Ownership structure and Information asymmetry - Business Journalz

Ownership structure and Information asymmetry - Business Journalz

Economics and Finance Review Vol. 2(5) pp. 62 – 70, July, 2012 Available online at http://www.businessjournalz.org/efr ISSN: 2047 - 0401 Ownership s...

383KB Sizes 0 Downloads 13 Views

Economics and Finance Review Vol. 2(5) pp. 62 – 70, July, 2012 Available online at http://www.businessjournalz.org/efr

ISSN: 2047 - 0401

Ownership structure and Information asymmetry Mehdi haji ghadirzade (Corresponding author) Department of Accounting, Neyshabor Branch, Islamic Azad University ,Neyshabor, Iran. E-mail: [email protected] Mansour Garkaz Department of Accounting, Ali Abad Katoul Branch, Islamic Azad University ,Golestan, Iran. E-mail: [email protected] Alireza mehrazin Department of Accounting, Neyshabor Branch, Islamic Azad University ,Neyshabor, Iran.

ABSTRACT In the present research, the relationship between ownership structure and information asymmetry has been studied. The study aims at focusing on the question: “Does ownership structure have an important and significant effect on information asymmetry?” Therefore, 356 estimated declared profits of companies from 2010 to 2012 were evaluated. Ownership structure was considered as independent variable, and suggested price difference for sale and purchase was considered as dependent variable. To test hypotheses, Pearson correlation procedure has been used. The results show that percentage of possession centrality of the companies accepted in Tehran stock exchange has a significant effect on degree of information asymmetry. In addition, the percentage of governmental, management, and institutional possession has a significant effect on information asymmetry. Keywords: information asymmetry, declared profit, suggested price stock sale and purchase, ownership structure. 1. INTRODUCTION For a long time in the past, economists believed that all groups of a stock company try for a common goal, but over the past thirty years, many cases have been presented by economists dealing with profit contrast among groups, and the way these companies encounter with these contrasts. These issues are called “theory of agency” in management accounting. According to Jenson and Mckling’s definition, agency relationship is a contract based on which the owner assigns his agent, and gives him the right of making decision (Jenson et al, 1976). In agency relationships, the owners’ goal is maximising his capital, therefore to gain such a goal, the owner monitor the agent’s job, and evaluate his information asymmetry . In this case, the presented question is whether companies ownership structure difference has an effect on their information asymmetry . If the company owners include different groups such as government, financial institutes, banks, and other companies, how will the suggested stock sale and purchase price difference be? And which of these different possession patterns is more effective in informational symmetry improvement? Through finding answer of this question, more appropriate measures can be taken to improve informational symmetry, and decision makers and investors will pay attention to the companies’ owners’ array to achieve an optimised performance for economical units. Therefore, investigating the relationship between ownership structure and information asymmetry is necessary for consumers to evaluate the managers’ performance more comprehensively. In addition, informational symmetry exists provided that managers and market have the same information about the company. Therefore, managers and market tolerate lack of confidence about the company similarly. However, in the case of information asymmetry , managers have better and more information than market because of having personal and secret information of the company, that is, they have access to information earlier than market. Special information of the company is conveyed to market through information divulging procedures over time. Market has some lack of confidence about the company before information divulging. The company information asymmetry is equal to all lack of confidence about the company, because it is more likely that managers and market have the same amount of knowledge about the effect of market variables on the company’s value. Market reflection to declared

62

Economics and Finance Review Vol. 2(5) pp. 62 – 70, July, 2012 Available online at http://www.businessjournalz.org/efr

ISSN: 2047 - 0401

profit can be the first criterion of the company’s informational symmetry through information divulging. Information asymmetry can be obtained by information setting, public reports’ multiplicity, and the number of the company’s transactions, and it can also be affected by the managers or market’s behavior. For example, when a public report is done for any company, when other factors are fixed, it is possible market get more information about the company’s real situation, and information asymmetry decreases. 2. REVIEW OF THE LITERATURE Empirical researches conducted recently have taken declared profit effect on the suggested price domain of stock sale and purchase into account. According to Oushman and Mures (1983) who conducted a research in 25 companies, they found a change in suggested price domain in stock sale and purchase in time of declared profit. Yan (1988) presents the factors based on which important movements in stock price around the time of declared profit have been observed, and it can convince traders to collect more information. He found out that both interest variety of interest and market reflection to unexpected interest has a positive relationship with changes stock sale and purchase suggested price domain before declared profit. Therefore, the suggested price domain in the previous day, the day of reporting, and the day after has an increasing trend. Eker (2002) reported a slight difference stock sale and purchase suggested price domain and increase in transaction volume around the time of declared profit. Having investigated the companies working in Torento stock exchange, Libi, Mateo, and Rab (2003) found out that before declared profit, stock sale and purchase suggested price domain increases. Having studied the companies working in Paris Euronext, Louji (2004) observed the size of the suggested price domain of stock sale and purchase in the time of declared profit. Having selected 1063 companies from the U.S stock exchange, Kini and Mian (2007) studied the relationship between possession disperse and the suggested price difference of stock sale and purchase and did not find any significant relationship. Having selected 260 companies, Haflin and Shaw (2003), found a positive relationship between stock percentage in stock holders and suggested price difference and stock sale. They contend that the more concentrated ownership structure is, the more personal information stockholders have access to. As a result, transaction parties will be exposed with the danger of inappropriate choice in the transactions in which they are dealing with. It causes stock transaction parties increase the suggested price difference of stock sale and purchase to decrease the risk of inappropriate choice. As a result, when the suggested price difference of stock sale and purchase increases, stock transaction in the market decreases, and cashability of stock market decreases. Jeng and Ken (2007) conducted a research about the relationship between ownership structure and interest informational content in Korea. Korean companies’ ownership structure is easily realized to be dealt with a main owner in management. Based on their findings, although institutional investors and main stockholders may have no motivation for supervision because of lack of skill, existence of some beneficiaries, and their relationship with management, with increasing in presence of institutions interest informational content increases. 3. METHODOLOGY The present study is going to investigate the question: “what is the role of ownership structure of the companies accepted in Tehran stock exchange in their information asymmetry ?” Therefore, this research is practical. This is a pseudo-experimental research with expo-facto approach. 3.1 Hypotheses Accounting declared profit or any published information decreases information asymmetry to some extant, if dispersed among investors properly. But if it is done inappropriately, we can see its negative impacts on

63

Economics and Finance Review Vol. 2(5) pp. 62 – 70, July, 2012 Available online at http://www.businessjournalz.org/efr

ISSN: 2047 - 0401

transaction volume, abnormal return, sale and purchase suggested price domain, … Investors, with less knowledge, can not analyse information as appropriately as experienced investors. As a result, if reported interest does not have appropriate quality, it can increase information asymmetry in financial markets investors. Investors announce their prices to sell and buy stock by going over to burse officials. These prices can highly fluctuate over the next few days, because every investor present their own desired price to burse officials. Transaction happens when the buyer and seller come into consensus over a single price. Regarding what has been discussed, this research contains four hypotheses as follows: Hypothesis 1: there is a significant relationship between possession centrality percentage with information asymmetry. Hypothesis 2: there is a significant relationship between governmental possession percentage with information asymmetry. Hypothesis 3: there is significant relationship between governmental stockholder percentage and lack informational symmetry. Hypothesis 4: there is a significant relationship between institutional possession percentage and information asymmetry. 3.2 Statistical population and selecting companies In this research, statistical population contains the companies accepted in Tehran stock exchange from 2010 to 2012. In fact, estimated declared profit for each share of accepted companies in Tehran stock exchange has been selected as statistical population, and after considering some limitations and conditions, qualified cases have been considered as samples. Among reasons for using these companies, are daily price of stock for these companies, the accepted companies’ commitment for estimated annual interest per share report, registering the suggested prices of selling and purchasing of Tehran stock exchange accepted companies stock. Regarding the longevity of studying, samples should be selected in a way they have an active market during study so that hypotheses can be tested. To do so, the following plan to select samples has been used: 1. Companies should be included in the active companies of the burse, or at least, they should be active during the days of the research. To do so, a decision has been made to determine a criterion to determine whether the companies are active or not until 21 days before, and 21 days after announcing estimated interest per share which is the basis of this research. (the companies which were active at least 11 days of this 21 day period and their stock have been transacted have been selected.) The period has been selected 21 days, because the shorter the period, the slighter the effects of other variables will be. But if so, many companies will be omitted, so a 21 day period has been selected. 2. The companies should not have capital increase in the studies period. 3. They should be accepted in Tehran stock exchange to the end of 2009. 4. They should not be included in investment and middling companies. 5. Required information should be accessible during the period. Regarding the mentioned limitations, 354 estimated declared profits which belong to 118 companies, have been selected as samples. 3.3 Data collection procedures In the present study, to collect data and information, first corpus procedure has been used. In the corpus part, theoretical considerations of the research were collected from Persian and English books and magazines, and then the research data were collected from the selected companies data, financial forms, descriptive notes, weekly reports, and stock exchange monthly magazine by the software of Denasanj, Sahra, and Tadbirpardaz. 3.4 Testing the research variables 3.4.1 Ownership structure Ownership structure components in the present research are as follows:

64

Economics and Finance Review Vol. 2(5) pp. 62 – 70, July, 2012 Available online at http://www.businessjournalz.org/efr

ISSN: 2047 - 0401

Possession centralization percentage (own): it is measured by proportion of owned stock of 5 big investors to total of the number of spread stock for company i in year t. Governmental possession percentage (STA): it is measured by proportion of owned stock of government to the total number of spread stock for company i in year t. Management possession percentage (MGR): it is measured by proportion of owned stock of managers to the total number of spread stock for company i in year t. Institutional possession percentage (INSIDER): it is measured by proportion of company possession which is assigned to semi-governmental units and charities and investment institutes (real persons), to the proportion of total number of spread stock for company i in year t. 3.4.2 Information asymmetry To calculate lac of informational symmetry, the following method is used. This model has been used by Chiang and Vikentaj to determine the suggested domain of stock sale and purchase. The mentioned model is as follows:

In which: SPREAD= the suggested price difference domain of stock sale and purchase (Ask Price) AP = the suggested price average of stock sale of the company i in the studied period (Bid Price) BP= the suggested price average of company i in studied period To do calculations, first the best sale and purchase suggested price for each share is extracted for 21 days before and after estimated declared profit. (By the best suggested price, we mean the highest suggested price for buying share, and by the best suggested price for selling stock, we mean the lowest suggested price for selling each share), then by calculating their average, the suggested price domain of stock sale and purchase is calculated. 3.5 data collection procedures Regarding the type of data and analysis statistical procedures, in the present research, procedure of “compound data” has been used. To study the relationship of ownership structure and information asymmetry in companies, dependent and independent variables are considered from two different aspects. On the one hand, these variables are tested in different companies, and on the other hand, in the time period of 2010 to 2012. To determine regression, the following relationship has been used: SPREAD= f(OWN, STA, MGR, INSIDER) + it SPREAD= the suggested price difference domain of stock sale and purchase OWN= possession centrality percentage STA= governmental possession percentage MGR= management possession percentage INSIDER= institutional possession percentage The above relationship will be reformulated as follows: SPREAD = α + β (Ownership) + ειτ 4. THE RESEARCH FINDINGS Of other analysis processes of data is studied variables description. Using descriptive statistics is to purify the collected data and to know studied population more and more. The aim of description is to extract basic points and to mix data by language of numbers. To describe this data, central deviation and distribution have been used. Each of these variables has 354 observations. It shows that all variables have normal distribution. It has been reported before and after omitting extreme observations.

65

Economics and Finance Review Vol. 2(5) pp. 62 – 70, July, 2012 Available online at http://www.businessjournalz.org/efr

ISSN: 2047 - 0401

The amounts of central deviation and distribution of variables before omitting extreme observations have been shown in table (1-4). Table (1-4) description of the studied variables central deviation and distribution indexes before extreme observations A-SPREAD

B-SPREAD

INSIDER

MGR

STA

OWN

title

1/01

1/150

64/5

1/88

41/5

55/1

average

1/10

1/100

57/9

1/70

44/6

60/2

mean

1/412

1/050

10/925

0/413

7/548

9/2

Standard deviation

1/124

1/141

119/35

0/17

56/97

84/64

variance

4/215

1/165

01/42

8/19

9/01

9/52

Skewness coefficient

04/44

06/06

001/10

85/65

000/249

00/12

kurtosis

(OWN: possession centrality, STA: governmental possession, MGR: management possession, INSIDER: institutional possession, B-SPREAD: the suggested price difference domain of stock sale and purchase before declared profit, A-SPREAD: the suggested price difference domain of stock sale and purchase after declared profit.) After omitting extreme observations, we elaborate more on descriptive statistics. Table (2-4) describing the studied variables central deviation and distribution indexes after omitting extreme observations A-SPREAD

B-SPREAD

INSIDER

MGR

STA

OWN

title

1/01

1/150

57/38

1/67

45/32

56/29

average

1/10

1/100

52/4

1/74

44/6

54/8

mean

66

Economics and Finance Review Vol. 2(5) pp. 62 – 70, July, 2012 Available online at http://www.businessjournalz.org/efr

ISSN: 2047 - 0401

1/412

1/050

11/306

0/421

7/796

9/121

Standard deviation

1/124

1/141

127/82

0/177

60/77

83/19

variance

2/364

3/249

9/86

8/19

9/65

9/45

Skewness coefficient

12/06

16/21

112/67

85/65

111/439

00/06

kurtosis

(OWN: possession centrality, STA: governmental possession, MGR: management possession, INSIDER: institutional possession, B-SPREAD: the suggested price difference domain of stock sale and purchase before declared profit, A-SPREAD: the suggested price difference domain of stock sale and purchase after declared profit.) 4.1 Hypotheses testing Through the following hypotheses testing as well as specific tests relevant to each of these four variables, in all cases, and using f tests and –Durbin-Watson, Regression model of research has been proved in terms of being linear of relationships among variables and independence of observations from each other. In all models, significance level of f test is lower than 0.05 and Durbin-Watson statistic is between 1.5 and 2.5. In all hypotheses, by Kalmogorove Smirnove tests, normality of data has been proved. 4.1.1 The first hypothesis testing: The first hypothesis is “there is a significant relationship between possession centrality and information asymmetry ”. To test the hypothesis, a Pearson correlation was run between the obtained possession centrality for the companies, and the degree of their lack of informational centrality with error probability of 0.05. In this test, obtained sig is 0.008, because it is higher than 0.05, as a result, H0 accounting for lack of significant relationship between these two variables, is rejected, and H1 accounting for significant relationship between possession centrality and information asymmetry , is accepted. Table (3-4) the first hypothesis testing results Slight correlation Sig level coefficient .1000

t statistic

Coefficient

variable

4.586

1.201

From beginning

1.4

0.008

16.0

1.045

Possession centrality

Sig level of correlation coefficient

6.501

F statistic

1.125

0.008

(P-VALUE) statistic F

1.120

152

Observation number

0.954

Determination coefficient Modified Determination coefficient Durbin-Watson statistic Statistical results

1.118

H1 hypothesis accepted

67

Economics and Finance Review Vol. 2(5) pp. 62 – 70, July, 2012 Available online at http://www.businessjournalz.org/efr

ISSN: 2047 - 0401

4.1.2 The second hypothesis testing Second hypothesis is “there is a significant relationship between governmental possession and information asymmetry ”. To test the hypothesis, a Pearson correlation was run between the obtained possession centrality for the companies, and the degree of their lack of informational centrality with error probability of 0.05. In this test, obtained sig is 0.128, because it is higher than 0.05, as a result, H0 accounting for lack of significant relationship between these two variables, is accepted. H0: r = 0 H1: r ≠ 0 Table (4-4) the second hypothesis results slight Sig leve

t statistic

coefficient

variable

0.1826

-4.4991

1.4468

From beginning

1.151

0.128

5.9201

1.181

Sig level of correlation coefficient

8.200

F statistic

1.5601

0.128

(P-VALUE) statistic F

1.5120

152

Observation number

0.825

Governmental possession Determination coefficient Modified determination coefficient Durbin-Watson statistic Statistical results

1.048

Hypothesis H0 accepted

4.1.3 The third hypothesis testing Third hypothesis is “there is a significant relationship between management possession and information asymmetry ”. To test the hypothesis, a Pearson correlation was run between the obtained possession centrality for the companies, and the degree of their lack of informational centrality with error probability of 0.05. In this test, obtained sig is 0.071, because it is higher than 0.05, as a result, H0 accounting for lack of significant relationship between these two variables, is accepted. H0: r = 0 H1: r ≠ 0 Table (5-4) the third hypothesis results slight Sig level

t statistic

coefficient

variable

0.031

5.9413

.11063

From beginning

0.172

0.071

2.36

.1544

Sig level of correlation coefficient

5.148

F statistic

0.345

0.071

(P-VALUE) statistic F

.1341

152

Observation number

2.008

Governmental possession Determination coefficient Modified determination coefficient Durbin-Watson statistic Statistical results

.1071

Hypothesis H0 accepted

68

Economics and Finance Review Vol. 2(5) pp. 62 – 70, July, 2012 Available online at http://www.businessjournalz.org/efr

ISSN: 2047 - 0401

4.1.4 The fourth hypothesis testing The fourth hypothesis is “there is a significant relationship between institutional possession and information asymmetry ”. To test the hypothesis, a Pearson correlation was run between the obtained possession centrality for the companies, and the degree of their lack of informational centrality with error probability of 0.05. In this test, obtained sig is 0.084, because it is higher than 0.05, as a result, H0 accounting for lack of significant relationship between these two variables, is accepted. H0: r = 0 H1: r ≠ 0 Table (6-4) fourth hypothesis results slight Sig level

t statistic

coefficient

variable

0.059

5.944

.1619

From beginning

.1358

0.084

0.917

.1169

Sig level of correlation coefficient

4.268

F statistic

1.521

0.084

(P-VALUE) statistic F

1.516

152

Observation number

.0784

Governmental possession Determination coefficient Modified determination coefficient Durbin-Watson statistic Statistical results

.1084

Hypothesis H0 accepted

5. CONCLUSION The results of hypotheses testing show that ownership structure of the companies accepted in Tehran stock exchange has no effect on information asymmetry which is in line with some researches done abroad. In addition, information asymmetry increases after declared profit which supports some researches conducted out of Iran. The findings of some researches conducted out of Iran show that information asymmetry after declared profit is more than that of before declared profit. Based on researches by Ushman et al (1983) done on 25 companies, found out the suggested price difference domain of stock sale and purchase at the time pf declared profit. Velker (1995) showed that there is a reverse relationship between divulging quality and the suggested price of stock sale and purchase. It implies the positive relationship between information asymmetry and interest management. The announced report quality means potential background of interest growth and probability degree of achieving future interests (Esmaeili). Therefore, the more reported interest is, the more its impacts on the investors decisions, and consequently the quality of divulged information increases, and it means that information asymmetry decreases because each share value depends on not only each share interest but also our expectation from the future of the company and profiting power for the future years and degree of confidence for future interests which affects on investment decisions (Esmaeili, 2006). Fransis, Landa, and Alexander (2008) came into conclusion that companies with good interest, have higher information divulging level and low level of information asymmetry . Besides, Ajinka et al (2005) found out that the more conservative the predictions are, the higher the level of information divulging which leads to decrease in lack of information symmetry. The findings of the present research are not in line with those of the researches done by Esmaeili (2006), Velker (1995), Ferancis, Nanda, and Alexander (2008) and Anjenika et al (2005). In terms of the fact that interest has informational content, the present study supports Vatanparast (2005). Selecting a population of 1063 companies accepted in the U.S stock exchange, Kini and Mian (2007) studied the relationship between possession distribution and the suggested price difference of stock sale and purchase and found no significant

69

Economics and Finance Review Vol. 2(5) pp. 62 – 70, July, 2012 Available online at http://www.businessjournalz.org/efr

ISSN: 2047 - 0401

relationship. Heflin and Shaw (2003) selected a population of 260 companies and found out that there is a positive relationship between block stock percentage for investors and the suggested price difference and stock selling. They believe that in companies which have more centralised ownership structure, big investors have access to private information which is in line with the findings of the present research. REFERENCES Aghion, P., J. Van Reenen & L. Zingales. (2009). "Innovation and institutional ownership", Working Paper, Chicago Booth School of business. Bollen, N., Smith and Whaley. (2004). "Modeling the Bid-Ask Spread: Measuring the Inventory-Holding Premium". Journal of Financial Economics. Vol, 32. 159-163. Cai Jie, Liu Yixin, Qian Yiming. Information Asymmetry and Corporate Governance; 2008. available at: http://ssrn.com. Copeland, T., Galai, D. (1983). "Information effects on the spread". The Journal of Finance. No. 38(5). 1457– 1469. Easley, D., O’Hara, M. (1992). "Time and the process of security price adjustment", The journal of Accounting reaserch, 47(2), 577-605. Glosten, L., Milgrom, P. (1985). "Bid-ask spread transactions prices in a specialist market", Journal of FinancialEconomic, No. 14(1), 70-100. Ho, T., Stoll, H,R. (1981). "Optimal dealer pricing under transactions and return uncertainty", Journal of Financial Economics, No. 9(1), 47-73. Kanagaretnam Kiridaran, Lobo Gerald J., Whalen Dennis J. Does Good Corporate Governance Reduce Information Asymmety Around Quarterly Earnings Announcements?. Journal of Accounting and Public Policy2007; 26(4): 497-522. Krinsky, I., Lee, J. (1996). 'Earnings announcements and the components of the bid-ask spread", Journal 0f Finance, 51,23-35. Kyle, A. (1985). "Continuous auctions and insider trading", Econometrica, 53(6), 1315-1335. LaFond, R., and R. Watts, 2008, "The Information Role of Conservatism", The Accounting Review (forthcoming). Availble at URL: http://www.ssrn.com. Libby, T., Mathew, R., Robb, S. (2002). "Earning announcement and information asymmetry: An intra-data analysis", Contemporary accounting research, 449-472. Navissi, F. & V. Naiker. (2006). "Institutional ownership and corporate value", Managerial Finance, Vol. 32, PP. 247-257. Velury, U. & D. S. Jenkins. (2006). "Institutional ownership and the quality of earnings", Journal of Business Research, Vol. 59, PP. 1043-1051.

70