Loading...
06/01/2012 - Report: Payroll Audit - City Council - Audit Committee Performance Audit: Payroll Policies & Procedures June 2012 City Internal Auditor’s Office City of College Station File#: 11.02 Audit Executive Summary: Payroll Policies and Procedures Why We Did This Audit Over 25,000 paychecks for $40 million are processed by the City each year to pay over 1,000 fulltime, part-time, and seasonal city employees. Based on a citywide risk assessment, an in-depth examination of the City’s payroll processes was included in the fiscal year 2012 audit plan. What We Recommended  Security measures for changing an employee’s direct deposit information should be strengthened.  Implementation of an automated timekeeping system should be considered in order to increase efficiency, reduce errors, and approve accountability. Regardless, the implementation of automatic trigger points related to sick leave usage should be considered.  The ability to add employees and change pay rates should be limited to only certain employees with a definite business need.  The City should develop a procedure to ensure that employees no longer actively employed by the City are timely removed from the payroll system. A special re- instatement process should be considered for returning seasonal workers.  Documentation authorizing pay rate changes should sufficiently demonstrate that all employee pay rates are authentic and have been approved by management. What We Found During this audit, we examined the City’s payroll policies, procedures, and practices to determine whether or not adequate controls were in place to prevent fraud, waste, or abuse of city resources. Although we did not discover any evidence leading us to believe that there was material fraud, we found areas where internal controls could be strengthened to reduce the City’s exposure to monetary risk. We also found indicators of abuse of the City’s sick leave policies. Under current city processes, the opportunity to create fictitious (i.e. ghost) employees exists. While no ghost employees were uncovered in our review, the City does expose itself to this risk because of the following: (1) there are inadequate controls over the changing of direct deposit information, (2) the process to ensure that terminated employees are timely removed from the payroll system needs improvement, and (3) the City’s manual process for recording time provides less accountability than a properly implemented automated system. In addition, we found 41 employees that had the ability to unilaterally alter employees’ paychecks to virtually unlimited amounts. Furthermore, most of these employees had no need for this system access. This finding was considered a major control weakness that needed to be timely communicated to management. Therefore, an interim audit report was issued on February 24, 2012. Finally, we found that the City’s sick leave policy has estimated costs of approximately $1.37 million annually, with City employees using approximately 9 days of sick leave per year. In addition, we found indicators that some employees may be abusing the City’s sick leave policy. Payroll Policies & Procedures Table of Contents Introduction ................................................................................................................. 1 Audit Objectives .................................................................................................... 2 Scope and Methodology ........................................................................................ 2 Findings and Analysis .................................................................................................... 4 No Evidence of Ghost Employees Were found but Risk Exists ............................. 4 Audit Tests Did Not Reveal any Ghosts on the Payroll ..................................... 4 Risk of Redirection of Pay into Unauthorized Accounts Exist .......................... 5 There is a Risk that Former Employee Ghosts Can be Created ........................ 7 Payroll Information System Controls Could be Strengthened ............................10 Timekeepers’ System Access to Alter Paychecks is Too Broad .......................11 High-risk Employees Showed No Evidence of Material Fraud .........................13 Department Timesheet Entry Practices Could be Improved ...........................13 Increased Accountability Could Further Curb Sick Leave Abuse ........................14 Sick Leave Policies Generally Align with Best Practices ..................................14 Some Indicators of Sick Leave Abuse Exists ....................................................15 Automated Timekeeping Systems Improve Accountability .............................18 Audit Tests Performed to Verify Compliance with Policy ....................................19 Receiving Inappropriate Pay after Termination was Verified .........................19 Examination of Donated Leave Yielded No Material Findings .........................21 There is Insufficient Documentation to Verify All Pay Rates ..........................21 Recommendations ....................................................................................................... 24 Appendix A: Interim Audit Report ................................................................................. 27 Appendix B: Management’s Responses ......................................................................... 29 Payroll Policies & Procedures 1 Introduction The City Internal Auditor’s Office conducted this performance audit of the City’s payroll policies, procedures and practices pursuant to Article III Section 30 of the College Station City Charter, which outlines the City Internal Auditor’s primary duties. A performance audit is an objective, systematic examination of evidence to assess independently the performance of an organization, program, activity, or function. The purpose of a performance audit is to provide information to improve public accountability and facilitate decision-making. Performance audits encompass a wide variety of objectives, including those related to assessing program effectiveness and results; economy and efficiency; internal control; compliance with legal or other requirements; and objectives related to providing prospective analyses, guidance, or summary information. Over 25,000 paychecks for $40 million are processed by the City each year to pay over 1,000 fulltime, part-time, and seasonal city employees. The results of a citywide risk assessment conducted in October 2007 identified payroll as a potential audit topic for the fiscal year 2010 audit plan. At this time, a payroll audit focusing primarily on overtime and compensatory time was conducted. On September 22, 2011, the City Council approved the City Internal Auditor’s audit plan for fiscal year 2012, which included a more comprehensive examination of the City’s payroll processes. In December 2011, leave policies and procedures were reviewed and a preliminary analysis of payroll data from fiscal year 2009 through 2011 was conducted. During the preliminary review, several aspects of risk within the payroll process were identified. As a result, audit methodologies were developed to evaluate internal controls, identify fraud risks, and assess potential abuse of city policy and possible waste of city resources. Payroll Policies & Procedures 2 Audit Objectives This audit evaluated internal controls, identified fraud risks, and assessed potential abuse of city policy and possible waste of city resources. This report answers the following questions:  Are internal controls sufficient to reduce the risk of payroll fraud to an acceptable level? Is there any evidence of ghost employees on the payroll?  Are payroll and personnel policies, procedures and practices in alignment with best practices in order to reduce the risk of fraud, waste, and abuse? Scope and Methodology This audit was conducted in accordance with government auditing standards (except for the completion of an external peer review),1 which are promulgated by the Comptroller General of the United States. Audit fieldwork was conducted from March 2012 through May 2012. Most audit tests were performed using payroll data from January 2008 to April 2012, which comprised 110,580 payroll checks for approximately $176,578,000. The audit methods included:  Reviewing the work of auditors in other jurisdictions and researching professional literature to identify best practices regarding payroll related policies and procedures.  Interviewing staff responsible for performing various payroll related duties and oversight functions.  Reviewing applicable city policies and procedures and relevant state and federal laws and regulations. 1 Government auditing standards require audit organizations to undergo an external peer review every three years. A peer review is planned for 2013. Payroll Policies & Procedures 3  Examining direct deposit documentation for all active employees to verify compliance with city policy and to identify any indicators of possible fraudulent activity.  Performing global analytics on leave accrual and check history data from January 2008 to April 2012 using specialized auditing software to test for potential fraud and abuse.  Drawing a statistical sample of 379 paychecks to verify that the pay rates found on employees’ paychecks corresponded to the authorized rates found in employees’ personnel files. The sample size was randomly selected from the total population of paychecks from January 2010 through April 2012 in order to reach a 95 percent confidence level and a 5 percent confidence interval. Payroll Policies & Procedures 4 Findings and Analysis No Evidence of Ghost Employees Were found but Risk Exists Ghost employees are individuals listed in the payroll register, who are not providing services, but who are receiving a payroll check. Generally, there are two types of ghost employees. The first are completely fictitious employees added onto the payroll, the second are former employees that remain on the payroll. Although we did not find evidence of ghost employees on the payroll during our audit tests, the City is susceptible to this type of fraud. Therefore, some improvements could be made in the City’s internal controls to further mitigate the risk of fictitious employees being added to the payroll. Audit Tests Did Not Reveal any Ghosts on the Payroll In general, ghost employees are very difficult to discover once they are in existence—which is why controls to prevent ghost employees are so important. Nevertheless, there were two types of investigations that we conducted in an attempt to discover ghost employees on the payroll. First, we investigated any accounts wherein multiple employees on the payroll were paying into the same bank account; and second, we investigated whether any employees were depositing their paychecks into accounts that were under the name of someone other than that employee. There was evidence to explain why multiple employees were depositing their checks into a single account. We found 17 accounts wherein multiple employees—146 employees in total—were paying into a single account. Two Prosperity Bank accounts have been established to collect association dues for 40 firefighters and 73 police officers. Through employees’ personnel files, we were able to locate personal identification documentation (e.g. a copy of a driver’s license or passport) for the remaining 33 employees. In addition, most of these accounts were jointly owned by married couples that both work for the City. We were able to clarify why some employees are depositing their paychecks into accounts that were not in their name. We found 14 accounts wherein the employee being paid was not the owner of the bank account the check was being paid into. Through Payroll Policies & Procedures 5 employees’ personnel files, we were able to locate personal identification documentation that verified that these individuals had actually worked for the City. In addition, we found evidence in 9 of the 14 cases where the employee and the owner of the bank account were related. For the five employees we could not verify family relations, we identified if the owner of the bank account is a current or former employee. Although there was one instance where the account holder was a former employee, further examination revealed that no payments have been made to this employee in the last four years. Risk of Redirection of Pay into Unauthorized Accounts Exist When employees change their direct deposit information, they are asked to submit a payroll direct deposit form along with a voided check. A completed payroll direct deposit form should include the employee’s name, bank name(s), account number(s), routing number(s), social security number, and signature. Besides the fact that the information on payroll direct deposit forms are needed by Payroll so that the City can pay the employee, a fully completed form has the additional benefit of reducing the risk of being fraudulently used by third parties since the form requires private information such as a social security number, and voided checks that usually have the account owner’s name printed on it. However, in practice, employees have not always been required to submit a completed direct deposit form with a voided check. All city employees are setup with direct deposits, which reduces the risk of fraud. Currently no city employees receive manual checks. Instead, paycheck amounts are directly deposited into employees’ personal bank accounts. Requiring all employees to be setup with direct deposits not only provides a more efficient and effective method for paying employees, but it also helps prevent payroll fraud. In order for a ghost employee scheme to work, three things must happen: (1) the ghost must be added to the payroll, (2) timekeeping and wage rate information must be collected, and (3) a paycheck must be issued to the ghost and the check must be delivered to the perpetrator or an accomplice. By requiring all employees to be setup with direct deposit, several common techniques used to have ghost employee paychecks delivered to the perpetrator are eliminated. Payroll Policies & Procedures 6 Therefore, to perpetrate a ghost employee scheme at the City, the potential fraudster would have to develop a method to redirect direct deposit pay to an unauthorized account. Changing of employee direct deposit information is an informal practice. After examining the payroll direct deposit forms for all active employees, we found that less than 50 percent of them were fully completed, and about 40 percent of the submitted forms had not included a voided check. Of particular note, we found two employees who changed their direct deposit account without submitting a direct deposit form. Instead, they submitted a piece of paper with a note to change their direct deposits and a direct deposit slip. Additionally, we found several employees who changed their direct deposit by submitting a direct deposit form that only contained their name, bank info, and last four digits of their social security number—which is available off of any employee’s ID badge. Finally, we found payroll direct deposit forms wherein the employees requested that an additional bank account be added to their paycheck deposits; but rather than also listing the bank account that the employees were already depositing into, they simply wrote, “keep the rest the same.” Current employee direct deposit change practices create a security risk. The fact that employees are able to change their direct deposit information without submitting a fully completed payroll direct deposit form along with a voided check creates a security risk wherein the employee’s wages could be stolen by an individual who submits a fake direct deposit form. For example, an individual could redirect a portion of an employee’s pay check into the individual’s own account by submitting an incomplete payroll direct deposit form that contains only the employee’s name and the individual’s own bank account information. Additionally, an employee’s pay stub does not necessarily show all of the accounts his money is being directed into (see Figure 1 on the next page), so it is possible that the employee will not notice that some of his funds have been redirected into another account. And if a fraudster were to coincide the redirection with an anticipated change in the victim’s take-home pay (such as a change in insurance, retirement, or income), the victim might not notice the redirection since he was already expecting his amount of payment to change. Potentially, a fraudster could steal a large amount of money from Payroll Policies & Procedures 7 fellow employees by only taking a small amount each pay period from many individuals. Figure 1: Employee Pay Stub Example The above employee has his money deposited into three accounts; $18 each into two checking accounts, and the remainder into a savings account. However, on the pay stub, only the savings account, one of the two checking accounts, and the total amount deposited is listed. Notice that 18.00 + 1,353.63 = 1,371.63—which is 18 lower than the 1,389.63 listed in the “total” section. There is a Risk that Former Employee Ghosts Can be Created A former employee ghost is created when an employee remains on the payroll after he or she has ended employment with the organization. For example, in many cases a supervisor creates a former employee ghost by waiting for one of his subordinates to quit, and then not informing Human Resources (HR) of the subordinate’s departure. The supervisor then redirects the employee’s paycheck Payroll Policies & Procedures 8 into his own account. In the City, the process of (1) accounting for employees’ time and (2) removing employees from the payroll creates opportunity for some city supervisors to perpetrate a former employee ghost scheme. The City has a manual process for accounting for and recording employees’ time. The exact process for accounting for and recording employees’ time worked varies amongst city departments. Although departmental timekeeping processes may differ, they all involve manual tracking of time on paper timesheets that are distributed to employees through department timekeepers or supervisors. Not only is the manual tracking of time through paper timesheets a cumbersome process, it also increases the likelihood of errors, fraud, and abuse. We found that some departments provide better controls and accountability in their timekeeping process than others. However, the risk of fraud, errors, and abuse is still prevalent because manual timekeeping through paper timesheets primarily relies on adequate supervisor oversight to ensure accurate recording of time. Supervisors not only approve time entries recorded manually, but they often take custody of these documents that can be altered without an audit trail. Improved processes that ensure timely removal of inactive employees could reduce the City’s risk exposure. Most municipalities employ a wide variety of part-time, seasonal and temporary employees to fill various jobs throughout the year, such as lifeguards at community pools and election-day workers at voting locations. In these situations, the proper protocol is to remove these employees from the payroll immediately after their work is complete. Allowing inactive employees to remain on the payroll creates an increased risk of ghost employees since these inactive employees can be used as the first step in creating a ghost employee. Therefore, it is important that inactive employees be timely removed from the payroll. We examined employee paychecks between fiscal years 2009 and 2011 and discovered 195 employees2 on the payroll who, on average, had 101 days between paychecks (see Table 1 on the next page). All but six of the employees identified in Table 1 were part-time, seasonal or temporary employees. 2 The analysis was based on employee identification number. Of the 195 employees, 49 had multiple jobs. Payroll Policies & Procedures 9 Table 1: Employees on the payroll with 20 or fewer checks (FY09-11) Department Employees1 Avg. # of checks Pay Check Gaps2 PARD (Recreation) 79 5 52 PARD (Special Facilities) 63 9 32 City Secretary 49 2 305 Other3 4 3 14 1Employees on the payroll from FY09 through FY11 receiving 20 or fewer pay checks 2Average length of days between pay periods 3Includes 1 employee from Police, Parks Operations, Sanitation, and Legal In addition to the analysis summarized in Table 1, we found 97 employees3 who remained on the payroll for an average of 526 days after receiving their final paycheck. Furthermore, we found 5 employees who remained on the payroll for 1,760 days after receiving their final paycheck. During the course of this audit, city management recognized this issue and began removing inactive employees from the payroll. We applaud their proactive approach to this important matter, and encourage them, once they have completed this first step of removing currently inactive employees, to continue on to the second step of creating a process to ensure that future inactive employees will be regularly and timely removed from the payroll once they become inactive. There is a risk that a ghost employee could be created. As previously mentioned, for a ghost employee scheme to work, three things must happen: (1) The ghost must be added to the payroll. In College Station, when an employee ends employment, the HR department relies upon the departing employee, or the departing employee’s department, to inform the City of the departure. In addition, there appears to be indicators that the process of timely removing employees from the payroll is not always effective. Therefore, we conclude that it is possible under the current control environment for a supervisor to create a former employee ghost by not informing HR of his subordinate’s departure. 3 This analysis excluded employees with multiple positions. Payroll Policies & Procedures 10 (2) Timekeeping and wage rate information must be collected. Manual timekeeping processes and practices allow some department supervisors custody over the paper timesheets of their employees. Consequently, these supervisors have the ability to receive or create the paper timesheets of former employees who have not been timely removed from the payroll, manually record time (not actually worked), and then approve the timesheet for processing. (3) A paycheck must be issued to the ghost, and the check must be delivered to the perpetrator or an accomplice. We found incomplete direct deposit change forms that were processed without identifying criteria. In addition, we found that these forms could be submitted without the employee being required to appear in person (e.g. scan, inter-office mail, or email). Therefore, an employee’s supervisor has the ability to change a former employee’s direct deposit bank account to the supervisor’s own account by submitting an incomplete payroll direct deposit form that contains only the former employee’s name and the supervisor’s own bank account information. Payroll Information System Controls Could be Strengthened Payroll is one of the areas at greatest risk of fraud and theft. Because of this, organizations need to be very careful about setting up proper controls in their information systems to minimize risk. Best practices in payroll information systems require a segregation of powers and responsibilities. This includes: 1. All changes to the employee master file should be reviewed and approved by a supervisory-level employee in Human Resources (HR) prior to being recorded in the system. No one employee should be able to record modifications to the employee master file. The modifications should be initiated by one employee and reviewed and authorized in the system by a separate employee. 2. Employees responsible for modifying the employee master file should not have access to the payroll system, be involved in the payroll process, distribute payroll checks or make hiring or termination decisions. Payroll Policies & Procedures 11 It is important to note that we issued an interim audit report that addressed the audit findings found in this section of the report on February 24, 2012. We felt that the findings relating to payroll system controls represented a significant risk to the City. Therefore, we thought it necessary to inform management of these findings prior to the release of this audit report. Not only did the interim audit report address many of the issues identified in this section of the report, but it also provided management with two key recommendations. Management concurred with these recommendations and started immediately to address them. As a result, several if not all of the issues identified in this section may have already been corrected. A copy of the interim audit report can be found in Appendix A at the end of this report. Timekeepers’ System Access to Alter Paychecks is Too Broad Although the City has sufficiently separated the responsibilities of employees, the ability—or power— to perform incompatible functions within the City’s information system exists. As a result, we found 41 employees that have access to alter paycheck amounts to virtually unlimited amounts. The City has properly segregated responsibilities. Responsibilities in the City have been segregated such that changes to an employee’s pay are initiated at the department level. Employee action forms are completed by department personnel and require signature approval by employees’ supervisors and department heads. These forms are submitted to HR to authorize HR staff to make changes to the employee’s pay within the City’s information system. Once HR has processed an employee action form, they notify Payroll of the change. Payroll staff must then go into the system to approve the change. System settings allowed 41 employees to unilaterally change paychecks to unauthorized amounts. Even though the responsibilities have been properly divided, many employees still retain the power to unilaterally change paycheck amounts. As of February 24, 2012, 41 employees had the power to alter paycheck amounts—and most of them have no need for this type of system access. These employees consist of a system administrator in the Department of Information Technology, 3 Department of Finance Payroll Policies & Procedures 12 payroll administrators, 9 Human Resources employees, and 28 department timekeepers. The fact that so many employees can alter paycheck amounts constitutes a significant control deficiency. Although the limit per entry in the dollar amount field is $100,000, the system allows virtually an unlimited amount of line entries. For example, a user could use code “16” (on-call pay type) and enter $100,000 (instead of $15) on multiple lines to create a multi-million dollar check. Security settings should be changed to prevent department timekeepers from making changes to any employee’s pay. This can be done by a system administrator simply changing the “User department dollar amount entry” from “Y” to “N” in the “Time Entry” screen in the AS400. Doing so will eliminate users changing the dollar amount in the hours entry screen. The payroll system uses hundreds of codes, many of which are used for purposes they were not originally designed. There are 255 pay type codes in the system and 22 of these codes can be used by department timekeepers to change paycheck amounts. In fiscal year 2011, $10.1 million in payroll payments were made from 9 of these pay codes—which is approximately 24 percent of total payroll. Table 2 below shows the dollar amount paid to employees from these 9 codes in fiscal year 2011. Table 2: FY11 Pay Codes Used with High Risk Exposure Pay Type Description Amount Employees Number of Payments Salary Adjustment $ 9,756,661 272 6,231 Lump Sum Payment 181,365 146 146 On-call Pay 145,816 162 7,115 Workers Compensation Payments 21,252 5 23 PD Field Training (Patrol) 13,935 12 505 PD Field Training (Communications) 7,104 8 264 Taxable Awards 4,159 22 24 Retro Adjustment Pay 2,079 26 27 Tax Adjustment Retro Pay 30 1 2 Total: $ 10,132,401 In addition to this, the pay codes are often used for purposes for which they were not originally designed. For example, the salary adjustment code is used for more purposes than just paying salaried employees their regular salaried pay check. There are instances Payroll Policies & Procedures 13 where this code was used to make paycheck reversals and retroactive pay increases. In addition, several exempt employees received compensation out of this code. Also, several un-identified salary adjustment amounts were made that did not correspond with the calculated salary amount of employees. Finally, the code was used to adjust fire fighter salaries so that they get paid an even amount each pay period (otherwise they would receive a large check one pay period and a smaller one the next because of the 24 hour shifts they work). High-risk Employees Showed No Evidence of Material Fraud We examined the detailed paycheck history of fiscal year 2011 of all 41 high-risk employees previously identified as having access to alter paycheck amounts. Specifically, we paid special attention to any payroll entries made by these employees from the pay codes identified in Table 2. Based on our review, we found only one possible anomaly. In pay period 7 of 2011 her paycheck amount was increased by $182.58. The paycheck amount was decreased by the same amount (through a reversal) in pay period 9 of the same year. We identified this instance as high-risk because (1) the employee does her own hours, and (2) this type of reversal transactions could be used in order to provide a pay advance that could be returned later. Determining whether or not this is the case would be difficult to confirm. In addition, we concluded that the amount was immaterial. Department Timesheet Entry Practices Could be Improved Process controls at the department level could be strengthened. All timesheets should be reviewed and approved by department supervisors and delivered directly to department timekeepers by supervisors. After timesheet data is entered into the system, a separate employee from the one who performed time entry should verify the accuracy of entries by reconciling physical timesheets to the system generated hours proof report. In addition, department timekeepers’ timesheets should be delivered to Department of Finance payroll staff by their supervisors to be audited by central payroll personnel. In College Station, each timekeeper has received instructions on how to review and enter work hours into the system. These instructions are found on the “Time Sheet Auditing and Entry” form. If all department timekeepers followed the instruction form as well as the Payroll Policies & Procedures 14 best practices described in the previous paragraph, timesheet entry practices could be improved citywide. However, based on our review, the Police Department was the only department that did not seem to have any control issues in this area. Increased Accountability Could Further Curb Sick Leave Abuse All employers who offer sick leave to their employees find themselves in the difficult position of trying to balance the competing interests of having a flexible, friendly sick leave policy for employees in genuine need of sick leave, while also having serious, objective protocols for curbing sick leave abuse among those who would abuse it. To help balance these competing interests, best practices have been developed; they include: sick leave that is accrued over time, periodic evaluations of sick leave usage, and “trigger points.” A trigger point is a mandatory action to be taken when an employee reaches the trigger point. For example, some cities require their employees to certify each sick day taken after reaching the trigger point of five sick days taken in a year. Other organizations require employees to review their sick leave usage with their supervisor after reaching the trigger point. Sick Leave Policies Generally Align with Best Practices In College Station, most employees accrue 96 hours (12 days)4 of sick leave each year, and there is no limit on the amount that can accrue. Sick leave may be used for illness, visiting a physician, or caring for an ill family member. It may also be used as funeral leave. When an individual ends employment with the City, sick leave is not paid out (whereas accrued vacation time is paid out). When employees use sick leave, they record it on their time sheets so that the City can keep track of how much sick leave has been used. At the discretion of each department, employees may also be required to provide specific notification, such as a physician statement for each absence from work. City policy also states that evidence of sick leave abuse may constitute grounds for disciplinary action, up to termination. 4 Fire Department employees assigned to shift earn a maximum of 144 hours of sick leave each year. Payroll Policies & Procedures 15 Although sick leave policies generally align with best practices, trigger points could help increase accountability. City sick leave policies align with best practices in having employees accrue sick leave over time (rather than giving a block of leave at the beginning of each year). The City has also done well in creating a policy that allows supervisors to review an employee’s sick leave usage. However, the City could further align its policies with best practices by instituting trigger points, including a mandatory (rather than discretionary) sick leave review trigger point. Mandatory sick leave review policies help supervisors be more thorough, objective, and fair; and help employees more responsibly use their sick leave since they know they will be held accountable. Some Indicators of Sick Leave Abuse Exists Although we do not believe that abuse of sick leave is endemic in the City, we found some indicators of sick leave abuse occurring along the margins. For example, we found a disproportionate number of employees exhausting all or most of their sick leave prior to their term of employment ending. We also identified specific examples where employees almost certainly abused sick leave policies. The City averages about nine sick days per employee annually. Over the past four years, city employees have used on average over 64,000 hours of sick leave per year—which equates to almost 9 days per year per employee. Table 3 below describes city employees’ sick leave usage from 2008 through 2011. Table 3: City employees’ sick leave usage Year Employees1 Sick Hours Avg. days2 2008 911 61,914 8.5 2009 925 67,979 9.2 2010 906 65,653 9.1 2011 910 62,351 8.6 Averages: 913 64,474 8.8 1 The number of employees who accrued some form of sick leave in a given year. 2 Average number of days a city employee takes sick leave in a given year. It is very difficult to compare the City’s level of sick leave usage to other organizations. This is because other organizations have different sick leave policies that will skew their numbers either higher or lower. For example, if another organization did not allow the use of sick Payroll Policies & Procedures 16 leave for funerals, as the City does, their sick leave numbers would be lower when compared to College Station. With this caveat in mind, it is valuable to note that the Bureau of Labor Statistics reports that in the private sector “workers participating in plans with a fixed number of paid sick-leave days per year used an average of 4 days of their paid sick leave annually.” Sick leave absences cost the City over a million dollars a year in lost productivity. Sick leave costs have three primary components at the City—hourly sick leave, salaried sick leave, and personal sick days.5 Over the past four years, the amount of sick leave used by employees has resulted in average annual costs of approximately $1.37 million. Some experts argue that salaried sick leave should not be figured into sick leave costs because these employees are not paid for their time but for their results, regardless of time worked. If we just looked at sick leave used by hourly employees, the City’s cost of sick leave would average approximately $1.02 million per year. Figure 2 below provides a description of these costs. Figure 2: Cost of sick leave (in millions of dollars) 5 Employees that have 12 months of accrued sick leave may have one workday converted from a sick day to a personal day, which functions similar to vacation leave (with the exception that employees are not paid for personal days upon termination). $- $0.2 $0.4 $0.6 $0.8 $1.0 $1.2 $1.4 $1.6 2008 2009 2010 2011 Avg. Hourly Sick Salaried Sick Personal Day Payroll Policies & Procedures 17 There are indicators that some employees may be abusing sick leave policy. As previously mentioned, when employees end employment with the City, they receive a cash payout for any vacation time they did not use, but they do not receive any payout for sick leave that went unused. It appears from our investigation that some employees may have abused this policy by over-using sick leave in order to save up more vacation days. For example, one employee accumulated approximately 2,430 sick hours (304 days) during his employment with the City, and by the time he ended, he used all of his sick leave. Simultaneously, this employee had saved up 467 vacation hours, and therefore was able to cash out $11,293 of unused vacation time when he left city employment. Table 4 below shows employees who ended their employment with the City in the last 4 years and received a vacation payout. These employees are broken into groups based on the percentage of the sick leave they used during their employment. Table 4: City employees’ sick leave usage Percent Used Employees % of Employees Vacation Payout <0.50 89 22.4% $ 246,600 0.500 - 0.549 16 4.0% 76,400 0.550 - 0.590 23 5.8% 71,600 0.600 - 0.649 15 3.8% 45,900 0.650 - 0.690 12 3.0% 43,500 0.700 - 0.749 11 2.8% 35,600 0.750 - 0.790 19 4.8% 52,000 0.800 - 0.849 22 5.5% 54,800 0.850 - 0.890 24 6.1% 18,500 0.900 - 0.949 36 9.1% 52,500 0.950 - 1.000 130 32.8% 167,500 Totals: 397 100% $ 864,900 Of particular note in the above table is the fact that there are a disproportionately large number of employees who used more than 95 percent of their sick leave and received a vacation payout. This, of course, does not mean that there is rampant sick leave abuse throughout the City, but it does seem to indicate that there are some city employees abusing sick leave policies. Payroll Policies & Procedures 18 Automated Timekeeping Systems Improve Accountability Employers with sick leave abuse problems often note that there is a spike in sick leave before or after weekends and holidays. By looking at how large these spikes are, managers can then get a fairly good idea of how much, or little, sick leave abuse is occurring in the City generally. We intended to develop a comprehensive analysis of citywide sick leave usage by the day to look for indicators of this type of abuse. However, we encountered the following challenges in the data hindering our ability to perform the type of analysis necessary to make reliable conclusions: 1. We found some inconsistencies in how departments recorded sick leave usage. For example, we found examples of when sick leave was not recorded on the specific day it was taken. Although the majority of sick leave appears to be recorded on the specific day it occurs, the exceptions we found made us question the reliability that sick leave was recorded accurately all the time. 2. City employees manually record their time worked on paper timesheets on a biweekly basis. This manual process is more susceptible to mistake of fact—that is, without even knowing it, the employee may enter incorrect information on the timesheet. Supervisory review is the primary method used by the City to prevent mistakes from occurring. However, this review is still susceptible to human error—especially when supervisors typically review and approve timesheets of several employees at the end of the bi-weekly period. 3. Because city employees provide many different kinds of services to citizens, there is a large variety of schedules among employees. For example some employees work 24, 12, 11, or 9 hour shifts; while others work flex schedules, shift work, weekends, or holidays. An automated timekeeping system would help mitigate the above stated problems. It would reduce inconsistency and increase reliability of the times employees worked because the information is automatically time stamped each day. It would reduce the chance of human error because many of the areas where error could occur would become automated. Also, such a system would be able to efficiently and effectively perform sophisticated analytics because it could automatically take into account the variation among employee Payroll Policies & Procedures 19 work schedules. Finally, automated timekeeping systems create an audit trail of each change a system user makes to the system; thereby enhancing accountability. Audit Tests Performed to Verify Compliance with Policy Several audit procedures were developed to test compliance with city payroll policies. First, we analyzed whether any employees were receiving inappropriate pay after their term of employment ended. Second, we examined whether sufficient controls were in place to ensure that employees were not abusing the City’s donated leave policy for catastrophic illness. Although we found a few exceptions when performing our audit tests, we determined that these exceptions were immaterial and did not constitute a significant risk to the City. Finally, we compared the pay rates found in the payroll system to the pay rates recorded in the employees’ personnel files to ensure that employees were receiving their authorized pay rate. When performing this audit work, we were unable to verify approximately 13 percent of the pay rates because of insufficient documentation. Receiving Inappropriate Pay after Termination was Verified According to city policy, employees should not receive a vacation payout if their term of employment was less than six months. Also, no employee should receive pay after their term of employment has ended, and salaried employees should not receive a compensatory time payout upon leaving the City. We found a few exceptions of employees being paid after their termination dates. However, the total amount was under $1,000 and possible explanations for these occurrences are feasible. We also found two employees who received a compensatory time payout during a classification change of salary to hourly—due to the fact that the City previously allowed salary employees to accrue compensatory time. Although we believe that these employees should not have received this extra compensation, the risk of this happening in the future is minimal because current city policy does not allow salaried workers to earn compensatory time. Payroll Policies & Procedures 20 No employees who worked less than six months were found to receive a vacation payout. According to city policy, employees who end employment with the City may receive a payout on any vacation leave they have saved. However, employees who have worked for the City less than six months are ineligible for vacation payouts. We examined all pay checks from January 2008 to April 2012 to determine if an employee who worked for the City for less than six months received a vacation payout. We found no exceptions in the period reviewed which leads us to conclude that the City has sufficient safeguards to prevent the violation of this policy. An immaterial amount of pay was received by a few employees after termination. We investigated whether any employees continued to be paid after their termination date.6 Specifically, 17 days after their termination date, since employees may still receive their last pay check up to 17 days after they end employment. We discovered only four employees who received pay more than 17 days after they ended employment; and their combined pay after termination only totaled $831. We were able to determine that one of these four employees was an employee that ended employment then re-entered the City’s workforce, which may explain the discrepancy in paycheck and termination dates. Because of the small amount of money at risk here, we decided it would not be the best use of our time to investigate the specifics of why the other three employees received pay checks after their termination. We believe it could be for a number of reasons, such as a second job, incorrect dates entered, or re-employment. Two salaried employees received a compensatory time payout. The Fair Labor Standards Act requires employers to pay time-and-a-half to hourly employees who work more than 40 hours a week. Compensatory time is an alternative to time-and-a-half that is available to public sector employers. With compensatory time, employers may choose to compensate an employee with paid time-off rather than paying the employee time-and-a-half. Because only hourly employees should be receiving compensatory time, as part of this payroll audit, it was necessary for us to make sure than no salaried employees have received compensatory time payouts. 6 Our scope of review was employees who received a paycheck between January 2008 and April 2012. Payroll Policies & Procedures 21 In our investigation, we found two employees that, while classified as salaried employees, received compensatory time. Because these employees were reclassified as hourly employees, the City decided to give them a payout on the compensatory time they had earned while salaried employees. Together, these employees received a payout of $10,980. Unless these employees were always mistakenly misclassified as exempt workers, the extra compensation these employees received was above what the City was legally obligated to provide. However, it should be noted that these payouts occurred in 2009, and the City has recently changed its policies so that salaried employees cannot receive compensatory time. Examination of Donated Leave Yielded No Material Findings City policy permits employees to donate sick leave to qualifying regular fulltime and regular part‐time employees. The purpose of allowing employees to donate leave is to assist employees and their families when a catastrophic event forces the employee to exhaust all leave time, lose compensation from the City, and the situation presents a hardship to the employee and the employee's family. An eligible catastrophic event is considered to be more severe than a “serious medical condition” as defined by the Family Medical Leave Act (FMLA). However, before an employee can use donated leave, they must have exhausted all of their personal leave first. Therefore, we developed audit procedures to determine if any employees received donated leave prior to all other types of leave being exhausted. In the last three fiscal years, our analysis revealed that 28 employees were found to have received donated leave, of which all had in fact exhausted their own personal leave (sick and vacation) prior to the usage of donated time. This leads us to conclude that the City has sufficient safeguards to prevent the violation of this aspect of the policy. There is Insufficient Documentation to Verify All Pay Rates Typically, authorization of employees’ pay rates are documented through employee actions forms (EAF). These forms are completed by department personnel and require signature approval by employees’ supervisors and department heads. EAFs are submitted to Human Resources (HR) to authorize HR staff to make changes to employees’ pay (e.g. new hire, promotion, etc.). However, we found that EAFs are not always used when submitting end-of-year pay raises or other rate changes that affect several employees at one time—and the Payroll Policies & Procedures 22 method of documentation used to support these changes was insufficient for us to verify authentication of pay rates. Documentation used to support end-of-year pay raises was insufficient for us to verify pay rates. Pay rate changes were documented electronically through excel spreadsheets for the last fiscal year’s end-of-year pay raises. Because this method of documentation can be easily altered, we found it inadequate for our purposes of verifying pay rates. Some departments provided HR with hard copies of their changes with some documentation of approval. However, we were only able to verify that this occurred in a few instances. Furthermore, documentation of the approval of these spreadsheets was contained in emails, but HR personnel did not keep a record of these emails. Regardless if HR had documentation of these emails on file, the process of keeping records of pay rate changes through excel spreadsheets was insufficient for verifying the accuracy of pay rates in the City’s payroll system. We were unable to verify approximately 13 percent of the pay rates in our sample. We drew a statistical sample of 379 paychecks from January 2010 through April 2012 to verify that the pay rates found on employees’ paychecks corresponded to the authorized rates found in employees’ personnel files. Our results are summarized in Table 5 below. Table 5: Pay Rate Verification Results Description Number Percent Sufficient documentation to verify pay rate:  Verified through EAF 284 74.9%  Verified through non-EAF document 19 5.0%  Verified but EAF incomplete1 26 6.9% Total Verified: 329 86.8% Insufficient documentation to verify  Check date after 10.10.11 38 10.0%  Check date prior to 10.10.11 11 2.9%  Entire employee file missing 1 0.3% Total Unverified: 50 13.2% 1This includes 1 EAF without an effective date, 1 EAF missing 2 signatures, and the remaining EAFs missing 1 signature. Of particular note, are pay rates on paychecks that we were not able to verify that fell before or after October 10, 2011. This date is significant because this is when end-of-year pay raises were Payroll Policies & Procedures 23 processed, which explains why we couldn’t find sufficient documentation for at least 38 of the exceptions identified in Table 5. The sample size of 379 paychecks was randomly selected in order to reach a 95 percent confidence level and a 5 percent confidence interval. One advantage of statistical sampling is the ability to draw inferences to the entire population. For example, we are 95 percent confident that there is insufficient documentation in employees’ personnel files to verify the pay rates on between 8 and 18 percent of paychecks between January 2010 and April 2012. There were 58,314 paychecks issued during this period for approximately $95 million in payroll payments. Therefore, we can infer that between 4,665 and 10,497 paychecks from January 2010 to April 2012 would have pay rates that could not be verified because of insufficient documentation.7 7 58,314 multiplied by 8% yields a lower limit of 4,665, and an upper limit of 10,487 when multiplied by 18%. Payroll Policies & Procedures 24 Recommendations In addition to the two recommendations issued in the interim audit report on February 24, 2012, the City’s payroll process needs a few slight improvements encompassed in the following audit recommendations. Implementing these recommendations would strengthen internal controls to further prevent any payroll related fraud, waste, or abuse. 1. Security measures for changing an employee’s direct deposit information should be increased. Pay redirection fraud and former employee ghost fraud can both be mitigated using the same control. This is because, ultimately, both frauds require the redirection of an employee (or former employee’s) pay check. We recommend that management select one of the following options: A. Require that employees attach a copy of current, legal identification (such as a driver’s license) to their direct deposit form when requesting a change to their direct deposits. This would reduce the risk of fraudulent redirection of an employee’s pay check since access to an individual’s current, legal identification is usually quite limited. However, it should be noted that some supervisors may have copies of their employees’ driver’s licenses on file, so this security measure would not prove a particularly high hurdle for those supervisors. Ultimately, the unfortunate fact is that security risks can never be completely eliminated, they can only be reduced. B. A second option is to require that employees physically appear at the payroll office to submit their direct deposit form, and require the employee to show photo ID when submitting it. This is the most secure method for preventing pay check redirection. But unfortunately it could also prove to be somewhat inconvenient for those employees who do not work in the same building as the payroll office. Nevertheless, since most employees change their direct deposit information infrequently, this increased burden may be worth the enhanced security. 2. The City should consider the implementation of an automated timekeeping system. The advantages of an automated timekeeping system are that it eliminates the use of paper timesheets Payroll Policies & Procedures 25 and the cumbersome, time consuming, manual process; while also reducing many of the risks associated with human error and fraud. An automated timekeeping system also provides more detailed records than a manual system and increases accountability. Finally, many automated timekeeping systems come with user-friendly analytics that will help supervisors and managers become stronger leaders with less time and training required. Specific to this report, an automated timekeeping system will greatly reduce the risks and inefficiencies associated with the following issues: (1) ghost employees, (2) process controls for department time keepers, (3) sick leave usage and management, and (4) timesheets and time keeping. 3. The City should develop a procedure to ensure that inactive employees are timely removed from the payroll system. Since the City has already begun removing current, inactive employees from the payroll, we recommend that the City now create an official policy and process that ensures future inactive employees will be timely removed from the payroll once they become inactive. Furthermore, some department personnel explained that the justification for not timely removing inactive seasonal workers was due to the time- consuming pre-employment processes. Therefore, the City should also consider developing a special re-instatement process for returning seasonal employees instead of simply not removing these employees from the payroll. 4. Official documentation should be filed for every pay rate change. Documentation authorizing pay rate changes should sufficiently demonstrate that all employee pay rates are authentic and have been approved by management. Before any employee’s pay rate change is entered into payroll, an official, department approved, document should be filed with Human Resources. The official document can be in either paper or electronic form; but, no matter the format, it should be a document that is not easily manipulated after receiving departmental approval. For example, an electronic spreadsheet is insufficient because the entered rates are easily changed. This official document should be kept on file (either in hard copy or electronically) with Human Resources until, at the very least, the employee receives another pay rate change. By following this protocol, the City can reduce the risk of unauthorized changes in pay rates and also verify pay rates after they are in place. Payroll Policies & Procedures 26 5. Management should consider instituting trigger points. Because there was evidence of sick leave abuse among some City employees, the City should implement trigger points in its sick leave policy. The trigger point should mandate that once any employee has used a predetermined level of sick leave in a year, a specific action must occur. In most organizations this action is either a mandatory review of sick leave usage by the employee’s supervisor, or the requirement that the employee’s illnesses for the rest of the year be verified by his physician. By instituting trigger points, supervisors will be able to better manage sick leave usage, and employees will more responsibly use their sick leave since they know they will be held accountable. Payroll Policies & Procedures 27 Appendix A: Interim Audit Report Payroll Policies & Procedures 28 Payroll Policies & Procedures 29 Appendix B: Management’s Responses Payroll Policies & Procedures 30 Payroll Policies & Procedures 31 Payroll Policies & Procedures 32 Payroll Policies & Procedures 33 Payroll Policies & Procedures 34