Enterprise values and enterprise policy interdependence

Enterprise governance results in enterprise policy, which is influenced by enterprise values. The aim of our research was to identify how enterprise values influence enterprise policy innovation. The study focuses on four value factors and their relationships as well as on their influence on enterprise policy innovation. The results show that enterprises should pay attention to the chain of value factors identified in our study (V3 ! V2 ! V4 ! V1). Namely, factors ‘extremes of the values of enterprise’s key stakeholders’ (V3), ‘analysis of the values of enterprise’s key stakeholders’ (V2), and ‘importance of the examination of the values of enterprise’s key stakeholders’ (V4) have an indirect impact on factor ‘enterprise policy innovation (EPI) via factor ‘values of enterprise’s key stakeholders’ (V1), which influences the factor ‘enterprise policy innovation’ (EPI) directly. The identified chain of value factors has a strong effect on factor ‘enterprise policy innovation’ (EPI) via factor ‘values of enterprise’s key stakeholders’ (V1). The implications happen via enterprise’s management and the basic-realisation process. ARTICLE HISTORY Received 13 October 2016 Accepted 22 October 2018


Introduction
With his belief-value-cultural research, Hofstede established in 1980 that enterprise values influence enterprise culture. Hofstede researched the IBM enterprise culture based on the beliefs and values as well as on geographical location of their 117,000 employees in 72 countries. He first determined four universally held human values and later added two more dimensions, i.e., the fifth dimension (Hofstede & Bond, 5. Short-term/long-term orientation (initially called the Confucius connection) (Hofstede & Bond, 1988). 6. Indulgence/restraint (based on the World value survey data of Minkov, 2007).
Based on Hofstede's early research, House, in 1991House, in conceived and in 1994House, in -1997 collected together with his team the data for the GLOBE research (Global Leadership and Organizational Behaviour Effectiveness Research) in 62 countries (see Chhokar et al., 2008;House et al., 1999House et al., , 2004. They established that enterprise values (and practices) help distinguish enterprise culture, which reflects both societal values and culture.
Based on his research findings, Ja. Belak (2010, p. 152) includes 10 key success factors in the MER-model of integral management, which is used as the basis for this research (hereinafter the MER-model) and, therefore, limits the research. Belak notes that these key success factors are the external and internal compliance of the enterprise, enterprise's credibility, competitiveness, entrepreneurship, synergy, culture, philosophy, ethics, ecology, and effectiveness (Ja. Belak, 2010, pp. 149-178). As we can see, the author includes enterprise culture and enterprise ethics in his model. According to Poto can and Mulej (2007), enterprise culture is influenced by enterprise values and it influences enterprise ethics. According to the MER-model (Ja. Belak, 2010;Ja. Belak & Duh, 2012;Ja. Belak et al., 2014;Duh, 2015), enterprise culture and enterprise ethics are enterprise key success factors (Je. Belak, 2016;Duh et al., 2016). Consequently, we can conclude that since enterprise values directly influence enterprise culture and indirectly enterprise ethics, enterprise values are one of the important enterprise's factors, too (see e.g., Me sko Stok et al., 2009;Strukelj, 2015) and they should be balanced (Malba si c et al., 2015).
In this paper, we research enterprise values and their influence on enterprise policy innovation. Theoretical discussions on this theme to date are mainly qualitative and extremely rare (e.g., Je. Belak, 2013;Malik, 2011;Shaw, 2006;Strukelj, 2015;Strukelj & Suligoj, 2014). We found mostly partial studies, which served as the basis for our research model development (see Section 5). Quantitative research linking enterprise values and their influence on enterprise policy innovation has not been found. Given that we have established that our research is statistically significant, this expands the set of factors which enterprise's key stakeholdersmainly owners/governorsshould pay attention to when perceiving factors that affect enterprise policy innovation. The paper addresses enterprise policy and enterprise values interdependence and demonstrates that enterprise values influence enterprise policy. The research question reduces the research gap on whether (see Sections 2-5; Figure 1 shows our research model) and to what extent (see Sections 6-8; Figure 2 shows the results of structural model analysis) different enterprise values factors influence enterprise policy innovation (see Section 9 for conclusions and recommendations for future research). This limits our research to just one of the 'soft' factors which influence enterprise policy (other factors are, for example, culture and ethics/sustainability ( Strukelj & Suligoj, 2014)). The research findings of this study show that enterprise owners/governors should pay close attention to the chain of value factors we have identified (V3 ! V2 ! V4 ! V1) because the factors 'extremes of the enterprise's key stakeholders values' (V3; Allport et al., 1931Allport et al., , 1951Allport et al., , 1960Kanungo, 2001;Malba si c et al., 2015;Me sko Stok et al., 2009;Russo, ed., 2000;Ulrich, 1990), 'the analysis of the enterprise's key stakeholders values' (V2; Je. Belak, 2013;Kluckhohn, 1961;Minkov, 2007;Schwartz, 1992;Ulrich, 1990) and 'the importance of the examination of the enterprise's key stakeholders values' (V4; Kav ci c et al., 2015;Lencioni, 2002;Mulej & Dyck, eds., 2014;Rokeach, 1979;Strukelj, 2015) impact enterprise policy innovation (EPI; Duh, 2015;Ulrich, 1990;Strukelj, 2015;Strukelj and Suligoj, 2014) indirectly through factor 'the values of enterprise's key stakeholders' (V1; Ja. Belak, 2010; Duh,  2015; Strukelj & Suligoj, 2014;Ulrich, 1990), which in turn directly impacts the factor EPI.
The paper has been divided into two parts. In the first part, we explain the MERmodel and enterprise policy (innovation) (Section 2), give an overview of the theoretical foundations for our research (Section 3) and introduce general values (Section 4). Then, we present our research model regarding enterprise values and discuss their importance for enterprise policy (Section 5). In the second part of this paper, we present the research methodology (Section 6) and the results of our survey (Section 7). This is followed by the discussion of the results (Section 8). Finally, we draw conclusions of our research and provide recommendations for further research work.

The MER-model of integral management and enterprise policy (innovation)
The MER-model is the result of the collaboration of researchers from many European countries as well as from Canada and the U.S.A. It combines European and Anglo-Saxon insights. It is based on the understanding of integral management (Ja. Belak & Duh, 2012, p. 9 and 16) as the one which enables the holistic governance and management of an enterprise in all its dimensions. These dimensions are (1) the objective perspective (it is applicable for all types of enterprises in the broader sense of the term), (2) the space perspective (market, operational and cognitive space dimension) and (3) the time perspective (for longer as well as shorter time periods). According to the MER-model (Ja. Belak & Duh, 2012, p. 13), enterprise governance directly results in enterprise policy, i.e., in the general, long-term statements of the enterprise's essential characteristics, and contains enterprise's mission, purpose/aims and basic goals. Enterprise's policy is directly realised in its strategic management and indirectly in its operational management and business practice. Innovation (Mulej et al., 2013, p. 256) is any novelty that its users consider beneficial in practice. One of its content criteria is enterprise policy and management process innovation (supportive of co-operation), to which we limit our study, with variations: radical or incremental consequences and with or without on-job-duty for its creation. Enterprise policy innovation ( Strukelj, 2015) represents non-technological innovation which includes requisitely holistic innovation of enterprise governance directions by taking into account all essential and only the essential viewpoints and their interdependence. Authors of all modern established integral management models (e.g., Ja. Belak, 2010;Duh, 2015;Duh & Strukelj, 2011;M€ uller-Stewens & Lechner, 2005;Wheelen et al., 2018) emphasise that there are some key factors which influence the enterprise's policy innovation. These are (1) different types of environment, e.g., natural, social-political and other social, technical-technological and economic environments as well as supply and sales markets (they serve for the analysis of the environment and the resulting opportunities/threats determination); (2) the analysis of the enterprise and the resulting strengths/weaknesses determination; (3) the system of values, culture and ethics/sustainability of the enterprise; (4) the expected behaviour of enterprise's key stakeholders (owners/governors and top managers). These factors are also included in the MER-model (Ja. Belak, 2010;Ja. Belak & Duh, 2012;Ja. Belak et al., 2014;Duh, 2015;Duh & Belak, 2014) and when they change sufficiently crucially, the enterprise's policy must be adaptedtheir innovation affects enterprise policy innovation (Ja. Belak, 2010;Duh, 2015;M€ uller-Stewens & Lechner, 2005;Wheelen et al., 2018). Enterprise policy innovation is also influenced by (5) the current enterprise policy (Duh, 2015, p. 133;Ulrich, 1990, p. 90). Therefore, the information about all of the five factors stated above forms the research-prognostic basis for enterprise policy planning/innovating. By using the MER-model (Ja. Belak, 2010, p. 15-16), governors and managers (institutional dimension) focus on process and instrumental dimensions as well as on enterprise itself, on its environment and its key success factors. The limits of this research are the process dimension of the MER-model and the selected key success factorenterprise values. Enterprise governance results in enterprise policy; enterprise policy is influenced by enterprise values, i.e., the values of key, decisive enterprise stakeholders. They define the 'reach/range' of enterprise policy (Duh, 2015, p. 122). Enterprise values, i.e., the internalised criteria for identifying/evaluating the desired behaviour of enterprise, influence its culture (Poto can & Mulej, 2007), which is classified among the key success factors of an enterprise in the MER-model (Ja. Belak, 2010;Ja. Belak et al., 2014;Duh, 2015;Duh & Belak, 2014;Duh et al., 2016). We can thus conclude that enterprise values also influence enterprise success (Me sko Stok et al., 2009) and are thus considered in this research as enterprise key success factor.

Theoretical foundations
In 1961, Kluckhohn and Strodtbeck presented their Values Orientation Theory. They suggested researching five universal problems to mankind stating that all societies are aware of several possible solutions to these problems, but they rank them in different order of importance. Their value-based solutions reflect the researched societies' values orientation profiles. Value orientation theory (Kluckhohn & Strodtbeck, 1961)  individualistic). 4. Activity orientation (doing orientationmotivation for behaviour is external; becoming orientationmotivation for behaviour is living for the moment; being orientationmotivation for behaviour is internal). 5. Human nature orientation (goodparticipatory management style is prevailing; badautocratic management style is prevailing; mixtureneither of the two extreme management styles prevails).
In his work Study of Values, Rokeach (1979) researched personal values. He determined 36 universally held values for every person such as honesty, wisdom, courage, etc., which is much more than Hofstede's six universally held values determination (Hofstede, 1980(Hofstede, , 1998(Hofstede, , 2001Hofstede et al., 2010;Hofstede & Bond, 1988). Similarly, Schwartz (1992) developed his values theory which was based on research carried out in 54 different countries. This theory proposes 10 universal values which reflect the individual's unique experiences and the normative influence of the culture, and seven values which are universal across cultures, i.e., they are invariant and are universally recognised guiding principles. Smith and Bond (1998) came to similar results as Hofstede (1980Hofstede ( , 2001 and Schwartz (1992) although they used different methodological approaches in their research. They indicated that universally applicable values theory will be developed soon.
Taking into account these starting points, researchers have shown that the establishment of universal sets of basic human values is possible (see also Allport et al., 1931 and onwards;Hofstede, 1980Hofstede, , 1998Hofstede, , 2001Rokeach, 1979;Schwartz, 1992). Because societal values influence enterprise values (House et al., 2004), the Values Orientation Theory (Kluckhohn & Strodtbeck, 1961), which is still used worldwide (Russo, 2000), may prove a useful tool for enterprise values research, which we introduce in Section 4.

Enterprise values
Future enterprise policy, and thus the resulting management and practice, is influenced by the existing enterprise vision and policy, by enterprise strengths, weaknesses, opportunities and threats. It is also influenced by soft variables such as enterprise values, culture, ethics and norms and the related interests of key stakeholders ( Strukelj & Suligoj, 2014). Enterprise (stakeholders') values can be used as the framework/criteria for enterprise policy assessment. Lencioni (2002, pp. 114-115) defined the following different types of enterprise values: (1) core values, which are deeply ingrained principles that guide enterprise development, management and practice; they also influence the enterprise's culture; (2) aspirational values, which are values that are currently lacking and should be stimulating and stimulated; (3) permission-to-play values, which reflect the minimum standards required of any employee (behavioural and social values); and (4) accidental values, which usually reflect the interests of employees and arise spontaneously; they can be either positive for an enterprise (they create an atmosphere of inclusivity) or negative (they foreclose new opportunities). According to Lencioni, only core values, which should be aggressively authentic and which often reflect the values of an enterprise's founders, are the most important ones and should be both stimulating and stimulated. In the MER-model, enterprise's (core) values are researched based on Ulrich's (1990) methodology. Ulrich (1990, p. 53) proposes to analyse the following indicators: profit for the payment of owners/profit for dividends, profits for new investments, risk relation, sales growth, quality of products/services sold, geographical expansion, ownership relationship, innovation directions, relation to government, social goals consideration, co-workers goals consideration, and management (leadership) style. Enterprises should choose among these indicators to develop their (core) values and to personalise them i.e., to make them authentic. Enterprises should also be socially responsibly oriented (Berger et al., 2007;Carroll, 1979 and later;Dankova et al., 2015;Duh & Strukelj, 2011;Epstein & Rejc Buhovac, 2014;EU, 2011;Graeme et al., 2016;ISO, 2010;Mulej et al., ed., 2013Mulej et al., ed., , 2015Mulej & Dyck, eds., 2014;Sarotar Zi zek & Mulej, 2013). This requires innovative behaviour diffusion (Ritala & Sainio, 2014) that can be spread through social responsibility (Mulej et al., ed., 2013(Mulej et al., ed., , 2015 or, in other words, socially responsible values can help enterprises develop their innovation, due to their openness to collaboration.

Research model
To support the existing research, we present the results of an empirical research conducted in SMEs in Slovenia as an example of a transition economy. Enterprise policy influences its management and practice ( Can cer & Sarotar Zi zek, 2015;Duh, 2015;Kav ci c et al., 2015); therefore, it is important that its values are socially responsible (Malik, 2011;Mulej et al., ed., 2013Mulej et al., ed., , 2015Mulej & Dyck, eds., 2014).
Malik (2011), Epstein & Rejc Buhovac (2014) and Strukelj (2015) stated that enterprise's values influence its policy. They suggest that these values influence enterprise policy innovation (EPI). This is why our research aim was to establish which factors define enterprise values. We selected the factors according to the MER-model and researched how much they influence EPI.
We also aimed to establish the extent to which the values affect EPI. As can be seen from the discussion so far, different factors influence enterprise values. These factors are consistent with the MER-model (Ja. Belak, 2010;Ja. Belak & Duh, 2012;Duh, 2015;Duh & Belak, 2014), which includes the findings of Ulrich (1990) and St. Gallen school of integral management (M€ uller-Stewens & Lechner, 2005) and are, as systems theories recommend (e.g., Malik, 2011;, supplemented with subjectively selected other important findings. Based on these research cognitions, we defined the factors of enterprise values research as: values of enterprise's key stakeholders (V1), which are the values of owners/governors and top managers (Ja. Belak, 2010;Duh, 2015;Strukelj & Suligoj, 2014;Ulrich, 1990), the analysis of the values of enterprise's key stakeholders (V2), which represents a research of values by identifying the circumstances that cause certain values of key stakeholders, to better understand their values (Je. Belak, 2013;Kluckhohn, 1961;Minkov, 2007;Schwartz, 1992;Ulrich, 1990), the extremes of the values of enterprise's key stakeholders (V3), which means identifying extreme values that can positively/negatively affect the enterprise (Allport et al., 1931(Allport et al., , 1951(Allport et al., , 1960Kanungo, 2001;Malba si c et al., 2015;Me sko Stok et al., 2009;Russo, ed., 2000;Ulrich, 1990), and the importance of the examination of the values of enterprise's key stakeholders (V4), which we must determine to identify the awareness why the research of the enterprise's key stakeholders' values is crucial; only when the importance of examining the values of enterprise's key stakeholders is perceived as important, appropriate attention to the values will be given (Kav ci c et al., 2015;Lencioni, 2002;Mulej & Dyck, eds., 2014;Rokeach, 1979;Strukelj, 2015).
The main research hypothesis (H) of this paper shall therefore read as follows: Values affect enterprise policy innovation (EPI).
According to Strukelj (2015), not all factors directly affect values that influence the EPI. We prove that the factor extreme of the values of enterprise's key stakeholders (V3) includes extreme values, which are included in the scheme for studying the values of enterprise's key stakeholders in the MER-model. They illustrate the limits, the maximum or minimum values of each enterprise key stakeholder of an enterprise (i.e., owners, also top managers) (Ja. Belak et al., 2014;Duh, 2015;Ulrich, 1990). These extreme values determine whether the enterprise will analyse the values of its key stakeholders at all because they differ from the average substantially (Ja. Belak, 2010;Duh, 2015). We therefore assume that these values have a significant impact on the analysis of the values of the enterprise's key stakeholders. This serves as the basis for our hypothesis 1: The extremes of the values of enterprise's key stakeholders (V3) affect the analysis of the values of enterprise's key stakeholders (V2).
The awareness of the importance of examining the values through analysis is growing (Minkov, 2007;Rokeach, 1979;Schwartz, 1992). When enterprises analyse the values of their key stakeholders, i.e., when they determine the consistency and/or the diversity of their key stakeholders' values sufficiently, the values become more important in the enterprise (Kanungo, 2001;Malba si c et al., 2001). The enterprise's key stakeholders become more attentive to the values and the studying of these values becomes more important for them (Allport et al., 1931(Allport et al., , 1951(Allport et al., , 1960Strukelj, 2015). This is the basis for our hypothesis 2: The analysis of the values of enterprise's key stakeholders (V2) affects the importance of the examination of the values of enterprise's key stakeholders (V4).
When enterprises examine the values of their key stakeholders, they emphasise this more (Je. Belak, 2013). The study of values is therefore becoming more important and this further affects the (core) values of the enterprise's key stakeholders (Kluckhohn & Strodtbeck, 1961;Lencioni, 2002). If, namely, the values of the enterprise are becoming more important and the enterprise examines them more closely, the enterprise's key stakeholders are also more attentive to them (Russo, ed., 2000;Mulej et al., ed., 2013Mulej et al., ed., , 2015. Therefore, their values (i.e., the values of the people who have the greatest influence on what kind of values will be applied in the enterprise) slowly change (Mulej & Dyck, eds., 2014). This is the foundation of our hypothesis 3: The importance of the examination of the values of enterprise's key stakeholders (V4) affects the values of enterprise's key stakeholders (V1).
As societal values influence enterprise values (House et al., 2004), enterprise's key stakeholders' values (V2) influence enterprise policy (Duh & Belak, 2014;Malik, 2011). The values of enterprise's key stakeholders are those which build a profile of enterprise values, i.e., those values which the values of the enterprise are developed from (Lencioni, 2002). Namely, due to the impact they have on the enterprise, the values of enterprise's key stakeholders help shape the values of the enterprise in accordance with their own values (Kav ci c et al., 2015;Me sko Stok et al., 2009). This is the basis for our hypothesis 4: The values of enterprise's key stakeholders (V1), which represent a system of enterprise values, affect enterprise policy innovation (EPI).
To our knowledge, no research into this issue has been carried out in this manner to date. Therefore, we designed a research model (Figure 1) in which the described hypotheses were included.

Research instrument
The factors of our proposed model (V1, V2, V3, V4) form the values of the enterprise and affect enterprise policy innovation (EPI). Enterprise policy defines general, long-term statements of the enterprise's essential characteristics (Ja. Belak & Duh, 2012, p. 13), which determines responsible or opportunistic enterprise orientation. Enterprise policy innovation (EPI) is considered as something new or improved in enterprise policy, and is to be beneficial in practice. It is therefore important to determine what and to what extent affects EPI (i.e., innovation of enterprise's mission, purpose/aims and basic goals).
As suggested by Straub (1989), we developed our research instrument (i.e., questionnaire) in multiple stages to include both values and EPI. First, we performed extensive literature review to determine which indicators had already been determined. The questionnaire used in our research was thus based on the theory of the MER-model (the indicators were proposed or developed by Ulrich (1990) and included in the MER-model) (Duh, 2015). Some indicators were adopted from Strukelj (2015; the author also follows the MER model theory). The indicators for factor 'values of enterprise's key stakeholders' (V1) were mainly developed by Ulrich (1990, p. 53) and are included in the MER-model (Duh, 2015, p. 122-124). Some of them were developed by Strukelj (2015, Attachment 2, pp. 40-41). The indicators for factors 'analysis of the values of enterprise's key stakeholders' (V2) and 'extremes of the values of enterprise's key stakeholders' (V3) were developed by Ulrich (1990, p. 53) and are included into the MER-model (Duh, 2015, p. 122-124). The indicators for factor 'importance of the examination of the values of enterprise's key stakeholders' (V4) were developed by Strukelj (2015, Attachment 2, pp. 40-41). Strukelj (2015, Attachment 2, pp. 48-49) also developed indicators for the dependent factor of the research model, i.e., enterprise policy innovation (EPI) by combining recognitions from the MER model (Ja. Belak, 2010;Ja. Belak et al., 2014;Duh, 2015), which incorporates research from Ulrich (1990, p. 90), and from integral management model developed in St. Gallen (Malik, 2011;M€ uller-Stewens & Lechner, 2005).
We measured EPI by: 1. determining the importance of individual fields of environment development (thus the importance of changes in opportunities and threats) for EPI (measuring how development in natural, socio-political and other social, technical-technological, and economic environments as well as development in supply and sales markets influence EPI); and 2. determining the importance of the individual strategic assessments for EPI (measuring how current enterprise policy and changes in strengths/weaknesses, enterprise's values, culture and ethics, and expected behaviour of enterprise's key stakeholders influence EPI) ( Strukelj, 2015).
Second, the adopted instrument was then validated to ensure construct validity and reliability within enterprise values (V1, V2, V3 and V4) and enterprise policy innovation (EPI) context. All indicators of factors were measured on a 7-point Likert scale, ranging from 'strongly disagree' to 'strongly agree'. Demographic information was included as well. The instrument was pilot tested with a group of 30 enterprise key stakeholders. The instrument's reliability was evaluated and the Cronbach's alpha values indicated a satisfactory level of reliability (exceeding value 0.5) (Hinton, 2004). As part of the pilot test, the comments and suggestions on the questionnaire items and their wording were taken into account. Based on the results of pilot testing and the respondents' comments, revisions and additions were made to the instrument. Final scales and indicators are listed in Table 2.

Sample and procedure
Our hypotheses were tested empirically by using a questionnaire which was completed by owners and/or top managers who develop enterprise policy. Enterprises were selected based on the following two criteria: 1. enterprises had to be Slovene enterprises with at least 10 years of business operations, and 2. the questionnaire had to be completed by the owner or a top manager of the enterprise with at least five years of working for that enterprise.
To get our sample, we used random stratified sampling method. We obtained our survey data on enterprises from Slovenian business directory Bizi.si (www.bizi.si). 1 We performed the search for enterprises per their field of activity in line with the Slovenian Standard Classification of Activities Fields categorisation (http://www.stat. si/doc/pub/skd.pdf). Based on the share of each field of activity in all fields of activity, we selected the share of potential enterprises to be surveyed and randomly selected them (33% from the first third, 33% from the second third, and 33% from the last third of enterprises on the list) (i.e., a randomised stratified sample). The initial email was sent to 4,200 enterprises to verify if they matched our selection criteria and to explain the purpose of the study. Seven hundred eighty-eight enterprises agreed to  participate in the research. We sent the second e-mail to these enterprises with a link to our online questionnaire. Seven hundred thirty-four questionnaires were completed properly and were included in the analysis.

Model estimation
Covariance-based structural equation modelling (SEM) and component-based SEM or partial least squares (PLS) approach can be employed to estimate the parameters in a hierarchical model. According to Chin (1998), PLS has several major strengths: it is a predicative technique suitable for situations with less theory development; it places minimal demands on measurement scales; it avoids factor indeterminacy problems and inadmissible solutions; it avoids identification problems of recursive models; it makes no assumptions about the data; it requires no specific distributions for measured variables; it assumes the errors are uncorrelated; it works well with small samples and it is better suited for analysing complex relationships and models (Sternad et al., 2011; see also Sebjan & Tominc, 2015). Due to all the benefits of PLS as shown above, we decided to use PLS. The empirical data were analysed in two stages involving a PLS technique and using Smart PLS 2.0 M3 (Ringle & Will, 2005). In the first stage, all measurement scales were examined for their psychometric properties, while the second stage focused on hypotheses testing and analysis. Legend: a Indicators were developed by Ulrich (1990, p. 53) and are included into the MER-model (Ja. Belak, 2010;Duh, 2015, p. 122-124). b Indicators were inverted before statistical data processing in SmartPLS. c Indicators were developed by Strukelj (2015, Attachment 2, pp. 40-41). dIndicators were proposed by Ulrich (1990, p. 90), are included into the MER-model (Ja. Belak, 2010;Duh, 2015, p. 134 ), and developed by Strukelj (2015, Attachment 2, pp. 48-49). Source: Own research.

Descriptive statistics
A total of 734 properly completed questionnaires were analysed. Survey respondents represented different groups of enterprise size, including micro enterprises (32.95%), small enterprises (54.90%), medium-sized enterprises (10.00%) and large enterprises (2.15%). The size of the examined enterprises was based on the quantitative criteria defined in the Slovenian Companies Act (ZGD, 2015) as follows: number of employees (headcount), annual turnover and total balance sheet.
There were 69.20% of male and 30.80% of female respondents of our questionnaire and there were more men than women in all size classes of enterprises examined, which is consistent with data of the Statistical Office of Republic of Slovenia. Out of 44,256 people who acted as enterprise managers at the end of 2010, 14,006 (31.6%) were women (SURS, 2012).
Overall, we can identify two groups of questionnaire respondents: 534 were either owners or co-owners of enterprises (i.e., 72.75%) and 200 were top managers of enterprises (or 27.25%). As regards the owners, the largest group of respondents were the sole (100%) owner-managers of their enterprises (346 persons or 47.14%) and the rest were the co-owners of enterprises (188 persons or 25.61%).

Measurement model
Discriminant validity between constructs was assessed following Fornell and Larcker's (1981) recommendation that the square root of AVE for each construct should exceed the bivariate correlations between that construct and all other constructs. The interconstruct correlation matrix (see Table 1; source: own research) shows that the principal diagonal elements (square root AVE) exceed non-diagonals elements in the same row or columns (bivariate correlations), demonstrating that the discriminate validity of all scales is also adequate.
All the scales were evaluated through pilot testing; psychometric properties (measurement model) of these scales were assessed via evaluation of reliability, convergent validity and discriminant validity of each measurement scale. We used standard PLS algorithm for model estimation (Henseler et al., 2009). We examined two measures of reliability: Cronbach's alpha (a) and composite reliability (CR). As shown in Table 2, each of our 5 scales had Cronbach's alpha exceeding 0.85 and composite reliability exceeding 0.90, thus assuring adequate reliability for our measurement scales. Fornell and Larcker's (1981) assessment criteria were adopted for convergent validity: all item factor loadings should be significant and should exceed 0.70, and the average variance extracted (AVE) for each construct should exceed 0.50. Table 2 (source: own research) lists item factor loadings, all of which were significant at p < 0.001 and were higher than recommended level of 0.70; all values AVE exceeded 0.50 as well. The remainder of our measurement scales shows strong evidence for convergent validity.

Structural model and hypothesis testing
The next step in the analysis was to examine the path significance and magnitude of each of our hypothesised effects and the overall explanatory power of the proposed model. The hypotheses testing results are based on bootstrapping (with 500 subsamples) to test the statistical significance of each path coefficient using t-tests, as recommended by Chin (1998). Results of this analysis are shown in Figure 2 (source: own research).
The structural model demonstrates predictive power as the variance explained (R 2 ) in key endogenous constructs. All of R 2 can be described as 'weak', except for EPI where R 2 can be described as 'medium' by Cohen (1988). The findings show that our model explains part of variance in the endogenous variables, with an average R 2 of 0.08 (Table 3; source: own research).
The examination of fit indexes reflecting the predictive power of estimated inner and outer model relationships is an important part of model evaluation. It can be measured by evaluating goodness-of-fit (GoF) coefficient (Tenenhaus et al., 2005). The general criterion for evaluating GoF is to calculate the geometric mean of the average AVEs and the average R 2 of endogenous variables (Henseler & Sarsted, 2013). Based on the categorisation by Cohen (1988), GoF criteria for small, medium, and large effect sizes would be 0.1, 0.25 and 3.6. GoF for our model is 0.25, which indicates medium fit of the model to the data.
Also, the blindfolding approach proposed by Wold (1982) was followed to calculate the cv-communality and cv-redundancy indexes. The cv-communality index (H 2 ) measures the quality of the measurement model, where the cv-redundancy index (i.e., Stone-Geisser's Q 2 which Tenenhaus et al. (2005) call F 2 ) measures the quality of the structural model. As shown in Table 3, the measurement model (H 2 ¼ 0.60) shows better quality than the structural model (F 2 ¼ 0.06).

Discussion
The aim of our research was to identify if values have an influence on enterprise policy innovation (EPI). Our starting point was the MER-model where four factors of values were identified. In our study, we observed a relationship between different factors of values and their impact on enterprise policy innovation (EPI). Based on the results of our PLS analysis (see Figure 2), the factor 'extremes of the values of enterprise's key stakeholders' (V3) is strongly positively related to the factor 'analysis of the values of enterprise's key stakeholders' (V2) (b ¼ 0.315; p < 0.001), which confirms our first hypothesis (H1). Research has also shown that factor V2 is also strongly positively related to the factor 'importance of the examination of the values Table 3. Number of indicators, explained variance (R 2 ), AVE and blindfolding results of cv-communality (H 2 ) and cv-redundancy (F 2 ).  (Tenenhaus et al., 2005). Source: Own research. of enterprise's key stakeholders' (V4) (b ¼ 0.175; p < 0.001). Because of that, H2 is empirically confirmed. Our findings also suggest that factor V4 has a strong positive effect on the factor 'values of enterprise's key stakeholders' (V1) (b ¼ 0.214; p < 0.001). These findings provide empirical support for hypothesis H3. No other relationships among the four factors under research were statistically significant.
According to the MER-model (Ja. Belak, 2010;Ja. Belak et al., 2014;Duh, 2015;Duh & Belak, 2014;based on Ulrich, 1990), the method of values analysis consists of two phases. In the first phase, the values of all enterprise's key stakeholders must be identified. It is highly significant that the owners and also top managers (in larger enterprises) state their values. The theoretical foundations of the above and similar researches have been incorporated in the indicators, which are included into scheme for the study of enterprise values and their impact on enterprise policy (Section 3). It is necessary to point out that some of the indicators in our model were adopted with the methodology of the dialectical systems theory ; see also Malik, 2011). Based on the aims of our research, the following indicators were selected: Compliance with environmental objectives: this indicator was chosen to assure the 'triple bottom line' (3BL) also known as 3P requirements (people, planet, profit) (Hacking & Guthrie, 2008;Strukelj & Suligoj, 2014). Behaviour in the enterprise and external behaviour: these two indicators are important because they address internal and external relationships (Duh, 2015). Short-term goals and long-term goals: according to the MER-model, these two indicators are important because short-term planning is needed for planning of the basic-realisation process and long-term planning is key for the development of the enterprise (Ja. Belak, 2010).
All indicators included in the measurement model constitute the factor 'values of enterprise's key stakeholders' (V1) (see Table 2, construct V1). When examining the measurement model, we have eliminated the indicator 'profit for dividends'. This indicator is included in the MER-model but our research did not confirm this indicator.
According to the MER-model, in the second phase of values' analysis, the enterprise's key stakeholders discuss their values in order to identify the reasons why enterprise's key stakeholders developed their values in the way they did, why their extreme values (i.e., limits, maximum or minimum values)if they have themarose and why they are (not so) good for the enterprise. In this second phase, all differences in the value systems of enterprise's key stakeholders should be communicated and coordinated so that the enterprise can form a harmonised values system (i.e., value system of this enterprise). According to our research, this part of values analysis comprises three phases. First, the 'extremes of the values of enterprise's key stakeholders' (V3) encourage the 'analysis of the values of enterprise's key stakeholders' (V2). Second, this raises the awareness regarding the 'importance of the examination of the values of enterprise's key stakeholders' (V4). Thirdly, all of this is reflected in the values of enterprise's key stakeholders themselves, which were defined in this research as the factor 'values of enterprise's key stakeholders' (V1).
Our main research hypothesis was that values affect enterprise policy innovation (EPI). We can confirm this hypothesis since the factor 'values of enterprise's key stakeholders' (V1) strongly positively influences the factor EPI (b ¼ 0.376; p < 0.001) and is very significant in determining this factor (EPI), explaining 14.6% of its variance. Our findings suggest that the chain of value factors identified in our study (V3 ! V2 ! V4 ! V1) statistically significantly influences EPI. However, no statistically significant relationship exists among other factors of values (V2, V3 and V4) and the factor EPI. Our research results are consistent with the MER-model since a 'coherent system of values of the enterprise represents a limiting framework of enterprise policy and of the process of its concept idea, design and selection' (Duh, 2015, p. 122).
Further influence happens through enterprise's management and through the basic-realisation process. Therefore, enterprises should be aware of the whole chain of values that have such a significant influence on their business (results).
Considering the number of the enterprises involved in the survey (734), values clearly demonstrate a significant impact on EPI with 14.6% variance explained. Therefore, the enterprises should monitor their values and the values of their key stakeholders very carefully. Namely, as our research has established, values have a significant impact on EPI. The MER-model proves that enterprise policy has a direct impact on strategic management, and an indirect effect on basic-realisation processes through operative management (Ja. Belak, 2010;Ja. Belak & Duh, 2012;Ja. Belak et al., 2014;Duh, 2015). For more detailed interventions, further research such as indepth interviews of enterprise's key stakeholders is needed. In this way, researchers could discover many other in-depth viewpoints that are also important for understanding the impact of enterprise values on enterprise policy which statistical analysis does not disclose.

Conclusions and recommendations for future research
Values are internalised criteria for the evaluation of what is right and what is wrong (Poto can & Mulej, 2007). Therefore, values may be used for judging the correctness of human behaviour, which can be socially responsible or not (Mulej et al., ed., 2013(Mulej et al., ed., , 2015. Personal values are transferred to the enterprise's values, culture, ethics and norms (and vice versa) and therefore influence enterprises' development and performance (Duh et al., 2010;Me sko Stok et al., 2009). These facts further emphasise the importance of the theoretical model developed in this paper, which in practice can serve as a basis for the realisation of the greater enforcement of (or the analysis of) the values of enterprise's key stakeholders.
The aim of our research was to identify if values influence enterprise policy innovation (EPI). Four factors of values were identified, i.e., V1, V2, V3, and V4. Based on these four factors, we set our research hypotheses (H1-H4), which we confirmed with an empirical study. The research findings of this study show that enterprises should pay close attention to the chain of value factors we identified (V3 ! V2 ! V4 ! V1) because factors 'extremes of the values of enterprise's key stakeholders' (V3), 'analysis of the values of enterprise's key stakeholders' (V2) and 'importance of the examination of the values of enterprise's key stakeholders' (V4) impact enterprise policy innovation (EPI) indirectly through the factor 'values of enterprise's key stakeholders' (V1), which in turn directly impacts the factor EPI. We thus statistically confirmed our main research hypothesis that values affect EPI, too.
The conducted survey shows that, from the viewpoint of the established average values found, the owners/top managers of the surveyed enterprises are not aware of the importance of the enterprise's orientation towards all enterprise's key stakeholders as well as of the importance of analysing and examining o the values of enterprise's key stakeholders (Table 2). That is, indicator V257 (1.661), which determines whether the surveyed enterprises compare the value systems of each enterprise's key stakeholder and their aspirations, expectations and more important interests, and indicator V258 (1.700), which explains whether the surveyed enterprises analyse what leads to differences of the identified values of enterprise's key stakeholders, have the lowest average value However, the surveyed enterprises are highly focused on sales markets; for example, they place greater importance on the analysis of sales growth (V241; average value 4.774) and on the quality of products/services sold (V242; average value 4.640), which are all important values of the enterprises studied. Based on the conducted research, we find that enterprises need to raise their awareness about the importance of focusing on the values of the enterprise. This can be achieved either directly or indirectly through the requirements of the sales market and through the inclusion of appropriate values as one of the required quality factors of products/services sold. This is also confirmed by the fact that the conducted survey shows ( Table  2) that in the construct 'enterprise policy innovation (EPI), all indicators EPI was measured with have an average value of over 4, meaning that the respondents consider all of them as quite important. Indicator V496 (4.117), which shows how important the surveyed socio-political and other social environment for EPI appears, and indicator V495 (4. 293), which examines the importance of natural environment for EPI, have the lowest average values. Indicator V500 (5.298), which explains the importance of the sales markets for EPI, and indicator V509 (5.108), which explains the importance of the analysis of the enterprise and the resulting strengths/weaknesses for EPI, have the highest average values. Our data also show that the surveyed enterprises do not attribute such great importance to values, culture, social responsibility or ethics (indicator V511, average value 4.800), as they do to the sales market (V500) or to the question of how to find the way to prepare the enterprise to a situation which will bring mainly strengths and not weaknesses (V509). Thus, on the basis of these findings, we can conclude that the requirements for appropriate values, culture, ethics and social responsibility will have to be expressed/reflected in the sales markets of enterprises; enterprises which do not take values, culture, ethics and social responsibility into account, will be less competitive on the market. To avoid this, owners/governors and managers of enterprises should focus on these success factors.
The developed model presented here can serve as a basis for further theoretical research as well. As mentioned before, our research shows the results from Slovenia as the EU transition economy. A similar research based on the model presented here should be conducted in other EU transition and non-transition economies. The findings from a comparative analysis of such studies would help those concerned draw conclusions and make recommendations concerning future EU enterprise governance,