FACTA UNIVERSITATIS Series: Economics and Organization Vol. 15, N o 1, 2018, pp. 29 - 41 https://doi.org/10.22190/FUEO1801029M © 2018 by University of Niš, Serbia | Creative Commons Licence: CC BY-NC-ND Original Scientific Paper KEY PREDICTORS OF INTERNAL SUPERVISION OF FINANCIAL STATEMENTS 1 UDC 657.6 Sunčica Milutinović 1 , Ivana Medved 1 , Dragomir Dimitrijević 2 1 University of Novi Sad, Faculty of Economics in Subotica, Serbia 2 University of Kragujevac, Faculty of Economics, Serbia Abstract. Investors and other external stakeholders have more confidence in the reliability and quality of financial statements if they have information that entities continuously carry out preventive and comprehensive internal supervision of financial statements. The research was carried out with the aim of identifying factors that have a direct impact on the internal supervision of financial statements. As the key predictors of the internal supervision of financial statements in the Republic of Serbia, the category of entity, the internal supervision of the compliance of financial statements with the regulations, and the internal supervision of the preparation of financial statements have been identified. Two regression models are also defined that best describe the relationship between the observed phenomena. Key words: Supervision, financial statements, compliance, preparation, regulations. JEL Classification: M41, M42, M48. INTRODUCTION Interpretation of supervision is best described by Isaac (1951, 7-9), who after making an etymologically sharp demarcation between the terms of audit, supervision, surplus and inspection, concluded that they have to complement each other in order to achieve adequate supervision of entity. Therefore, Isaac incorporates audit, supervision and inspection under the concept of "supervision". In this paper, supervision will be observed under this comprehensive term, but more closely monitored – from the aspect of supervision of financial statements. Received August 10, 2017 / Revised November 11, 2017 / Accepted November 20, 2017 Corresponding author: Sunčica Milutinović University of Novi Sad, Faculty of Economics in Subotica, Segedinski put 9-11, 24000 Subotica, Serbia E-mail: suncica@ef.uns.ac.rs 30 S. MILUTINOVIĆ, I. MEDVED, D. DIMITRIJEVIĆ Although the supervision of financial statements significantly improves their quality and supervision is the best basis for risk management in the financial reporting domain, absolute assurance in the reliability and quality of financial statements is not possible, because supervision is a process whose planning, creation, implementation, monitoring and reporting is under the influence of a human factor. Since people make the most important element of the supervision environment, assurances can be qualified exclusively as "reasonable" (Dmitrović & Milutinović, 2011). This is the initial deficiency of the supervision over financial statements. Within the framework of this paper, internal supervision (of the preparation of financial statements, and of the compliance of financial statements with regulations) is the process carried out by internal bodies in entity, such as supervisory boards, internal auditors, and others. In the scope of this paper, external supervision (of the preparation of financial statements, and of the compliance of financial statements with regulations) is the process carried out by the government's supervisory and regulatory bodies in the field of accounting, which include: Department for Financial System of the Ministry of Finance, The National Commission for Accounting, and Tax Administration (Milutinović & Medved, 2017, pp. 102-104). External supervision in the questionnaire is defined as an oversight performed by an external body outside entity, without being an external auditor. The research with the aim of identifying the predictors of internal supervision of financial statements in the Republic of Serbia, according to the authors’ knowledge, has not been carried out so far in domicile conditions. The specific goal of the research in this paper could be set in the form of two research questions: What factors predict the probability that internal bodies in entity will monitor the preparation of financial statements? What factors predict the probability that internal bodies in entity will monitor the compliance of financial statements with regulations? The results of the research are significant for the reason that the creators of the national financial reporting system and the regulatory framework can be in this way informed about the "situation on the ground" through the results presented in this paper. Timely feedback from practice is a key determinant of well-created both legislation and internal regulation, and this research is not significant only for the above-mentioned creators, but also for members of the accounting profession, supervisory authorities and management bodies in entity. Originality of paper is reflected in the fact that similar studies have not yet been implemented in domestic practice. The importance of the research is undeniable, given the fact that the area of financial reporting is very dynamic in terms of frequent changes of regulations, both professional and legislation regulation. In order to timely observe novelties and effects of changes in this area, constant research is necessary, especially in the domestic practice as it unjustifiably lacks this type of research. On the other hand, research results are an important source of information before the planned improvement of the financial reporting system and the improvement of existing regulations in the Republic of Serbia. The limitation of the paper of an objective nature is the lack of similar research in the country and the region in order to do parallel analysis and get a wider picture of this problem. The limitation of the paper of a subjective nature is a limited number of initial factors that have been taken into account in the research in order to identify the predictors. Key Predictors of Internal Supervision of Financial Statements 31 1. LITERATURE REVIEW The responsibility for the preparation and publication of financial statements, in accordance with the Directive 2013/34/EU (Official Journal of the EU, 2013), is on members of administrative, management and supervisory bodies of entity in the form of the so-called collective responsibility. On the other hand, according to the Law on Accounting (Official Gazette of RS, 2013), the legal representative, the managing authority, the supervisory body and the person responsible for keeping business books and drawing up financial statements are responsibility holders in domicile conditions. Taking into account that both in professional and legislation regulation responsibility holders are the supervisory authorities, the focus in the paper is on the internal supervision – supervision carried out by the internal bodies of entity – to the internal supervision of the preparation of financial statements, on the one hand, and the internal supervision of the compliance of financial statements with the regulations, on the other hand. There were several studies in past four years in Asia about factors affecting internal audit or internal control over financial reporting. Alzeban and Gwilliam (2014) investigated factors that may influence internal audit effectiveness within public sector organizations in Saudi Arabia, a nation just beginning to use internal audits. Their study employs several measures of internal audit effectiveness, including the department’s perceived ability to plan, to improve the organization’s productivity, to assess the consistency of results with established objectives and goals, to implement internal audit recommendations, to evaluate and improve risk management, to evaluate internal control systems, and to provide recommendations for improvement. Furthermore, their study provides evidence as to the manner in which competence, size of the internal audit department, relationship between internal and external auditors, independence of internal audit, and extent of management support for the internal audit function, contribute to perceived effectiveness of the internal audit function in the Saudi public sector. Dejnaronk, Little, Mujtaba and McClelland (2015) were the first ones that examined the factors influencing the effectiveness of internal audit function in Thailand. Their study included seven influencing variables, which were useful in Thai companies: objectivity, independence, top management support, risk-based audit plan, compliance with IIA standards (International Standards for the Professional Practice of Internal Auditing by The Institute of Internal Auditors), communication and reporting, and quality assurance and improvement program. Three influencing factors found are compliance with IIA standards, risk-based audit plan, and quality assurance and improvement program. Yazawa (2015) examined the incentive factors for the (non-)disclosure of material weakness in internal control over financial reporting in Japanese companies. The results indicate that companies that do not disclose material weakness are characterized by longer management tenure, large auditors, lower audit fees, larger boards of directors, fewer outside directors, and greater involvement of their main banks. In addition, non-disclosure of material weakness at the company level is associated with longer management tenure, a larger management shareholding, and greater main bank involvement, whereas non-disclosure of material weakness at the account-specific level is associated with large auditors, lower audit fees, higher non-audit fees, and greater main bank involvement. These results suggest that the assessment and audit process of internal control systems in Japan is sensitive to (non-)disclosure incentives for managements and auditors. 32 S. MILUTINOVIĆ, I. MEDVED, D. DIMITRIJEVIĆ Some of the studies conducted in America are summarized below. Gramling, O’Donnell and Vandervelde (2013) examined how different types of audit evidence accumulated during the audit influence auditor judgment of internal control over financial reporting operating effectiveness and about whether an identified control deficiency is a material weakness in internal control over financial reporting. They experimentally examined audit managers’ and partners’ assessments of internal control over financial reporting operating effectiveness and judgments of whether a control deficiency is a material weakness to determine the influence of the presence of: (1) a material weakness unrelated to the deficiency being assessed, and (2) a known misstatement associated with the identified control deficiency. Results suggest that the presence of either an unrelated material weakness or a known misstatement influences the assessed operating effectiveness of an internal control and the likelihood of a material weakness assessment. Lin, Wang, Chiou and Huang (2014) find that CEOs with higher ownership, longer tenure, and younger age are significantly associated with a material internal control weakness disclosure under Sarbanes-Oxley Section 404 (SOX 404). Their results demonstrate that entrenchment and age may affect CEOs’ behavior in response to the SOX 404 internal control requirements. This study points out that CEO entrenchment and age are likely to affect the strength of internal control mechanisms. Christ, Masli, Sharp and Wood (2015) find that companies that use a rotational staffing model for the internal audit function (systematically rotate their internal auditors into management positions outside of internal audit) have significantly lower financial reporting quality than companies that do not. Clinton, Pinello and Skaife (2014) investigated consequences of ineffective internal control for an important class of financial statement users and suggested the required reporting on the effectiveness of internal control is beneficial to understanding the properties of analysts’ forecasts. They find that analysts provide less accurate forecasts and there is greater forecast dispersion for firms with ineffective internal control. They also find that firms that disclose internal control problems have less analyst coverage and that analyst following declines after the material weakness in internal control is disclosed. The latest research was conducted in Africa. Bruwer, Coetzee and Meiring (2017) determined the empirical relationships that exist between two of the elements of a sound internal control system, namely internal control activities and managerial conduct, and the perceived sustainability of South African small, medium and micro enterprises (SMMEs). They find out that there exist positive statistically significant relationships between general management skills and the perceived sustainability of South African SMMEs. Study indicates that internal control activities and managerial conduct had a limited influence on the perceived sustainability of South African SMMEs. 2. RESEARCH METHODOLOGY AND HYPOTHESES The research was directed at all employees in the accounting and/or auditing positions in the private sector of the Republic of Serbia, regardless of their professional title and years of experience. The sample meets all the requirements of a representative sample and consists of 338 (first regression model) or 336 respondents (second regression model). A random sample in the research is considered an ideal sample. The sample in this study was selected randomly, because every member of the basic set had the same chances, i.e. the same probability of being selected into the sample. In addition, the selection of each Key Predictors of Internal Supervision of Financial Statements 33 member of the sample was independent, i.e. the selection of one member did not influence the selection of any other member of the sample. The study was conducted during 2014- 2015 and the research was completed in 2016. The research faced the same problems that usually appear in research of financial reporting: research cannot be conducted in conditions of a supervision led experiment, it is impossible to include all variables and identify all the factors that influence the research problem, the research sample is often small, and any research, including one in the area of financial research, comprises mistakes, both those that can be predicted before the start of the research and mistakes that cannot be predicted. However, these problems have not significantly affected the independence and regularity of the research, nor the credibility and presentation capacity of the obtained results. Two hypotheses have been tested in this paper. The first hypothesis (H1) that is the subject of proof in the paper reads: Internal supervision of the preparation of financial statements (ISPFS) is directly influenced by the category of entity, external and internal supervision of compliance of financial statements with regulations and external supervision of the preparation of financial statements. The first hypothesis will be proved using the following auxiliary hypotheses: H1a: The category of entity has a direct negative impact on internal supervision of the preparation of financial statements. H1b: Performing external supervision of compliance of financial statements with regulations has a direct positive impact on internal supervision of the preparation of financial statements. H1c: Performing internal supervision of compliance of financial statements with regulations has a direct positive impact on internal supervision of the preparation of financial statements. H1d: Performing external supervision of the preparation of financial statements has a direct positive impact on internal supervision of the preparation of financial statements. The second hypothesis (H2) that is the subject of proof in the paper reads: Internal supervision of the compliance of financial statements with regulations is directly influenced by the category of entity, external and internal supervision of the preparation of financial statements and external supervision of the compliance of financial statements with regulations. The second hypothesis will be proved using the following auxiliary hypotheses: H2a: The category of entity has a direct negative impact on internal supervision of the compliance of financial statements with regulations. H2b: Performing external supervision of the preparation of financial statements has a direct positive impact on internal supervision of the compliance of financial statements with regulations. H2c: Performing internal supervision of the preparation of financial statements has a direct positive impact on internal supervision of the compliance of financial statements with regulations. H2d: Performing external supervision of the compliance of financial statements with regulations has a direct positive influence on internal supervision of the compliance of financial statements with regulations. To test hypotheses in order to prove them, direct logistic regression will be used. Then, regression models that best describe the relationship between phenomena will be determined. Based on these models, values of dependent variables (ŷ1, ŷ2) will be evaluated and predicted for selected values of predictor (independent) variables (x1... xn). Dependent variables are: ŷ1 – internal supervision of preparation of financial statements (ISPFS hereinafter) and ŷ2 – internal 34 S. MILUTINOVIĆ, I. MEDVED, D. DIMITRIJEVIĆ supervision of compliance of financial statements with regulations (ISCFSR hereinafter). Predictor variables that were taken into consideration are: category of entity, external/internal supervision of compliance of financial statements with regulations and external/internal supervision of preparation of financial statements. 3. FINDINGS AND DISCUSSION 3.1. Predictors of internal supervision of preparation of financial statements Table 1 shows the predictors included in the first model as well as the values that are taken with the frequencies. Number of analyzed cases is 338. Table 1 Categorical Variables Codings #1 Frequency Parameter coding (1) (2) (3) (4) (5) The category of entity Large IFRS 38 .000 .000 .000 .000 .000 Medium-sized IFRS 44 1.000 .000 .000 .000 .000 Medium IFRS SMEs 15 .000 1.000 .000 .000 .000 Small IFRS SMEs 74 .000 .000 1.000 .000 .000 Small RB 2 29 .000 .000 .000 1.000 .000 Micro 138 .000 .000 .000 .000 1.000 External supervision of compliance of FS with regulations (ESCFSR) No 263 .000 Yes 75 1.000 Internal supervision of compliance of FS with regulations (ISCFSR) No 205 .000 Yes 133 1.000 External supervision of the preparation of FS (ESPFS) No 263 .000 Yes 75 1.000 Source: authors’ calculation The Goodness of Fit Test in Table 2 shows how well the model predicts results. The obtained results are very significant, since the significance takes values p<0.0005. The model with the selected set of variables shown in Table 1 provides better than the initial model (without any independent variables) obtained assuming that ISPFS is never performed. The hi- square indicator is 236.57 with 8 degrees of freedom. Table 2 Omnibus Tests of Model Coefficients #1 Chi-square df Sig. Step 1 Step 236.566 8 .000 Block 236.566 8 .000 Model 236.566 8 .000 Source: authors’ calculation 2 Rulebook on the Manner of Recognition, Measurement, Presentation and Disclosure of Positions in the Individual Financial Statements of Micro and Other Entities, Official Gazette of RS, No. 118 of December 30 th 2013, 95 of September 5 th 2014 (RB hereinafter). Key Predictors of Internal Supervision of Financial Statements 35 The Hosmer and Lemeshow Goodness of Fit Test in Table 3 is IBM SPSS's most reliable model prediction quality test. The model is supported because significance takes values p>0.05. The H-square for Hosmer and Lemeshow Test is 2.43 with a significance of 0.88. Table 3 Hosmer and Lemeshow Test #1 Step Chi-square df Sig. 1 2.428 6 .876 Source: authors’ calculation Table 4 shows the values of Cox&Snell R Square and Nagelkerke R Square. The model, or the set of variables, explains between 50.3% and 67.8% of the variance of the dependent variable. Table 4 Model Summary #1 Step -2 Log likelihood Cox&Snell R Square Nagelkerke R Square 1 221.294 .503 .678 Source: authors’ calculation Table 5 shows that the model, or the set of variables, correctly classifies 89.3% of all cases that is better than 58.9%, which represents value of this indicator before including predictor variables in the model. Positive and negative predictive values are calculated on the basis of data in Table 5. The positive predictive value shows that the model correctly selected 88.72% of respondents who answered that ISPFS performs in entity where they are employed. The negative predictive value shows that the model correctly selected 89.76% of respondents who answered that ISPFS does not perform in entity where they are employed. Table 5 Classification Table #1 Step Observed Predicted Internal supervision of preparation of FS Percentage Correct No Yes 1 Internal supervision of preparation of FS No 184 15 92.5 Yes 21 118 84.9 Overall Percentage 89.3 Source: authors’ calculation Table 6 shows a variable that significantly contributes to predictive capabilities of the model. The main factor influencing whether ISPFS will be performed is ISCFSR. Entities that perform ISCFSR are most likely to have ISPFS also. 36 S. MILUTINOVIĆ, I. MEDVED, D. DIMITRIJEVIĆ Table 6 Variables in the Equation #1 β S.E. Wald df Sig. Exp(β) 95% C.I.for EXP(β) Lower Upper S te p 1 Large IFRS -.443 .728 .370 1 .543 .642 .154 2.675 Medium-sized IFRS -.213 .949 .051 1 .822 .808 .126 5.186 Medium IFRS SMEs -.411 .666 .380 1 .538 .663 .180 2.449 Small IFRS SMEs -.176 .818 .047 1 .829 .838 .169 4.165 Small RB -.856 .629 1.856 1 .173 .425 .124 1.456 ESPFS .295 .632 .218 1 .641 1.343 .389 4.635 ESCFSR .699 .633 1.220 1 .269 2.012 .582 6.960 ISCFSR 4.021 .365 121.32 1 .000 55.765 27.265 114.056 Constant -1.792 .599 8.943 1 .003 .167 Source: authors’ calculation Based on the obtained values of the β coefficient, the regression equation of the first model is: 11 021.4792.1ˆ xy  (1) Legend: ŷ1 – the expected value of the dependent variable (ISPFS) x1 – predictive variable (ISCFSR) A direct logistic regression was conducted to assess the impact of several factors on the likelihood that ISPFS performs in entity. The model contained four predictors (independent variables) and was statistically significant, c² (8, N=338) = 236.57, p<0.001, indicating that the model distinguishes respondents who responded and those who did not respond that ISPFS performs in entity where they are employed. The model explains as a whole between 50.3% (Cox&Snell’s r²) and 67.8% (Nagelkerke’s r²) variance of dependent variables and accurately classifies 89.3% of cases. The strongest predictor of the responses that there is the performance of ISPFS in entity (see Table 6) was the performance of ISCFSR, whose quotient of probability is 55.77. It shows that respondents from entity that performs ISCFSR over 55 times more often answer with “yes” on the question about the performance of ISPFS in contrast to respondents from entity that does not perform ISCFSR, with all other equal factors in the model. 3.2. Predictors of internal supervision of the compliance of financial statements with regulations Table 7 shows the predictors included in the second model as well as the values that are taken with the frequencies. Number of analyzed cases is 336. The Goodness of Fit Test in Table 8 shows how well the model predicts results. The obtained results are very significant (p<0.0005) and the model with selected set of variables shown in Table 7 predicts better than the initial model (without any independent variables) obtained assuming that ISCFSR is never performed. The hi-square indicator is 248.54 with 8 degrees of freedom. Key Predictors of Internal Supervision of Financial Statements 37 Table 7 Categorical Variables Codings #2 Frequency Parameter coding (1) (2) (3) (4) (5) The category of entity Large IFRS 36 .000 .000 .000 .000 .000 Medium-sized IFRS 44 1.000 .000 .000 .000 .000 Medium IFRS SMEs 15 .000 1.000 .000 .000 .000 Small IFRS SMEs 74 .000 .000 1.000 .000 .000 Small RB 29 .000 .000 .000 1.000 .000 Micro 138 .000 .000 .000 .000 1.000 External supervision of the preparation of FS (ESPFS) No 263 .000 Yes 73 1.000 Internal supervision of the preparation of FS (ISPFS) No 199 .000 Yes 137 1.000 External supervision of compliance of FS with regulations (ESCFSR) No 263 .000 Yes 73 1.000 Source: authors’ calculation Table 8 Omnibus Tests of Model Coefficients #2 Chi-square df Sig. Step 1 Step 248.54 8 .000 Block 248.54 8 .000 Model 248.54 8 .000 Source: authors’ calculation The model is supported (p>0.05) according to results from Table 9. The H-square for Hosmer and Lemeshow Test is 2.84 with a significance of 0.9. Table 9 Hosmer and Lemeshow Test #2 Step Chi-square df Sig. 1 2.837 7 .900 Source: authors’ calculation The model explains between 52.3% and 70.8% of the variance of the dependent variable (see Table 10). Table 10 Model Summary #2 Step -2 Log likelihood Cox&Snell R Square Nagelkerke R Square 1 202.565 .523 .708 Source: authors’ calculation Table 11 shows that the model correctly classifies 89.9% of all cases, that is better than 60.4%, which represents value of this indicator before including predictor variables in the model. Positive and negative predictive values are calculated on the basis of data in Table 11. The positive predictive value shows that the model correctly selected 86.13% of respondents who answered that ISCFSR performs in entity where they are employed. The 38 S. MILUTINOVIĆ, I. MEDVED, D. DIMITRIJEVIĆ negative predictive value shows that the model correctly selected 92.46% of respondents who answered that ISCFSR does not perform in entity where they are employed. Table 11 Classification Table #2 Step Observed Predicted Internal supervision of compliance of FS with regulations Percentage Correct No Yes 1 Internal super. of compliance of FS with regulations No 184 19 90.6 Yes 15 118 88.7 Overall Percentage 89.9 Source: authors’ calculation Table 12 shows variables that significantly contribute to predictive capabilities of the model. The main factors influencing whether ISCFSR will be performed are:  The category of entity: the change in the category of entity for one (from larger to smaller entity, or from the one that uses IFRS to the other that does not use IFRS) reduces the likelihood of performing ISCFSR. The smaller entity has lower likelihood of performing ISCFSR, and vice versa. Entities that use IFRS have higher likelihood of performing ISCFSR than those which do not use IFRS.  Internal supervision of the preparation of financial statements: entities that perform ISPFS are most likely to perform ISCFSR. It means, respondents from entities with ISPFS answer to the greatest extent with "yes" on the question of whether ISCFSR is performed. Table 12 Variables in the Equation #2 β S.E. Wald df Sig. Exp(β) 95% C.I.for EXP(β) Lower Upper S te p 1 Large IFRS -1.592 .785 4.114 1 .043 .204 .044 .948 Medium-sized IFRS -2.075 1.019 4.143 1 .042 .126 .017 .926 Medium IFRS SMEs -1.355 .710 3.642 1 .056 .258 .064 1.037 Small IFRS SMEs -1.545 .867 3.175 1 .075 .213 .039 1.167 Small RB -1.882 .676 7.757 1 .005 .152 .040 .573 ESCFSR .777 .646 1.448 1 .229 2.175 .614 7.712 ESPFS .361 .650 .308 1 .579 1.434 .401 5.130 ISPFS 4.173 .382 119.61 1 .000 64.912 30.728 137.124 Constant -1.182 .607 3.785 1 .052 .307 Source: authors’ calculation Based on the obtained values of the β coefficient, the regression equation of the second model is: 43212 173.4882.1075.2592.1182.1ˆ xxxxy  (2) Legend: ŷ2 – the expected value of the dependent variable (ISCFSR) x1, x2, x3, x4 – predictive variables from Table 12. Key Predictors of Internal Supervision of Financial Statements 39 A direct logistic regression was conducted to assess the impact of several factors on the likelihood that ISCFSR performs in entity. The model contained four predictors (independent variables) and was statistically significant, c² (8, N=336) = 248.54, p<0.001, indicating that the model distinguishes respondents who responded and those who did not respond that ISCFSR performs in entity where they are employed. The model explains as a whole between 52.3% (Cox&Snell’s r²) and 70.8% (Nagelkerke’s r²) variance of dependent variables and accurately classifies 89.9% of cases. The strongest predictor of the responses that there is the performance of ISPFS in entity (see Table 12) was the performance of ISCFSR, whose quotient of probability is 64.91. It shows that respondents from entity that performs ISPFS almost 65 times more often answer with “yes” on the question about the performance of ISCFSR in contrast to respondents from entity that does not perform ISPFS, with all other equal factors in the model. CONCLUSION Both regression models are supported, statistically significant and accurately classified over 89% of cases. The variable that significantly contributes to the predictive possibilities of the first regression model is the internal supervision of compliance of financial statements with regulations, while the remaining three variables do not have a significant contribution to the model. Variables that significantly contribute to the predictive possibilities of the second regression model are the size of an entity and internal supervision of preparation of financial statements, while the remaining two variables do not have a significant contribution to the model. The above results provide an answer to the research questions raised at the beginning of this paper. The key factor that predicts the probability that internal bodies within an entity will perform supervision of preparation of financial statements is performance of internal supervision of compliance of financial statements with regulations. Key factors that predict the probability that internal bodies within an entity will perform supervision of compliance of financial statements with regulations are the category of entity and performance of internal supervision of preparation of financial statements. Based on the analyzed results of the dependent variable ŷ1, it can be concluded that the internal supervision of preparation of financial statements is under direct negative influence of the category of entity (the change in the category of entity for one – from the one that uses IFRS to the other one that does not use IFRS – reduces the likelihood of performing ISPFS), direct positive influence of both external and internal supervisions of compliance of financial statements with regulations and direct positive influence of external supervision of preparation of financial statements. According to the above, the first hypothesis was confirmed in its entirety. Based on the analyzed results of the dependent variable ŷ2, it can be concluded that the internal supervision of compliance of financial statements with regulations is under direct negative influence of the category of entity (the change in the category of entity for one – from the one that uses IFRS to the other one that does not use IFRS – reduces the likelihood of performing ISCFSR), direct positive influence of both external and internal supervisions of preparation of financial statements and direct positive influence of external supervision of compliance of financial statements with regulations. According to the above, the second hypothesis was confirmed in its entirety. Such outcomes have resulted in the following conclusions. 40 S. MILUTINOVIĆ, I. MEDVED, D. DIMITRIJEVIĆ Regulators and lawmakers should form team of experts constituted of professional accountants and auditors who would create a quality legal framework and develop instruments for the financial reporting process supervision. The Sarbanes-Oxley Act represents an excellent example of successful state regulation of financial reporting. In the Republic of Serbia, state supervision in this area is desirable, but it must be carefully thought out in order to avoid overlapping and interfering with the competencies of state authorities and internal bodies in entities. The first preferred solution would be to define a model by which the management of entity would be entrusted with the obligation to perform an assessment of the effectiveness of internal supervision, and that this assessment is subsequently verified by the external supervise bodies – Department for Financial System of the Ministry of Finance, The National Commission for Accounting and Tax Administration. The second preferred solution would be the introduction of practice of public disclosure on the functioning and success of the internal supervision system in entity, as has long existed in the countries of the Anglo-Saxon accounting model. This solution would increase the transparency of financial reporting, which would have a long-term impact on increasing its quality. On the other hand, the research results have shown that entities applying professional accounting regulations tend to organize internal supervision of compliance of financial statements with regulations, in contrast to entities that do not use IFRSs or IFRS for SMEs. This leads to the proposal of the third preferred solution that is the continuation of the convergence of legislation with professional regulation and the work on dissemination of the application of IFRSs and IFRS for SMEs among entities. Our results should be of interest to management, professional accountants and auditors, state regulators, lawmakers, standard setters, external supervise bodies and internal supervise bodies. Our research provides insights into key factors that management and internal supervise bodies are likely considering when assessing the need for implementation of internal supervision. After the expected changes in the legislation regulation, the research is planned to be repeated. The aim of the future research will be to obtain new regression models in order to test existing regression models for the purpose of passing a conclusion on the success of the implemented changes in the legal regulations in the Republic of Serbia. If the findings demonstrated in this research are in contrast to what was intended by the state regulators and lawmakers, they may want to consider the results of repeated research in order to test new legal regulations. REFERENCES Alzeban, A. & Gwilliam, D. (2014). Factors Affecting the Internal Audit Effectiveness: A Survey of the Saudi Public Sector. Journal of International Accounting, Auditing and Taxation. 23 (2014). 74–86. DOI: 10.1016/j.intaccaudtax.2014.06.001 Bruwer, J-P., Coetzee, P. & Meiring, J. (2017). Can Internal Control Activities and Managerial Conduct Influence Business Sustainability? A South African SMME Perspective. Journal of Small Business and Enterprise Development. DOI: 10.1108/JSBED-11-2016-0188 Christ, M.H., Masli, A., Sharp, N.Y. & Wood, D.A. (2015). Rotational Internal Audit Programs and Financial Reporting Quality: Do Compensating Controls Help? Accounting, Organizations and Society. 44 (2015). 37–59. DOI: 10.1016/j.aos.2015.05.004 Clinton, S.B., Pinello, A.S. & Skaife, H.A. (2014). The Implications of Ineffective Internal Control and SOX 404 Reporting for Financial Analysts. Journal of Accounting and Public Policy. 33 (2014). 303–327. DOI: 10.1016/j.jaccpubpol.2014.04.005 Key Predictors of Internal Supervision of Financial Statements 41 Dejnaronk, J., Little, H.T., Mujtaba, B.G. & McClelland, R. (2015). Factors Influencing the Effectiveness of the Internal Audit Function in Thailand. Proceedings of Conference: Business and Social Sciences Research Conference: Research for Development, Bangkok, Thailand, 10-11 th December 2015. ISBN: 978-1-922069-90-0 Dmitrović, Lj. & Milutinović, S. (2011). The Control Mechanisms in Accounting. Strategic Management. 16 (2). Gramling, A.A., O’Donnell, E.F. & Vandervelde, S.D. (2013). An Experimental Examination of Factors That Influence Auditor Assessments of a Deficiency in Internal Control over Financial Reporting. Accounting Horizons. 2 (27). 249–269. DOI: 10.2308/acch-50410 Isaac. A. (1951). Revision und wirtschaftsprüfung. Wiesbaden: Gabler Verlag. Lin, Y.Ch., Wang, Y.Ch., Chiou, J.R. & Huang, H.W. (2014). CEO Characteristics and Internal Control Quality. Corporate Governance: An International Review. 1 (22). 24–42. DOI: 10.1111/corg.12042 Milutinović, S. & Medved, I. (2017). Supervision of Financial Reporting Through the Prism of the Regulatory Framework and Practice in the Republic of Serbia. Industrija. 45 (1). 99-120. DOI: 10.5937/industrija45- 11653 Official Gazette of RS (2013). Accounting Law. No. 62/2013. Belgrade: Author. Official Gazette of RS (2014). Rulebook on the Manner of Recognition, Evaluation, Presentation and Disclosure of Positions in the Individual Financial Statements of Micro and Other Entities. No. 118/2013, 95/2014. Belgrade: Author. Official Journal of the EU (2013). Directive 2013/34/EU of the European Parliament and of the Council of 26 June 2013. Vol. 56. Brisel: Author. doi:10.3000/19770677.L_2013.182.eng Yazawa, K. (2015). The Incentive Factors for the (Non-)Disclosure of Material Weakness in Internal Control over Financial Reporting. International Journal of Auditing. 19 (2015). 103–116. DOI: 10.1111/ijau.12035 KLJUČNI PREDIKTORI INTERNOG NADZORA FINANSIJSKIH IZVEŠTAJA Investitori i ostali eksterni stejkholderi imaju više poverenja u pouzdanost i kvalitet finansijskih izveštaja ukoliko raspolažu informacijama da se u pravnom licu kontinuirano obavlja preventivni i sveobuhvatni interni nadzor finansijskih izveštaja. Istraživanje je izvršeno s ciljem da se prepoznaju faktori koji imaju direktni uticaj na odvijanje internog nadzora finansijskih izveštaja. Kao ključni prediktori internog nadzora finansijskih izveštaja u Republici Srbiji identifikovani su kategorija pravnog lica, sprovođenje internog nadzora usklađenosti finansijskih izveštaja sa regulativom i sprovođenje internog nadzora pripreme finansijskih izveštaja. Definisana su i dva regresiona modela koja najbolje opisuju vezu između posmatranih pojava. Ključne reči: nadzor, finansijski izveštaji, usklađenost, priprema, regulativa.