AN EMPIRICAL RESEARCH FOR THE DETERMINATION OF VARIABLES WHICH AFFECT THE EXPLANATION LEVEL OF ACCOUNTING AND REPORTING STANDARDS
Author :
DOI : 10.26728/ideas.47
Year-Number: 2017-6
Language : null
Konu :
Abstract
The extraordinary changes in technology nowadays have also influenced change in individuals’ consumption habits. Consequently, many changes have occurred in economic life and multinational investments have increased. Consequently, while businesses who continued their activities in commercial life have steered their investments towards new business areas needed to sustain their existence, they had to move towards capital diversification and form new partnerships. On the other hand, it is clear that capital is the basic requirement for businesses to sustain their existence. Even though businesses use their own resources to meet their capital needs, they require international capital for the continuation of their activities. In this sense, not only businesses, but also countries do need multinational investments to realize their economic goals and increase their levels of prosperity. As capital owners move their fund resources to investment areas it is unarguable that they consider different criteria. The most fundamental reference information among these is the financial information on a business. It is mandatory that the financial information required for investment has to be primarily composed of qualitative accounting information. The formation of qualitative accounting information is possible with accounting and financial reporting standard prepared with universal qualification. On the other hand, the explanation level of the information, required to be explained as prescribed by a business’ accounting and reporting standard, for the financial information users is very crucial. This information are explained by businesses in the footnotes of financial tables. An “explanation index” has been created in many studies in the literature to determine the footnote explanation level of a business. The objective of this study is the determination of variables which affect the explanation levels related to the presentation of information prescribed in the accounting and reporting standards of the sectors within the scope of the research. An “explanation index” has been created within this framework to determine the level of information required to be explained in the standards. This index has been accepted to be the dependent variable in the study and the relationship with the independent variables such as the age of business, size of business, active size of business, 4 large auditors, change of auditing firm and change of chief auditor, has been tested using the panel data regression analysis method. As a result of the study, the change in auditing firm and the change in the chief auditor have been observed to have a higher effect on the “explanation index” in comparison to the other variables
Keywords
Abstract
The extraordinary changes in technology nowadays have also influenced change in individuals’ consumption habits. Consequently, many changes have occurred in economic life and multinational investments have increased. Consequently, while businesses who continued their activities in commercial life have steered their investments towards new business areas needed to sustain their existence, they had to move towards capital diversification and form new partnerships. On the other hand, it is clear that capital is the basic requirement for businesses to sustain their existence. Even though businesses use their own resources to meet their capital needs, they require international capital for the continuation of their activities. In this sense, not only businesses, but also countries do need multinational investments to realize their economic goals and increase their levels of prosperity. As capital owners move their fund resources to investment areas it is unarguable that they consider different criteria. The most fundamental reference information among these is the financial information on a business. It is mandatory that the financial information required for investment has to be primarily composed of qualitative accounting information. The formation of qualitative accounting information is possible with accounting and financial reporting standard prepared with universal qualification. On the other hand, the explanation level of the information, required to be explained as prescribed by a business’ accounting and reporting standard, for the financial information users is very crucial. This information are explained by businesses in the footnotes of financial tables. An “explanation index” has been created in many studies in the literature to determine the footnote explanation level of a business. The objective of this study is the determination of variables which affect the explanation levels related to the presentation of information prescribed in the accounting and reporting standards of the sectors within the scope of the research. An “explanation index” has been created within this framework to determine the level of information required to be explained in the standards. This index has been accepted to be the dependent variable in the study and the relationship with the independent variables such as the age of business, size of business, active size of business, 4 large auditors, change of auditing firm and change of chief auditor, has been tested using the panel data regression analysis method. As a result of the study, the change in auditing firm and the change in the chief auditor have been observed to have a higher effect on the “explanation index” in comparison to the other variables
Keywords
- The relationship between the explanation rates of some companies operating in Sweden and theindependent variables such as being public and size of business has been researched in the studydone by Cooke (1989). As a result of the study, a meaningful relationship has been determinedbetween the explanation rate and independent variables such as being public and size of business.
- The effect of independent external audit on accounting information levels for 1987-1991 explainedby 122 firms active in İMKB has been tested in the study by Çürük (2004), named ‘Practical studyon the effect of independent external audit on accounting transparency: Turkey as an Example’’. Asa result of the study, the independent external audit has been determined to have no effect on the explanation level.
- Explanations related to the important accounting methods in the 2005 financial tables of 107 publiccompanies, among the top 500 largest industrial establishments in Turkey registered with İMKBhave been reviewed by Ulusan (2007) and the businesses’ accounting policy preferences have beentried to be determined. As a result of the study, (Ulusan, 2007:195) has determined as a result of thestudy that “in the simple linear regression analysis results, the selection of an accounting policy didnot explain the business management’s income separating (decreasing) accounting policy selectiontendency of the economic determinants, other than the size of business in companies which prepare consolidated financial tables”.
- In the study by Akman (2009); Explanation index has been calculated according to the informationformed from the 2004, 2005 and 2006 financial tables of companies selected from Germany,Australia, France, Holland, England, Italy and Turkey according to sampling mass and the relationbetween this index and the independent variables such as country specific cultural values andcompany size, profitability, being traded in more than one capital market has been tested usingregression analysis. According to the results of the study, the author has determined cultural valuesto be effective on companies’ financial explanation levels before UFRS and that the cultural values’ effect on the financial explanation continued after UFRS as well.
- Rajhi (2014) has tried in this study to determine the explanation level using the 10 accountingstandards for 2009, 2010 of 118 French companies whose shares get traded in the NYSE Euronextstock market. As a result of the study, the author has determined all the firms to have fulfilled the explanation obligations prescribed by UFRS.
- Raithatha and Bapat (2014), have tried in their study on the Bombay stock market in India todetermine the explanation level and the relationship between the explanation level of 29 accountingstandards for 2009 of 234 companies out of the 4200 registered total in Bombay stock market. Theaverage explanation level (compliance level) of companies in the manufacturing sector has beendetermined to be 73% and the average explanation level (compliance level) of companies in theservice sector has been determined to be 69%. The total average explanation level of all the sectorsis 71%. As a result of the statistical study done, they have determined the explanation level to be high in all the sectors in the case when auditing gets done by the largest 4 auditors.
- Demir and Bahadır (2014), have tried in their study done on 168 manufacturing companiesregistered with the Stock Exchange İstanbul for 2011, to determine the relationship betweenexplanation level, over 5 basic subjects and 215 sub policy headings and independent variables suchas the size of business, profitability, age of business, 4 largest auditors. As a result of the study, theyhave determined the profitability and auditing by 4 largest auditors to have effect on explanation level.
- The basic objective of the study is; the determination of the variables that affect the explanationlevel related to the presentation of information prescribed by the accounting and reporting standardswithin the scope of the research. A preliminary study has been carried out for the companies in thesectors to be included in the research for purposes of determining the research sample's mass.Within the scope of the research, consecutive information has been searched under the heading ofimportant accounting policies in the footnotes included in the independent audit reports of thefinancial statements of the companies in the sectors in Stock Exchange Istanbul. In addition, sincethe coherence of the research and the achievement of all companies and sectors operating in theIstanbul Stock Exchange will be very difficult in terms of time or technique, in the study 45different accounting policy explanations in the 10 accounting standards displayed in annex-1,included in the footnotes of financial tables for 2012-2013-2014 of 235 firms, shown in table 1,operating in 21 different sectors in the Stock Exchange İstanbul, on which information has beenaccessed using the Public Enlightening Platform, have been reviewed. The “explanation index”formulated according to the explanation levels of policies within the scope of the study has beenaccepted to be the dependent variable and the relationship with the independent variables such asthe age of business, size of business, active size of business, 4 large auditors, change of auditingfirm and change of chief auditor, has been tested using the panel data regression analysis method. Table 1. Sectors within the Scope of Research and Number of Companies
- Panel data analysis regression method has been applied in the research to put forth the direction andthe level of relationship in statistical sense, if any, between the explanation levels of accountingstandards of companies in sectors determined within the scope of the study and the determinedvariables such as age of business, sales, size of business, 4 large auditors, change of auditing firmand change of chief auditor, has been tested using the panel data regression analysis method. Sincethe sampling of the research is comprised of 3 years’ of simultaneous data between 2012, 2013,2014 for 235 companies, it has panel data characteristics. In other words, the data within the scopeof the study has the feature to contain a time series for each horizontal section data. Therefore,panel regression equation has been created in the application of data and the regression equation has been tested using the E-Views 7.1 statistical package program
- The dependent and independent variables used within the scope of the study are displayed in table 2below. On the other hand, the Explanation index (disclosure) method found in the literature hasbeen used in the study to create the dependent variable. The Explanation index tries to determinethe level of providing information concerning situations where a business has to make anexplanation (Cooke and Wallace, 1989: 47). In other words, it shows the account policies related toaccounting and reporting standards in financial tables, meaning the explanation levels for companyspecific situations in footnotes. Basically there are two methods in the calculation of theExplanation index, one is weighted and the other not weighted. The weighted method expressesranking to be done among the information that needs to be explained, meaning to assign weight tothis information. Each explanation item in this method is assigned a value from 0 to 1. In theunweighted method each footnote explanation obligation is accepted to have equal importance. If anexplanation is made for information that requires footnote explanation in this method, a value of 1assigned and value of 0 is assigned if explanation is not done. Nevertheless if the informationwhose explanation has been reviewed is not a usable information, it is not included n the evaluation(Cooke,1989:115). The unweighted method shall be used in the study similar to researchers whouse it often; (Cooke,1989; Soh,1996; Al-Modahki, 1996; Çürük, 2004; Esen and Sakin, 2009;Rajhi,2014). The equation which shows the model related to this method can be shown as below (Rajhi, 2014:4);
- Some data has to be collected in order to create the explanation index within the scope of theresearch. The data for the research has obtained within the scope of the work using the footnoteexplanations in the financial tables of independent audit reports for 2012-2013-2014 from PublicEnlightening Platform for companies specified in table 1 above. The accounting policies and subheadings provided in annex-1 below, related to the accounting and reporting standards explained bysectors under the sub heading of, summary of important accounting policies, located under theheading of “basis related to the presentation of financial tables” in footnotes have been reviewed.As the policy explana0ions regarding the accounting and reporting standards got reviewed, theexplanations related to footnote information for companies included in control lists created bysector according to years have been scanned. If an explanation is made regarding the accountingand reporting standards of the identified companies, a value of 1 has been assigned and value of 0has been assigned for unexplained information. This application used to create the data set has beenperformed for the 21 sectors and 235 companies, for 10 accounting and reporting standards for2012, 2013,2014 and the identified 45 sub accounting policy headings. The scores each companyhas received from the explanation items according the control lists created have been added and ithas been proportioned to the total scores and accounting and reporting standards’ explanation levels by sector have been determined.
- In regard to panel data analysis, a balanced panel is relevant if each unit has been observedthroughout the time periods and an unbalanced panel is relevant if periods of time are lost for someunits. A balanced panel has been used in this study. In this study an econometric model has beenestablished using annual data of 235 companies for which uninterrupted data for 2012-2014 hasbeen accessed. The data for the study has been obtained from the footnotes of financial tables in theindependent audit reports of companies registered in Stock Exchange İstanbul in the 21 sectorsexplained above within the scope of the study. The variables have to be stable when time andhorizontal section analyses are done together in panel data analyses so that false results would notappear. The unit root tests have to be done to find out if the variables used in analysis are stable or not.
- The performance of stability analysis for variables modifies the effectiveness of multiples obtainedas a result of model forecast. The existence or non-availability of unit root in the series subject tothe analysis changes the structure of the panel data analysis to be applied and the type of the testsused. In this test, 1 dependent variable subject to the analysis and Hadri test has been used toresearch the stability of 6 independent variables. As part of Hadri (2000), the hypothesis whichdefends the existence of unit root in any of the series that form the panel is tested against the emptyhypothesis which states the non-existence of unit room. The Hadri unit root test based on the KPSSunit root test basis in time series analyses, is a test based on error terms obtained as a result of theapplication of the least squares method on the fixed series based on Lagrange Çarpanı (LM) test or on fixed and trend. The hypotheses related to Hadri (2000) are as follows:
- Upon review of the results in table 3, it is observed that all the variables subject to the model werestable at the level and that the dependent and independent variables which formed the panel did notinclude unit root. The probability for the calculated test statistics are meaningful at 5% (p<0.05) andthe zero hypothesis which states the inexistence of unit root cannot be rejected. In regard to theanalyses where classic panel regression shall be applied in the next phase of the study, the fixedeffects model from the panel regression methods or the compatibility of the random effect modelhave to be determined. This situation is determined with the use of Hausman Test. If unit or timeeffects have been determined as a result of the analyses done, it has to be decided if these effectswere fixed or coincidence. Hausman (1978) specification test developed to test the definition errorin this aspect is used to make a choice between the forecasters in the panel data models (Uluyol andTürk, 2013:377). The difference between fixed effects model parameter forecasters and the randomeffects model parameter forecasters has been reviewed to see if it had statistical meaning. Hausman test results are displayed in Table 4;
- According to the results in Table 4, it shows chi squared distribution with k degree of freedomunder the “random effects are in effect” zero hypothesis. According to the findings obtained, thezero hypothesis cannot be rejected for the panel data model forecasted for the 2012-2014 period.Consequently, the selection of random effects model is appropriate and the regression multiple forecasters are effective.
- In general, if data has been withdrawn randomly from a main mass with a large horizontal sectionsize, random effects model is used and if a data set defined more specifically is involved then afixed effects model is used. While the correlation between the unit effects and the explanatoryvariables is assumed to be zero in the random effects model, the correlation is permitted to bedifferent than zero in the fixed effects model. Nevertheless, while the fixed time variables arepermitted to exist in the random effects model, the mentioned variables’ existence has beenrestricted in the fixed effect model (Tatoğlu, 2013: 79). In order to prevent a possible successivedependency issue in terms of the variables in the model used in the study, the EstimatedGeneralized Least Squares- EGLS method where random effects have been taken into considerationhas been used instead of the least squares (EKK) method. This method provides consistent andeffective multiples even when Heteroskedasticity and autocorrelation exist. As it is known,Heteroskedasticity, the “Error term” which is one of the fundamental assumptions of Gauss-Markovhypothesis and the least squares method is with fixed variance. Autocorrelation is the situationwhere there relationship between the values that follow the error term in the multiple regressionanalysis. This situation is accepted to be a deviation from an important assumption of the generallinear regression model (Sümer,2006:18). On the other hand, it eliminates the error terms to be withfixed variance (square of standard deviation) and removes the relationship that could occur between the values of the error term.
- Upon review of the regression model explained above to test the hypotheses according to theempirical findings, the long term regression multiples are not meaningful at 5% level statisticallybetween the age of business and the explanation level. As it can be seen in the results of theregression model, the probability value has been found to be (p<) 0.2592. This situation shows thatthe age of business has no effect on the firms’ explanation level. Therefore, the first hypothesis ofthe study (H1) has been rejected. In regard to the other assumption, the long term regressionmultiples are not meaningful at 5% level statistically between the active size of business and theexplanation level. As it can be seen in the results of the regression model, the probability value hasbeen found to be (p<) 0.1752. This situation shows that the active size has no effect on the firms’explanation level. Therefore, the third hypothesis of the study (H3) has been rejected. According tothe regression results, the long term regression multiples are meaningful at 5% level statisticallybetween the size of business and the explanation level. As it can be seen in the results of theregression model, the probability value has been found to be (p<) 0.0000. This situation shows thatthe size of business has effect on the firms’ explanation level. Therefore, the second hypothesisassumed in the study (H2) cannot be rejected. In the literature, Wallace and others, 1994; Hope,2003; Raithatha and Bapat 2014; Rajhi, 2014; Demir and Bahadır 2014 have determined a highlevel of meaningful relationship in the positive sense between the size of business and theexplanation level as a result of their conducted studies. This situation supports the assumed second hypothesis (H2) of the study.
- According to the regression results, the long term regression multiples are meaningful at 5% levelstatistically between the 4 large auditing firms and the explanation level. As it can be seen in theresults of the regression model, the probability value has been found to be (p<) 0.0000. Thissituation shows that the auditing of firms by the largest 4 auditing firms has effect on the firms’explanation level. In other words, the explanation level regarding the accounting policies is high infirms where the auditing function is performed by one of the largest 4 auditing firms. Therefore, theassumed fourth hypothesis (H4) in the study cannot be rejected. In the literature, researches done byGlaum and Street, 2003; Al-Shammari, 2011; Juhmani, 2012, they have found meaningfulrelationship in the positive sense between working with the 4 large auditing firms and theapplication level of these firms’ accounting standards. This situation supports the assumed fourth hypothesis (H4) of the study.
- The determination of the variables affecting the explanation level of information prescribed in theaccounting and reporting standards have been attempted in the study. Within this scope, and 45different accounting policy explanations in 10 accounting standards in the footnotes of financialtables for 2012-2013 of 235 firms in Stock Exchange, İstanbul active in 21 different sectors havebeen reviewed. The “explanation index” (DIC) formulated within the scope of the study accordingto the policies’ explanation levels, has been accepted as the dependent variable and the relationshipwith the independent variables such as the age of business, size of business, active size of business,4 large auditors, change of auditing firm and change of chief auditor, hypotheses have been testedusing the panel data regression analysis method. Upon general evaluation of the results of themodel created within the scope of the study, it was found that long term regression results were notstatistically significant between the age of company and explanation level. This is based on theassumption that the age of the company is old, in other words, because the corporate structure of thebusiness is strong, the level of disclosure is high. But the results of the analysis show that, the ageof the company has no effect on the level of disclosure of firms. Statistically, long term regressioncoefficients between the asset size and the explanatory level are not significant. The results of theanalysis show that the asset size has no effect on firms' level of disclosure. According to the modelresults, long term regression coefficients between the firm size and the explanatory level werefound to be statistically significant in the positive direction. This suggests that the firm size has aneffect on firms' level of disclosure. In the study, the effect of especially the change in auditing firmand the change in chief auditor on the “explanation index” has been observed to be higher incomparison to the other variables. While firms’ changing of the auditing firms they work with affectthe explanation index positively, the change of the chief auditor affects the explanation indexnegatively. On the other hand, it has been observed that the variables determined within the scopeof the research had effect on sectors where explanation level was high. For example, in situationswhere firms in these sectors get audited by largest 4 auditors, the explanation level rises. Thissituation may be a proof that the auditing activities are done more attentively by these 4 largeauditing firms. In addition, the results of auditing activities impact the business’ compliance levelswith international standards and also increase the level of having financial reports compared at aninternational dimension. Therefore, the independent auditing activity should not be accepted just toguide the preparation of business’ financial tables according to international standards. It can alsobe stated that the auditing firms work to transform the audit process in the business to a corporateunderstanding at the place they provide auditing service by providing training service to the related units and managers at the same time.
- Akman, N. (2009). The Effect of Culture on Financial Explanation: Before and After UFRS, Unpublished Doctorate Thesis, Başkent University SBE, Ankara.
- Al-Modahki, J. S. (1996). An Empirical Study of Accounting Disclosure Development in the Kingdom of Saudi Arabia, Doctorate Thesis, Exeter University, England
- Al-Shammari, B. (2011). “Factors Influencing the Extent of Mandatory Compliance withInternational Financial Reporting Standards: The Case of Kuwaiti Listed Companies”, Journal of International Business and Economics, vol. 11, no. 4: 11-31
- Cooke, T. E. (1989). Voluntary Corporate Disclosure By Swedish Companies. Journal of International Financial Management & Accounting, 1(2), 171-195.
- Cooke, T. Wallace, R. (1989). ‘Global Surveys Of Corporate Disclosure Practices And Audit Firms: A Review Essay’, Accounting And Business Research, Vol. 20, No. 77, Pp. 47- 57.
- Çürük, T. (2004). A Practical Study on the Effect of Independent External Audit on AccountingTransparency: Turkey as an Example. ODTÜ Development Periodical. Volume 31. Number: 2. s. 149-165.
- Demir, V. Bahadir, O. (2014). An investigation of compliance with International FinancialReporting Standards by listed companies in Turkey. Accounting and Management Information Systems, 13(1), 4.
- Esen, Ö. Sakin, T. (2009). ‘‘Compliance with Footnote Explanation Obligations: A study on theStatus of Real Sector Firms included in İMKB 100 Index’, Management Periodical, Year:20, Number:64, October, ss.112-124.
- Glaum, M. Street, D. L. (2003). “Compliance with the Disclosure Requirements of Germany’s NewMarket: IAS Versus US GAAP”, Journal of International Financial Management and Accounting, vol. 14, no. 1: 64-100
- Hadri, K. (2000). “Testing for Stationarity in Heterogeneous Panel Data”, Econometric Journal, 3,pp. 148-161,http://onlinelibrary.wiley.com/doi/10.1111/1368423X.00043/epdf?r3_referer=wol&tracking_action =preview_click&show_chec
- kout=1&purchase_referrer=www.google.com.tr&purchase_site_license=LICEN SE_DENIED (05.03.2016).
- Hausman, J. A. (1978). Specification tests in econometrics. Econometrica, 46(6):1251– 1272, http://www.econ.uiuc.edu/~econ536/Papers/hausman78.pdf (05.02.2016).
- Hope, O.K. (2003). Firm-Level Disclosures and the Relative Roles of Culture and Legal Origin. Journal of International Financial Management and Accounting, 14(3):218-248.
- Juhmani, O. I. H. (2012). “Factors Influencing the Extent of Corporate Compliance with IFRSs:Evidence from Companies Listed in Bahrain Stock Exchange”, Journal of International Business and Economics, vol. 12, no. 2: 67-79
- Raithatha, M. Bapat, V. (2014). Accounting Standards Compliance: Comparison betweenManufacturing and Service Sector Companies from India. International Journal of Economics and Finance, 6(9), 158, file:///C:/Users/hp/Downloads/39754-136122-1-SM.pdf (28.03.2016).
- Rajhi, M. (2014). International Financial Reporting Standards and Level of Compliance withMandatory Disclosure Requirements, International Journal Of Multidisciplinary Sciences AndEngineering, Vol. 5, No. 5, May, http://www.ijmse.org/Volume5/Issue5/paper1.pdf (28.03.2016)
- Soh, C.S. (1996). Factors Influencing Company Information Disclosure: An Empirical Study of Companies Listed on the KLSE, Unpublished PhD. Thesis, University of Warwick, UK
- Sümer, K. K. (2006). Model Forecasting under Heteroskedasticity via White’s HeteroskedasticityConsistent Covariance Metrix Estimation İstanbul University Economics FacultyEconometry and Statistics Periodical, 4, 12-28, http://eidergisi.istanbul.edu.tr/sayi4/iueis4m2.pdf (25.04.2016).
- Tatoğlu, Y. F. (2013). Advanced Panel Data Analysis, with Advanced Application, Beta Publication, İstanbul.
- Ulusan, H. (2007). ‘‘An Empirical Study on the Economic Determinants concerning the Selectionof Accounting Policy’’, Ankara University Political Science Faculty Periodical Volume: 62, No: 2, ss. 167-198, http://dergiler.ankara.edu.tr/detail.php?id=42 (20.02.2016).
- Uluyol, O.Türk, E.V.(2013). “The Impact of Financial Ratios on Firm Values: An Implementationin Stock Market İstanbul (BİST)”, Afyon Kocatepe University, İİBF Periodical, Volume XV, Number II
- Wallace, R.S. Naser, K. Mora, A. (1994). “The Relationship between the Comprehensivensess ofCorporate Annual Reports and Firm Characteristics in Spain.Accounting and Business Research,” 97: 41-53
