High-performance work practices,socioemotional wealth preservation,and family firm labor productivity *

Abstract

Despite growing research on the effect of high-performance work practices (HPWPs) on family firm performance, the implications of socioemotional wealth (SEW) preservation remain ambiguous. This stems from SEW preservation being used primarily as an explanatory construct and assessed indirectly rather than directly in empirical studies. To address this research gap, we draw upon organizational control and signaling theories to determine the “true” interaction between HPWPs and SEW preservation for labor productivity. Specifically, competing hypotheses are presented to determine if this interaction supports complementarity or substitutability. Using a sample of 124 Spanish family firms and a direct measurement of SEW preservation, our results provide support for substitutability, suggesting that family firms can realize higher labor productivity when HPWPs are fully implemented and commitment to SEW preservation is low, and vice versa. These findings have important implications for family firms, given HPWPs’ inverse relationship with SEW preservation regarding labor productivity.

JEL CLASSIFICATION

J24, L20, L21, L26, M12_M12, M54_M54, O15

Introduction

According to the organizational control literature, family firms can embody multiple forms of control, with those controls being primarily bureaucratic or clan, designed to enable firm performance (Moores & Mula, 2000; Ouchi & Price, 1978). However, the precise nature of how these different control mechanisms interact for family firms remains unclear. Specifically, management scholars have suggested that bureaucratic and clan control mechanisms act as substitutes (Songini & Gnan, 2015), whereas others indicate that these mechanisms are complementary (Poppo & Zenger, 2002). This study explores these possibilities within the context of family firms by considering the interplay of these control mechanisms for enabling family firm productivity. Formal human resource management (HRM) practices, such as high-performance work practices (HPWPs), can serve as a bureaucratic control mechanism (Snell, 1992), whereas family firms are prone to exercising clan or normative control via their prioritization of socioemotional wealth (SEW) preservation (Singal & Gerde, 2015). The preservation of SEW, or the non-financial value that controlling families extract from the firm, encompasses family control and influence, binding social ties, emotional attachment, family members’ identification with the firm, and renewal of family bonds through dynastic succession (Berrone et al., 2012). Such aspirations by controlling families tend to shape not only the strategies, activities, and outcomes of family firms but their norms and values as well (Gómez-Mejía et al., 2011; Singal & Gerde, 2015).

While research has focused on SEW preservation as a reference point for family firms in deciding which HRM practices to use (e.g., Cruz et al., 2011; Mullins, 2018), empirical studies, to date, paint an ambiguous picture as to the true nature of the interplay between HPWPs and SEW preservation for family firm outcomes. On one hand, family business scholars have found that family firms with higher levels of HPWPs experience greater firm performance (e.g., Tsao et al., 2009). However, others note that family firms can limit the effectiveness of formalized HRM practices including HPWPs (e.g., Sánchez-Marín et al., 2019). Much of this ambiguity is driven by SEW preservation being treated as an explanatory construct and measured indirectly rather than being directly assessed (Berrone et al., 2012; Miller & Le Breton-Miller, 2014). Such an approach can lead to “hazardous inference[s]” according to Miller and Le Breton-Miller (2014, p. 716). Hence, our understanding can be enhanced by taking a direct approach to evaluating SEW preservation in conjunction with HPWPs for family firm productivity. This becomes theoretically and practically crucial to understanding the complementarity or substitutability of HPWPs with SEW preservation dynamics. Essentially, researchers will become better aware of the benefits or drawbacks of family-centric mechanisms like SEW preservation when used in conjunction with HPWPs, so that family firms may direct their investments accordingly. For instance, when commitment to SEW preservation is high, would it be worthwhile to invest in HPWPs? Without knowing the complementarity or substitutability, family business researchers and practitioners may prescribe both, which could lead to favorable or unfavorable family firm outcomes.

In light of this, we draw upon both organizational control and signaling theories in making our arguments. Based on the organizational control literature, family firms often rely upon multiple mechanisms (i.e., bureaucratic and normative) to control nonfamily employee activities and behaviors, as a means of ensuring firm success (Gill, 2019; Ouchi & Price, 1978; Songini & Gnan, 2015). When used in combination, these mechanisms can interact in different ways to influence firm outcomes (Cardinal et al., 2017). While HPWPs comprise a formal bureaucratic system (Snell, 1992), SEW preservation represents an informal normative control because its socio-ideological aspects can directly shape employee emotions and thoughts and indirectly their behaviors (Gill, 2019), through communications and interactions among family business leaders, family employees, and nonfamily employees. Given this, we contend that HPWPs and SEW preservation concerns represent such distinct forms of control that the effectiveness of HPWPs may depend upon SEW preservation, and vice versa. To guide our exploration of this interaction, we focus on complementarity and substitutability to determine the true interplay between these mechanisms. Furthermore, given that the commitment to SEW preservation is less visible to nonfamily employees relative to HPWPs (Bowen & Ostroff, 2004; Hauck et al., 2016), we consider whether the signals emanating from each of these mechanisms are either consistent or conflicting to nonfamily employees (Connelly et al., 2011), thereby impacting family firm labor productivity. Hence, this study examines the following research question: Are HPWPs and SEW preservation concerns complements or substitutes for influencing family firms’ labor productivity?

Based on a review of those few empirical studies directly measuring SEW, it becomes apparent that family firms are heterogeneous based on their level of commitment to SEW preservation (e.g., Goel et al., 2013; Hernández-Linares et al., 2020; Schepers et al., 2014; Vandekerkhof et al., 2015). This suggests that any hypothesis building involving SEW preservation should account for this variation among family firms. Moreover, when it comes to evaluating SEW preservation, Dyer (2018) posits “that family and nonfamily firms are largely noncomparable entities” (p. 244). Accordingly, to ascertain the true effects of SEW preservation in connection with HPWPs, we further extend relevant past studies by concentrating exclusively on family firms.

Our study endeavors to make several important contributions to the existing family business and HRM literatures. First, with past empirical studies relying on indirect measures of SEW preservation (e.g., Pittino et al., 2016; Sánchez-Marín et al., 2019), our study is among the first to directly measure SEW preservation to determine its “real” interaction with formal HRM practices (i.e., HPWPs) in affecting family firm productivity. This multidimensional measure of SEW (Berrone et al., 2012) captures the true meaning of the SEW construct which also enhances theoretical development. Second, since family firms exhibit idiosyncrasies with the co-existence of financial and family-centered non-financial goals (Aparicio et al., 2017; Basco, 2017; Chrisman et al., 2012), such as the concern for SEW preservation (Gómez-Mejía et al., 2007) and associated signals sent to nonfamily employees, this investigation improves our understanding of the organizational conditions that foster or limit nonfamily employee collective contributions in the family business context as well as the impact of multiple signals in family firms. Third, with contextual factors being critical in determining the HPWPs-firm performance relationship (see Lengnick-Hall et al., 2009), this study introduces SEW preservation as a key contextual factor in shaping the efficacy of HPWPs within family firms. Without consideration for the non-financial dynamics such as SEW preservation and the interplay between such phenomenon with other practices, our understanding of HRM practices and their impact on firm outcomes in family firms would be limited. Hence, this study advances our knowledge about the strategic HRM activities in the form of HPWPs and their degree of compatibility with non-financial forces such as SEW preservation in family firms, in turn affecting firm outcomes. Fourth, we seek to contribute to the organizational control literature (e.g., Cardinal et al., 2017; Gill, 2019) by highlighting the implications of control mechanisms as signaling devices in determining nonfamily employee composition and motivation, which affects family firms’ labor productivity. Finally, this study attempts to build on the recent works examining family firm heterogeneity in connection with family firms’ labor productivity (e.g., Neckebrouck et al., 2018) by examining the complementarity or substitutability between HPWPs and SEW preservation in influencing labor productivity through different signals that nonfamily members may perceive. As a result, our study focuses exclusively on family firms, which represent the backbone of most economies (Neckebrouck et al., 2018; Prencipe et al., 2014). When family businesses are aware of the interplay between non-financial factors and strategic activities and the impact on family firm outcomes, they can make effective strategic decisions and attain success in the long run. Moreover, when both family and nonfamily employees are informed about the priorities of a family firm (and if they are consistent with each other), they are expected to direct their efforts toward achieving strategic goals, in turn reaching greater degree of productivity. In the remainder of this article, the theoretical framework, the hypotheses, methods, and discussion are presented.

Theoretical framework and hypothesis

Organizational control mechanisms: HPWPs and SEW preservation

Organizational control has been a critical function in the management of the firm (Cardinal et al., 2017). When it comes to exercising control, managers often rely upon different mechanisms “to direct attention, motivate, and encourage individuals to act in ways that support the organization’s objectives” (Cardinal et al., 2017, p. 559). While control is often conceptualized as being coercive, Adler and Borys (1996) posit that control can also be enabling such that it “provides needed guidance and clarifies responsibilities, thereby easing role stress and helping individuals be and feel more effective” (p. 61). In their review of the organizational control literature, Cardinal and colleagues (2017) note that formality is a key dimension of organizational control. Specifically, formal bureaucratic control has been described as “explicit and codified institutional mechanisms (…) such as written procedures and rules and directives” (Cardinal et al., 2017, p. 567). In contrast, informal normative control mechanisms are “unwritten, unofficial values, norms, shared values, and beliefs that guide employee actions and behaviors” (Cardinal et al., 2004, p. 414). In applying this enabling control perspective to family firms, we focus on two distinct control mechanisms at opposite ends of the control formality spectrum—formal HPWPs and informal SEW preservation. HPWPs and family firm goals such as SEW preservation can inform and guide nonfamily employees toward key priorities and associated responsibilities. Hence, they serve as control mechanisms to direct effort, prompt, and guide nonfamily employees to act in certain ways that can lead to firm success through productivity.

As a form of formal bureaucratic control (Snell, 1992), HPWPs have been found to be positively associated with a range of firm-level outcomes, including labor productivity and financial performance (e.g., Combs et al., 2006; Saridakis et al., 2017). Moreover, this combination of HRM practices has proven effective for family firm performance as well (e.g., Madison et al., 2018; Pittino et al., 2016; Tsao et al., 2009). In general, HPWPs generate long-term value creation through effective workforce management (Lepak et al., 2007; Wright et al., 1994). While there is a lack of consensus as to the precise configuration of HRM practices comprising HPWPs, the HRM practices commonly linked to this classification includes formal selection techniques, performance-based compensation, extensive training and development programs, and motivational job design approaches such as job enrichment and worker empowerment, among others (Huselid, 1995; Meuer, 2017). Collectively, HPWPs enable the realization of four key principles: egalitarianism and engagement, shared information, knowledge development, and a clear performance-reward linkage (Lawler et al., 1995; Nadler & Gerstein, 1992; Snell & Bohlander, 2013). According to Snell and Bohlander (2013), these interconnected principles can be described in the following manner. Regarding the principle of egalitarianism and engagement, HPWPs are designed to minimize power differences in the firm, thereby enabling cooperation and teamwork. Furthermore, through delegation, employees will be empowered to take on an active role in the firm’s decision-making processes. This necessitates the second principle of shared information. Empowered employees must possess critical firm information for the purpose of providing relevant suggestions and solutions that will enable firm success. However, empowered employees must be fully capable of making quality decisions, which emphasizes the third principle that being the need for firm employees to have continuous knowledge development. For the final principle, rewards must be tied to employee performance, such that employees are motivated to pursue those actions deemed beneficial for the firm.

In contrast to HPWPs, many family firms continue over generations in order to achieve primarily family-centered goals such as ensuring employment and job security for family members, even though these may not enhance financial performance (Gómez-Mejía et al., 2007; Morgan & Gómez-Mejía, 2014). Controlling families exercise considerable discretion or influence over family firms, as they are typically unencumbered by organizational restraints that tend to limit managerial authority (e.g., Carney, 2005). In view of this, the achievement of these non-financial goals, often referred to as the preservation of SEW, involves family control and influence, identification with the firm, binding social ties, emotional attachment, and renewal of family bonds to the firm through intra-family succession (Berrone et al., 2012). However, the failure to preserve SEW can result in diminished harmony, lowered status, and an inability to meet the family’s expectations (Gómez-Mejía et al., 2007). As such, the potential losses or gains in terms of SEW preservation constitute a reference point for the strategic and operational activities of family firms (Berrone et al., 2012; Gómez-Mejía et al., 2007).

The prioritization of SEW preservation by controlling families becomes integral to family firms’ exercising social control over nonfamily employees. Specifically, the values and priorities embodied in SEW preservation cultivate a unique culture for family firms that influences firm operations, which includes their employment relationship with nonfamily employees (O’Reilly & Chatman, 1996; Singal & Gerde, 2015). Given this, a high level of commitment to SEW preservation is likely to influence how family firms go about implementing their HRM activities. For instance, the staffing criteria in family firms might place greater value on kinship ties rather than on human capital or competence (Jaskiewicz et al., 2013; Jennings et al., 2018). Then again, because of their emphasis on SEW preservation, family firms may possess the necessary culture of understanding and empathy to accommodate both family and nonfamily employee abilities, motivations, and opportunities (Sun et al., 2007).

HPWPs and SEW preservation: complementarity versus substitutability hypotheses

While family firms may seek to use HPWPs as a means of enhancing its professionalization, and subsequently labor productivity (Stewart & Hitt, 2012), family firms are heterogeneous regarding their affinity toward SEW preservation (Berrone et al., 2012). In light of this, we anticipate that labor productivity in family firms will be determined based on how nonfamily employees respond to the messaging stemming from these distinct control mechanisms. According to Gill (2019), employees can experience either fulfillment or suffering in response to the simultaneous use of multiple modes of control. Specifically, fulfillment represents those positive psychological rewards that one derives from working (Nemiroff & Ford, 1976), whereas suffering captures “the distress a person experiences when they perceive a threat to any aspect of their continued existence” (Gill, 2019, p. 380). Moreover, what an employee experiences hinges on the compatibility of their personal characteristics (e.g., values) with the particulars of a control mode (Gill, 2019). Because the congruence among different control modes can shape employee compatibility (Gill, 2019), we examine the interplay between HPWPs and SEW preservation for influencing labor productivity by considering whether these elements of control either complement or substitute one another. To that end, we present competing hypotheses to unravel the nature of this interaction. Indeed, this may not be clear to both researchers and practitioners without a thorough conceptualization and examination of both potential effects (i.e., substitutability and complementarity) of this interaction on labor productivity.

While firms may have certain control mechanisms in place, signals to employees and their interpretation of such signals matter in turning these to organizational outcomes (Connelly et al., 2011). Therefore, it is crucial to draw upon both theoretical perspectives (i.e., organizational control and signaling) in examining the interaction effects of HPWPs and SEW and the influence on labor productivity. With the conditions of the past being instrumental in shaping whether employees experience fulfillment or suffering (Gill, 2019), we consider the role that employer prehire signaling may play in regard to HPWPs and SEW preservation for influencing the nonfamily employee experience. Based on signaling theory, which describes how information asymmetries between signalers and receivers can be mitigated (Connelly et al., 2011; Spence, 1973, 2002), HRM scholars note that current and potential employees often rely upon signals from employers in lieu of having complete employer information when it comes to making employment-related decisions (e.g., Rynes, 1991; Suazo et al., 2009). For instance, job applicants may look at a firm’s diversity management policy as a positive signal that its work environment is supportive of diversity, thereby making it more attractive as an employer (Williams & Bauer, 1994). Moreover, Suazo and colleagues (2009) posit that current employees may interpret the firm’s HRM practices as signals regarding the nature of their employment relationship with the firm. Hence, employer signals can profoundly affect the composition and motivation of its workforce and ultimately firm performance.

Extending this to the family firm context, research suggests that nonfamily employees pay close attention to any signals emanating from family firms to ascertain who they are as employers (Tabor et al., 2018). For instance, family firms are considered as having favorable work cultures in which nonfamily employees are able to grow along with family employees where the principles of benevolence and compassion abound (Cruz et al., 2010; Miller & Le Breton-Miller, 2005; Milton, 2008; Tabor et al., 2018). Their cultural uniqueness likely hinges upon the extent to which SEW preservation is prioritized. The family’s influence and visibility within the firm can send a signal to employees regarding the work culture. Moreover, through their interactions with current employees such as selection panelists and interviewers (e.g., Carless & Wintle, 2007; Gilmore et al., 1999), prospective employees are likely to form an impression of the family firm’s values and norms, which is shaped by their commitment to SEW preservation. While family firms prefer to hire familiar nonfamily members to ensure cooperation (Luo & Chung, 2005), there are certain types of nonfamily members that are drawn to this family-like work environment, particularly those whose values closely align with that of the family (Fang et al., 2013; Karra et al., 2006). According to Hauswald et al. (2016), those nonfamily members that are attracted to these environments are described as valuing conformity, security, and tradition and have a deep concern for the welfare of others.

With nonfamily members differing in their attraction toward these environments, it is plausible that HPWPs may serve to complement family firms having a strong emphasis on SEW preservation. As mentioned before, nonfamily members drawn to these family-like cultures appear to have dispositions that facilitate and reinforce coordination and cooperation among employees, leading to enhanced labor productivity. However, nonfamily members who may avoid these environments are likely to be more concerned with career advancement and progression (Chrisman et al., 2014; Vandekerkhof et al., 2015). Given this, family firms through the use of HPWPs can signal to these nonfamily members that career development and advancement opportunities are available. Moreover, in their study focusing on the benefits of a single HPWP—incentive compensation—for labor productivity, Chrisman et al. (2017) surmised that the use of incentives can enhance the quality of the workforce as it provides a signal to prospective employees that performance is rewarded, thereby increasing the family firm’s attractiveness as an employer. With HRM practices being more effective collectively rather than individually (Wright & McMahan, 1992), this signal is likely to be further amplified in attracting higher quality potential employees to family firms when a broader collection of HPWPs are used. Hence, family firms that prioritize both SEW preservation and HPWPs are better able to strengthen the capabilities of their workforces while simultaneously cultivating a work environment that exudes cooperation, thereby increasing their labor productivity.

Taken together, family firms with an emphasis on SEW preservation coupled with the use of HPWPs are likely to be perceived by nonfamily members as being able to provide fulfillment through employment. As such, it is likely that these two modes of control will be congruent, as both place a strong emphasis on employee welfare. These favorable pre-hire impressions may then translate into nonfamily employees experiencing fulfillment when employed by the firm. As such, these employees will expend greater effort leading to increased labor productivity. Therefore, we formally propose:

• Hypothesis 1. HPWPs and SEW preservation act as complements in influencing family firm labor productivity, such that a stronger presence of HPWPs coupled with a higher degree of SEW preservation is positively associated with family firms’ labor productivity.

Conversely, considering the potentially diverse SEW outcomes (i.e., both positive and negative) (Hernández-Linares et al., 2020; Kellermanns et al., 2012; Miller & Le Breton-Miller, 2014), HPWPs may have the opposite effect or may undermine the labor productivity of family firms placing a strong emphasis on SEW preservation. Family firms by implementing HPWPs may be desirous of projecting an image of professionalism when it comes to managing their workforces (Fang et al., 2012; Stewart & Hitt, 2012), and yet, continue to strongly maintain those family-centric values associated with prioritizing SEW preservation (Madison et al., 2018).

The use of HPWPs may signal to both family and nonfamily members that this firm is similar to nonfamily firms (Stewart & Hitt, 2012). Essentially, HPWPs can be instrumental in attracting nonfamily members to family firms because it communicates an employment relationship that is built on social exchange (Suazo et al., 2009). However, once employed by and then socialized into family firms (Singal & Gerde, 2015), nonfamily members, in particular, may experience frustration as the execution of these practices may send a different signal that conflicts with their initial impressions. In this case, a strong commitment to SEW preservation can generate a double-bind (i.e., ambiguous and paradoxical) intensive message (Litz, 2012) due to the mixed signals that nonfamily employees receive from both the emphasis on SEW preservation and HPWPs, while family members may be already aware of them. Specifically, potential discord may emerge between HPWPs that signal things like egalitarianism and are designed to enhance the capabilities, motivations, and opportunities for family and nonfamily members alike, whereas the more family-centric SEW preservation espouses the asymmetric treatment of family over nonfamily employees (Daspit et al., 2018; Madison et al., 2018). When it comes to these double-bind communications, employees face contradiction between the signals and experience constraint due to not being able to challenge the signal’s contradictory nature, resulting in confusion and frustration in family firms (Litz, 2012). Given this, when there is greater use of HPWPs coupled with higher levels of SEW preservation, the empowerment of nonfamily employees would contradict with the family’s desire to be altruistic to family members, which favors the exclusive treatment and promotion of family employees despite the value placed in binding social ties with all stakeholders (Barnett & Kellermanns, 2006; Daspit et al., 2018). Similarly, family control, emotional attachment, and a firm identity predominantly tied to the family identity, intra-family succession intentions can also limit nonfamily employees’ career development and prospects and result in negative perceptions (Memili & Barnett, 2008) despite the presence of HPWPs. Thereby, unlike family members, nonfamily employees may experience confusion when there is a strong presence of both HPWPs and a high degree of SEW preservation. This would indicate that these two control mechanisms are clashing such that these nonfamily employees would “seek to remove or separate themselves” from this situation (Gill, 2019, p. 387). As a result, these employees may feel that their career progression is impeded and are likely to engage in withdrawal behaviors including deciding to leave the family firm. This can create dysfunction due to increases in voluntary turnover, which will serve to hinder overall labor productivity. Hence, we hypothesize:

• Hypothesis 2. HPWPs and SEW preservation act as substitutes in influencing family firms’ labor productivity such that a stronger presence of HPWPs coupled with lower preservation of SEW (or higher SEW preservation along with weaker presence of HPWPs) is positively associated with family firms’ labor productivity.

Methods

Data collection

To test the proposed hypotheses, we used survey data collected in Spain during 2017. Our questionnaire was originally created in English and then translated into Spanish to facilitate respondents’ understanding because the vast majority of the CEOs were native Spanish speakers. Finally, it was translated back to English in order to check for consistency (e.g., Hernández-Linares et al., 2020). To identify family firms, we used a database provided by the Family Firm Institute in Spain containing firms classified as family firms based on ownership, management, and other governance criteria obtained from both the SABI/Amadeus database (e.g., Díaz-Fernández et al., 2017; Hernández-Linares et al., 2018a) and the network of Family Business Chairs in Spain.

The inclusion requirements for family firms were limited companies or limited liability companies, active in 2015, with sales more than 2 million Euros a year or 10 or more employees during any of the 3 years from 2011 to 2013. We then excluded companies with less than 50 employees resulting in a database of 8,542 firms. Due to budgetary restrictions for carrying out a mail survey, we randomly selected 1,355 firms for a telephone survey. Among the 1,355 initially selected firms, 110 had incorrect telephone numbers and 437 did not consider themselves as family firms, leading to a sample of 808 family firms with correct telephone numbers. We administered the questionnaire to CEOs because they are considered reliable sources of information on firm governance dynamics (Glick et al., 1990) and are often used as key informants in the family businesses literature (e.g., Kellermanns et al., 2008). In addition, the use of similar informants across organizations (e.g., CEOs) increases the validity of variable measurement (Glick, 1985) because the level of influence of all informants in their organizations may be considered uniform (Aragón et al., 2007). We received 136 completed questionnaires from CEOs for a final response rate of 16.83%, which is higher than the “10.12 percent rate typical for studies which target executives in upper echelons” (Geletkanycz, 1997, p. 622) and higher than similar studies on CEOs in Spanish family firms (e.g., Blanco-Mazagatos et al., 2016; Cruz et al., 2010). Next, the survey data was merged with archival performance data for years 2016 and 2017. The resulting sample, based on casewise deletion, included 124 family firms.

Measures

Dependent variables

We use two dependent variables measured in the 2017: labor productivity and profit per employee. Directly tied to employee’s attitudes and behavior (Firfiray et al., 2018), labor productivity is considered “the crucial indicator of workforce performance” (Datta et al., 2005, p. 139) and essential to understanding people management issues in family firms (Firfiray et al., 2018). In this study, labor productivity is measured as the log of sales in million Euros in 2017 divided by the number of equivalent full-time employees (Datta et al., 2005; Huselid, 1995). Profit per employee, a key financial metric in HRM research (Mattsson, 2019), is measured as the profit (in millions) in 2017 divided by the number of equivalent full-time employees.

HPWPs

HPWPs are measured using the scale from Chadwick and colleagues (2015, pp. 367–368), which is based on measures used in previous HRM studies (e.g., Snell & Dean, 1992; Wright et al., 2003). All survey questions were constructed with a 5-point Likert-type scale, where 1 indicated “strongly disagree” and 5 indicated “strongly agree.” The measurement includes HPWPs in staffing (3 items), training (3 items), performance appraisals (4 items), pay for performance (4 items), job design (2 items), and communication and participation (4 items).

Staffing was measured using three items: (1) “selects people according to highly refined selection criteria and procedures”; (2) “hires people by utilizing different kinds of selection tools (for example, interviews, aptitude tests, written exams, etc.)”; and (3) “invests money in order to select the right people.” Next, training was measured using three items: (1) “provides employees with a variety of training and development opportunities”; (2) “invests heavily in employee training and development”; and (3) “provides employees with structured formal training and development programs.” In addition to staffing and training, performance appraisal was measured using a four-item scale: (1) “has an effective formal performance appraisal system to evaluate employees’ performance and competencies”; (2) “appraises employees’ performance with objective and quantitative criteria (for example, management by objectives)”; (3) “utilizes the results of performance appraisal in deciding pay raises or promotions of employees”; and (4) “appraises employees’ performance based on their objective achievement.” Pay for performance was measured using four items: (1) “bases pay raise decisions on employee performance”; (2) “has wide range in pay within the same job grade”; (3) “extensively utilizes a company-wide profit-sharing and/or a gain-sharing program”; and (4) “utilizes a reward system based on seniority” (reverse-coded). Job design was measured using two items: (1) “provides employees with opportunities to work flexibly (for example, flexible work schedule)” and (2) “flexibly assigns the scope and responsibilities of jobs, based on employees’ skills and needs.” Finally, communication and participation were measured using four items: (1) “utilizes formal programs through which employees can participate in organizational activities (e.g., work council)”; (2) “provides employees with opportunities to participate in decision making and problem-solving related to the job”; (3) “shares various information with employees (e.g., business strategy and financial status)”; and (4) “listens to employees’ opinions through different kinds of formal or informal programs (e.g., attitude surveys).”

In line with Chadwick et al. (2015), we take the mean of the subscores for each subscale to measure the systemic HPWPs index.

SEW preservation

SEW preservation was measured using the 27-item scale proposed by Berrone et al. (2012) and applied in recent HR studies (e.g., Sánchez-Marín et al., 2019). The scale includes the following subscales: (F) family firm influence and control (6 items); (I) identification of family members with the firm (6 items); (B) binding social ties (5 items); (E) emotional attachment of family members (6 items); and (R) renewal of family bonds through dynastic succession (4 items). Cronbach’s alpha for the five FIBER dimensions was .796, which surpasses the threshold point of .7 for suggesting internal consistency (Nunnally, 1978). We further conducted a confirmatory factor analysis (CFA) on the five dimensions of SEW construct by using AMOS software, which is consistent with prior research using this type of analysis (e.g., Hernández-Linares et al., 2018a). The five SEW dimensions were included in the analysis model (see Appendix 1), despite the hypothesized model showed lower model fit than desirable, χ² = 627.923 (319), comparative fit index (CFI) = 0.786, incremental fit index (IFI) = 0.791, Tucker–Lewis index (TLI) = 0.765, and adjusted goodness-of-fit index (AGFI) = 0.689. Most of the 27 items (22) and all the subscales showed standardized factor loadings exceeding the 0.50 cut-off for practical significance (Hair et al., 2006), and 24 of them were significant at the 0.001 level (t > 2.0), suggesting convergent validity (Kohli et al., 1998). Although some standardized factor loadings were below the 0.5 threshold, we included all items in the scale due to two main reasons. First, because in our article SEW dimensions are not treated in a disaggregated order but as part of a second-order construct (SEW), all standardized factor loadings of the dimensions of SEW broadly exceed the 0.50 cut-off (Hair et al., 2006) and were significant at the 0.001 level (t > 2.0), providing evidence of convergent validity (Kohli et al., 1998). Second, while we note that another SEW measure (SEW-i) is available (Debicki et al., 2016), the literature has not yet embraced its empirical use (Hernández-Linares et al., 2020).

Control variables

To lower the effects of confounding variables on the dependent variable, we control for 10 variables. First, we controlled for different family firm aspects by including both the number of family employees and the number of female family employees involved in the firm (count) because family and nonfamily employees may be compensated, monitored, and disciplined differently. Furthermore, it is more difficult to attract qualified nonfamily employees when family firms pursue non-financial goals such as SEW preservation and favor family employees over nonfamily employees (Schulze et al., 2001; Verbeke & Kano, 2012); and because a recent study (Hernández-Linares et al., 2018b) points out that when women are seen as representative of the dominant coalition (the family), they tend to have an advantage relative to women nonfamily employees, which could influence the HR practices of family firms. In addition, we controlled for the level of family influence and control by including the percentage of firm owned by the family, since their influence on labor productivity has been empirically supported (e.g., Barbera & Moores, 2013). We also controlled for family commitment to family goals using a binary variable (yes/no) because the literature increasingly recognizes that goals vary among family businesses (e.g., Chrisman et al., 2012; Kotlar & De Massis, 2013) and that the commitment toward family-centered goals—often intangible, subjective, and with a longer time horizon—will hinder the adoption of professional management practices and employment of nonfamily employees (Chua et al., 2009; William et al., 2018).

Second, we controlled for different variables related to family firms in the year prior to survey. Thus, given that firm size has been found to be positively related to employee productivity (Terrón-Ibáñez et al., 2020) and more sophisticated HRM activities (Rodríguez-Ruiz et al., 2016), we included sales in 2016 in millions of Euros as a fifth control variable. Although sales is a common measure of firm size (e.g., Balkin et al., 2000; Boeker & Karichalil, 2002; Hambrick & Cannella 2004; Hillman et al., 2007) to avoid any bias, we also included assets in 2016 in millions as a complementary size measure. We also controlled previous performance (Pittino et al., 2016) by including the profits in 2016 in millions because it has been found to positively affect organizational human capital (Blanco-Mazagatos et al., 2018). Similarly, we controlled for debt to equity of the firm, measured in 2016 in millions (Jain & Shao, 2016), because firms “with greater amounts of debt likely face higher hurdles in gaining additional financing relative to firms with little debt” (Anderson et al., 2012). Next, and in line with recent studies (Sánchez-Marín et al., 2019), we controlled for firm age, measured as the number of years since the founding of the business (Muñoz-Bullón et al., 2018), since it has been recognized to influence both labor productivity (Way, 2002) and HR practices in family firms (Anderson & Reeb, 2003). Finally, consistent with prior research (Chadwick et al., 2015; Messersmith & Guthrie, 2010), we also controlled for industry effects because industry-specific variables (e.g., growth rate or capital intensity) have been found to affect the HPWP effectiveness (Datta et al., 2005). Specifically, 13 industry dummies based on two-digit NACE code industry sector were included in our analysis, using the Agricultural sector as the reference category.

Results

Given that our data were captured from a single informant, we attempted to ensure that the presence of common method variance (CMV) did not affect our findings, by artificially inflating observed relationships between focal variables. In line with other scholars (Arend, 2014; Koryak et al., 2018), we employed both procedural and statistical approaches. In particular, we employed four procedural steps. First, we guaranteed that the anonymity of the respondents was protected; then, the pretest ensured minimum ambiguity through the use of the correct wording of the questions (Arend, 2014). Second, to reduce possible desirability bias, we promised that we would keep all individual responses completely confidential and confirmed that our analyses would be restricted to an aggregated level that would prevent the identification of any organization. Third, we employed proximal separation of the items related to independent and dependent variables (Koryak et al., 2018). Fourth, we used data from two sources in our analysis—our survey and the SABI database. With regard to the statistical approach, we followed convention in conducting a Harman’s (1967) single-factor test (Hernández-Linares et al., 2020; Koryak et al., 2018), which is conducted by loading each of the study variables into an exploratory factor analysis. This test resulted in 27 factors with an eigenvalue exceeding 1.0 and explained 98.1% of the cumulative variance. This test provided evidence that CMV was not substantial (Podsakoff & Organ, 1986); however, given the large number of variables in the study, it cannot be ruled out completely.

Table 1 presents the means, standard deviations, and the bivariate correlations among the variables used in the regression analyses. This table shows that on average the family firms in our sample have been in business for 29 years and represent a variety of industries.

Table 1.

Means, standard deviations, and correlations.

		M	SD	Minimum	Maximum	1	2	3	4	5	6	7	8	9	10	11	12
1	Labor productivity (2017)	−2.2622	0.9645	−4.5104	0.0804
2	Profit per employee (2017)	0.0065	0.0113	−0.0149	0.0929	0.5179*
3	HPWPs	3.5585	0.6826	1.55	5	−0.0109	0.1164
4	SEW preservation	4.1278	0.5306	2.4815	4.9259	−0.0854	0	0.4537*
5	Number of family employees	1.0484	0.2504	1	3	−0.0026	−0.0839	0.1118	0.1865*
6	% of the firm owned by family	25.4278	25.2274	0	100	−0.1042	−0.0041	0.0227	0.0184	−0.0368
7	Family commitment to family goals	1.1694	0.6952	1	8	0.097	0.099	−0.1787*	−0.1473	−0.0008	−0.0718
8	Number of female family members involved	1.5484	1.2054	0	5	−0.1653	−0.2139*	0.0017	0.1569	0.1808*	0.3439*	0.005
9	Sales (millions; 2016)	24.7748	101.1340	0.9082	1099.0490	0.3342*	0.7506*	0.1225	0.0483	−0.0269	0.052	−0.03	−0.1462
10	Assets (millions; 2016)	26.9669	141.6915	0.3037	1564.4320	0.2652*	0.7291*	0.1201	0.0443	−0.0322	0.065	−0.02	−0.1419	0.9873*
11	Profits (millions; 2016)	1.7297	12.9919	−1.8059	144.6244	0.2345*	0.7343*	0.1209	0.0457	−0.0227	0.081	−0.02	−0.1283	0.9762*	0.9862*
12	Debt to Equity (millions; 2016)	2.5999	7.4655	0.0713	76.9629	−0.1721	−0.1065	0.0235	0.1121	−0.002	−0.0801	−0.03	0.0351	−0.0493	−0.0444	−0.0362
13	Firm age	29.4597	12.8243	1	69	0.4631*	0.3034*	0.0225	−0.1312	−0.0981	0.1206	−0.03	−0.0611	0.2807*	0.2607*	0.2288*	−0.1767*

HPWP: high-performance work practice; SEW: socioemotional wealth.

N = 124, industry dummies included but not reported.

*

p < .05 (two-tailed).

The proposed theoretical model was tested by using ordinary least squares (OLS) regression. Table 2 provides the estimates. To lower the effects of outliers, we estimated robust standard errors. Models 1–3 were estimated by using the labor productivity as the dependent variable, and Models 4–6 were estimated by using the profit per employee as the dependent variable. In Model 1, the control variables were included, and five of them were significant: family commitment to family goals (β = .139, p < .01); sales 2016 (β = .021, p < .01), assets (β = −.010, p < .01), and debt to equity (β = −.016, p < .01) from the year t − 1 (2016); and firm age (β = .017, p < .01). Similarly, 5 of the 13 industry sectors showed a significant impact on labor productivity (i.e., water supply, waste management, and decontamination; administrative activities and auxiliary services; education; health and social services activities; and other services). Similar results were obtained for profit per employee outcome, although firm age did not show a significant effect (β = .000, p = n.s.) and only an industry sector (hospitality) significantly influenced the profit per employee (β = .008, p < .05). The two independent variables were included in Model 2 and their interaction in Model 3. Compared to Model 2 where neither of the direct effects are significant, in Model 3, the inclusion of the interaction term increases variance explained by 1.8% (65.3% − 63.5%). In Model 3, the interaction effect of HPWPs and SEW preservation on labor productivity is negative and significant (β = −.377, p < .05). Hence, the results do not support Hypothesis 1 which proposed that HPWPs and SEW preservation would complement each other in influencing family firm performance, such that both HPWPs and SEW preservation would be positively associated with family firms’ labor productivity. However, results support Hypothesis 2, suggesting that HPWPs and SEW preservation would act as substitutes in influencing family firm performance, such that a stronger presence of HPWPs (and lower preservation of SEW) or higher SEW preservation (and weaker presence of HPWPs) would be positively associated with family firms’ labor productivity. Figure 1(a) supports the proposed effect, where with increasing HPWPs and higher levels of SEW preservation, labor productivity is lower, whereas lower levels of HPWPs coupled with higher levels of SEW preservation are positively associated with labor productivity.

Figure 1.

Moderation effects on (a) labor productivity (2017) and (b) profit per employee (2017).

Table 2.

OLS estimates.

Variables	Labor productivity (2017)			Profit per employee (2017)
	(1)	(2)	(3)	(4)	(5)	(6)
HPWPs		−0.0235	1.540**		0.000633	0.0205***
		(0.0977)	(0.745)		(0.00122)	(0.00725)
SEW preservation		−0.0263	1.281*		0.000951	0.0176***
		(0.129)	(0.646)		(0.00167)	(0.00564)
HPWPs × SEW preservation			−0.377**			−0.00480***
			(0.178)			(0.00166)
Number of family employees	0.0277	0.0408	0.137	−0.00223	−0.00266	−0.00143
	(0.136)	(0.140)	(0.128)	(0.00148)	(0.00166)	(0.00182)
% of the firm owned by family	−0.00464	−0.00457	−0.00397	−1.36e−05	−1.60e−05	−8.34e−06
	(0.00319)	(0.00318)	(0.00302)	(2.76e−05)	(2.78e−05)	(2.68e−05)
Family commitment to family goals	0.139***	0.134***	0.141***	0.00179**	0.00196**	0.00206***
	(0.0491)	(0.0481)	(0.0449)	(0.000774)	(0.000811)	(0.000767)
Number of female family members involved	−0.0259	−0.0245	−0.0469	−0.000671	−0.000719	−0.00100**
	(0.0514)	(0.0511)	(0.0512)	(0.000492)	(0.000476)	(0.000495)
Sales (millions, 2016)	0.0212***	0.0213***	0.0214***	0.000136***	0.000134***	0.000135***
	(0.00625)	(0.00629)	(0.00591)	(4.39e−05)	(4.11e−05)	(4.45e−05)
Assets (millions, 2016)	−0.0101***	−0.0101***	−0.0104***	−9.12e−05***	−9.14e−05***	−9.50e−05***
	(0.00326)	(0.00323)	(0.00307)	(2.73e−05)	(2.71e−05)	(2.94e−05)
Profits (millions, 2016)	−0.0402	−0.0405	−0.0385	0.000556**	0.000567**	0.000592**
	(0.0272)	(0.0277)	(0.0278)	(0.000232)	(0.000228)	(0.000233)
Debt to equity (millions, 2016)	−0.0164***	−0.0161***	−0.0164***	−7.94e−05*	−8.88e−05*	−9.26e−05**
	(0.00384)	(0.00413)	(0.00414)	(4.76e−05)	(5.06e−05)	(4.51e−05)
Firm age	0.0176***	0.0175***	0.0171***	8.51e−05	8.99e−05	8.53e−05
	(0.00571)	(0.00572)	(0.00542)	(5.81e−05)	(6.02e−05)	(6.15e−05)
Manufacturing industry	−0.135	−0.143	−0.0862	0.00160	0.00191	0.00264
	(0.380)	(0.376)	(0.334)	(0.00197)	(0.00203)	(0.00222)
Water supply, waste management, and decontamination	−0.887**	−0.874**	−0.901***	0.00148	0.00113	0.000786
	(0.379)	(0.373)	(0.336)	(0.00191)	(0.00204)	(0.00219)
Construction	−0.0835	−0.0895	−0.120	3.59e−05	0.000243	−0.000146
	(0.415)	(0.409)	(0.375)	(0.00170)	(0.00166)	(0.00193)
Wholesale and retail trade; motor vehicle and motorcycle repair	0.219	0.221	0.230	0.00163	0.00161	0.00172
	(0.394)	(0.386)	(0.348)	(0.00212)	(0.00212)	(0.00236)
Transport and storage	−0.0743	−0.0898	−0.0555	0.00209	0.00254	0.00298
	(0.486)	(0.494)	(0.426)	(0.00359)	(0.00350)	(0.00329)
Hospitality	−0.387	−0.384	−0.363	0.00837**	0.00830**	0.00857**
	(0.429)	(0.421)	(0.391)	(0.00397)	(0.00405)	(0.00409)
Information and communications	−0.281	−0.313	−0.281	0.00639	0.00752	0.00792
	(0.387)	(0.397)	(0.372)	(0.00461)	(0.00503)	(0.00482)
Professional, scientific, and technical activities	−0.268	−0.268	−0.216	0.00121	0.00129	0.00195
	(0.455)	(0.445)	(0.434)	(0.00290)	(0.00291)	(0.00304)
Administrative activities and auxiliary services	−1.112**	−1.119**	−1.103***	−0.00231	−0.00208	−0.00186
	(0.451)	(0.449)	(0.418)	(0.00214)	(0.00220)	(0.00236)
Education	−1.357***	−1.355***	−1.366***	−0.000701	−0.000689	−0.000834
	(0.480)	(0.473)	(0.427)	(0.00170)	(0.00199)	(0.00220)
Health and social services activities	−0.848*	−0.849*	−0.810*	0.00149	0.00157	0.00207
	(0.464)	(0.453)	(0.423)	(0.00267)	(0.00302)	(0.00316)
Artistic, recreational, and entertainment activities	0.0141	−0.0271	0.0541	0.00759	0.00899	0.0100*
	(0.430)	(0.430)	(0.383)	(0.00695)	(0.00663)	(0.00545)
Other services	−1.201**	−1.186**	−1.106**	0.00105	0.000645	0.00167
	(0.516)	(0.519)	(0.505)	(0.00172)	(0.00178)	(0.00206)
Constant	−2.689***	−2.502***	−7.977***	0.00218	−0.00389	−0.0736***
	(0.449)	(0.654)	(2.671)	(0.00284)	(0.00703)	(0.0244)
Observations	124	124	124	124	124	124
R 2	0.635	0.635	0.653	0.665	0.669	0.689

OLS: ordinary least square; HPWP: high-performance work practice; SEW: socioemotional wealth.

Robust standard errors in parentheses.

*

p < .1, **p < .05, ***p < .01.

We obtained similar results for profit per employee outcome. In Model 5 (as in Model 2), neither of the direct effects were significant. However, in Model 6 (like in Model 3), the inclusion of the interaction increases the change in R² by 2%. The interaction effect of HPWPs and SEW preservation on profit per employee outcome is also negative and significant (Model 6: β = −.0048, p < .05; change in R² = 2%). Therefore, results support Hypothesis 2; Figure 1(b) shows effects consistent with those in Figure 1(a) and supports the proposed effect in Hypothesis 2, where with increasing HPWPs and higher levels of SEW preservation, profit per employee is lower, whereas lower levels of HPWPs coupled with higher levels of SEW preservation are positively associated with profit per employee.

Overall, Hypothesis 2 on substitution effects between HPWP and SEW is supported; however, Hypothesis 1 on complementary effects is not supported.

Discussion and conclusion

Worldwide, family firms constitute the majority of businesses and provide employment to 60% of the workforce (Gersick et al., 1997; Neckebrouck et al., 2018). Taking a closer look at the country from which we draw our sample, family firms comprise 89% of all businesses in Spain, employing about two-thirds of the workforce (Instituto de la Empresa Familiar [IEF], 2015). Despite burgeoning research on the effect of HPWPs on family firm performance (e.g., Pittino et al., 2016; Sánchez-Marín et al., 2019; Tsao et al., 2009), our understanding of the interplay between HPWPs and SEW preservation, a key factor distinguishing family firms from nonfamily firms (Gómez-Mejía et al., 2007), has been largely inferential. Given this research gap, we drew upon organizational control and signaling theories and proposed a set of competing hypotheses centered on complementarity and substitutability in order to discover the true interaction between these important yet distinct control mechanisms. Specifically, on the complementarity side, we argued that the use of HPWPs (a formal bureaucratic control mechanism) in family firms coupled with a strong emphasis on SEW preservation (an informal normative control mechanism) would send a consistent, favorable signal for attracting higher quality nonfamily members for employment, thereby enhancing firm labor productivity. Conversely, on the substitutability side, we contend that while HPWPs will likely attract more nonfamily members concerned with career advancement, the family firms’ strong affinity toward SEW preservation will become apparent as family employees will receive preferential treatment over nonfamily employees. Hence, these nonfamily members will become confused and frustrated by these conflicting or inconsistent messages, thereby causing withdrawal behaviors and voluntary turnover which are detrimental to family firms’ labor productivity. Using a sample of 124 Spanish family firms for testing these hypotheses, our findings confirm a substitution effect between HPWPs and SEW preservation when it comes to family firms’ labor productivity.

In this and the paragraphs that follow, we discuss the implications of these findings for the extant literatures. First, our study clarifies the precise nature of the interaction between HPWPs and SEW preservation concerns among family firms. With prior research using indirect measures of SEW preservation indicating a negative relationship between these two mechanisms on labor productivity (e.g., Sánchez-Marín et al., 2019), our findings both confirm this perspective and extend the family firm literature in this space. Specifically, the evidence provided by this study demonstrates that full implementation of HPWPs coupled with high commitment to SEW preservation decreases labor productivity. However, these two mechanisms reveal a substitution effect such that full implementation of HPWPs along with low commitment to SEW preservation results in greater labor productivity. Likewise, limited implementation of HPWPs together with high commitment to SEW preservation enhances labor productivity. Moreover, these results suggest that examining either HPWPs or SEW preservation in isolation with regard to labor productivity paints an incomplete picture in terms of strategic HRM in family firms, as the effectiveness of one clearly depends on the degree of investment in the other. Hence, with our direct measurement of SEW preservation, we are better able to specify the true nature of this relationship for labor productivity.

Next, concerning HRM in family firms, the findings from our study suggest that nonfamily members pay close attention to the signals affecting their employment that emanate from family firms. As mentioned before, family firms implement HPWPs for mainly reputation or impression management in competitive labor markets to ameliorate the disadvantages that family firms face in hiring and retaining high-quality nonfamily members (Chrisman et al., 2014). While this may be enough to attract quality family and nonfamily members, our findings imply that post-hire nonfamily employees continue to evaluate family firm signals to ensure signal continuity or reliability with regard to their employment relationship. Essentially, the use of HPWPs signals to nonfamily members that the family firm has a commitment to professionalism in managing their workforce such that career advancement and growth will be integral to the employment relationship (Suazo et al., 2009). However, when these nonfamily employees discover that perception and reality do not line up due to a strong desire for SEW preservation, these nonfamily members are likely to withdraw from the family firm, in turn negatively affecting family firms’ labor productivity. Adding to signaling theory, this would indicate that signaling with respect to employment is an ongoing rather than a discrete event, particularly in family firms. Moreover, while organizational control mechanisms can guide employee behavior (see Cardinal et al., 2017; Gill, 2019), these mechanisms can also play an important role in the attraction of family and nonfamily employees to the firm through signaling, which extends the organizational control literature.

Finally, this study adds to the research on HPWPs and firm outcomes (e.g., Combs et al., 2006; Posthuma et al., 2013; Saridakis et al., 2017), by considering the implications of controlling families’ SEW preservation for determining the effectiveness of HPWPs among family firms. Past studies have found that the efficacy of HPWPs can depend largely on the firm’s context such as industry characteristics (Datta et al., 2005) and business strategy (Arthur, 1994); however, these studies do not distinguish between family and nonfamily firms. With this study, we focus our attention on a contextual factor that is idiosyncratic to family firms, that being their emphasis on SEW preservation. Furthermore, these findings build on the recent works examining the heterogeneity among family firms by exploring how idiosyncrasies within such firms may influence the firm members’ productivity (e.g., Neckebrouck et al., 2018). Hence, we are among the first to directly explore how the level of commitment to SEW preservation affects the effectiveness of HPWPs in family firms.

Practical implications

Our findings demonstrate to family firm managers that the effectiveness of HPWPs depends on the amount of emphasis placed on SEW preservation, and vice versa. Essentially, family firms strongly valuing SEW preservation should minimize the use of HPWPs which are contradictory, to ensure increased family firms’ labor productivity. On the flip side, family firms desiring to realize the performance benefits associated with implementing HPWPs should minimize their emphasis on SEW preservation. This ensures that any signals emanating from either of these mechanisms are both clear and reliable in seeking to attract and motivate employees (Connelly et al., 2011). Otherwise, any mixed signals particularly to nonfamily members can be detrimental to family firms’ labor productivity, which results from efforts to fully implement HPWPs while simultaneously maintaining a strong hold on SEW preservation. Hence, those family firm managers considering the adoption of HPWPs should first conduct a cultural audit to ascertain the family firm’s inclinations toward SEW preservation. Once the family firm’s status regarding SEW preservation is clarified, this assessment should be used as a guide in determining whether it is appropriate to introduce and use HPWPs.

Limitations and future research directions

Our study has some limitations which also provide future research opportunities. First, our study was cross-sectional; hence, the inferences are not causal (Tabachnick & Fidell, 1996). Longitudinal studies of how family firm factors can affect employees’ labor productivity in family firms (e.g., impact of HPWP in year t on labor productivity in year t + 1) are needed.

Second, the cultural setting and legal framework can also affect the observed relationships and generalizability, since our study was conducted on a sample of family firms in Spain. HPWPs in Spain tend to be used relatively less than those in the United States or the United Kingdom (Pruneda, 2015). This is because Spanish labor law has been undergoing profound changes in response to the economic and financial crises. The labor law reform implemented in 2012 aimed to loosen the criteria for “fair” dismissal on economic grounds by extending and clarifying the reasons for employee separations and by establishing severance payments in 20 wage days per seniority year (with the upper limit of one salary year) for (newly hired) permanent workers. In the past, severance payments for unfair dismissals on economic grounds were 45 wage days per seniority year (with the upper limit of 42 salary months) for ordinary open-ended contracts or 33 days (with the upper limit of 42 salary months) for ordinary open-ended contracts or 33 days (with a limit of two salary years) for open-ended contracts which were launched during the 1997 labor market reforms aimed to foster permanent employment (Eichhorst et al., 2017; García-Serrano & Malo, 2013; Pruneda, 2015). Currently, a distinctive feature of labor market in Spain is the high proportion of temporary workers (Latorre et al., 2016). Despite such developments and differences in terms of economic and legal frameworks, increased globalization tends to cause similarities in business conduct across countries (e.g., Carr, 2005). Hence, context may not significantly affect our findings. However, future studies can test or extend our work to other countries.

Third, among the variety of definitional criteria appearing in the literature (Hernández-Linares et al., 2017, 2018c), we have defined family firms by using objective and subjective criteria, which is common in SEW studies (Hernández-Linares et al., 2020; Schepers et al., 2014). However, future studies could use a broader definition of the “family firms” term, and analyze, for instance, if our results are corroborated when a family owns the majority of company’s shares but does not perceive the business as a family business.

Fourth, we measure SEW with FIBER scale (Berrone et al., 2012). As Jiang and colleagues state (2018), “a direct measurement of SEW phenomena is virtually non-existent, and construct development is just beginning” (p. 133). Despite the limitations of our measure, it is much richer than the distal proxies commonly used in empirical literature (Debicki et al., 2016). However, it would be very useful to replicate this study using other measures, such as SEW-i (Debicki et al., 2016).

Finally, while we follow past studies for operationalizing labor productivity (Datta et al., 2005; Huselid, 1995), this measure is a coarser measure of labor productivity. The labor productivity measure at the firm level is subject to measurement error driven by variation in labor productivity across departments. Furthermore, the measure may be more accurately operationalized by accounting for a variety of labor-related inputs and outputs and by using stochastic frontier-type analysis.

Turning to more future research directions, other family firm–specific antecedents of labor productivity in family firms can be investigated. Future research might also investigate whether labor productivity varies according to other unique family firm–specific factors such as family life cycle stages or family firm culture (Dyer, 1988; Gersick et al., 1997). Moreover, family firms may not always exhibit supportive and harmonious work environments, particularly when family relational conflict prevails. In some family firms, family relational conflict can harm decision-making process, firm development, and family firm performance (e.g., Au & Kwan, 2009; Beckhard & Dyer, 1983; Dyer, 1986; Eddleston et al., 2008). Therefore, future research can investigate labor productivity within the context of family relational conflict and other socio-psychological dynamics as well.

Moreover, some families may desire to build or expand SEW rather than preserve SEW, and HPWPs might facilitate this. For instance, some family firms may prefer to enhance their SEW. This can lead to the exclusive treatment of family business members over nonfamily employees through elevated family control and intra-family succession intentions. We expect that this may further reinforce the substitutability between HPWPs and SEW in influencing labor productivity. In our article, we focus on SEW preservation. However, future research may explore the interplay between the SEW growth priority and HPWPs in influencing family firm outcomes.

As discussed previously, HPWPs and SEW preservation can consequently affect employee turnover intentions and voluntary turnover through labor productivity in family firms; however, this needs to be more directly assessed. In the absence of support systems such as HPWPs and in turn productivity, both family and nonfamily employees, who are likely to be dissatisfied with their jobs, may choose employment alternatives outside the family business or pursue self-employment by starting up their own business ventures, endangering the transgenerational survival of family firms. Given the extensive and critical presence of nonfamily employees in family firms and the calls for studies investigating relationships with nonfamily employees (Barnett & Kellermanns, 2006; Chua et al., 2003; Daspit et al., 2018; Madison et al., 2018; Tabor et al., 2018), it is crucial to take a closer look at how HPWPs and SEW preservation influences the nonfamily employees’ perceptions, intentions, and actions in family firms.

Furthermore, we recommend qualitative and case study research (De Massis & Kotlar, 2014) that may examine both the owner-managers’ and employees’ potentially nuanced perspectives in terms of HPWPs, SEW preservation, and family firm outcomes. In addition, longitudinal studies can capture the changes in HPWPs, SEW preservation, and family firm outcomes over time owing to economic and legal developments (or downturns) as well as the shifts in priorities related to the financial and non-financial goals.

While this study has drawn upon the organizational control and signaling theories, future research may also apply different theoretical perspectives in exploring HPWPs, SEW preservation, and family firm outcomes. For instance, behavioral agency and prospect theories can shed light onto the potential agency dynamics and risk preferences concerning HPWPs and SEW preservation in family firms.

Conclusion

In conclusion, family firms must be mindful of their commitment to SEW preservation when it comes to implementing HPWPs. Essentially, each of these control mechanisms send signals that appeal in different ways to nonfamily members when evaluating family firms as employers. However, when signals emanating from these mechanisms clash, family firms are likely to end up with a frustrated and disgruntled workforce such that firm labor productivity levels decline. Hence, our findings indicate that family firms are likely to experience higher labor productivity when HPWPs are fully implemented along with low commitment to SEW preservation or when there is limited implementation of HPWPs along with high commitment to SEW preservation. In these scenarios, the signals become clear to nonfamily employees, thereby ensuring greater family firm success.