.962***	-4.392**	2.293	6.951***
	(1.777)	(1.854)		(1.799)
Duality		54.63
		(47.54)
TieCEOboard			-56.31
			(56.61)
Constant	-2,614***	-1,208***	-3,226***	-2,905***
	(456.6)	(428.4)	(410.0)	(430.1)
Observations	73	73	73	73
R-squared	0.534	0.370	0.471	0.503
Numberofcompanies	25	25	25	25

Standard errors in parentheses*** p<0.01, ** p<0.05, * p<0.1



Appendix 2

VARIABLE patent applications

mean	sd	min	max	p1	p5	p10	p25	p50	p75	p90	p95	p99
298.3	538.2	1	2,100	1	3	5	19	44	245	906	2,000	2,100

Consistent with previous results considering patents, founder`s involvement and CEO`s ownership, as well as duality has negative impact on patent applications again approving results of (Bannт, 2016). Nevertheless, positive effects of duality accompanied with ownership and family ties can be observed.Despite the fact, that it reaffirms our hypothesis H2a and H4ait contradicts our overall findings, mostly with ones connected to patents, we find the explanation of such phenomena in overconfidence on CEO, who own`s a share and plays a dual role or have ties in the board, so that the patent applications are not approved. Nevertheless, additional research should be done.

	(3)	(6)	(12)	(9)
VARIABLES	patentapp	patentapp	patentapp	patentapp
FounderCEO	-89.76*
	(46.38)
1.CEOCHAIROWN5	-7.273			-56.59*
	(31.29)			(31.38)
2.CEOCHAIROWN5	-18.77
	(187.8)
3.CEOCHAIROWN5	265.8			220.8*
	(201.6)			(129.3)
lrd	219.4***	213.4***	201.9***	41.18
	(61.67)	(53.26)	(51.33)	(36.06)
lcap	111.8***	118.5***	130.1***	40.57***
	(26.98)	(22.06)	(26.29)	(14.08)
ldeq	-9.495	-8.891	-11.52	-11.36
	(11.97)	(10.30)	(11.29)	(9.016)
age	0.973	1.233	1.732	3.417
	(2.205)	(2.011)	(1.931)	(3.963)
Duality		-86.59***
		(22.34)
TieCEOboard			114.9**
			(45.52)
industry				-292.8**
				(143.4)
Constant	-1,223***	-1,269***	-1,346***
	(216.1)	(218.8)	(231.7)
Observations	103	103	103	104
R-squared	0.620	0.614	0.612	0.200
Numberofcompanies	19	19	19	19

Standard errors in parentheses*** p<0.01, ** p<0.05, * p<0.1

Размещено на Allbest.ru