Managerial Property, Strategic Resources and Diversification of Enterprises, Case of Tunisian Businesses Rated

The question of the relationship between shareholder structure and diversification of the enterprise has caused many debates. We propose to shed a particular light in applying this question in the context of companies listed on the BVMT, which have specific characteristics in terms of relations between shareholders and leaders. Our study focuses on 51 companies. We are studying the relations between character of the company level and type of diversification. Our results show that family businesses are more diverse than the non-family ones. In addition, we identify a negative relationship between the presence of an actionnarial block and level of diversification for the non-family businesses. In contrast, this relationship is not found for the family businesses. Our results allow as well to emphasize the special features of family businesses. However, our results also show that these features do not lead to a difference between family businesses and non-family businesses in terms of the choice of type of diversification (connected or not connected). This article aims to analyze the impact of the managerial property and strategic resources on diversification in the Tunisian companies listed on the stock exchange of movable values of Tunis (BVMT). The data have been collected from a sample of 51 Tunisian enterprises holding data on four successive years. The results obtained show the existence of an impact statistically significant of the managerial property on the strategy of diversification of the undertaking. These results have also shown that the non redeployable resources positively affect the connected diversification. Indexing terms/


Introduction
The shareholder structure is one of the mechanisms of governance that has an impact on the creation of value of such a firm. Their influence on the financial performance of the company has been the subject of several empirical studies as well as other further researches. However, the study of the impact of such structure on the strategic decisions taken by leaders has been less analyzed.
Note that this strategy of diversification has been stated since the end of the fifties by Ansoff (1957). However, the expansion of the literature is triggered only after the work of Rumelt (1974). In his research, Rumelt was the first to highlight the relationship between diversification and performance. According to him, diversification is neither an objective nor a plan; each firm which diversifies must study: the type of business that it intends to undertake, its dependence of the forces and existing skills, the need to develop new activities and finally the degree of appropriate diversity. As well, diversification strategies continue to be put in place in the area of banking. Banks choose this concept in order to expand their range of products and services to meet the financial needs of the clientele. Two distinct disciplines are interested in this concept: the economy and the strategy. Economists tend to see the firms such as players of a game with several actors.
The interest is related to the game and its results and not the performance of individual firms. They are interested, as well, in the relationship between diversification and the sectoral concentration on the one hand and the behavior of prices and the power on the market, of the other hand. Researchers in strategic management have concentrated, on their side, on the strategies of the firms, and particularly on the human and physical assets and their relations with the environment. Rumelt (1974) asserts that the degree and the type of link between the segments of a firm determines its performance.
Contrary to the disciplinary perspective, the cognitive perspective adopts a favorable vision to diversifying. They can be seen as a means of creating value, including, whether the company has the necessary resources that allow it to cope with the uncertainty and complexity generated by this type of strategies.
Given that the work in Corporate Governance Under this angle of analysis are rare, to our knowledge, other fields than finance, such as the strategy, can contribute to a better understanding of this phenomenon. In addition, researches combining several theoretical perceptual appear to be lacking, while a " multithéories " approach of corporate governance is necessary to better understand the impact of governance on the value creation process (Daily & al., 2003). Nothing, all the same, , who explains the strategies of diversification by three different theoretical frameworks : the theory of the Agency, the " resource-based view " and the forces of the market.
With regard to the previous reflections, the main objective of this paper is to achieve two summaries of the literature leading to the construction of a theoretical model. The latter is designed to help explain the role of the system of corporate governance and its mechanisms in the implementation of strategies of diversifying. As well, we shed light on the debate of implementation of strategies of diversification in the enterprises (and precisely the Tunisian businesses) and improve our knowledge of the links system of corporate governancediversifications and mechanism of corporate governance-diversifications.
These analyses present a number of shortcomings; future research have been developed in order to partially fill them, and complete the series of previous proposals, as a second element of the model. The previous analyses have highlighted some of the shortcomings of the past studies, formulate perspectives of research to fill it in. These research perspectives combined with proposals that enable the development of a new conceptual model. From a sample of 51 Tunisian businesses between 2008 and 2011, we study the determinants of diversification of Tunisian businesses listed on the BVMT in US pressing on a theoretical framework and combining the approach by the resources and the theory of the agency. Our work will be organized as follows : In section I, we conduct a brief review of the literature about the determinants of diversification, as well as the measures of diversification and we propose a number of assumptions. In section II, we present the methodology chosen for empirically analyze the determinants of diversification and we interpret the obtained results. Then, we conclude.
We will try to answer an important question that arises in the field of finance, namely the relationship between diversification and the other two key concepts which are always linked: the performance and risks. To do this, we will present the results of some earlier researches that have treated these relations. In effect, the objective of this research is to study the impact of the structure of the shareholdings and strategic resources on the decisions of diversification taken by Tunisian companies.

A summary of Theoretical Literature
Our work is going to be divided into two parts. The first part is devoted to the presentation of the theories, as well as the development of our thesis. The second part will develop our hypotheses of research and will endeavor to verify it empirically. Subsequently, a synthesis of the main results of our different studies will be presented in the conclusion.

The relationship between the managerial property and diversification
In the case where the leaders hold a significant share of capital, they can use their power to choose the diversification of activities in the company. But, when the share of the managerial ownership is increasing at a very high level, the leaders can decide arbitrarily the diversification. However, they must be attentive to the risk created by a disproportionate diversification, because its investments will have adverse effects on the financial situation of the company in case of failure of diversification. In the contrary case, when the leaders hold a significant share of the capital of the company, they can adopt an aggressive strategy of diversification for increasing their private profits to the detriment of the other shareholders. Before applying the financial activity, this concept has been used in the theory of portfolio.
It has been defined as the distribution of investments on different types of investments, that is to say the fact to diversify its portfolio (of assets, activities, sectors, etc.). This strategy has been, moreover, very debated in literature. Several studies dealt with its role in the context of exhibitioner, by analyzing the portfolios of shares (studies of Markowitz). Thus diversification of a portfolio of securities allows you. To reduce the total risk of a portfolio of actions for a determined rate of yield, or to improve the performance of a given level of risk. Portfolio theory suggests that diversification allows you to reduce the risk through the imperfect correlation of the different markets and different regions. Shapiro (1978), Boot and Schmeits (2000) have made the same conclusion by asserting that volatility, probability of bankruptcy and risk decreased by resorting to a strategy of diversification. Beston (1994) declares on his side, that risks tend to decline when the banks are authorized to carry out new activities such as insurance, operations on the financial products… in addition to traditional activities. However, by diversifying, the complexity of banking institutions is accentuated and some risks are intensified.
According to the approach developed by Jensen and Meckling (1976), among the major problems of diversified firms are the problems related to the theory of agency. The diversified banks are much more exposed to the risks induced by the agency theory than the specialised banks. Fauver & al (2004) have identified, on their side, the risk of exchange rate and the political risk as risks that could affect diverse societies. This is particularly true for the financial institutions whose activity is regarded as one of the most regulated and where the banks can be affected by a change of laws and regulations in the countries where they operate. In addition, the relationship between diversification and the performance has also been studied. Several studies argue that there is a strong positive relationship between diversification and performance of firms (Chandler (1997) and Teece (1980). Jensen and Ruback (1983) have argued that diversification allows banks to explore new markets, to seize the best opportunities for investment and create a synergy between the different regions and sectors, thus increasing the value of the bank and its profitability.
Finally, Morck and Yeung (1991) have found that the international diversification positively affects the value of the firm. However, other studies reveal that diversification has a negative impact on the performance. Thomas (1999), Doukas & al (2001 and Liu and Qi (2001) argue that the organizational complexity of diversified firms leads to the increase of the costs of agency, which is translated into a reduction of the performance of businesses. According to Grant (1987), Sambharaya (1996), Qiand and Li (2002), international diversification can have negative consequences on the performance of firms because of the asymmetry of information, the lack of surveillance, the strong delegation and decentralization that they entail.
As stressed at the beginning, diversification is not the only solution to be explored by the regulatory authorities. The triggering or exasperation factors are only the expression of profound structural dysfunctions whose origin is a behavioral deficit and bad systems of governance at all levels of decision-making of a globalized and deregulated system. In this context, Amihud and lev (1981) have studied the relationship between the structure of property, diversification and performance of enterprises. They have noticed that the firms whose leaders are less controlled adopt diversification strategies that are non beneficial to the interests of the shareholders. Therefore, we expect a relationship between the managerial property and diversification. The Hypothesis 1 is presented in the following manner: Hypothesis 1 : The Managerial property has an impact on the strategy of diversification.

The relationship between the strategic resources and diversification
In the approach by resources, as we have discussed previously, the company adopts a strategy in order to use non-employees resources with low costs of transactions. The study of  demonstrates that the criteria of flexibility of resources determine the mode of development of the enterprise. There are generally two types of resources to know: the redeployable resources. Such as the current assets; and the indebtedness that has a positive impact on Related and non Related Diversification. This has been confirmed by the Studies of  and Lee (2005). Concerning the non redeployable resources; such as tangible capital assets, they influence only connected Diversification,  has shown the impact of strategic resources on diversification of Korean companies. He says that the capital asset cannot be assigned in other activities. Therefore, it affects the DR (redeployable resources) rather than DU (non redeployable resources). This idea was supported by Lee (2005). We expect that the non redeployable resources will have a positive impact on the related diversification, while the redeployable resources will have a positive impact on the related diversification and non connected diversification. Where the assumptions: Hypothesis 2 : The non redeployable resources have a positive impact on connected Diversification. Hypothesis 3 : The redeployable resources have a positive impact, on the related and non connected diversification.

Conceptual Framework
Based on the arguments presented above, the conceptual model proposed in this study is presented in Figure 1. Figure 1 Conceptual Model of study Before the put in place of the assumptions, we will present the methodology for the estimation of the variables and we will conduct empirical analysis.

Research Methodology
In the framework of this section, we will explain first of all, the choice of the sample. In the second place, we will present the dependent and independent variable measures of the study. In last place, we will analyze the model specification of performance.

Sample and Data Collection
In order to observe the impact of the managerial property and strategic resources on diversification of Tunisian businesses; we have used econometric techniques of estimation of panel data. Our study has been carried out on a basis of a sample of 51 Tunisian companies listed on the stock exchange of Tunis. The data relating to this sample were collected from the financial reports of these companies. The study is spread over a period of 4 years from 2008 to 2011.

Measurements of the variables
There are several methods to measure the importance of diversification: the number of industries in which the company is present, the ratio of specialization, the index of Berry-Herfindahl, the entropy measure, the index of Utton and the classification of Rumelt. The measure of diversification must take into account the number of industries in which the company is present, the relative weight of an industry among the whole of the industries in which the company undertakes, and the relationship between these industries. Among these methods, the index of Berry-Herfindahl, the entropy measure and the classification of Rumelt are the most used.
The measure of diversification aims to highlight the relationship between performance and the strategy of diversification. The Berry-Herfindahl index and the classification managerial ownership strategic resources ownership structure debt of the company Business performance of Rumelt have advantages and disadvantages. The index of Berry-Herfindahl is simple to calculate and objective, but it does not distinguish between Related and non Related diversification. It does not allow the establishment of an exact ratio between diversification and performance. On the other hand, the classification of Rumelt captures the subtlety of the diversification strategy, but it is subjective and complex to measure.
The entropy index, proposed by Jacquemin and Berry, accumulates the merits of these two methods. The entropy measure is constituted by the following three elements: (1) The number of industries in which the firm is involved; (2) The distribution of the total turnover of the firm between these industries; (3) the intensity of the relationship between these industries (Palepu, 1985).
It enriches the index of Berry-Herfindahl by measuring the report between the activities in which the firm is committed and it also simplifies the method of Rumelt, which distinguishes in too much details the strategies of diversification. The main property of this index is to allow the decomposition of the total diversification into two components inter and intra-class, when we can have data at two levels of nomenclatures (industry and the sector). Nonconnected Diversification quantifies the distribution of activities of the firm between several non-linked sectors, and related measure diversification the distribution of activities of a firm between the industries in the same sector.
The objective of our empirical investigation is to study the relationship between the types of diversification, on one part, and the managerial Property and strategic resources, on the other part, (see, Joon Tae Lee, 2007). To test these interactions, we consider the following model: With,


The total diversification measured as we have seen previously by the entropy The related diversification measured by the following formula: The non-connected Diversification measured ) 1 (


: the tangible capital asset of the company i in year t measured by the report between the tangible capital assets and the total assets of the company.


: The rate of indebtedness of the company i in year t measured by the ratio between the total debt and the total assets.


: the circulating assets of the company i in year t measured by the report between the circulating assets and the total active.


: The size of the company i in year t measured by ln (assets). We used the logarithm to avoid the problems of heteroscedasticity.


: represents a constant term of the model,  Ε i,t : is an error term while are unknown parameters to be estimated.

The measures of the dependent variables  The total diversification
We retain the total diversification as a representative variable of diversification of the Tunisian company side. This measure has been retained by Jun ( ,1997 in the framework of its studies about the impact of strategic resources on diversification of the Korean company. In our study, we retain the total diversification (DT) as a representative variable of diversification of the business. In fact, total diversification is calculated as follows:

 The connected Diversification
We retain the related diversification as a variable representative of diversification of the Tunisian company side. This measure has been retained by Jun (1997), Hong andWhang(1997) in the framework of their studies concerning the impact of the strategic resources on diversification of the Korean company. In our study, we used the variable (DR) as a representative variable of the related diversification of the business. In effect, this index is calculated as:

Non-connected Diversification
We retain the non-connected diversification as a representative variable of diversification of the Tunisian company side. This measure has been retained by Jun (1996), Hong andWhang (1997) in the framework of their studies concerning the impact of strategic resources on the diversification of the Korean company. In our study, we used the variable (DU) as a representative variable of the non-related diversification of the company. This index is calculated as follows: This variable is measured, as we have seen previously, by the part of the capital held by the leaders. This variable has been used by Denis & al (1997) who have studied the impact of the Managerial property on diversification in the United States.  KDIR : The share of the capital held by the leaders  The representative variables of resources : There are two types of resources: the redeployable resources and non-redeployable ones. We retain in our study the part of tangible capital assets such as representative variable of non-redeployable resources.
In this framework, , who has used this variable in his study concerning the relationship between the strategic resources and diversification of Korean companies, indicates that the capital asset cannot be assigned in other activities. Therefore, it affects the DR rather than DU. In other words, it allows the company to diversify in sectors related to the activity of the firm. This variable is measured under the following formula:  IMC: Tangible Capital Assets / Total Assets Note that the debt and the assets are retained as distinctive variables of redeployable resources.
 Debt : We retain the rate of indebtedness as a representative variable of redeployable resources. This measure has been used by Hong and Whang (1997) in their study about the impact of strategic resources on diversification of Korean companies. In this study, we used the rate of indebtedness as a representative variable of redeployable resources, in order to analyze the effect of these resources on diversification of the Tunisian rated company. We calculate the report between the total debt and the total active.  Circulating assets : We retain the circulating assets as a representative variable of redeployable resources. This measure has been used by  and Lee (2005) in the framework of their research about the impact of strategic resources on diversification of Korean companies. In this study, we retain the percentage of the circulating assets in the total assets as representative variable of redeployable resources, in order to analyze the effect of these resources on diversification of the Tunisian rated company. We calculate the report between the assets and the total assets of the company.

Control Variable
 We consider the size (in terms of total assets) of the company as a control variable. This variable is measured by the logarithm of the total assets (Gedajlovic andShapiro, 1998 and2002).

Application and results
We analyze in this section the impact of the structure of ownership and the strategic resources on diversification of Tunisian businesses. It is more specifically about the managerial property, redeployable and non-redeployable resources. The methodology adopted consists of analyzing the influence of the managerial property and strategic resources on diversification of the company measured by the index of entropy.
The analysis of the presence of a problem of multi-collinearity between the explanatory variables shows that the correlation coefficients between the independent variables taken two by two are not statistically significant with a threshold of 10%. The values taken by these correlation coefficients have not exceeded a maximum of 0.4. This value is less than the fixed threshold by Gujarati (2003) beyond which we speak about a problem of multi-collinearity.
As well, we can confirm that the multi-collinearity between the variables poses no problem in the framework of this research. We present the results of regressions in the following table :  In what follows, we will interpret the results in function of the explanatory variables. -

The impact of the Managerial property on diversification
The share of the capital held by the leaders of the company plays an important role in the decision concerning the diversification strategy adopted by the company. In effect, the leader will consider each attempt of diversification of activities before embarking in this adventure. In this case, it is likely to say that the leaders who have a low share of the capital decide to diversify the activity in the sectors, which are not related to the main activity of the company to serve their own interests. In these conditions, the DR (connected Diversification) will decrease and the DU (non-connected diversification) will increase. But beyond a certain level of managerial property, the situation is reversed. In effect, when the leaders increase their shares in the capital of the company, they will, therefore, adopt a reasonable and prudent diversification strategy. The DR will consolidate and the DU will reduce. This will alleviate the problem of agency and conflicts of interest between the leaders and the shareholders of the company, since there will be convergence of interests. The results of our estimate affirm the hypothesis (H1) relating to diversification of the business. In effect, the managerial property has an impact on diversification strategy this corroborates with the studies of Amihud andlev (1981,1999) who showed that the leaders tend to engage the company in diversification strategies and mergers conglomerate mergers in the aim to diversify their activity and to reduce their risk. Then the hypothesis (H1) relative to the managerial property; according to which the managerial property has an impact on the strategy of diversification; is accepted according to our research.
-The approach by Resources  The impact of the non-redeployable resources on diversification Our results regarding the analysis of the approach by the resources show that the variable "BMI" negatively affects the DU and positively the DR. We have formulated the hypothesis (H2) according to which the non-redeployable resources positively affect the DR.
This confirms the studies of  and Lee (2005) who found a positive influence and statistically significant relationship between the tangible capital assets and the related diversification. This study was carried out on Korean firms. Then, the hypothesis (H2) according to which the non-redeployable resources positively affect the DR is accepted according to our research.
 The impact of the redeployable resources on diversification We have formulated the hypothesis according to which the redeployable resources (Assets, debts) would influence positively both the DR and the DU. The debts incurred by the rated Tunisian company could be used for the growth of the society in different sectors. However, according to our results, the debts negatively only the affect DR and has no effect on the DU. This result confirms, in part, the studies of Hong and Whang (1997) who have not found a relationship between the debts and the DU. Therefore, one can judge that the debts have a negative impact on connected Diversification. While the circulating assets negatively affects the threshold of 10% the DU and has no influence on the DR. This corroborates in part the results of Lee (2005) who found a negative impact between the assets and the diversification into non related (OF). Thus, the hypothesis (H3) according to which the redeployable resources has a positively impact both on the related non-connected diversification is rejected according to our research.
The size of the company This variable is considered as an indicator of economies of scale. If these savings exist in the activities of the business, we have to expect a positive relationship between diversification and the size of the company. The variable of the control size only affects the DU; this influence is negative.
Although, the three studies are not directly comparable, we bring our results to those of Claessens & al. (1999) andDenis &al. (1997) (Table 6). Indeed, the comparison can be useful for understanding the structure of relationship of ownership-diversification in different countries.

Discussion and Conclusions
We have tried in this research to study the impact of the managerial property and strategic resources on diversification of Tunisian businesses through the entropy index. We based our study on a sample of 51 Tunisian businesses listed on the BVMT. Our estimate shows that the share of capital held by the leaders has a negative impact on the strategy of diversification. This confirms the analysis of Galbraith (2004) : "The Myths of the authority of the investor, the shareholder administrator, the ritual meetings of the Board of Directors and the annual general assembly of the shareholders continue. But for any observer of the large corporation modern anonymous. The reality is unavoidable. The power on the company belongs to the management team and bureaucracy which controls its stain and its remuneration" Similarly, our regression shows that the hypothesis (H2) according to which the non-redeployable resources positively affect the DR is confirmed.
The question of the link between structure of ownership of shares and the strategy of diversification has given rise to many jobs tasks, yet few have been devoted to the influence of the family character of the company while the latter leads to a particular feature due to the nesting between family and business. Here, we have proposed to shed a new light on this link between diversification and ownership structure, working in the particular context of companies listed on the Stock Exchange. Our results allow to highlight several characteristics.
First, it appears a significant and positive relationship between the level of diversification and the character of the company: family businesses are on average more diverse than the non-family businesses (Berger and Ofek, 1996;Dennis & al.,1997). the obtained results in the two sub-samples are opposed, putting also highlight on the particularity of family businesses. In effect, for the latter, the variables characterizing the shareholding (share of the main shareholder, the presence of a majority) do not allow to explain the level of diversification. In contrast, in the sample of non family, the level of diversification is significantly and negatively influenced by the part of the principal shareholder and by the presence of a majority. Finally, the type of diversification (related or not) is not linked to the character of the company. These results confirm the interest to continue the work on the Tunisian businesses, and are important features in terms of management. In the future, it is planned to continue this work by adding data on the risk, performance, and variables to refine the description of the enterprises (the identity of the officer, Generation...), in order to better understand and explain the reasons why their strategy of diversification is different from that of non listed ones. This is contrary to the results obtained in the previous research by Anderson and Reeb (2003b) and Gomez-Mejia & al, (2010). A first explanation of this difference may reflect the influence of the behavior regarding the risk. Anderson and Reeb (2003b) had assumed that the family businesses would be more diversified than the non-family members because they are seeking more to limit the risk. Their results do not confirm this hypothesis. In contrast, our results confirm it.
However, this explanation related to the risk has been widely minimized by Gomez-Mejia & al. (2010). They show that the relationship between the managers and shareholders in the context of risk depends strongly on the context of decision-making. In the framework of the Tunisian businesses, many factors related to the preservation of the socio-emotional value such as the desire to maintain the control, to protect the members of the administration, perpetuate conservative strategies, to avoid the use of the debt, and book to the members of board of directors the key positions within the company contribute to limit recourse to diversification. Therefore the desire to limit the risk cannot be a unique explanatory factor. In fact, the connected diversification appears to present less risk in terms of the managerial complexity. The existence of links between the activities in terms of resources and skills reduces the complexity, therefore the risk of failure and it also reduces the risk of againstfinancial performance (Markides and Williamson, 1994;Robins and Wiersema, 1995). Our results contradict our hypothesis.
This allows to extend the discussion related to the first hypothesis. It can be assumed that the risk aversion may not be the major factor of influence in the strategy of diversification. This contributes, as well, to put in before our other explanations of the differences between our results and those ofAnderson and Reeb (2003b) and Gomez Mejia & al, (2010). However, this interpretation of the influence of the aversion to risk is complex in the case of the analysis of the type of diversification. In fact, the connected diversification presents a low risk with regard to the managerial complexity. However, the non-connected diversification allows you to reduce the risk by allowing them to distribute on several different activities. As well, the lack of difference in terms of the type of diversification between businesses cannot truly conclude the relationship to the risk. On the other hand, it should be noted that if the specificities of firms influence the strategic choices in terms of the level of diversification, this is not the case as regards to the type of diversification.