Corporate Ownership Structure And Corporate Performance

1.Introduction

Corporate Governance is an issue of public importance of late. One aspect of the issue is the structure of the corporate ownership structure vis- a-vis corporate performance.

1.2 Background

Share-holder activism has evoked interest as to how shareholders are really being treated by the management of the companies. It depends on their ownership structure impacting on their performance in terms of profit and distribution of profits amongst shareholders. In 1990 a California based pension fund company known as California Public Employees Retirement Systems (CalPERS) started questioning the practice of those listed companies buying back shares from the share holders at higher prices led to the draining of companies’ capital and reduction in value of their shares. Contemporary companies with widely dispersed shareholders also followed suit by questioning such unethical practice.

1.3. Purpose

The purpose of this research is to empirically examine if there is a relationship between ownership structure and corporate performance among companies listed on Johannesburg Stock Exchange (JSE). We will explore these relationships by comparing respective variables against each other using regression models. In addition, we propose to look into company specific factors in order to evaluate their impact on performance

1.4 Problem Statement

South Africa is also seized of the issue of corporate ownership especially after the lifting of Apartheid   The exploitation the small and widely dispersed investors by the companies management prevalent in South Africa needs to be studied. This worldwide phenomenon. of the principal-agent theory and stake-holder theory needs to be briefly discussed. Berle and Means (1932)[i]  were the first to discuss the development of agency theory. Adam Smith in his book “The wealth of Nations” (1776) had already identified the agency problem by postulating that company directors would not take much care of other people’s money as their own , as observed by Letza, Sun and Kirkbride (2004)[ii].

The present state of affairs only reconfirms to reflect that it is truism to say that the share holders own the company. . Principal-agency relationship is a problem because the agents may not always act in the best interest of the principals. Therefore decisions taken by the agents are not always consistent with principals’ welfare.  The stake-holder theory is in stark contrast to the principal-agent theory which mainly focuses on the interests of shareholders.  The stake-holders theory suggests that if managers (agents) are over concerned with share holders, the tendency is to concentrate on short-term profits rather than long term interests. If the share holding is on long-term basis this problem will not arise. Corporate Governance is a broader concept in that ‘what is optimal for shareholder often is not optimal for the rest of the society.’ (Blair 1995:13)[iii].

1.5 Research Questions

This study intends to explore the relationship between ownership structure and corporate performance. The study can be said to achieve its purpose if it can provide answers to the following:

Who actually control listed companies in South Africa?

Are the ownership structures of South African companies especially large ones widely dispersed or concentrated?

How does ownership structure affect corporate performance?

What are the actual control and ownership rights of state of South Africa in listed companies of South Africa.?

What are the actual control and ownership rights of directors and management in South African listed companies and their effect on corporate performance?

What effect do corporate specific factors, play in ownership structure and corporate performance?

1.6 The Rationale-Why the research is important

Economic prosperity of a country depends on how fairly and widely wealth is distributed amongst its people. In the present era of globalization, though the world has become small, the players have become large by virtue of mergers and acquisitions aimed at achieving competitive advantage which includes large scale economies. Hence while large scale operations are unstoppable, interests of shareholders should not be lost in the process. The shareholder activism will be meaningful only if the decision makers in activists group are informed of the realities that whether or not performance of companies is synonymous with the concentration of ownership. This research’s findings should stimulate debate whether concentration of ownership is desirable for the sake of better performance or better performance is possible even without concentration of ownership. And whether policy makers can really do any thing about whatever be the phenomenon is.

2 Literature Review
Our initial expectations of these relationships, are that companies with widely held/dispersed ownership will be associated with poor corporate performance and companies with significant holding by majority shareholders will reflect good performance as there will be expropriation of minority shareholders. Companies with major shareholders as insiders will also reflect good performance as there will be minimal agency problems. On the other hand, state ownership is negatively related to corporate performance as state often tends to pursue its political and social agendas as a priority at the disadvantage of profit and value maximization.

Post-apartheid South Africa being an infant country, data available from the relatively shorter period of corporate history may not give easy insights to the corporate ownership and structure. Hence this study will enquire into the situations prevailing in the rest of the world. For the purpose of this proposal a sample observation of two countries alone is given.

2.1 Sweden

Sweden: In the paper entitled Agency conflicts, ownership concentration, and legal share holder protection [iv] published on line 17 March 2005, it has been deduced that share holder protection measures prevent diversion of corporate resources by managers in collusion with large shareholders.

Swedish industry is dominated with few large multinational corporations like ABB, Ericsson, Electrolux, Stora Enso, Volvo etc. The fifth generation of Wallenberg Dynasty dating back to middle of 19^th century is said to be active owners of these industries.

Findings in the study conducted by Steen Thomsen, Torben Pedersen, Hans Kurt Kvist [v]are evident of the conflict of interests between block holders and minority investors. In continental Europe. The percentage of block holders’ ownership can be said to be very high from minority share holder value view point.

2.2 U.S.A.

In U.S.A. : Adams, Santos [vi] in their paper [vii] “Identifying the effect of managerial control on firm performance” have concluded that managerial control by share ownership has positive effects on performance over at least some range, contrary to the belief that managerial control is purely detrimental, in certain Banks in the U.S.A.

Yet another study of U.S. authors Demsetz and Villalonga [viii](September 2001) have found that statistically there is no significant relation between ownership structure and firm performance. “This finding is consistent with the view that diffuses ownership, while it may exacerbate some agency problems, also yields compensating advantages that generally offset such problems. Consequently, for data that reflect market-mediated ownership structures, no systematic relation between ownership structure and firm performance is to be expected.”(Demsetz, Villalonga)

2.3 History of Corporate Ownership in South Africa

Ayogu[ix] (2001) in his economic research paper on Corporate Governance in South Africa, states that in a few cases for which data is available for ownership structure and concentration in South African companies, many of the listed companies belonging to local and multinational investors,  ultimate ownership have not been reported. He says” It is certainly possible that in many of the countries, politicians have used nominee companies to stake their interests in corporations. Hence, the widespread incidence of “crony capitalism” and lackluster enforcement of the Rule of Law in many African nations.”

La Porta et al (1998) says that in South Africa where wealth is concentrated, Anglo alone controlled 60.1% of the listed firms in Johannesburg Stock Exchange as on 1987. According to Goldstein (2000) in 1991 top 5 share holders controlled 84.9% and in 1997 66.4% in South Africa.  Despite the acute concentration as a moral hazard, Corporate Governance in true spirit prevailed when NEDCOR made a hostile bid to take over STANBIC because of investor activism even though Old Mutual, the second largest institutional investor controlling NEDCOR intervened and threatened to offload STANBIC shares if the merger move was stalled. “Yet, the management fought undeterred. Going beyond a mere metaphor, this event may have taken the term, management entrenchment, to a new height.”(Ayogu 2001)

The six largest groups accounted for two thirds of market capitalization in 1983 accelerated by withdrawal of foreign investors as a sequel to Soweto riots.  Though by pyramidal structure ownership is diffused, control remains concentrated in the founding families of Oppenheimer, Rupert, Gordon, Mennel and Hersov vide table below. (Goldein 2000 page 10) (see Table 1in Appendix)

2.4 Black economic empowerment (bee)

Black Economic  Empowerment is another step forward post-apartheid as a result of which black  controlled ownership rose from 0.6% in 1995 to 9.3% in 1997(see table 2 in appendix  ) Black economic empowerment is unique to South Africa in that it is a declared objective of the post apartheid regime. From a mere 0.6% in 1995 when blacks were given power to rule, it has grown to 9.3% by 1997 in terms of blacks’ control in the listed companies of South Africa. Now in 2006, in retrospect let us see the broad-based economic empowerment of blacks in South Africa in the light of the Codes of Good Practices on Broad-Based Black Economic Empowerment published by the South African Department of Trade and Industry.[x]. Admittedly the systematic disempowerment of blacks under Apartheid, led to an economic lopsidedness obscuring the majority of country’s citizens. Still in the second decade of democracy economic disparities on racial lines are still in evidence.

As per the said code, a black person is either an African, coloured or   Indian South African who must be a citizen of the Republic of South Africa by birth or descent or must have been naturalized prior to coming of South Africa Act of 1993 or naturalized after the said Act if he had been denied of naturalization because of apartheid.  Because of this definition of black person subjected to some form of discrimination, ironically people of Chinese origin do not qualify to be black person even though they had been discriminated as Asians. There are such people of Chinese origin in the 4^th or 5^th generation living in South Africa. This code prescribes mandatory ownership percentages in terms of voting rights or cash flow rights for black people in listed companies among other items, As on 2005, about 185 companies haven found to be complying with BEE codes for black

2.5 Definition of Variables

2.5.1 Return on Assets
This ratio is an important indicator of how efficiently the company turns assets at its disposal into earnings. It tells an investor how much profit a company generated for each R1 in assets. Viewed differently it also paint a good picture to reflect the asset intensity of a business. There is a general agreement amongst researchers and financial practitioners that a ratio of below 5% should be regarded as more-asset-intensive and anything above 20% as less assets-intensive. In calculating this ratio, caution is always made that companies with seasonal periods usually experience a fluctuating ROA and this should be smoothed out by using the average total assets as compared to using total fixed assets at year end.Though database derived, data for ROA is calculated by taking Net Income for the year and diving that by Average Total Assets for the year.

2.5.2 Return on Equity

ROE is the most significant ratio which even a lay man shareholder can easily understand. It reflects the earnings of the shares held by the shareholder.  It discloses how much profit a company has earned in relation to the value of shareholdings stated on the Balance Sheet as paid up share capital and retained earnings. ROE, if used properly, it discloses company’s capability on its profitability, asset management and financial leverage.  We break down the calculation of RoE into three parts as follows.

Return on equity = (PAT/Sales)*(Sales/Fixed Assets)*(Fixed Assets/Equity)

Or

Return on Equity= Profit margin* Fixed Asset Turnover* Leverage

In this manner, investors not only acquire a better sense whether the company will generate sufficient return on equity but also perceive the ability of the management to run the organization for a higher growth. In order to avoid distortion due to presence of debts in the balance sheet, ROI (Return on Invested Capital) is calculated by clubbing equity with debts divided by net profit after tax.  So an investor can prefer ROI to ROE when Debt Equity ratio is higher than industry’s average.

2.5.3 Net Profit Margin

The Net profit margin is the ratio indicating how much profit a company makes from every Rand it generates in revenue i.e sales. In other words it calculated taking into the account of the of cost of sales, administration costs, the selling and distribution costs and all other costs.  The remaining value out of sales revenue, is the net profit that is left to pay interest, tax, dividends and so on i.e. appropriation of profits.

2.5 4 Share Price

A share is a unit which represents the holder’s interest in the Incorporated company. Because the company is a separate legal entity from the company membership, it can be sold in specified units called shares.The share price is the value of the share at a given time. It is arrived at by demand and supply on stock markets. The price at which share can be traded. The share prices are quoted on the Stock exchange based on the share’s par value and other key factors like company’s earnings and future prospects.

2.5.5 Tobin Q

The ratio of market price of shares to replacement cost of assets is called Tobin’s Q named after the Economic Nobel Price winner James Tobin who propounded this ratio. The Tobin Q should be ideally higher than 1 as market price of share should be more than the replacement value as an  indicator of company’s strength and solvency. Replacement cost is the one that would be incurred in setting up an identical firm from the scratch. The idea is that the firms cannot trade too far above replacement cost, because competitive pressure from other firms entering the same industry will drive down the market value of the existing firms

2.5.6.Economic Value Added

It is actually a Trade mark of Stern Stewart widely being used by companies and consultants as a measure of performance and inevitable part of investor’s analysis of a company. This EVA (Economic Value Added) is also known as ‘economic profit’. The underlying idea behind this EVA is Cash is King, some expenses are really investments in disguise and equity capital is expensive. It is calculated as follows.

Economic Profit = NOPAT-(WACC x Invested Capital)  Where WACC stands for Weighted Average Cost of Capital and NOPAT stands for Net Operating Profit After Taxes.

EVA gains relevance because investors are only concerned with economic reality (earnings) and not accounting earnings.

3 Research Methodology

It is proposed to adopt the three recognized methods of research namely, literature review, analytical study and a questionnaire. Apart from the brief literature review of studies for countries outside South Africa not less than 25 companies of South Africa will be selected from among those listed on JSE by available means of internet search engines to access the trust worthy websites for necessary data for five year period from 2001 to 2005, excluding regulated companies and financial institutions.

This study will attempt to use the powerful methodology for ultimate owners’ identification through the tracing of control chains advocated by La Parta et. al (1999) “this method has rapidly gained recognition and is almost universally used in modern finance research that involves investigation of the ownership structures. It proposes intuitively simple procedures to distinguish widely-held and ultimately owned companies at different threshold levels to identify ultimate owners and to capture the discrepancy between ownership and control rights. In addition, the use of a universal methodology simplifies an international comparison of empirical results” (Chernykh 2004). [xi]

Empirical models used for previous studies on ownership structure and firm performance will serve as useful guidelines for the present study.

• Analytical Study

This is a powerful and mostly used research method for testing a hypothesized association between two or more variables. There is always a belief that certain events do not happen on their own or without cause. It is for this reason amongst others that one could argue that corporate performance can be linked to certain events or situations that exists. Henceforth, our study plans to establish just that and we will have corporate structure as independent variable and corporate performance defined by selected ratios as dependent variables, primarily to establish that causality between them.

The analytical study on these variables, will be expressed using multiple- regression-analysis (MRA).

Multiple regression analysis is a statistical method for modeling simultaneously the effects of multiple independent variables on a dependant variable.  The concurrent testing  of independent variables enables  estimation of  their independent and combined effects; It ascertains  more clearly  their direction and strength.and also  ensures that there are no spurious effects; We can also forecast , and explain a dependent variable; and also eliminate  the probability of Type I errors. Further, with multiple regression we can design main, interacting, or curvilinear effects and ascertain the incremental improvement in a model resulting by the addition or deletion of independent variables.  Multiple regression can also absorb any combination of nominal, ordinal, or interval level independent variables. Lastly, multiple regression can be applied for the analysis of data collected using various research models, including experimental, quasi-experimental, and non-experimental ones.

Social science researchers adopt the linear regression model otherwise known as ordinary least squares multiple regression, or simply multiple regression. If an inappropriate model is used for the dependent variable at hand will result in a number of undesirable consequences. The effects of independent variables might become distorted. (that is, “biased”). More over “parameter estimates might be “inefficient” (vary more from sample to sample) or “inconsistent” (have sampling distributions whose variability does not decrease with larger samples.”(Buehler, Cheryl 2001)[xii]

3.2 Corporate Ownership Structure

There are different methods that could be used to define ownership structure, but it is proposed to pick the following for the purpose of this study:

Institutional Owners (local and foreign companies, pension funds, unit trusts, and nominee companies representing companies)

Government Ownerships (shares held by investment agency of the South African government – Public Investment Commission (SA)

Insiders Ownership (Directors, management, employees and their nominee companies)

The list of Major Shareholders and a list showing combined shareholding by top three shareholders, will each be grouped as follows for comparison with corresponding performance variables for each year from 2001 to 2005.

Major Shareholder with 5% – 10%

Major Shareholder with 11% – 19%

Major Shareholder with 20% – 39%

Major Shareholder with 40% – 49%

Major Shareholder with 50% – 74%

Major Shareholder with 75% – 100%

3.3 Corporate Performance

For each group of shareholding mentioned above, we will work out the average for each year from 2001 to 2005, make analytical comparison and interpretations to establish any relationship and correlation between them.

Sourced from a McGregorBfa database the following corporate performance variables will be used for the purpose of this study:

Return on Assets (ROA).

Return on Equity (ROE)

Net Profit Margin (NPM)

Share / stock prices

Tobin Q

Economic Value-Added

.3 4 Data sources and sample selection

Most of the data relating to company performances, required for this study will be extracted from McGregorBFA Station Online database and will be verified by data published in a biannual magazine Profile’s JSE Handbook and by reference to individual websites of selected companies and archives of Johannesburg Stock Exchange. Data on shareholding will be obtained from profile’s Stock Exchange handbook, a biannual publication of Sharedata.

A biannual magazine Profile’s JSE Handbook will be used to get a list of top three shareholders for comparison with corporate performance variables.

We will draw our sample of companies for the purpose of this study, from those listed on the main board of JSE but use the following criteria to select our sample:

1. Companies that have 30 June as financial year end
   2. Should have been effectively listed on the main board of JSE for the period 2001 to                2005.
   3.         Exclude Financial and regulated companies (banks, insurance, telecommunications and venture capital companies)
   4.         The required data should be available for the whole period 2001 to2005.
   We decided on 30 June as a basis for selections after we looked at distribution of companies’ financial year end and it became clear that majority of them have 30 June as their financial year-end. Choosing any other year-end, could have resulted in lesser population from which to select the sample and even less representivity of the different sectors of the stock exchange.

Selecting companies with the same year-ends, will increase the comparability since most of the variables are measured at year-end. (See Table 3)

Applying our selection criteria listed above, we arrived at our sample of 75 companies by doing the following:

Companies with 30 June as Year End                                                 150
Delisted and Merged Since 2001                                                                        (33)

Companies remaining in 2005                                                               117

 Banks and Insurance                                                                                (11)

Venture Companies                                                                                               (  7)

Telecommunications                                                                                    (0)
Companies with Incomplete Data                                                                       (24)
Remaining Companies – Sample Size

3.5 Questionnaires

Our initial review of our primary sources, Database from McGregorBFA and biannual magazine Profile’s JSE Handbook, indicates that we will not get a complete data to reflect ownership structure in the required format. The ownership data will be collected by distributing questionnaires to companies included in the sample. The questionnaire will ask each company to indicate in percentages and in numbers, shares held by each of the categories of shareholders, namely, Institutional Shareholders; Government Shareholder; and Insider Shareholders.

4  Preliminary Results of Study (see table 4)

Contrary to our expectations, it is clear from the results in appendix 4 that concentrated ownership is negatively associated with performance when using accounting profit ratios like ROA, ROE and NPM. However, the results paint a different picture when one looks at economic profits. These preliminary results encourage me to proceed with the research by analyzing the data statistically to establish correlation and robustness of the results. The 75 companies selected out of 352 companies at 30 June 2005 is considered representative of the ownership structure and performance of the rest of the companies

[i] Berle A.A. and Means, G.C. (1932) The Modern Corporation and private Property. New York: Macmillan

[ii] ‘Shareholding versus stockholding: a critical review of Corpoarate Governance’, Corporate Governance: An international Review, Vol 12 No 3:242-62

[iii] Blair M.M. (1995) Ownership and Control: Rethinking Corporate Governance for the twenty-firstCentury. Washington, D.C: The Brookings Institution.

[iv] By Mike Burkart, Finance Dept, Stockholm School of Economics, Sweden and Fausto Panunzi , Department di Science Econmiche, Universita di Bologna, Italy

[v] Department of International Economics and Management, Copenhagen Buisnes School, Denmark and Statistics Group, Copenhagen Business School, Denmark

[vi] Adams B Renee, Dept of Finance, Stockholm School of Economics, Sweden and Santos Joao A.C, Research Department, Federal Reserve Bank of Newyork,Nrwyaork,USA

[vii] Journal of Accounting and Economics 41(2006) 55-85

[viii] Demsetz Harold, Department of Economics, University of California and Villalonga Belen, Harvard Business School, Harvard University, Boston

[ix]  Ayogu D. Melvin, University of Cape Town, South Africa in his Economic Research paper No 66 on Corporate Governance in South Africa (2001)

[x] http://way2bee.cliffedekker.com

[xi] Chernykh Lucy, Ultimate Ownership and Corporate performance in Russia(Auust 2004)

[xii] Buehler, Cheryl 2001, Social Work research “introduction to multiple regression for categorical and limited dependent variables”

Tables

Table No  1

Table No 2

* Owned by Barlows, in which Old Mutual holds a 23 per cent controlling stake

Table No 3

Table No 4

                        Source: Profile’s Stock Exchange Handbook

Table No 5