Auditing in a Computerized Environment PDF

CHAPTER 7 AUDITING IN A COMPUTERIZED ENVIRONMENT With the rapid development in technology n recent years, computer information systems (CIS) have become feasible, perhaps essential, for use even in small scale business operations. Almost all entities now use computers to some extent...

CHAPTER 7 AUDITING IN A COMPUTERIZED ENVIRONMENT With the rapid development in technology n recent years, computer information systems (CIS) have become feasible, perhaps essential, for use even in small scale business operations. Almost all entities now use computers to some extent in their accounting systems. This widespread use of computers has offered new opportunities for professional accountants and has also created some challenging problems to auditors. Regardless of the extent of computerization or the methods of data processing being used, the establishment and implementation of appropriate internal control systems rests with management and those charged with governance. The auditor’s responsibility is to obtain an understanding of the entity’s internal control system to be able to assess control risk and determine the nature, timing and extent of tests to be performed.  Characteristics of Computer Information Systems (CIS) Computer information systems have essential characteristics that distinguish them from manual processing systems.  Lack of visible transaction trails In a manual system, it is normally possible to follow a transaction through the system by examining the source documents, entity’s records and financial reports. In a CIS environment, data can be entered directly into the computer system without supporting documents. Furthermore, records and files may not be printed and can not be read without using the computer. The absence of these visible documents supporting the processing of transactions makes the examination of evidence more difficult.  Consistency of Performance CIS performs functions exactly as programmed. If the computer is programmed to perform a specific data processing task, it will never get tired of performing the assigned task in exactly the same manner. Because of this capability of the computer to process transactions uniformly, clerical errors that are normally associated with manual processing are eliminated. On the other hand, an incorrect program could be very devastating because it will result to consistently erroneous data processing.  Ease of Access to Data and Computer Programs  In a CIS environment, data and computer programs may be accessed and altered by unauthorized persons leaving no visible evidence. It is important, therefore, that appropriate controls are incorporated to the system to limit the access to data files and programs only to authorized personnel.  Segregation of duties Proper segregation of duties is an essential characteristic of a sound internal control system. However, because of the ability of the computer to process data efficiently, there are functions that are normally segregated in manual processing that are combined in a CIS environment. As a particular example, in manual processing the function of recording cash disbursements is incompatible with the responsibility for reconciling disbursements. Since one of these functions serves as a check up the other, assigning both functions to one employee would enable that employee to commit and conceal errors or irregularities. A properly programmed computer, on the other hand, has no tendency or motivation to commit irregularities or conceal its errors. Hence, what appears to be an incompatible combination of functions may be combined in a CIS environment without weakening the internal control provided appropriate compensating controls are put in place.  Systems generated transactions Certain transactions may be initiated by the CIS itself without the need for an input document. For example, interest may be calculated and charged automatically to customer’s account balances on the basis of pre-authorized terms contained in a computer program.  Vulnerability of data and program storage media In a manual system, the records are written in ink on substantial paper. The only way to lose the information is to lose or to destroy physical records. The situation is completely different in a CIS environment. The information on the computer can be easily changed, leaving no trace of the original content. This change could happen inadvertently and huge amount of information can be quickly lost.  Internal Control in a CIS Environment Many of the control procedures used in manual processing also apply in a CIS environment. Examples of such control procedures include authorization of transactions, proper segregation of duties, and independent checking. The elements of internal control are the same; the computer just changes the methods by which these elements are implemented.  A variety of controls are performed to check accuracy, completeness and authorization of transactions. When computer processing is used in significant accounting applications, internal control procedures can be classified into two types: general and application controls.  General Controls General controls are those controls policies and procedures that relate to the overall computer information system. These controls include: 1. Organizational controls Just as in a manual system, there should be a written plan of the organization, with clear assignment of authority and responsibility. In a CIS environment, the plan of the organization for an entity’s computer system should include segregation between the user and CIS department, and segregation of duties within the CIS department. a. Segregation between the CIS department and user departments. CIS department must be independent of all departments within the entity that provide input data or that use output generated by the CIS. The function of CIS is to process transactions. However, no transactions will be processed unless it is initiated by the user department. Therefore, all changes in computer files must be initiated and authorized by the user department. b. Segregation of duties within the CIS department Functions within the CIS department should be properly segregated for good organizational controls. The entity’s organizational structure should provide for definite lines of authority and responsibility within the CIS department. A sample of an organizational structure within the CIS department is presented below:  CIS Director Systems Other Operation Development Functions Computer System Librarian Operator Analyst Data Entry Control Programmer Operator Group Position Primary Responsibilities CIS Director Exercises control over the CIS operation. System Analyst Designs new systems, evaluates and improves existing systems, and prepares specifications for programmers. Programmer Guided by the specifications of the systems analyst, the programmer writes a program, tests and debugs such programs and prepares the computer operating instructions. Computer Operator Using the program and detailed operating instructions prepared by the programmer, computer operator operates the computer to process transactions. Data Entry Operator Prepares and verifies input data for processing. Librarian Maintains custody of systems documentation, programs and files. Control Group Reviews all input procedures, monitors computer processing, follows up data processing errors, reviews the reasonableness of output, and distributes output to authorized personnel. Optimal segregation of duties dictates that each of the above tasks be assigned to different employees. However, some entities may not have enough resources to maintain a large CIS department. In small entities, with limited number of personnel, the functions of systems analyst and programmer may be combined. As a minimum, the functions of system development and computer operations must be segregated. Systems analyst and programmer should not be allowed to use the programs they developed and they should not be allowed to operate the computer. Also, computer operators who run the program should not participate in program design. A number of computer related frauds have resulted when these functions were combined. 2. Systems development and documentation controls Software development as well as changes thereof must be approved by the appropriate level of management and the user department. To ensure that  computer programs are functioning as designed, the program must be tested and modified, if needed, by the user and CIS department. Moreover, adequate systems documentation must be made in order to facilitate the use of the programs as well as changes that may be introduced later into the system. 3. Access Controls Every computer system should have adequate security controls to protect equipment, files and programs. Access to the computer should be limited only to operators and other authorized employees. Additionally, appropriate controls such as the use of passwords must be adopted in order to protect data files and programs from unauthorized alteration. 4. Data recovery controls One of the characteristics of the CIS is the vulnerability of files and programs. Computer files can be easily lost and the lost of these files can be disastrous to an entity. The survival of an entity affected by such disaster depends on its ability to recover the files on a timely basis. A data recovery control provides for the maintenance of back up files and off site storage procedures. Computer files should be copied daily to tape or disks and secured off-site. In the event of disruption, reconstruction of files is achieved by updating the most recent back-up with subsequent transaction data. When magnetic tapes are used, a common practice in file retention called Grand-father, father, son practice requires entity to keep the two most recent generations of master files and transaction files in order to permit reconstruction of master files if needed. 5. Monitoring controls Monitoring controls are designed to ensure that CIS controls are working effectively as planned. These include periodic evaluation of the adequacy and effectiveness of the overall CIS operations conducted by persons within or outside the entity.  Application Controls The processing of transaction involves three stages: the input, processing and output stage. The input stage involves capturing of a mass of data; the processing stage involves converting the mass of raw data into useful information; and output  stage involves preparation of information in a form useful to those who wish to use it. To ensure that all relevant data are captured as input to the system, and to ensure that the data are accurately processed during their conversion into meaningful financial information, controls or other mechanisms must be incorporated into the system. Application controls are those policies and procedures that relate to specific use of the system. These are designed to provide reasonable assurance that all transactions are authorized, and that they are processed completely, accurately and on a timely basis. These include 1. Controls over input A large number of errors in a computer system are caused by inaccurate or incomplete data entry. Input controls are designed to provide reasonable assurance that data submitted for processing are complete, properly authorized and accurately translated into machine readable form. Examples of input controls include: Key verification This requires data to be entered twice (usually by different operators) to provide assurance that there are no key entry errors committed. Field check This ensures that the input data agree with the required field format. For example, all SSS number must contain ten digits. An input of an employee’s SSS number with more or less than ten digits will be rejected by the computer. Validity check Information entered is compared with valid information in the master file to determine the authenticity of the input. For example, the employees’ master file may contain two valid codes to indicate the employee’s gender “1” for male and “2” for female. A code of “3” is considered invalid and will be rejected by the computer. Self-checking digit This is a mathematically calculated digit which is usually added to a document number to detect common transpositional errors in data submitted for processing. Limit check Limit check or reasonable check is designed to ensure that data submitted for processing do not exceed a pre-determined limit or a reasonable amount.  Control totals These are totals computed based on the data submitted for processing. Control totals ensure the completeness of data before and after they are processed. These controls include financial totals, hash totals and record counts. As an example, assume the following data regarding the entity’s disbursements for the day. Voucher No. 142 P 20, 000 Voucher No. 143 P 5,000 Financial total = P 40,000 (P15,000 + P20,000 + 5000) Hash total = 426 (141 + 142 + 143) Record count = 3 2. Controls over processing Processing controls are designed to provide reasonable assurance that input data are processed accurately and that data are not lost, added, excluded, duplicated or improperly changed. Almost all of the input controls that were mentioned earlier are also part of the processing controls because such controls are usually incorporated in the client’s computer program to detect errors in processing of transactions. 3. Control over output  Output controls are designed to provide reasonable assurance that the results of processing are complete, accurate and that these output are distributed only to authorize personnel. A person who knows what an output should look like must review the CIS output for reasonableness. Control totals are compared with those computed prior to processing to ensure completeness of information. Finally, CIS outputs must be restricted only to authorize employees who will be using such output. The effectiveness of the general CIS controls is essential to the effectiveness of CIS application controls. Thus, it may be more efficient to review the design of the general controls first before reviewing the application controls.  Test of Control in a CIS environment Like manual processing environment, test of control in a CIS environment involves evaluating the client’s internal control policies and procedures to determine if they are functioning as intended. Regardless of the nature of the client’s data processing system, auditors must perform tests of controls if they intend to rely on the client’s internal control. The auditor’s objective and scope of the audit do not change in a CIS environment. However, the use of the computer changes the processing and storage of financial information and may affect the organization and procedures employed by the entity to achieve adequate internal control. Accordingly, the methods employed by the auditor in testing the control may also be affected. Testing the reliability of general controls may include observing client’s personnel in performing their duties; inspecting program documentation; and observing the security measures in force. In testing application controls, the auditor ma either: 1) Audi around the computer; or 2) Use Computer-Assisted Audit Techniques Auditing Around the Computer Auditing around the computer is similar to testing control in a manual control structure in that it involves examination of documents and reports and determine the reliability of the system. When using this approach, the auditor ignores the client’s data processing procedures, focusing solely on the input documents and the CIS output. Input data are simply reconciled with the output to verify the accuracy of processing. Auditing around the computer is based on the assumption that if the input reconciles with the output, then the computer program must have processed the transaction accurately. Hence, the auditor obtains knowledge about the reliability of the system without directly examining the computer program of the system. Auditing around the computer can be used only if there are visible input documents and detailed output that will enable the auditor to trace individual transactions back and forth. This is also known as “black box approach” because it does not permit direct assessment of actual processing of transaction.  Computed Assisted Audit Techniques (CAATs) When computer accounting systems perform tasks for which no visible evidence is available, it may be impracticable for the auditor to test manually. Consequently, auditor will have to audit directly the client’s computer program using CAATs. This is also called “white box approach” CAATs are computer programs and data which the auditor uses as part of the audit procedures to process data of audit significance contained in an entity’s information systems. Some of the commonly used CAATs include test data, integrated test facility and parallel simulation. 1) Test data The test data technique is primarily designed to test the effectiveness of the internal control procedures which are incorporated in the client’s computer program. The objective of the test data technique is to determine whether the client’s computer programs can correctly handle valid and invalid conditions as they arise. To accomplish this objective the auditor prepares test data (fictitious transactions) that consist of valid and invalid conditions. The auditor enters the test data into the system and have the data processed by the entity’s computer program. Since the auditor is the one who creates the test data, the auditor knows what the output should look like assuming the client’s computer program is functioning effectively. The auditor then compares the processing results with his predetermined output. If the output generated by the client’s program is the same as the auditor’s expected output, the auditor may conclude that the client’s program is reliable.  TEST DATA Auditor’s Test Data Processed Using client’s program Compare Auditor’s Output Manually Expected Output 2) Integrated test facility (ITF) A disadvantage of the test of data technique is that the auditor does not have an assurance that the program tested is the same program used by the client throughout the accounting period. In order to overcome this advantage, the test data technique can be extended to an integrated test facility (ITF). When using this technique, the auditor creates dummy or fictitious employee or other appropriate unit for testing within the entity’s computer system. Unlike test data, which is run independent of the client’s data and ITF integrates the processing of test data with the actual processing of ordinary transactions without management being aware of the testing process. The resultant output, relating to the dummy unit, is then compared with the predetermined results to evaluate the reliability of the client’s program. By processing test data simultaneously with client’s data, ITF provides assurance that the program tested by the auditor is the same program used by the client in processing transactions.  INTEGRATED TEST FACILITY Auditor’s Client’s Data Test Data Processed Using Client’s Program Compar Auditor’s Output e Expected Manuall Output y When using ITF, the auditor must be alert to the danger of contaminating the client’s master files. Thus, care must be taken to reverse or eliminate the effects of all audit test transactions in order to avoid contamination of client’s computer files. 3) Parallel simulation In contrast to the test and ITF techniques, which require the auditor to create test inputs to be processed using the client’s computer program that simulates key features or processes of the program under review. The simulated program is then used to reprocess transactions that were previously processed by the client’s program. The auditor compares the results obtained from the simulation with the client’s output to be able to make inference about the reliability of the client’s program. PARALLEL SIMULATION Client’s Client’s Data Data Processed Processed Using Client’s Using Auditor’s Program Program Compar Output e Output Manuall y Parallel simulation can be accomplished by using generalized audit software or purpose written programs. Generalized audit software consist of generally available computer packages which have been designed to perform common audit tasks such as performing or verifying calculations, summarizing and totaling files, and reporting in a format specified by the auditor. Purpose written programs, on the other hand, are designed to perform audit tasks in specific circumstances. These programs may be developed by the auditor, the entity being audited or an outside programmer hired by the auditor.  CAATs for Advanced Computer Systems Advanced computer systems sometimes do not retain permanent audit trails, thus requiring capture of audit data as transactions are processed. Such systems may require audit procedures that are able to identify and capture audit data as transactions occur. Some of the commonly used CAATs include: 1) Snapshots This technique involves taking a picture of a transaction as it flows through the computer systems. Audit software routines are embedded ate different points in the processing logic to capture the images of the transaction as it progresses through the various stages of processing. Such a technique permits an auditor to track data and evaluate the computer processes applied to the data. 2) Systems control audit review files (SCARF) This involves embedding audit software modules within an application system to provide continuous monitoring of the systems transaction. The information is collected into a special computer file that the auditor can examine.