REPORT TO THE NATIONS - 2018 GLOBAL STUDY ON OCCUPATIONAL FRAUD AND ABUSE - Amazon S3
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
FOREWORD With the publication of the 2018 Report to the Nations, I am struck by how this study, like the Association of Certified Fraud Examiners itself, is in many ways a tribute to the vision and dedication of our founder and chairman, Dr. Joseph T. Wells, CFE, CPA. When Dr. Wells created the ACFE, he did so because he recognized there was a fundamental flaw in how organiza- tions were attempting to prevent, detect, and investigate fraud. His goal in founding our association was to establish a body of Bruce Dorris, J.D., CFE, CPA knowledge and training that would help anti-fraud professionals President and CEO, reduce the incidence of fraud and white-collar crime. Association of Certified Fraud Examiners One thing Dr. Wells came to recognize in the early days of the ACFE was that the anti-fraud profession suffered from a glaring weakness: we simply did not know enough about the crimes we were trying to fight. So, with the aid of ACFE researchers John Warren and Andi McNeal, he set out to address the problem by commissioning the first Report to the Nation in 1996. The 2018 report is the 10th edition of Dr. Wells’ study, and the Re- port to the Nations remains the most comprehensive and widely quoted source of occupational fraud data in the world. Based on information from real fraud cases as reported by CFEs from around the globe, the report continues to be a tremendous re- source for those interested in how occupational fraud is commit- ted, how it is detected, who commits it, and how organizations can protect themselves from it. On behalf of the ACFE, I am proud to present the 2018 edition of the Report to the Nations. Bruce Dorris, J.D., CFE, CPA President and CEO, Association of Certified Fraud Examiners 2 Foreword Report to the Nations
CONTENTS Key Findings 4 Collusion by Multiple Perpetrators 42 Perpetrator’s Criminal Background 42 Introduction 6 Perpetrator’s Employment History 43 The Cost of Occupational Fraud 8 Behavioral Red Flags Displayed by Perpetrators 43 Projecting Total Fraud Losses Based on Imperfect Data 8 Spotlight: The Red Flags of Fraud 44 The Fraud Costs We Know 8 Non-Fraud-Related Misconduct by Perpetrators 46 How Occupational Fraud Is Committed 10 Human Resources-Related Red Flags 46 Categories of Occupational Fraud 10 Case Results 47 Spotlight: Corruption 13 Internal Action Taken Against Perpetrator 47 Duration of Fraud Schemes 14 Criminal Prosecutions and Civil Suits 48 Spotlight: Concealing Fraud 14 Spotlight: When Victim Organizations are Fined 50 Detection 16 Spotlight: Recovering Fraud Losses 51 Initial Detection of Occupational Frauds 16 Methodology 52 Tip Sources 17 Analysis Methodology 52 Median Loss and Duration by Detection Method 18 Survey Participants 54 Spotlight: Hotlines and Reporting Mechanisms 19 Regional Focus 56 Victim Organizations 20 Asia-Pacific 56 Type of Organization 20 Canada 58 Size of Organization 21 Eastern Europe and Western/Central Asia 60 Spotlight: Fraud in Small Businesses 22 Latin America and the Caribbean 62 Industry of Organization 24 Middle East and North Africa 64 Anti-Fraud Controls at the Victim Organization 26 Southern Asia 66 Perpetrators 33 Sub-Saharan Africa 68 Perpetrator’s Position 33 United States 70 Perpetrator’s Tenure 34 Western Europe 72 Spotlight: Fraud Committed by Owners and Executives 34 Index of Figures 74 Perpetrator’s Department 36 Fraud Prevention Checklist 76 Perpetrator’s Gender 39 Perpetrator’s Age 41 Glossary of Terminology 78 Perpetrator’s Education Level 41 About the ACFE 79 Contents Report to the Nations 3
KEY FINDINGS 2,690 $7 billion+ Median duration of a fraud scheme real cases of IN TOTAL LOSSES occupational fraud from $130,000 125 countries MEDIAN LOSS PER CASE MONTHS in 22% 23 OF CASES CAUSED industry categories LOSSES OF Corruption $1 million+ was the most common scheme in every global region Tips are by far the most common �� initial detection method Asset Misappropriation schemes $800,000 are the most common and least costly median loss �� internal management tips �� $114,000 40% audit review median loss 15% 13% 89% of cases employees Organizations with hotlines provide over half detect fraud by tips more often of tips, and nearly 1/3 come from financial statement outside parties 46% fraud schemes OF CASES 30% are the least common 10% DETECTED of cases BY TIP OF CASES and most costly DETECTED BY TIP HOTLINES NO HOTLINES SMALL BUSINESSES LOST ALMOST TWICE AS MUCH PER SCHEME TO FRAUD $104,000 MEDIAN LOSS $200,000 MEDIAN LOSS 100+ EMPLOYEES
INTERNAL CONTROL WEAKNESSES WERE RESPONSIBLE FOR NEARLY Owners/executives HALF OF FRAUDS accounted for a small percentage of cases Losses caused by men were 75% larger than losses caused by women 1 ALL 18 ANTI-FRAUD CONTROLS median losses ANALYZED WERE ASSOCIATED 19% $$$$$$$ are far greater WITH LOWER FRAUD LOSSES of cases $74,000 when fraudsters collude AND QUICKER DETECTION 2 but caused a $$$$$$$$$$$$$$ $150,000 median loss of $850,000 3+ $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ $339,000 DATA MONITORING/ANALYSIS and SURPRISE AUDITS were correlated with the largest reductions in fraud loss and duration FRAUDSTERS WHO HAD BEEN WITH THEIR COMPANY LONGER ���52% Data monitoring/ analysis ��� 58% 85% STOLE TWICE AS MUCH MORE THAN 5 YEARS’ TENURE $200,000 lower losses faster detection ���51% lower losses Surprise audits ��� 54% faster detection of fraudsters displayed at least one behavioral red flag MEDIAN LOSS LESS THAN 5 YEARS’ TENURE $100,000 Yet only 37% of victim organizations of fraud MEDIAN LOSS implemented these controls Only 4% Over the past 10 years, occupational fraud referrals to prosecution declined 16% OF PERPETRATORS TOP REASON FOR HAD A PRIOR NON-REFERRALS WAS FRAUD CONVICTION -16% FEAR OF BAD PUBLICITY A MAJORITY OF THE VICTIMS RECOVERED NOTHING 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Key Findings Report to the Nations 5
INTRODUCTION Fraud in general poses a tremendous threat to organizations of all types and sizes, in all parts of the world. Among the various kinds of fraud that organizations might be faced with, occupational fraud is likely the The goal of the Report to the Nations largest and most prevalent threat. Occupational is to compile detailed information fraud1—fraud committed against the organization by about occupational fraud cases in its own officers, directors, or employees—constitutes an attack against the organization from within, by the five critical areas: very people who were entrusted to protect its assets The methods by which occupational and resources. Since we began tracking data on fraud is committed occupational fraud cases in 1996, we have reviewed thousands of cases in which insiders collectively The means by which occupational frauds stole billions of dollars from their employers, and are detected those cases were merely a drop in the bucket. There are millions of business and government organiza- The characteristics of the organizations tions operating throughout the world and every one that are victimized by occupational fraud of them, in some way, is vulnerable or potentially vulnerable to fraud committed by their employees. The characteristics of the people who Most of those employees will never steal or abuse commit occupational fraud the trust that has been placed in them, but the ones The results of the cases after the frauds who do can cause enormous damage. It is because have been detected and the perpetrators of this risk that we continue to study these frauds identified and publish the Report to the Nations. This study contains an analysis of 2,690 cases of derstand occupational fraud and the impact it has on occupational fraud that were investigated between organizations. January 2016 and October 2017. The data present- ed herein is based on information provided by the The data we have gathered provides a broad and Certified Fraud Examiners who investigated those deeply representative picture of occupational fraud’s cases. Their firsthand experience with these frauds impact. The cases in this study include frauds com- provides an invaluable resource for helping us un- mitted against organizations in 23 major industry categories. Victim organizations range from small local 1 Occupational fraud is defined as the use of one’s occupation for per- businesses to multinational corporations with thou- sonal enrichment through the deliberate misuse or misapplication of the employing organization’s resources or assets. sands of employees. These frauds were committed by 6 Introduction Report to the Nations
individuals who worked in virtually every part of the agers, researchers, and the public at large. Readers organization, from entry-level employees to C-suite will find a wealth of information about the methods, executives. causes, costs, and indicators of occupational fraud, as well as important information on how to prevent and The cases we studied occurred in 125 countries detect it. This study is the 10th edition of the Report throughout the world, which helps us develop a global to the Nations, and occupational fraud remains an view of the costs, methods, victims, and perpetrators of enormous threat to the global economy, just as it was these crimes. Figure 1 shows the number of cases from when we published the first edition in 1996. But in the nine key geographical regions, along with the medi- time since that first report, we have seen organiza- an loss per fraud in each of those regions. (Readers tions make tremendous strides in terms of awareness, should note that the number of cases per region largely along with significant advancements in their ability to reflects the geographical distribution of ACFE mem- combat these crimes. Such advances are only possi- bership, so this data should not be taken to mean that ble when there is a true understanding of the nature fraud is more or less likely in any particular region.) and extent of the threat that must be dealt with. We publish this 2018 Report to the Nations hoping it will Our hope is that the findings in this report will be of advance the collective understanding and awareness value to anti-fraud professionals, organizational man- of occupational fraud risk for all those concerned. FIG. 1 Countries with reported cases and median loss for each region2 United States CASES: 1,000 (48%) Sub-Saharan Africa CASES: 267 (13%) Asia-Pacific CASES: 220 (11%) MEDIAN LOSS: $108,000 MEDIAN LOSS: $90,000 MEDIAN LOSS: $236,000 Latin America Middle East Western Europe CASES: 130 (6%) and the Caribbean CASES: 110 (5%) and North Africa CASES: 101 (5%) MEDIAN LOSS: $200,000 MEDIAN LOSS: $193,000 MEDIAN LOSS: $200,000 Eastern Europe and Southern Asia CASES: 96 (5%) Western/Central Asia CASES: 86 (4%) Canada CASES: 82 (4%) LOSS: $100,000 $150,000 $200,000 MEDIAN MEDIAN MEDIAN LOSS: LOSS: 2 Geographical information was provided in 2,092 of the cases submitted. See pgs. 56–73 for lists of cases submitted by country. Introduction Report to the Nations 7
THE COST OF Accordingly, providing a measure of the cost of fraud is an incredibly important endeavor. It is also an OCCUPATIONAL incredibly difficult one, given the number of unknown factors required to make such an estimate. No one FRAUD knows the amount of frauds that go undetected or unreported, and even for those frauds that do come to light, the full amount of loss might never be calculated. How much money do Such limitations mean that any attempts to quantify organizations lose as a the global amount of fraud will be imperfect. result of fraud? Projecting Total Fraud Losses Based on Imperfect Data Even with these limitations on projecting the total Anti-fraud professionals know just amounts lost to fraud, however, we know that such a how devastating a fraud can be to projection has an important place in the fight against its victims. But they are not the only fraud. Consequently, we asked survey participants, ones who benefit from insight into based on their professional experience, what per- centage of revenues they believe a typical organiza- the amount of damage that fraud tion loses to fraud each year. The median response causes organizations and their stake- provided by these CFEs is that organizations lose 5% holders. Business leaders need to of their annual revenues to fraud. While this number is only a general estimate based on the opinions of understand how much is at stake the CFEs who took part in our study, it represents the as they assess their risks and make collective observations of more than 2,000 anti-fraud resource-allocation decisions. Reg- experts who together have investigated hundreds ulators need to determine where to of thousands of fraud cases. To place their estimate in context, if the 5% loss estimate were applied to focus their enforcement efforts. In- the 2017 estimated Gross World Product of USD vestors and customers need to make 79.6 trillion, it would result in a projected total global informed decisions about where fraud loss of nearly USD 4 trillion.3 To be clear, this number is only an estimate and, given the limitations to direct their own money. And the described above, it is unlikely we will ever be able media desires context and direction to calculate the true cost of fraud on a global scale. for helping to raise awareness of the But we can be certain that the amount of damage is issue to the general public. incredibly large, and this estimate, provided by an- ti-fraud professionals who work to prevent and detect fraud on a daily basis, helps give us some insight into just how big the problem may be. The Fraud Costs We Know Determining total fraud losses—whether globally, re- gionally, by industry, or even within a specific organiza- tion—is outside the primary scope of our study. Instead, we focus on analyzing known data to better under- 3 See https://www.cia.gov/library/publications/the-world-factbook/geos/ xx.html (retrieved March 22, 2018). 8 The Cost of Occupational Fraud Report to the Nations
stand the risks posed by occupational fraud. To FIG. 2 How much does an occupational fraud cost that end, we examined the losses incurred in the the victim organization? actual cases of fraud reported to us to learn about Less than $200,000 how fraud affects its victims. 55% The total loss caused by the cases in our study $200,000–$399,999 exceeded USD 7.1 billion. While we do not know 4 11% the total number of cases of fraud that occurred globally during our study period, it is safe to as- $400,000–$599,999 sume that the 2,690 cases included in our study 7% represent only a tiny fraction of the frauds com- mitted against organizations worldwide during $600,000–$799,999 that time. Thus, the USD 7.1 billion in known loss- 3% es—while staggering on its own—does not come $800,000–$999,999 close to representing the total amount lost to fraud. The true global cost of fraud is likely mag- 2% nitudes higher, especially when factoring in the $1 million or more indirect costs, such as reputational harm and loss of business during the aftermath of a scandal. 22% The mean, or average, loss due to the frauds in our study was USD 2.75 million,5 which is also an Throughout this report, we further examine enormous amount when considering how much these losses through different lenses, based damage such a loss represents to most organi- on the specific schemes, victim organizations, zations. However, due to the presence of several perpetrators involved, and other factors. very large frauds in our data, this amount likely We hope that our research into and analysis of these cases does not illustrate the typical fraud case. Conse- helps shed additional light on the way that fraud impacts quently, throughout this report we use median the global business community and its stakeholders. loss calculations, rather than mean, to provide a more accurate representation of how fraud typically affects organizations. The median loss for all cases in our study was USD 130,000. Figure 2 shows the loss distribution of the cases. While 55% caused less than USD 200,000 in financial damage, more than one-fifth resulted in a loss of at least USD 1 million. 4 The total losses represented in our study were actually signifi- cantly higher than USD 7.1 billion. However, our survey results included a few cases with losses so large that including them in the total loss figure may have enabled them to be identified. To avoid compromising the confidentiality of our survey participants, we winsorized the top and bottom 1% of the data used in this total loss calculation (i.e., assigned all cases in the top 1% and bottom 1% the same value as the 99th percentile and 1st percentile, respectively). While including those cases would increase the total loss amount figure substantially, we believe it prudent to both ensure those cases remain unidentified and conservatively report loss amounts. 5 As with the total loss figure, the top and bottom 1% of the data were winsorized for purposes of the average loss calculation. The Cost of Occupational Fraud Report to the Nations 9
HOW OCCUPATIONAL FRAUD IS FIG. 3 How is occupational fraud committed? COMMITTED 89% What methods do fraudsters use to commit their schemes? PERCENT OF CASES One of the goals of the first Report to the Nation was to examine the methods by 38% which fraudsters commit their schemes, and we have continued this line of study in every subsequent report. Over the last two decades, even with tremendous technolog- 10% ical development and numerous changes in the global business and regulatory environ- ments, our research shows that occupation- Asset Financial misappropriation Corruption statement fraud al fraud falls into several time-tested cate- gories. The taxonomy of these categories is illustrated in the Occupational Fraud and $114,000 Abuse Classification System, also known as the Fraud Tree, as depicted in Figure 4. $250,000 MEDIAN LOSS Categories of Occupational Fraud Of the three primary categories of occupational fraud, asset misappropriations are by far the most common, occurring in 89% of the cases in our study. However, they are also the least costly, causing a median loss of USD 114,000. Corruption schemes are the next most common form of occupational fraud; 38% of the cases in our study involved some form of corrupt act. These schemes resulted in a median loss to the victim organi- zations of USD 250,000. The least common and most costly form of occupational fraud is financial statement $800,000 fraud, which occurred in 10% of the cases and caused a median loss of USD 800,000. 10 How Occupational Fraud Is Committed Report to the Nations
FIG. 4 Occupational Fraud and Abuse Classification System (the Fraud Tree)6 Corruption Asset Misappropriation Financial Statement Fraud Net Worth/ Net Worth/ Conflicts of Illegal Gratuities Economic Net Income Net Income Interest Bribery Extortion Overstatements Understatements Purchasing Invoice Timing Timing Schemes Kickbacks Differences Differences Sales Fictitious Understated Bid Rigging Revenues Revenues Schemes Concealed Overstated Liabilities and Liabilities and Expenses Expenses Improper Improper Asset Asset Valuations Valuations Improper Improper Disclosures Disclosures Cash Inventory and All Other Assets Theft of Cash Theft of Cash Fraudulent Misuse Larceny on Hand Receipts Disbursements Asset Requisitions Billing Payroll Expense Check and Register and Transfers Skimming Cash Larceny Reimbursement Payment Schemes Schemes Disbursements Schemes Tampering False Sales and Shipping Shell Ghost Mischaracterized Refunds Employee Forged Maker False Voids Sales Receivables Company Expenses and Other Purchasing and Receiving Non- Overstated Accomplice Falsified Forged Write-Off Expenses Endorsement False Refunds Unrecorded Vendor Wages Schemes Unconcealed Larceny Personal Commission Fictitious Lapping Purchases Schemes Altered Payee Understated Expenses Schemes Multiple Authorized Unconcealed Reimbursements Maker 6 The definitions for many of the categories of fraud schemes in the Fraud Tree are found in the Glossary of Terminology on pg. 78. In previous reports, the category check and payment tampering was referred to simply as check tampering. However, to better reflect the increasing shift toward electronic payment methods, we have changed the category title. Report to the NationsFraud How Occupational Committed Report HowIsOccupational to Committed Fraud Is the Nations 11
When assessing an FIG. 5 How often do fraudsters commit more than one type of occupational organization’s fraud risks fraud? and designing anti-fraud controls, it is important to remember that fraudsters typically seize whatever Financial statement fraud opportunity arises when committing their schemes. Thus, many frauds— Asset including nearly one-third misappropriation Corruption of the cases in our study, as illustrated in Figure 5—involve more than one form of occupational fraud. Asset Misappropriation Sub-Schemes Within the category of Asset misappropriation only 57% asset misappropriation, Asset misappropriation and corruption 23% our research shows that there are several dis- Corruption only 9% tinct sub-categories of schemes. The heat map Corruption, asset misappropriation, and financial statement fraud 4% in Figure 6 illustrates the Asset misappropriation and financial statement fraud 3% relative frequency and cost of each of these scheme Financial statement fraud only 1% types. The schemes falling Corruption and financial statement fraud 1% in the darkest area of the heat map—check and pay- ment tampering,7 billing, and theft of noncash as- FIG. 6 What asset misappropriation schemes present the greatest risk? sets—rank among the most common and the costliest Check and payment tampering $150,000 (12%) scheme types and thus typically pose the greatest Billing risk to organizations. $100,000 (20%) Noncash $98,000 (21%) 7 In previous reports, this category was referred to simply as check tampering. However, to better Cash larceny reflect the increasing shift toward $75,000 (11%) electronic payment methods, we Payroll $63,000 (7%) have changed the category title to Skimming check and payment tampering. $50,000 (11%) Expense reimbursements Register disbursements $31,000 (14%) $29,000 (3%) Cash on hand $20,000 (15%) LESS RISK MORE RISK 12 How Occupational Fraud Is Committed Report to the Nations
CORRUPTION Corruption represents one of the most significant fraud risks for organizations in many industries and regions. Understanding the specific factors involved in corruption schemes can help organizations effectively prevent, detect, and investigate them. Percent of cases involving corruption Industries with highest proportion of 40% 36% 60% CORRUPTION CASES: WESTERN EASTERN EUROPE AND EUROPE WESTERN/CENTRAL ASIA CANADA 30% 51% UNITED STATES 62% 49% 53% ASIA-PACIFIC MIDDLE EAST AND ____NORTH AFRICA SOUTHERN ASIA ENERGY 51% SUB-SAHARAN AFRICA LATIN AMERICA AND THE 49% CARIBBEAN 51% MANUFACTURING 70% of corruption cases were perpetrated by someone in a POSITION OF AUTHORITY 82% of corruption cases were committed by males 50% GOVERNMENT AND PUBLIC ADMINISTRATION 38% MANAGER WHILE ONLY 27% EMPLOYEE 18% 3% OTHER 32% 50 % OWNER/ of corruption cases were EXECUTIVE committed by females OF CORRUPTION CASES TOP RED FLAGS were detected by a tip in corruption cases 43% Living beyond means 34% Unusually close association with vendor/customer 23% Financial difficulties 21% “Wheeler-dealer” attitude How Occupational Fraud Is Committed Report to the Nations 13
Duration of FIG. 7 How does the duration of a fraud relate to median loss? Fraud Schemes 27% Examining how long frauds tend to last can also provide insight into how they affect 19% their victims. The median PERCENT OF CASES duration for all of the fraud 13% cases in our study was 16 11% months. However, it stands 10% to reason that the longer a 8% fraud goes undetected, the 6% 5% larger the scheme will grow. Figure 7 shows that frauds that last over 60 months are more than 20 times as costly 6 months 7–12 13–18 19–24 25–36 37–48 49–60 More than or less months months months months months months 60 months as those that are caught in the first six months. Our data also $30,000 indicates that fraudsters tend $75,000 to start small and increase $125,000 their frauds rapidly over $200,000 MEDIAN LOSS the first three years. Thus, it is incredibly important for organizations to implement $400,000 proactive fraud detection $425,000 mechanisms to catch frauds $500,000 quickly and minimize their damage (see pg. 18). $715,000 CONCEALING FRAUD An act of fraud typically involves not only the commission of the scheme itself, but also efforts to conceal the misdeeds. Understanding the methods fraudsters use to cover their crimes can help organizations better design prevention mechanisms and detect the warning signs of fraud. TOP 8 CONCEALMENT METHODS USED BY FRAUDSTERS 55% 48% 42% 34% 31% 30% 29% 27% Created fraudulent Altered physical Created fraudulent Altered transactions Altered electronic Destroyed physical Created fraudulent Created fraudulent physical documents documents transactions in the in the accounting documents or files documents electronic journal entries accounting system system documents or files 14 How Occupational Fraud Is Committed Report to the Nations
We also examined the duration FIG. 8 How long do different occupational fraud schemes last? of the cases reported to us Payroll based on the type of scheme 30 months involved. Figure 8 shows the result of this analysis. The Check and payment tampering 24 months payroll schemes in our study tended to last the longest, Financial statement fraud with a median duration of 30 24 months months, while schemes involv- ing cash on hand and register Expense reimbursements disbursements were both 24 months typically uncovered one year Billing after they began. 24 months Cash larceny 24 months Corruption 22 months Skimming 18 months Noncash 18 months Cash on hand 12 months Register disbursements 12 months How to Conceal: what to Conceal: 3 ONLY % ����� � Create, Alter, or Destroy? physical or electronic evidence? OF CASES 80% 80% Altered existing DID NOT Created fraudulent evidence evidence involve 21% 63% 12% ��� any attempts to conceal the fraud Manager-level 43% fraudsters are more likely to alter evidence. All of these unconcealed Deleted or Owners/executives are ELECTRONIC PHYSICAL cases were destroyed evidence more likely to create BOTH or delete evidence. EVIDENCE EVIDENCE committed by owners/ executives How Occupational Fraud Is Committed Report to the Nations 15
DETECTION How are fraud schemes initially detected? Understanding the methods by which occupational frauds are detected is critical for both investi- gating schemes and implementing effective prevention strategies. We asked survey participants to tell us how the frauds they investigated were initially detected, which helps us understand how organizations are most likely to discover frauds in the future. This data also shows how organizations can take steps to detect fraud proactively, rather than passively. Initial Detection of Occupational Frauds Figure 9 shows that the leading detection methods are tips, internal audit, and man- agement review. This finding is not surpris- ing, as these have been the three most com- mon means of detecting occupational fraud in every edition of the report since 2010. Collectively, these three detection methods were cited in 68% of the cases in our current study. Tips were by far the most common means of detection at 40% of cases—more than internal audit (15%) and management review (13%) combined. 16 Detection Report to the Nations
Tip Sources Since tips are the most common detection method, but this data suggests organizations should also it is important to understand where those tips come consider promoting reporting mechanisms to outside from. Figure 10 shows that slightly more than half parties, especially customers and vendors. Addition- of all tips (53%) were provided by employees of the ally, 14% of tips came from an anonymous source, victim organizations. Meanwhile, nearly one-third demonstrating that a significant portion of those who (32%) of the tips that led to fraud detection came reported fraud did not want their identities known. from people outside the organization: customers, Whistleblowers often have a fear of being identified vendors, and competitors. Active cultivation of tips or retaliated against, which is why it is important that and complaints, such as the promotion of fraud they be able to make reports anonymously where hotlines, is often geared primarily toward employees, such practice is legally permissible. FIG. 9 How is occupational fraud initially detected? FIG. 10 Who reports occupational fraud? Tip 40% Internal audit 15% Management review 13% Employee By accident 53% 7% Other 6% Account reconciliation Customer 5% 21% Document examination 4% External audit Anonymous 4% 14% Surveillance/monitoring 3% Vendor 8% Notified by law enforcement 2% Other IT controls 5% 1% Competitor Confession 3% 1% Shareholder/owner 2% Internal source External source Other Detection Report to the Nations 17
Median Loss and Duration by Detection Method In addition to determining the most common meth- passive detection methods because those mechanisms ods of detection, we also analyzed the median loss might or might not involve proactive efforts specifically and duration of fraud schemes based on how they to identify fraud, depending on the circumstances. were uncovered. Our results indicate that there is a correlation between the way in which occupational Our findings show that median duration and median fraud schemes are detected and the severity of the loss were relatively low in frauds that were detected fraud. More importantly, the data points to steps or- by active methods. Frauds detected passively tended ganizations can take to detect fraud proactively and, to last much longer and have larger median losses. in doing so, mitigate losses. For instance, frauds detected actively by IT controls tended to last five months and cause a median loss of Figure 11 portrays the median loss and median dura- USD 39,000, compared to schemes detected passive- tion for all cases, based on the method by which they ly through notification from law enforcement, which were detected. We grouped each of these detection tended to last two years and cause a median loss of methods into three categories: active, passive, or almost USD 1 million. The key takeaway from this data potentially active or passive. Active detection methods (shaded teal) involve a deliberate search for miscon- is that organizations can reduce the impact of fraud duct from someone within the organization or an inter- by pursuing internal controls and policies that actively nal control designed to detect fraud. Passive detection detect fraud, such as thorough management review, methods (shaded black) refer to cases in which the account reconciliation, and surveillance/monitoring. organization discovers the fraud by accident, confes- Organizations that do not actively seek out fraud are sion, or unsolicited notification by another party. We likely to experience schemes that continue for much classified tips and external audit as potentially active or longer and at a higher cost. FIG. 11 How does detection method relate to fraud duration and loss? 30 Active detection method $186,000 Potentially active or passive detection method MEDIAN MONTHS TO DETECTION 24 months 25 Passive detection method $150,000 20 $250,000 24 months $935,000 23 months 24 months 15 $130,000 $126,000 18 months 18 months $110,000 10 $108,000 14 months $52,000 12 months 11 months 5 $50,000 $39,000 6 months 5 months 0 n ls ing ion it n e w Tip nt it tio sio ud lic ud ro vie ide iat or po ina nt la la es re nit c cil co na ac nf na by am nt mo on Co IT r er me te ed ex By ec e/ Int Ex tifi ge nt tr nc me No un na illa co Ma cu rve Ac Do Su 18 Detection Report to the Nations
HOTLINES AND REPORTING MECHANISMS The presence of a hotline or other reporting mechanism affects how organizations detect fraud and the outcome of the case. Fraud losses were 63 $200,000 LOREM IPSUM % of victim organizations had hotlines 46% OF CASES DETECTED BY TIP HOTLINES 30% OF CASES DETECTED BY TIP NO HOTLINES Corruption is particularly likely Organizations with hotlines detected fraud by tip more often 50% SMALLER at organizations $100,000 DOLOR SIT to be detected by tip with hotlines Asset Financial ��� ��� ��� Corruption Misappropriation Statement Fraud than those without Organizations without hotlines were 50% DETECTED 38% DETECTED 38% DETECTED more than TWICE AS LIKELY to detect BY TIP BY TIP BY TIP fraud by accident or by external audit Telephone hotlines are most popular, but whistleblowers use various reporting mechanisms NOT ALL TIPS COME THROUGH HOTLINES Web-based/ When a reporting mechanism is not Telephone hotline Email online form used, whistleblowers are most likely to report to: 42% 26% 23% DIRECT SUPERVISOR 32% EXECUTIVE 15% Mailed letter/form Other Fax FRAUD INVESTIGATION TEAM 13% 16% 9% 1% COWORKER 12% INTERNAL AUDIT 10% Detection Report to the Nations 19
VICTIM ORGANIZATIONS How are different kinds of organizations affected by occupational fraud? To better understand the victim organizations in our study, we asked participants to provide information about T Y P Ethe O F Vorganizations’ I C T I M O R G A N I Z A T type, I O N — size, FREQUENCY AND MEDIAN LOSS and industry, as well as the mechanisms that the organizations had in place to prevent and detect fraud at the time the scheme occurred. Type of Organization FIG. 12 What types of organizations are victimized by occupational fraud? As shown in Figure 12, more than 70% of 42% the frauds in our study occurred at for-prof- it organizations, with 42% of the victim PERCENT OF CASES 29% organizations being private companies and 29% being public companies. The private 16% companies in our study suffered the greatest 9% median loss, at USD 164,000. Not-for-profit 4% organizations were the victim in only 9% of frauds and had the smallest median loss of Private Public Government Not-for-profit Other company company USD 75,000; however, for many not-for-profit entities, financial resources are extremely limited and a loss of USD 75,000 can be particularly devastating. MEDIAN LOSS Level of Government Organization $75,000 Resources and operations vary greatly by level of government, meaning that fraud can $117,000 $118,000 $120,000 affect these organizations differently. Con- sequently, we broke down the government fraud cases in our study based on the level $164,000 of government agency involved. While there was not a large variation in the percentage FIG. 13 What levels of government are victimized 38+31+265F of schemes that occurred at local, state/ by occupational fraud? provincial, and national levels, the frauds at national-level agencies tended to be much larger, causing a median loss approximately National: 38% Local: 31% twice as large as the losses experienced by ($200,000*) ($92,000*) local and state/provincial governments (see State/provincial: 26% Other: 4% ($110,000*) ($58,000*) Figure 13). *Dollar amounts are median loss. 20 Victim Organizations Report to the Nations
Size of Organization The size of an organization’s staff can directly affect both the opportunity for fraud and the ability to enact certain anti-fraud mechanisms. Larger entities typically have more resources to invest in their anti-fraud programs, as well as a greater ability to separate duties among staff members to help prevent fraud; however, the large staff size can also mean more potentially dishonest employ- ees who might attempt schemes and more complex processes and transac- tions, which can increase the risk of fraud. To provide some insight into the relative risks of fraud for organizations of various sizes, we analyzed the cases reported to us based on the number of employees at the victim organization. Figure 14 shows that small organiza- tions (those with fewer than 100 employees) both experienced the greatest percentage of cases in our study (28%) and suffered the largest median loss (USD 200,000). See “Fraud in Small Businesses” on pgs. 22–23 for more information about how fraud affects these organizations. FIG. 14 How does an organization’s size relate to its occupational fraud risk? 28% 26% 24% 22% PERCENT OF CASES
FRAUD IN SMALL BUSINESSES Fraud can be especially devastating to small businesses. These organizations typically have fewer resources to both prevent and recover from a fraud, and they often require an increased level of trust in employees due to a lower ability to implement robust anti-fraud controls. SMALL BUSINESSES LOSE ALMOST TWICE AS MUCH PER SCHEME TO OCCUPATIONAL FRAUD 0 EMPLOYE + EMPLOYE 10 E 00 E S < 1 S Median loss: Median loss: $200,000 $104,000 Frauds detected by tip: 29% 44% Frauds caused by lack of internal controls: 42% 25% Frauds perpetrated by an owner/executive: 29% 16% 22 Victim Organizations Report to the Nations
32% Corruption 43% 29% Billing 18% Check and 22% payment tampering 8% Expense 21% reimbursements 11% Skimming 20% Small businesses face DIFFERENT RISKS 8% 20% Cash on hand 14% 16% than larger organizations Noncash 22% Financial 16% statement fraud 7% Cash larceny 14% 9% Payroll 13%
Industry of Organization In addition, we examined the cases reported to us based on the industry of the victim organization. The greatest num- ber of cases in our study occurred in the banking and financial services, manufacturing, and government and public administration sectors. Readers should note that this data likely represents the industries that most often employ CFEs, rather than the industries that are most susceptible to fraud. However, information about occupational fraud in various industries can be useful for benchmarking purposes. FIG. 15 How does occupational fraud affect organizations in different industries? 17+83+R Banking and 366 Cases financial services M E D I A N LO S S : $110,000 10+90+R Manufacturing 212 7% Cases M E D I A N LO S S : $240,000 9+91+R Government and 201 14% Cases public administration M E D I A N LO S S : $125,000 7+93+R 5+95+R Health care 158 Cases M E D I A N LO S S : $100,000 Retail M E D I A N LO S S : $50,000 108 Cases 5+95+R Insurance 101 Cases M E D I A N LO S S : $153,000 4+96+R Education 97 Cases M E D I A N LO S S : $68,000 4+96+REnergy 94 Cases M E D I A N LO S S : $300,000 4+96+R 4+96+R Construction 90 Cases M E D I A N LO S S : $227,000 Other M E D I A N LO S S : $70,000 84 Cases 4+96+R 83 Cases Transportation and warehousing M E D I A N LO S S : 4+96+R 76 Cases Food service and hospitality M E D I A N LO S S : $90,000 3+97+R Technology 68 Cases M E D I A N LO S S : $150,000 3+97+R 3+97+R or social services 60 Cases Religious, charitable, M E D I A N LO S S : $90,000 58 Cases Services (professional) M E D I A N LO S S : $258,000 $140,000 2+98+R 51 Cases Arts, entertainment, and recreation M E D I A N LO S S : 2+98+R Telecommunications 50 Cases M E D I A N LO S S : $100,000 2+98+R Real estate 35 Cases M E D I A N LO S S : $180,000 1+99+R 1+99+R 32 Cases Agriculture, forestry, fishing, and hunting M E D I A N LO S S : $136,000 Utilities M E D I A N LO S S : $150,000 29 Cases $88,000 1+99+R Services (Other) 28 Cases M E D I A N LO S S : $82,000 1+99+R Mining 27 Cases M E D I A N LO S S : $208,000 1+99+R Communications and publishing 24 Cases M E D I A N LO S S : 1+99+R Wholesale trade 24 Cases M E D I A N LO S S : $110,000 $525,000 24 Victim Organizations Report to the Nations
Most Common Schemes by Industry Understanding the frequency of specific fraud schemes within different industries can help organizations assess and design controls to guard against the schemes that pose the most significant threats. Figure 16 provides a heat map showing the relevant risk for each category of occupational fraud in every industry that had at least 50 report- ed cases in our study. Boxes are shaded from light to dark red based on the respective level of occurrence, with darker boxes indicating higher-frequency schemes. FIG. 16 What are the most common occupational fraud schemes in various industries? Financial statement fraud Expense reimbursements Register disbursements Check and payment Cash on hand Cash larceny Corruption tampering Skimming Noncash Payroll Billing INDUSTRY Cases Banking and financial 338 11% 14% 23% 12% 36% 7% 8% 11% 2% 3% 9% services Manufacturing 201 27% 8% 15% 12% 51% 18% 10% 28% 5% 3% 7% Government and public 184 15% 11% 11% 9% 50% 11% 5% 22% 7% 2% 11% administration Health care 149 26% 7% 13% 13% 36% 16% 11% 19% 17% 1% 12% Retail 104 20% 10% 19% 9% 28% 8% 12% 34% 5% 13% 13% Education 96 23% 19% 19% 6% 38% 18% 6% 19% 6% 0% 14% Insurance 87 20% 9% 3% 18% 45% 8% 7% 11% 3% 1% 11% Energy 86 20% 2% 10% 12% 53% 10% 3% 27% 7% 2% 10% Construction 83 37% 12% 8% 19% 42% 23% 16% 23% 14% 1% 13% Transportation and 79 25% 8% 8% 9% 46% 15% 8% 28% 3% 3% 13% warehousing Food service and 75 17% 16% 20% 11% 29% 12% 12% 24% 7% 0% 23% hospitality Technology 62 26% 5% 10% 8% 42% 21% 16% 32% 8% 0% 6% Religious, charitable, or 58 40% 9% 22% 19% 34% 29% 10% 19% 22% 3% 17% social services Services (professional) 54 26% 17% 15% 26% 17% 30% 13% 13% 15% 0% 15% Arts, entertainment, 50 14% 20% 36% 6% 32% 12% 8% 18% 4% 8% 28% and recreation LESS RISK MORE RISK Victim Organizations Report to the Nations 25
As noted in Figure 16 on pg. 25, corruption poses Effectiveness of Anti-Fraud Controls a significant risk to several industries, with the most Demonstrating the return on investment in anti-fraud common occurrence of corruption schemes in the initiatives can be a difficult task, as it is nearly impos- energy, manufacturing, and government and public sible to measure the amount of fraud prevented by a administration sectors. Skimming schemes were also specific control. However, many anti-fraud profession- notably more common in the arts, entertainment, and als find themselves needing to make a business case recreation and the food service and hospitality indus- to justify additional fraud prevention and detection tries than elsewhere, while payroll schemes occurred initiatives. To provide some visibility into the relative ef- more frequently in the religious, charitable, or social fectiveness of various anti-fraud controls, we compared services and the health care sectors. Interestingly, the the losses experienced by the victim organizations that cases that occurred in religious, charitable, or social had specific controls in place against the losses experi- services organizations also tended to involve the most enced by those that had not implemented each control. crossover between scheme types, meaning the perpe- The results of this analysis are provided in Figure 18 trators in these cases used many different schemes to on pg. 28. Interestingly, the presence of every control defraud the victims, rather than limiting their frauds to we analyzed was correlated with lower fraud losses. one specific area. For example, the use of proactive data monitoring and analysis and surprise audits was associated with a Anti-Fraud Controls more than 50% reduction in fraud losses. at the Victim Organization We similarly analyzed the duration of fraud schemes The presence of a robust system of anti-fraud con- based on the presence or absence of each anti-fraud trols can be a powerful deterrent, as well as a proac- control (see Figure 19 on pg. 29). Data monitoring and tive prevention and detection mechanism, in the fight analysis and surprise audits were correlated with the against fraud. Thus, organizations can benefit from most significant reductions in fraud duration; as these knowing which anti-fraud controls are commonly two controls were also associated with some of the used by their peers, as well as which tend to be the largest loss reductions, our data indicates that they are most effective. To help explore this information, we among the most useful tools in the fight against fraud. provided survey respondents with a list of 18 enti- ty-level, anti-fraud controls and asked which, if any, were present at the victim organization at the time the fraud occurred. As noted in Figure 17, 80% of the organizations had a code of conduct and underwent external financial statement audits, while 73% had internal audit departments, and 72% had company management certify the financial statements. On the other end of the spectrum, 19% of organizations had policies requiring job rotation or mandatory vacation, and only 12% provided rewards for whistleblowers. 26 Victim Organizations Report to the Nations
FIG. 17 What anti-fraud controls are most common? Code of conduct 80% External audit of financial statements 80% Internal audit department 73% Management certification of financial statements 72% External audit of internal controls over financial reporting 67% Management review 66% Hotline 63% Independent audit committee 61% Employee support programs 54% Anti-fraud policy 54% Fraud training for employees 53% Fraud training for managers/executives 52% Dedicated fraud department, function, or team 41% Formal fraud risk assessments 41% Surprise audits 37% Proactive data monitoring/analysis 37% Job rotation/mandatory vacation 19% Rewards for whistleblowers 12% Victim Organizations Report to the Nations 27
FIG. 18 How does the presence of anti-fraud controls relate to median loss? $250,000 PERCENT REDUCTION $200,000 56% 52% 50% 51% 50% 50% 46% 47% 43% 38% 41% $150,000 38% 35% 33% 29% 20% 23% 12% Median loss without controls $100,000 Median loss with controls $50,000 Co Pr Su de oa Ex rp Ma cti of te Ho ris An v rn co na ed ea tlin Int al Ma ti- nd ge er ud at Fr e au fra uc na m Fo am na au its dit ud Em en t ge rm Fr la dt 0 De on t of po Ex plo me au ud al r r dic udit e Jo a ito i te lic dt Ind n v ini fra ye nt it d i br te Re rn e at rin y ra ng es ce w ud ep rn al e ed ot w g/ i pa n rti al up ar en fo at a ris ing an fra fic rtm co re d ion po d k aly sf en ud ncia at fo nt as mp ents rt en / or of ion ma sis rm ro ta se de pr t loy wh ls fin ud nd ss og an pa of ov ee a m ist it c at r ag rtm fin am er s or leb om er an en y fin s l s/e low cia v sta mi t, f an a xe ca tte ls er t un cia e cu tio ta s me e cti lr tiv te n ep on nt me es s or ,o nt tin rt s g ea m Percent Control Control not Percent Control of cases in place in place reduction Code of conduct 80% $110,000 $250,000 56% Proactive data monitoring/analysis 37% $80,000 $165,000 52% Surprise audits 37% $75,000 $152,000 51% External audit of internal controls over financial reporting 67% $100,000 $200,000 50% Management review 66% $100,000 $200,000 50% Hotline 63% $100,000 $200,000 50% Anti-fraud policy 54% $100,000 $190,000 47% Internal audit department 73% $108,000 $200,000 46% Management certification of financial statements 72% $109,000 $192,000 43% Fraud training for employees 53% $100,000 $169,000 41% Formal fraud risk assessments 41% $100,000 $162,000 38% Employee support programs 54% $100,000 $160,000 38% Fraud training for managers/executives 52% $100,000 $153,000 35% Dedicated fraud department, function, or team 41% $100,000 $150,000 33% External audit of financial statements 80% $120,000 $170,000 29% Job rotation/mandatory vacation 19% $100,000 $130,000 23% Independent audit committee 61% $120,000 $150,000 20% Rewards for whistleblowers 12% $110,000 $125,000 12% 28 Victim Organizations Report to the Nations
FIG. 19 How does the presence of anti-fraud controls relate to the duration of fraud? 25 58% 54% 50% 50% PERCENT 50% 50% REDUCTION 50% 50% 50% 20 50% 50% 50% MEDIAN MONTHS TO DETECTION 48% 46% 44% 38% 40% 33% 15 Median duration without controls 10 Median duration with controls 5 Pr Su oa Int rp Ma cti er Ex ris Ma na ve na Ho te ea An la na ge Fr rn tlin poli da ti- ud Fr au ud al ge me Fo ta fra e au aud i au dt it d fica t me Re rm s mo nt ud dt Ind f con Co atio dit ra wa ep nt al Jo 0 ce nit ra ini ep de of fr a br re rd De rti ini Ex or ng r en Em tm vie s i o cy ot ng nt te ing dic fo de fo e er rn plo w ris t nt rw fo /an at re ion nt na a rm ed la ye ka n/ du mp his aly au lc o ud m es fra an ct ss f fi on dit tle sis loy an it o es up ag s ud n tro blo co da ee an sm po e f fi de ls mm t rs/ w s cia o en rt na ov er r pa ex y ls pr itt t nc s er va r ec og tm ta ee ial fin ca ut te ra en sta tio ive an me ms t, n cia s te fu n ts me nc lr ep tio nt or s n, tin or g te am Percent Control Control not Percent Control of cases in place in place reduction Proactive data monitoring/analysis 37% 10 months 24 months 58% Surprise audits 37% 11 months 24 months 54% Internal audit department 73% 12 months 24 months 50% Management certification of financial statements 72% 12 months 24 months 50% External audit of internal controls over financial reporting 67% 12 months 24 months 50% Management review 66% 12 months 24 months 50% Hotline 63% 12 months 24 months 50% Anti-fraud policy 54% 12 months 24 months 50% Fraud training for employees 53% 12 months 24 months 50% Fraud training for managers/executives 52% 12 months 24 months 50% Formal fraud risk assessments 41% 12 months 24 months 50% Rewards for whistleblowers 12% 9 months 18 months 50% Independent audit committee 61% 12 months 23 months 48% Code of conduct 80% 13 months 24 months 46% Job rotation/mandatory vacation 19% 10 months 18 months 44% Dedicated fraud department, function, or team 41% 12 months 20 months 40% External audit of financial statements 80% 15 months 24 months 38% Employee support programs 54% 12 months 18 months 33% Victim Organizations Report to the Nations 29
Background Checks FIG. 20 Was a background check run on the perpetrator prior to hiring? Effectively preventing fraud Did the check reveal begins with ensuring that existing red flags? the organization hires ethical Yes 10% employees. As part of our study, we examined whether No 48% No 90% the victim organizations ran a background check on the perpetrator prior to hiring him or her, as well as whether the background check revealed any potential indicators of the employee’s dishonesty. As Yes 52% noted in Figure 20, 52% of the organizations ran background checks, while 48% did not. Of the organizations that did run a check before hiring the perpetrator, 10% were alerted to a red flag regarding the perpetrator but chose to hire FIG. 21 What types of background checks were run the person anyway. on the perpetrator prior to hiring? We also asked about the Employment history 78% types of background checks Criminal checks 75% used by the victim organiza- Reference checks 55% tions in our study. Figure 21 Education verification 50% shows that these organiza- Credit checks 36% tions were most likely to look into the individual’s employ- Other 4% ment and criminal history, with three-quarters or more of the background checks covering these areas. 30 Victim Organizations Report to the Nations
Internal Control Weaknesses that Contributed to Fraud Understanding the factors that can lead to fraud is the foundation of preventing future occurrences. Conse- quently, we asked survey respondents what they perceived to be the primary internal control weakness that contributed to the fraud they reported. In 30% of cases, a simple lack of controls was the main factor that enabled the fraud to occur, while another 19% of cases occurred because the perpetrator was able to override the controls that had been put in place. FIG. 22 What are the primary internal control weaknesses that contribute to occupational fraud? Lack of internal controls 30% Override of existing controls 19% Lack of management review 18% Poor tone at the top 10% Lack of competent personnel in oversight roles 8% Lack of independent checks/audits 4% Other 6% Lack of employee fraud education 2% Lack of clear lines of authority 2% Lack of reporting mechanism
We also analyzed these FIG. 23 How do internal control weaknesses vary by scheme type? control weaknesses based on the category of fraud Lack of internal controls 32% involved in the scheme (see 25% Figure 23). Not surprisingly, 29% a poor tone at the top was much more likely to be the Lack of management review primary factor in financial 19% 15% statement fraud and cor- 15% ruption cases than in asset misappropriation cases. Override of existing internal controls However, it is interesting to 18% 21% note that a lack of internal 14% controls is more common in asset misappropriation and Poor tone at the top financial statement frauds, 9% 18% while corruption schemes 23% are more likely than other schemes to involve an Lack of competent personnel in oversight roles override of existing con- 8% 6% trols. In addition, a lack 5% of management review is more commonly the reason Other for asset misappropriation 5% 7% schemes than other forms 8% of fraud. Lack of independent checks/audits 4% 3% 5% Lack of employee fraud education 2% 3%
PERPETRATORS What does a typical fraudster look like? We asked survey respondents to provide a broad range of information about the fraud perpetrators they investigated, including the offenders’ conditions of employment, basic demographics, prior misconduct, and behavior that might have been warning signs of fraudulent activity. Our goal is to identify common characteristics and risk profiles for those who commit occupational fraud, which can help organizations better recognize fraud perpetrators or those at risk for engaging in fraudulent activity. Perpetrator’s Position FIG. 24 How does the perpetrator’s level of authority relate to occupational fraud? As seen in Figure 24, there is a strong correlation between 44% the fraud perpetrator’s level of authority and the size of the 34% fraud. While owners/execu- PERCENT OF CASES tives only committed 19% of the frauds in our study, the schemes committed by these individuals 19% resulted in a median loss of USD 850,000, which was nearly six times larger than the median loss caused by managers, and 17 times larger than the median loss 3% caused by low-level employees. A significant correlation between Employee Manager Owner/executive Other authority and fraud loss has been found in every edition of $50,000 the report dating back to 1996. This correlation likely reflects $150,000 $189,000 the fact that high-level fraudsters tend to have greater access to MEDIAN LOSS an organization’s assets than low-level personnel. They may also have greater technical ability to commit and conceal fraud, and they might be able to use their authority to override or conceal their crimes in ways that low-level employees cannot. $850,000 Perpetrators Report to the Nations 33
You can also read