The Relationship between Corporate Performance and Ownership Structure: Evidence from Turkey

This study investigates the effects of ownership structure on the performance of the listed companies in BorsaIstanbul Stock Exchange 30 Firms (BIST 30). The main hypothesis of our study is that there is a significant relationship between companies' performances and their ownership structures. The statistical population includes 19 non-financial companies in the period of years between 2008 and 2013. The results show that, the concentration of the large shares of companies one or a few shareholders has a negative effect on related firm’s performance.


I. Introduction
In finance literature, it is known that there are argues that there is a relationship between corporate performance and ownership structure of the firms, however institutional and economic influences differ from both nationwide and internationally. The major objective of this study is to provide an empirical evidence to respond to the argument that, ownership structures differ according to companies' financial performances.
There are generally two basic types of ownership structure in Turkish economic environment, private and publicly ownership. When looking private sector, there is a tendency to foreign weighted ownership structure from domestic one in the last decade. In this study, we have taken into account the concentration of the shares in one hand or a group. In other words, the performance differences between sub samples such as private and public or domestic and foreign owner/s, and etc. weighted firms are not tested separately, because it is subject to another research in the literature.
The structure of the paper is as follows: After the Introduction, Section 2 is about to explain theoretical foundation and reviews the ownership structures and their significance in the corporate governance. Section 3 presents the research method, including hypotheses, research variables, model buildings, population and sample selection, data collection method, and statistical test. The findings of the study and hypotheses and discussion of the article is reported in section 4. Finally, 5 th Section has concluding remarks of the study.

II. Literature Review
According to Craswell at al. (1997), the impact of the ownership structure on corporate performance occurs in at least three ways. First, there are studies which test for ex post performance effects. Demsetz and Lehn (1985) do not find evidence of a linear relationship between three measures of ownership concentration and measures of profitability. 1 Morck et al. (1988)   Research Hypotheses: The Main Hypothesis is "there is a significant relationship between ownership structure and the performances of the BIST 30 companies".
So "There is a significant relationship between the size of the largest share of the firm equity and the firm performance" Research Variables: We examine the effects of ownership structure on firm value among listed Turkish non-financial BIST 30 companies. Data used for this analysis come from two sources. First, the ownership structures were taken from annual reports of related firms. The other sources of data are based on the financial ratios of firms announced to public.
The dependent variable of this study is "the value of the firm" which is represented by the performances of the companies. The performance was measured by Tobin's Q = (Market Value / Book Value).
In order to assess the relationship between corporate performance and ownership structure, as a measure of performance, Craswell at al. (1997) indicated that US researchers have utilized Tobin's Q (=Market Value / Book Value).
The statistical data could be managed via three ways: cross sectional, time series and panel data approach. With the panel data method researchers can do cross sectional observations within different time periods. In this study, the panel method (Brooks, 2008) was exerted. By employing panel data, a group of data which cover a great number of cross sectional variables (N) that is obtained during a time period (T) is collected. The applicability of this technique is limited. It can be employed only when the number of time series observations, T, per cross sectional unit, i, is at least as large as the total number of such unit, N.
The simplest types of fixed effects models allow the intercept in the regression model to differ cross-sectionally but not over time.
We can think of μ i as encapsulating all of the variables that affect Y i t cross-sectionally but do not vary over time-for example, the sector that a firm operates in, a person's gender, or the country where a bank had its headquarters, and etc. This model could be estimated using dummy variables, which would be termed the least squares dummy variable (LSDV) approach (Brooks, 2008).
The fixed effects approach is a sensible one, given the data analyzed here, since there is an unusually large number of years compared with the number of firms (19), resulting in a total of (114 =19 firm x 6 years) firm years observation.
The data employed in the study are obtained from firms' annual reports and BIST 30 Index statistics.
The analysis is conducted for the whole sample period of 2006-2013.
In order to control the effects of extraneous variables on the performance of the companies, seven control variables were also selected as follows: For testing these hypotheses, t test was employed. In this test (with 95 percent probability) if we couldn't reject H o , it means that the considered coefficient isn't significant and its rejection means the opposite.

IV. Empirical Results
The descriptive statistics of the variables and the results of the normality test are given in  Table 2 reports that ownership coefficient on Tobin's Q is negative and significant at 5 percent level (P<0.05). 1 percent increase in the size of the largest share of the firm equity will decrease 0.044871 in Tobin's Q.