Tackling external fraud

Transcription

1 Good practice in tacking externa fraud Xxxxxxxxxxx GOOD PRACTICE GUIDE Tacking externa fraud

2 Good practice in tacking externa fraud Xxxxxxxxxxx The Nationa Audit Office scrutinises pubic spending on behaf of Pariament. The Comptroer and Auditor Genera, Tim Burr, is an Officer of the House of Commons. He is the head of the Nationa Audit Office which empoys some 850 staff. He and the Nationa Audit Office are totay independent of Government. He certifies the accounts of a Government departments and a wide range of other pubic sector bodies; and he has statutory authority to report to Pariament on the economy, efficiency and effectiveness with which departments and other bodies have used their resources. Our work saves the taxpayer miions of pounds every year: at east 9 for every 1 spent running the Office.

3 Good practice in tacking externa fraud Xxxxxxxxxxx Contents Foreword 5 Introduction 6 Part 1: Understanding and managing the risks 10 of fraud Taking a strategic approach to tacking externa fraud 11 Assessing the scae of the threat 13 Understanding the types of fraud risks 16 Focusing resources on the most effective anti-fraud measures 17 Setting targets and monitoring performance 17 Responsibiities for tacking fraud 18 Part 2: Deterring and preventing externa fraud 20 Changing pubic attitudes to fraud 21 Changing staff attitudes to create an anti-fraud cuture 23 Contros to prevent fraud 24 Strengthening interna contros and checks 26 Part 3: Detecting and investigating externa fraud 28 and imposing sanctions Detecting fraud 29 Investigating cases of fraud 33 Imposing sanctions 36 The recovery of money defrauded 39 Evauating the effectiveness of sanctions 41 Working with others in tacking fraud 41

4 4 Good practice in tacking externa fraud Xxxxxxxxxxx

5 Good practice in tacking externa fraud Foreword 5 Foreword Reports pubished by departments, HM Treasury, the Nationa Audit Office and the Committee of Pubic Accounts show that much good work is being done across departments to prevent and detect fraud against pubic funds. But arge sums of pubic money are sti ost each year through externa fraud. These osses arise from a wide range of different types of fraud. At one end of the scae are peope who see an opportunity to make a sma gain. At the other extreme a reativey sma number of organised crime groups carry out premeditated systematic attacks for arge sums of money. The aw abiding pubic have the right to expect government departments and agencies to safeguard pubic funds and to crack down on those committing fraud. A types of fraudsters weigh up the potentia gains against the risk of getting caught and the sanctions they may face. Government departments and agencies need to make fraud as unattractive as they can. Not acting against fraud can undermine the reputation, integrity and professionaism of the organisation and perceptions about the quaity of the services it provides eading to a oss in pubic confidence. A number of departments are carrying out fraud risk assessments to show the scae of the probem and are assessing how these risks can be reduced. Tota eimination of fraud is unikey ever to be achieved, but it is important for momentum to be maintained, and good practices deveoped and shared. The purpose of the guide is to demonstrate and expain some of the good practices used by organisations in tacking externa fraud. It incudes checkists to hep you assess your current practices. The guide shoud be of interest to pubic sector managers who are responsibe for expenditure programmes and for protecting revenue, incuding poicy staff, financia managers and interna auditors. Smaer departments and agencies wi need to consider how practices used in some of the arger departments can be adapted and appied to their own circumstances. We are gratefu for the hep provided by the departments and other organisations in producing this guide. John Kingman Second Permanent Secretary HM Treasury Caroine Mawhood Assistant Auditor Genera Nationa Audit Office

6 6 Good practice in tacking externa fraud Introduction Introduction 1 This guide concentrates on externa fraud which is where third parties, such as businesses, individuas or organised crime groups, stea money from a department or agency, either by obtaining payments to which they are not entited or keeping monies they shoud pay over to the department. Frauds may be opportunistic attempts by individua customers or businesses to obtain a financia advantage. The sums invoved in any one such case may be sma, but these can mount up to significant osses of pubic money if there are a ot of cases invoved. At the other end of the scae, departments or agencies may suffer from more systematic and premeditated attacks by organised crime groups. These may be fewer in number but the osses in each case are substantia. In some cases fraudsters may work in cousion with the department s staff. As we as diverting money that shoud be spent on pubic services fraud can undermine the position of honest citizens and businesses and support the activities of those invoved in other serious crime. 2 The Fraud Act incudes three casses of fraud: Fraud by fase representation; Fraud by faiing to discose information; Fraud by abuse of position. In a three casses of fraud, the Act requires that for an offence to have occurred, the person must have acted dishonesty, and that they had to have acted with the intent of making a gain for themseves or anyone ese, or inficting a oss (or risk of a oss) on another. There are aso offences of fraud specific to particuar departments activities. 3 A government departments and agencies have a responsibiity to deveop antifraud poicies to show those seeking to defraud the government that such action is unacceptabe and wi not be toerated 2. The annua Statement on Interna Contro summarises the processes used to identify and manage risks incuding fraud 3. In addition the Proceeds of Crime Act , the Money Laundering Reguations (2003 5, ) and reated egisation have paced responsibiities on reguated entities and every person carrying out reevant business within the meaning of the Reguations, to report where they know, or suspect, or have reasonabe grounds to suspect, that money 1 Fraud Act Managing Pubic Money, Annex Managing Pubic Money, Annex Proceeds of Crime Act The Money Laundering Reguations 2003 (Statutory Instrument 2003, Number 3075) 6 The Money Laundering Reguations 2007 (Statutory Instrument 2007, Number 2157)

7 Good practice in tacking externa fraud Introduction 7 aundering may be taking pace. Most government departments are not cassified as reevant businesses but it is good practice to assess the risks and potentia exposure; and, where appropriate, departments shoud regard themseves as reevant businesses and act accordingy. These reporting responsibiities aso appy to the Nationa Audit Office as externa auditor. Money aundering invoves not ony the direct proceeds of crime, but any deaings with crimina property. It incudes possessing, or in any way deaing with, or conceaing, the proceeds of any crime. Any faiure to report suspicions of money aundering is now a crimina offence. Reguated entities must take appropriate measures to ensure that key staff are aware of the provisions of the Act and the Reguations, and appoint a Money Laundering Reporting Officer to report their knowedge or suspicions to the Serious Organised Crime Agency. 4 In response to the Fraud Review of , the Government provided 29 miion of new funding to impement its recommendations, incuding the setting up of: a Nationa Fraud Strategic Authority to define and co-ordinate deivery of a nationa strategy that sets out a vision of success in tacking fraud; the steps needed to reaise it; and measuring the impact that these efforts are having. The Fraud Loss Measurement Unit of the Nationa Fraud Strategic Authority to provide robust estimates for repeat measurement of osses before and after action has been taken; and a strategic assessment of the risks and possibe osses from future threats. The Nationa Fraud Reporting Centre and Inteigence Bureau to be the hub in which knowedge about fraud is coated and managed. It wi provide an outward facing service to the pubic; the business community and poice forces; receiving reports of fraud offences and incidents, adding inteigence and providing packages of anaysed information to target investigations and other action. These may incude crimina investigations conducted by the poice, confiscation investigations or other forms of action taken by government departments, agencies or by industry. The Nationa Lead Force for Fraud to provide resources and support for poice forces in tacking fraud. The City of London Poice wi take on the ead roe. 5 Departments and agencies face a wide range of different risks from externa fraud which are demonstrated in Figure 1. There are aso many other types of fraud perpetrated by third parties, such as fraud by contractors. In some departments or agencies externa fraud is a sizeabe and continuing probem for their main business but in others it may ony occur occasionay. 7 Fraud Review Fina Report, Juy Review%20Fina%20Report%20Juy% pdf

8 8 Good practice in tacking externa fraud Introduction Figure 1 The diversity of externa frauds faced by the pubic sector Benefit caimants who fai to decare a earnings, income or capita, or who concea famiy circumstances, to obtain benefits to which they are not entited. Peope who caim exemption from paying for prescriptions to which they are not entited. Dentists who caim for treatments which they have not carried out. Peope who evade vehice excise duty. Peope or businesses who caim grants to which they know they are not entited. Peope who work in the hidden economy and do not pay income tax or nationa insurance contributions on their earnings. They may aso caim means tested benefits from the Department for Work and Pensions to which they are not entited. Peope and businesses who register with HM Revenue & Customs but fai to decare a transactions/income or assets on which tax is due. Staff couding with criminas to defraud the department or agency. For exampe in 2005 an empoyee in a department sod detais of departmenta records to an externa accompice to aow them to frauduenty caim 1.25 miion of payments from the department. Serious criminas obtaining arge sums, for exampe: through evading tobacco, acoho and hydrocarbon oi duties; setting up what appear to be egitimate companies but intend to carry out frauds, such as to stea VAT; committing organised fraud against the benefit system through stoen, forged or counterfeit instruments of payment; or through creating fictitious benefit caims. 6 Departments shoud consider whether they need to deveop a package of measures specificay taiored to each type of fraud. There wi not be a one size fits a approach. But there is much vaue in promoting a wider understanding of how others tacke fraud and good practices which are successfu esewhere. Smaer departments and agencies shoud consider whether they can adapt and appy practices used by arger departments in tacking externa fraud. 7 The guide shows how a number of departments are tacking fraud by taking an integrated strategic approach which is summarised in Figure 2. It aso expains why they are taking this approach and what it invoves. The strategic approach accords with the HM Treasury s guidance The Orange Book: Management of Risk 8 and Managing the Risk of Fraud: A Guide for Managers 9 which provide guidance on the identification and management of risk, and guidance on controing identified fraud risk to acceptabe eves of exposure. 8 The Orange Book: Management of Risk Principes and Concepts, HM Treasury, October Managing the Risks of Fraud: A guide for managers managing_the_risk_fraud_guide_for_managers.pdf

9 Good practice in tacking externa fraud Introduction 9 Figure 2 The main eements of an integrated strategic approach to tacking fraud Understand the size of the probem Understand the types of risk Assign responsibiities for tacking fraud Monitor performance and review pans as necessary Prepare pans for managing fraud risks Set targets Impement measures to tacke fraud Prevent and deter fraud Detect, investigate and impose sanctions 8 The guide is structured as foows: Understanding and managing the risks of externa fraud (Part 1); Preventing and deterring externa fraud (Part 2); Detecting and investigating fraud and imposing sanctions (Part 3). Questions at the beginning of each Part are to hep you assess your organisation s practices. If you are not using a particuar practice you wi need to consider whether that is appropriate given your circumstances. 9 We hope that the guide is a usefu source of reference for pubic sector managers in demonstrating the experience and good practice of others. It does not seek to provide everything you need to know to tacke externa fraud. To do so woud require many voumes. However, the guide provides references to usefu sources of information and gives inks to where these are avaiabe on websites. The case exampes in the guide are for iustrative purposes ony. There may be many other exampes of good practice in other departments or agencies.

10 10 Good practice in tacking externa fraud Xxxxxxxxxxx Part 1 Understanding and managing the risks of fraud NHS fraud is not a victimess crime. Every pound ost to fraud deprives the NHS of vauabe resources it needs to provide patient care. Dermid McCausand, Managing Director, NHS Counter Fraud Service In ooking at this Part of the guide, ask yoursef whether your organisation: takes a strategic approach to tacking fraud risk; assesses the size of the threat from externa fraud and, where significant, undertakes a separate risk assessment; knows the size of the fraud threat/types of fraud committed/who is committing them/how often/and how much is invoved; has a package of measures in pace to tacke osses from fraud where these are significant; has targets to stabiise or reduce fraud; has aocated responsibiities for tacking, and ownership of, fraud risks to ensure that risks are managed, pans are impemented and progress monitored. This Part of the guide ooks at how you can tacke some of these issues and gives exampes of how others approach these issues. As you read through the guide you wi need to consider how appropriate the practices are to your circumstances.

11 Good practice in tacking externa fraud Understanding and managing the risks of fraud 11 Taking a strategic approach to tacking externa fraud Protecting tax revenues is an essentia part of the Government s commitment to prosperity and fairness. Its strategy has deivered a major improvement in revenue coection over the ast few years. Protecting tax revenues, HM Revenue & Customs strategy document, Budget Report Some organisations have taken a strategic approach to understanding and managing the risks of fraud because this: Fits in with good corporate governance. A major eement of good corporate governance is a sound assessment of the organisation s business risks. Fraud risk shoud be managed in the same way as managing any other business risk and shoud therefore be approached systematicay at both the organisationa and operationa eve. Figure 3 shows exampes of how departments and agencies have reduced externa fraud and Figure 4 provides the main eements of a strategy to tacke externa frauds. Figure 3 Exampes of how externa fraud has been reduced In the Department for Work and Pensions estimated that fraud against Jobseeker s Aowance, Income Support and Pension Credit has faen from 900 miion to 430 miion since a reduction of just over 50 per cent. In the 2008 Budget report, HM Revenue & Customs estimated that the set of strategic measures introduced to protect revenues over the previous 5 years had reduced tax osses in priority areas VAT, the main excises, the main direct taxes and Nationa Insurance Contributions by over 5 biion a year. Figure 4 The main eements of a strategic approach to tacking externa fraud Heps with deveoping a range of measures which appy proportionate and we targeted pressure at a eves of the probem. Can hep achieve a cost effective approach in tacking fraud by focusing on areas of greatest risk and where efforts may have the greatest impact. A strategic approach can provide a rationa and robust basis in bidding for additiona resources to tacke fraud. Set targets and monitor performance Strategic approach Assess the size of the fraud threat Can be a hepfu way of communicating to staff what the organisation is trying to do and what is expected from them. Some organisations have aso pubished their strategies as a way of informing the pubic that they have a we thought out approach to tacking externa fraud. This can aso send a deterrent message to potentia fraudsters that they are unikey to succeed in attempts to commit fraud against the organisation. As exampes, HM Revenue & Customs, the Department for Work and Pensions and the NHS Counter Fraud Service have pubished strategies. Focus resources on the most effective anti-fraud measures Assign responsibiities for tacking fraud Understand the types of risk: Who? How? When? Why?

12 12 Good practice in tacking externa fraud Understanding and managing the risks of fraud 1.2 In taking a strategic approach, some departments and agencies have taken an across the board approach to ooking at externa fraud, and some have ooked at individua fraud risks and produced a strategy for each. Others tacke fraud within the context of an overa strategy to combat osses from a types of non-compiance. The overa compiance approach recognises that there is a oss continuum ranging from inadvertent customer error at one end of the spectrum to fraud at the other with shades of grey in between A of these approaches can be equay vaid depending on a department s circumstances and the stage they are at in deveoping their approach (Figure 5). However, a common feature is that the departments and agencies deveop fraud risk assessment toos to identify the fraud risks, their ikeihood and impacts, and how to manage them. These toos need to be reviewed reguary to assess whether they remain appropriate or require updating to respond to the threat from new fraud risks. Figure 5 Exampes of counter fraud strategies The Department for Work and Pensions has deveoped a counter fraud strategy for a wefare benefits. First set out in 1999, the strategy proposed action on four fronts: Getting it right aiming to get benefit payments correct from day one; Keeping it right ensuring payments are adjusted as circumstances change; Putting it right detecting when payments go wrong and taking prompt action to correct them with appropriate penaties to prevent a recurrence; Making sure the strategy works by monitoring progress, evauating the strength of preventive measures and adjusting them in the ight of experience. The Department seeks to ensure that their goas of supporting peope in their efforts to find work, whist providing appropriate financia support where necessary, are not compromised by their efforts to prevent frauduent abuse of the benefits system. The NHS Counter Fraud Service s strategy consists of seven objectives: the creation of an anti-fraud cuture; maximum deterrence of fraud; successfu prevention of fraud which cannot be deterred; prompt detection of fraud which cannot be prevented; professiona investigation of detected fraud; effective sanctions where fraud is proven; redress for money defrauded. HM Revenue and Customs approach to tacking fraud has a number of key components: the deveopment of reiabe estimates of the size of tax osses as far as possibe; estabishing the nature and economics of the activity and behaviour that cause the osses, through the anaysis of inteigence and other data; a comprehensives set of responses based on: 1 Support: buiding services that are taiored to the needs and circumstances of taxpayers; 2 Prevention: ensuring that basic processes and design make the system as secure as possibe from fraud and avoidance; 3 Identifying and tacking those who set out to obtain an unfair tax advantage: using improved risk assessment and checks, with sanctions as appropriate taiored to behaviour. Monitoring effectiveness against key outcomes to measure the success of the strategies.

13 Good practice in tacking externa fraud Understanding and managing the risks of fraud 13 Assessing the scae of the threat It wi never be possibe to measure 100 per cent of fraud; there wi aways remain undiscovered fraud. But better measurement is crucia to a propery designed and effective strategic response to fraud and to supporting better management of fraud risks. Fraud Review: Fina Report 1.3 Assessing the scae of oss from fraud is an important first step in deveoping a strategy for tacking externa fraud. An estimate highights the scope for potentia savings which can then hep to determine the reative priority that shoud be given to tacking fraud in the context of a the other cas on an organisation s resources. Such estimates then estabish a baseine against which performance can be judged. If repeated at intervas, estimates can hep an organisation assess how we they are doing and whether the threat is changing. There may be circumstances where an organisation decides it is not practicabe to produce overa estimates. Nevertheess they may be abe to use a range of techniques such as carrying out in-depth research into an area where fraud is suspected to gain a better understanding of the scae and nature of the threat (Figure 6). Figure 6 HM Revenue & Customs approach to measuring taxpayer compiance Where it is not possibe to produce reiabe estimates of the amount of tax ost to fraud HM Revenue & Customs may use the foowing techniques: audit based studies random enquiry programmes to provide estimates of noncompiance and to faciitate research into understanding the risks of non-compiance in specific taxpayer and customer groups; modeing techniques appied to compiance data to expain or predict taxpayer non-compiance. 1.4 Some may say that: It is too difficut to produce estimates of fraud and that it is not worth attempting to do so; The resources used to produce an estimate coud be better used on tacking fraud, for exampe, by carrying out more investigations. These issues are deat with beow. Producing reiabe estimates 1.5 A number of departments have produced estimates of fraud or osses from fraud and error. For exampe, HM Revenue & Customs has produced estimates of overa osses on VAT and of particuar types of oss such as missing trader fraud. This work is described in two documents Measuring Indirect Tax Losses 2007 and Deveoping Methodoogies for Measuring Direct Tax Losses which are avaiabe on the HM Revenue & Customs website. 1.6 Departments have used operationa research and statistica methods to produce such estimates. Two main methods used are statistica modeing and samping. Statistica modeing 1.7 Statistica modeing has been used to produce overa estimates of fraud or oss notaby on revenue activities. This invoves comparing eves of actua receipts or expenditure with the tota eve of receipts or expenditure that might be expected using other sources of data on the eve of activity under review (Figure 7). Figure 7 Exampes of statistica modeing Estimated evasion rates by the BBC on the teevision icence fee are cacuated using a statistica mode. In summary, the mode estimates the eve of evasion by subtracting the number of icences currenty in force from the estimated number of properties for which a icence shoud be hed. Repeating the modeing process at reguar intervas has shown an overa downward trend in the evasion rate in recent years with the atest figures showing a rate of 5.1 per cent for HM Revenue & Customs produce a top-down estimate for osses on VAT. This entais comparing the tota eve of expenditure in the economy that is theoreticay iabe for VAT with actua VAT receipts and assuming that the difference represents the tota revenue oss. The theoretica tax iabiity is a goba measure based mainy on data from the Office for Nationa Statistics. Bottom up estimates can be used in combination with the top eve estimates as a means of vaidating the eve of osses. Methods to produce bottom up estimates incude surveys and use of administrative and operationa data.

14 14 Good practice in tacking externa fraud Understanding and managing the risks of fraud 1.8 Points to consider for statistica modeing are: The data required may be incompete. The mode may use therefore a number of assumptions which mean that the resuts are subject to a margin of error. It is important to take this into account when making decisions on actions to reduce osses. Other work may be needed to give an insight into those committing the fraud or the type of action that might deter them. This may incude more in depth modeing work (see Figure 12 on the BBC). Further research may be needed into the causes of increases or decreases in the eve of osses and the extent to which this is due to anti-fraud measures impemented. Samping 1.9 Estimates of oss can be generated by checking a representative sampe of cases to see whether fraud is invoved, and extrapoating the resuts to the whoe popuation to estimate the tota eve of fraud oss in the area of expenditure or revenue (Figure 8). When checking individua cases it can be difficut to determine whether any discrepancy is due to fraud or error (reckessness, careessness or ignorance) because of the judgements that need to be made. In its work in estimating fraud in individua expenditure streams, the NHS Counter Fraud Service is deciding whether fraud is invoved by using the concept of fraud and burden of proof appicabe in civi aw that is whether someone knowingy or reckessy obtained resources to which they were not entited and the baance of probabiity. This is to ensure that a behaviour which can egay be determined as fraud is measured. Figure 8 The Driver and Vehice Licensing Agency use samping to estimate the amount of evasion of Vehice Excise Duty The Department for Transport commissions a roadside survey of over one miion passing vehices annuay in June to identify unicensed vehices. Statistica weightings are then appied to the observed evasion rate in traffic to cacuate the estimated evasion in the overa stock of vehices, and then the estimated Vehice Excise Duty revenue oss. The resuts are used to report performance against targets for the Department and the Driver and Vehice Licensing Agency A key consideration in producing an estimate of fraud oss on an area of expenditure/revenue is the eve of accuracy required. A greater degree of precision produces more reiabe estimates (essentia for assessing any rea change in the eve of fraud over time) but at additiona cost because the size of the sampe required increases (Figure 9). Figure 9 Generating precise estimates of fraud requires arger sampe sizes and incurs higher measurement costs The degree of precision of a fraud oss estimate depends on: the size of the sampe checked. For exampe, the NHS Counter Fraud Service set sampe sizes so that they can determine the eve of fraud in each area to within to +/- one per cent; whether the sampe is stratified according to the type of risk; use of skied reviewers abe to detect where fraud has occurred, the amount of oss suffered and the nature of the fraud; the quaity contro and vaidation arrangements to ensure the review process is correcty and consistenty appied.

15 Good practice in tacking externa fraud Understanding and managing the risks of fraud For some organisations, producing a nationa estimate may be sufficient. In others, it may be necessary to produce estimates which are aso broken down by region. This wi have important impications for the samping exercise and its costs, as separate sampes within each region increase the tota sampe that must be checked (Figure 10). Figure 10 Generating regiona estimates of fraud oss invoves arger sampe sizes Where a department needs estimates of the fraud oss within each of its regions, it wi have to take a separate sampe of transactions in each. To produce estimates that are sufficienty precise to revea any important differences between regions in the rate of fraud oss requires the samping of a much greater tota number of transactions than when generating ony a nationa eve estimate. Regiona samping wi revea which regions suffer the greatest and owest eves of fraud oss and whether there are significant changes between regions over time. The Department for Work and Pensions random sampe of over 40,000 cases a year covered each of neary 100 districts, three times a year, spit between each type of benefit caimant. Costs of estimating fraud 1.12 As indicated above, the costs of measurement vary according to: the frequency of the estimating exercise; the sampe sizes checked; the work invoved in checking each case samped; the work invoved in vaidating the resuts For smaer departments and agencies, a one-off estimate or one produced at intervas may be sufficient. Accepting ess precision by using smaer sampe sizes may be one way forward. Athough the resuts wi be ess reiabe, these wi indicate whether further work is desirabe. Others may require continuous measurement exercises to produce ongoing estimates of fraud oss. Whie this invoves greatest cost, it does mean that a department is abe to track changes over time in the estimated fraud oss, and the types of fraud committed. Figure 11 outines the Department for Work and Pensions continuous measurement approach. Figure 11 The Department for Work and Pensions continuous measurement approach The Department for Work and Pensions spent over 9 miion in as part of their ongoing measurement of fraud and error in Income Support, Jobseeker s Aowance, Pension Credit and Housing Benefit. The atest estimate shows that for the period October 2006 to September 2007, fraud osses were 210 miion on Income Support, 60 miion on Jobseeker s Aowance, 110 miion on Pension Credit and 140 miion on Housing Benefit. Tota osses across a benefits were around 800 miion or 0.6 of expenditure. The Department produces estimates of the incidence and magnitude of fraud, customer and officia error in these benefits every six months. The Department has carried out a programme of reviews for other benefits such as Disabiity Living Aowance produced in 2005 and Incapacity Benefit in The Department s aim is to produce snapshot estimates of the amount fraud and error. The size and risk of potentia oss determine which benefits are reviewed and how frequenty.

16 16 Good practice in tacking externa fraud Understanding and managing the risks of fraud 1.14 Costs can be spread over severa years by carrying out a roing programme of estimates. For exampe, the NHS Counter Fraud Service has set out to measure fraud across the Primary care services incuding the pharmaceutica, optica and denta services. Three exercises have been carried out across each of these areas. Exercises are aso underway ooking at fraud within the procurement process. Another aternative is to carry out a oneoff measurement exercise (with possibe foow up severa years ater) to confirm the significance of the eve of fraud. This can be a usefu approach where the eve of fraud is thought to be ess significant. Understanding the types of fraud risks 1.15 A department or agency wi be unabe to deveop an appropriate response based ony on the estimates of fraud. They aso ideay need to know: the types of fraud perpetrated against them, for how ong and the financia oss invoved; who the fraudsters are, their characteristics and behaviours, how often they carry out the frauds, which types of frauds they commit, how they do it, and whether they are opportunistic or organised Examination of detected fraud cases either from investigation or from the random sampes of cases examined to produce estimates of fraud oss, can give an insight into these (Figure 12). Larger departments which face serious threats aso have inteigence anaysts and/or commission research into the threats. At the other end of the spectrum, there are some departments and agencies that may have few or no recent instances of externa fraud. Checking a sampe of cases, or carrying out research into the possibe threats, wi hep to confirm whether the risks from fraud are ow. Figure 12 Exampes of work organisations have done to understand and manage fraud risks faced The BBC has commissioned market research programmes, over a number of years, which have shown that evaders of the teevision icence fee vary both in their behaviour and their attitude to the icence fee. Overa the research indicates that non-payers are more ikey to be younger peope and ess we off. They aso tend to regard the icence as unfair or just a tax you get nothing for. Further research was carried out to mode the evader popuation by matching evasion rates for postcode areas with commerciay avaiabe data on income and ife stye. This information heped to identify the characteristics of those most ikey to evade. Areas with high evasion rates are most ikey to have, for exampe, a higher than average proportion of younger peope, ow income househods, students and singe parent famiies. The Department for Work and Pensions coects information about the types of fraud, the characteristics of the customers and the way the fraud and customer errors are detected. These hep the Department target resources to detect and prevent fraud. For exampe, the Department estimates that the most common benefit frauds in the period October 2006 to September 2007 were caiming as a singe person, but iving with a partner as husband and wife ( 92 miion); fraud committed by customers iving abroad ( 93 miion) and undecared earnings ( 77 miion). The Department recognises that these frauds are high risk, and it targets investigations at this type of fraud. For Income Support, the Department generate separate estimates of fraud and error for each of their three main cient groups Lone Parents, Pensioners and Disabed peope/others. The measurement reviews aso provide estimates of the main causes of fraud and error and the proportion of benefits overpaid due to each cause. For exampe, this enabed the Department to estimate that in the period October 2006 to September 2007 the faiure of one parent caimants to discose they were iving together with a partner was the cause of neary haf of the amount overpaid for this reason across a benefits. HM Revenue & Customs uses centraised risk assessment and anaysis of buk third party data to identify groups posing a high risk of understating taxabe profits. It aims to deveop campaigns to everage the effects of its anti-fraud work beyond the imited poo that can be directy investigated on a one to one basis. The resuts show that its approach has reduced understated income.

17 Good practice in tacking externa fraud Understanding and managing the risks of fraud 17 Focusing resources on the most effective anti-fraud measures 1.17 There is no singe package of measures which can be appied universay by departments and agencies to tacke fraud. Measures need to be taiored to the type and size of threat faced. In deciding which measures to use and the extent to which to use them some departments have assessed the savings that coud be achieved by targeting their resources in a better way. Savings coud arise in three ways: The direct effects from recovering amounts defrauded. Where the measures invove reaocating resources into existing activities the department can ook at the current costs/ savings as a basis for estimating the return from increasing the eves of counter fraud activity. Where new measures are proposed, it is good practice to piot these beforehand to test and refine their operation, assess their ikey effectiveness and the type of savings that can be achieved. The preventive effect, through improved future compiance from those previousy detected committing fraud. For exampe, HM Revenue & Customs assumes that the VAT yied wi increase immediatey from businesses previousy detected committing VAT fraud, but that this additiona yied wi graduay reduce if no further checks are subsequenty made. The deterrent effects on others that become more compiant as they earn of the greater efforts being taken to crack down on fraud. In practice it can be very difficut to assess these deterrent effects with any accuracy and Departments do not aways seek to do so. Setting targets and monitoring performance 1.18 Some departments have set targets to stabiise or reduce fraud over a period of time (Figure 13). Focusing targets on the overa eve of fraud or oss is a good way of assessing performance, and generay a better measure than the amount of fraud or oss detected. The atter is difficut to interpret if the fu scae of fraud or oss is not known. Other measures of performance are usefu compements to estimates of tota fraud oss, such as changes in regiona eves of oss, the cost of tacking fraud compared to the return obtained and the rate of recovery of detected frauds. Figure 13 Exampes of Departmenta targets to reduce fraud and error The Department for Work and Pensions has a strategic objective to pay our customers the right benefits at the right time. Supporting this objective is the aim to drive down eves of fraud and error to deiver a reduction in benefit expenditure overpaid to 1.8 per cent and underpaid to 0.7 per cent by HM Revenue & Customs has a strategic objective to improve the extent to which individuas and businesses pay the tax due and receive the credits and payments to which they are entited. Key outcomes are: increase tax and nationa insurance contributions actuay received reative to the amounts that shoud be received; reduce the eve of incorrect tax credit payments as a resut of error and fraud as a percentage of finaised entitement; and maintain take-up of entitements to tax credits and chid benefit.

18 18 Good practice in tacking externa fraud Understanding and managing the risks of fraud 1.19 Performance data on outcome targets may not be avaiabe unti ong after the period measured due to the amount of work invoved in samping cases, checking, cacuation and vaidation of the resuts. To monitor performance in-year, managers may rey on output resuts to indicate whether the outcomes are ikey to be achieved. For exampe, managers may monitor: the resuts of operationa checks on transactions; fraud investigation activity and outcomes (see Part 3, Figure 31); number and types of sanctions imposed (see Part 3, paragraph 3.39 and Figure 39); rate of recovery of defrauded amounts detected (see Part 3, paragraphs ). Responsibiities for tacking fraud 1.20 The responsibiity for tacking fraud and managing fraud risks start at the top of the organisation within the senior management board. At this eve, ownership of fraud risks is assigned and responsibiities aocated for managing individua fraud risks. Athough everybody in the organisation has a roe to pay in tacking fraud (paragraphs 2.5 to 2.7), some departments have aso set up centra units or foca points with responsibiity for tacking externa fraud. These have coordinated work on deveoping the department s strategies ensuring their impementation, monitoring resuts and providing advice and guidance. Fraud can be a moving target as the scae and nature of the risks change, so that reguar monitoring of the situation is needed to identify and respond to new threats. A foca point for tacking fraud can hep. HM Revenue & Customs, Department for Work and Pensions and NHS Counter Fraud Service each have centra units (Figure 14). Where fraud numbers and osses are significant, departments aso have teams of professionay trained investigators or enforcement officers dedicated to investigating cases of fraud. Figure 14 Exampes on how departments have assigned responsibiities The NHS Counter Fraud Service remit To have overa responsibiity for a work to counter fraud and corruption within the Department of Heath and the NHS with particuar priority for countering fraud in Famiy Heath Services. The Department for Work and Pensions The Fraud and Error Strategy Division is a centra unit which advises on fraud poicy. Cases sent for investigation are either referred to the Fraud Investigation Service or Customer Compiance, both of which sit within Jobcentre Pus. The Fraud Investigation Service investigates potentia frauduent attacks against a benefits made by the Department or administered on its behaf. It concentrates on those cases that are ikey to resut in a sanction. Customer Compiance was set up in Apri 2006 to dea with cases where a fu crimina investigation is not deemed appropriate but where action is needed to: identify eves of incorrectness; put it right and recover any overpayment; and ensure future compiance. The approach invoves a robust face to face interview with customers in receipt of benefit. The approach is intended to target investigation resources more efficienty on deaing effectivey with the cases of fraud most ikey to resut in a crimina sanction Regardess of the arrangements in pace, departments and agencies need to ensure that someone is fuy responsibe for ensuring that the pans for tacking fraud are impemented in the way intended and that sufficient resources are in pace. That individua shoud aso be responsibe for performance against targets. There is no point in having a we thought out strategy if it is not then put into effect.

19 Good practice in tacking externa fraud Xxxxxxxxxxx 19

20 20 Good practice in tacking externa fraud Xxxxxxxxxxx Part 2 Deterring and preventing externa fraud 2.1 Deterrence invoves convincing potentia fraudsters that frauds against a department or agency are not worthwhie. Prevention measures aim to stop frauds entering departments systems. Effective mechanisms for deterring and preventing fraud are essentia eements in combating fraud. Reaisticay however, some fraudsters wi never be deterred and not a frauds wi be prevented. In these cases, prompt detection and professiona investigations are needed (Part 3). Measures to deter and prevent fraud can be costy and departments need to ensure they are we designed for greatest effectiveness. Figure 15 sets out the main eements for deterring and preventing fraud. In ooking at this Part of the guide ask yoursef whether your organisation: seeks to infuence customers and the wider genera pubic s attitudes to fraud; sends a strong message to potentia fraudsters that they are ikey to be caught and sanctions wi be imposed. For exampe are there press reeases on peope/businesses prosecuted and are there any targeted or wider campaigns regionay or nationay? considers the fraud proofing of new programmes; ensures fraud contros are appied consistenty and their use is monitored. What is the roe of Interna Audit in this? considers strengthening contros where new fraud risks appear or where fraud starts to escaate; has an anti-fraud cuture where staff understand the standards of conduct required and their persona responsibiities in preventing fraud; appying contros and reporting cases of suspected fraud.

21 Good practice in tacking externa fraud Deterring and preventing externa fraud 21 Figure 15 The main eements for deterring and preventing fraud Gain support of the pubic Create an antifraud cuture Get the message to fraudsters that they wi be caught Deterrence and prevention Strengthen contros in response to emerging threats Fraud proof new programmes Compy with existing contros Changing pubic attitudes to fraud Nationa and oca media campaigns continue to pay an effective part of the Department s strategy for tacking benefit fraud. As we as targeting hard hitting messages directy at actua and potentia fraudsters, they aso remind the pubic that benefit fraud is theft of honest taxpayer s money. Combining cear messages that fraudsters wi be caught and punished with continued awareness of the abuse of taxpayers money heps increase both deterrence and the socia unacceptabiity of benefit fraud two vita eements in maximising the fight against fraud. David Barr, Head of Fraud and Error Strategy Division, Department for Work and Pensions 2.2 Departments have sought to infuence the attitude of customers and the wider genera pubic to fraud by deterring those who might consider committing fraud and by making fraud sociay unacceptabe. The aim shoud be to get pubic support in the efforts to tacke fraud. Some deterrence messages Departments have used to deter potentia fraudsters from attempting fraud are: strong contros wi stop them from succeeding; foowed by it is ikey they wi be caught; evidence of their fraud wi then be discovered; they wi thus face penaties; and amounts defrauded wi be recovered.

22 22 Good practice in tacking externa fraud Deterring and preventing externa fraud 2.3 Departments have used a variety of methods to pubicise the success of their work, such as issuing press reeases and putting information on their websites of cases prosecuted. These are cost effective actions which can be used by smaer departments and agencies. In the United States of America the Interna Revenue Service issues periodic Tax Fraud Aerts on their website warning the pubic of the risks and costs of buying into tax evasion schemes, as we as providing information on the atest schemes, scams and cons. The Aerts set out in more detai the main strands of the Service s enforcement programme and the consequences of non compiance. For exampe, for empoyment taxes, the reevant Tax Fraud Aert provides detais of ega requirements, empoyer and empoyee responsibiities, exampes of tax evasion schemes, and data on how noncompiance has been deat with in the courts incuding specific significant exampes. To strengthen their message, some departments have used media campaigns (Figure 16). Before introducing a nationa campaign, piots can be used to test and improve the ikey effectiveness. Figure 16 Exampes of Departmenta campaigns to change pubic attitudes and to deter fraudsters The approach used by the Department for Work and Pensions The campaign has run in severa phases, each with a different message and using different communication media. This aims to: reinforce honest behaviour by customers; create a cimate of intoerance to benefit fraud among the wider pubic and undermine its socia acceptabiity; deter potentia fraudsters. From September 2001, the campaign focused on dishonest caimants. Messages of deterrence and detection aimed to raise the fear of getting caught and portray the ikey consequences. Scenarios used in teevision advertisements showed benefit fraudsters being caught or punished, or both. In addition, the campaign used radio and regiona press advertisements, the atter featuring rea newspaper headines from fraud prosecutions. In June 2003, the next phase used the sogan We re on to you. It featured a spotight that foowed fraudsters in reaistic scenarios, such as at work, to show them that they woud be found out if they were continuing to caim benefits to which they were no onger entited, and to warn potentia cheats that benefit fraud is a serious crime. In August 2003, a new phase used sma posters on ampposts and in pub washrooms. Aiming to ook ike an offer of cash in hand work, the message stated that continuing to caim benefits whie working was benefit fraud and woud ony earn the person a crimina record. In November 2003, this phase was adapted. Sma mock advertisements were paced in pubs, cubs and shop windows. The main message Do you want to earn s and sti sign on? was designed to catch the eye of those considering cash in hand work, whie continuing to caim benefits, to fund Christmas ceebrations, and then to deter them from faiing to report to the Department that they were now working. In 2006, the Department started it s No ifs, no buts phase of the campaign. This phase focuses on the benefit fraudster and targets their views that there may be excuses for committing benefit fraud which the campaign seeks to overturn. No if, no buts has been used in teevision and poster advertising, in oca press and magazines, buses, and door drops, as we as on the Department s website. The Department arranges for independent evauations of each phase of its campaigns. Research for the Department suggests that between 2005 and 2006 there was an increase from 75 per cent to 83 per cent of peope who strongy agree that it is wrong to caim benefits to which they are not entited. The approach used by HM Revenue& Customs to encourage sef-empoyed peope to register for tax HM Revenue & Customs ran a series of advertising campaigns at a cost of 2 miion specificay to raise awareness amongst the sef-empoyed to register for tax. It ran three advertising campaigns during the period June 2005 to March One of the main messages was it is a simpe process to register and we are there to hep, ring our hepine. The advertising campaigns focused on particuar groups who may have faied to register for tax such as e-traders and andords. The initia campaign costing 1.5 miion resuted in: An estimated additiona 5,000 peope caing the hepine to register. These are peope who may have otherwise joined or remained in the hidden economy. HM Revenue & Customs estimates that it wi coect 23 miion in tax over three years from these peope; A further 3,300 peope caed the hepine to request the Thinking of working for yoursef? and if these peope register HM Revenue & Customs woud coect an estimated 15 miion in tax over three years.

23 Good practice in tacking externa fraud Deterring and preventing externa fraud To maximise the deterrent effect, departments have: researched fraudster behaviour and risk taking/ aversion to determine which messages wi be most effective in changing their behaviour; designed media messages to achieve maximum effect; used reevant media to ensure potentia fraudsters are aware of these messages; refreshed messages reguary to maintain a strong deterrent effect; deveoped performance indicators to evauate the effectiveness of the approach. It can however be difficut to make a direct ink between the campaign and reductions in fraud eves, because of other anti-fraud measures aso in force; fed back the evauation into renewed campaigns to deter fraudsters. Changing staff attitudes to create an anti-fraud cuture Creating an anti-fraud cuture invoves having a cear statement of ethica vaues, promoting staff awareness, recruiting honest staff and maintaining good staff morae Managing the Risk of Fraud: A Guide for Managers, HM Treasury 2.5 Creating an anti-fraud cuture, in which a staff understand the standards of conduct required, their persona responsibiities in preventing fraud and the importance of contros, is vita in preventing externa fraud, as Figure 17 iustrates. Pubicising internay the organisation s strategic approach to tacking fraud and what it is trying to achieve can be a good way of reinforcing the antifraud cuture. Figure 17 The NHS Counter Fraud Service have emphasised the vaue of an anti-fraud cuture in preventing fraud The creation of an anti-fraud cuture is one of the primary aims of the NHS Counter Fraud Service. Loca Counter Fraud Speciaists, who are empoyed by heath bodies, are aso responsibe for creating an anti-fraud cuture and are we paced to target those areas within their heath bodies where fraud awareness may sti be ow. The NHS Counter Fraud Service has impemented a number of initiatives to raise awareness of the issue and the roe that NHS staff can pay in preventing and deterring fraud, incuding: deivering fraud awareness presentations to key audiences both within the NHS and to externa agencies; pubishing the quartery magazine Insight which spreads good practice and seeks to strengthen further the anti-fraud cuture; obtaining media coverage in newspaper artices, radio broadcasts and teevision programmes. These can pay an important roe in raising fraud awareness and deterring those who may commit or seek to commit fraud; running awareness campaigns such as Fraud Awareness Month to promote the roe of the Loca Counter Fraud Speciaist, informing staff how they can report suspicions of fraud and generate referras; deveoping cose working reationships with stakehoders across the NHS and signing counter fraud agreements with reguatory and professiona bodies representing over 1.4 miion NHS staff and professionas; providing Loca Counter Fraud Speciaists with resources to assist them in raising awareness on a oca eve and in promoting the Fraud and Corruption reporting ine

24 24 Good practice in tacking externa fraud Deterring and preventing externa fraud 2.6 Training can hep raise staff awareness of the risks of externa fraud and the importance of compiance with interna contro procedures and security checks to prevent such frauds. And cose monitoring of staff compiance with these contros heps ensure their consistent appication. Training may take severa forms such as: fraud awareness workshops for a wide range of staff; targeted persona mentoring for staff working in areas found to be vunerabe to fraud; coser manageria supervision with feedback to staff on their compiance with security procedures. 2.7 A staff survey or focus group may be used to test staff attitudes to security, and their compiance with contros to prevent fraud. The findings from such research can hep identify opportunities to improve prevention and to strengthen interna contros, identify any messages that need to be reinforced, revea any areas where compiance with prevention contros is insufficient and generate further information about the frauds identified by staff. Contros to prevent fraud It is aways possibe to have contros which prevent fraud, but such contros need aso to enabe Departments to give a timey service to honest customers, without unacceptabe burdens. Designing effective contros depends on understanding the scae and nature of the risks and the costs. Caroine Mawhood, Assistant Auditor Genera, Nationa Audit Office

25 Good practice in tacking externa fraud Deterring and preventing externa fraud There are a range of contros (for exampe, physica checks, reconciiation, supervisory checks and cear roes and responsibiities) that address risk, incuding fraud. Departments and agencies need to consider which contros are most appropriate in their particuar circumstances. The consistent appication of interna contros can be highy effective in preventing fraud osses. Interna Audit shoud provide assurance on the operation of those contros and their effectiveness in preventing fraud. Interna contros can impose both interna and externa costs from their operation. Contros need to be designed which are proportionate to the risk, whie enabing the organisation to deiver the services to its customers to meet their needs. 2.9 Two key aspects to prevention are: fraud-proofing new programmes and systems; consistent appication of existing contros and strengthening of these where needed. Fraud proofing new programmes and systems 2.10 Organisations need to recognise their responsibiity when designing and impementing new poicies, programmes and systems to buid good contros in to manage fraud where there are vunerabiities or to fraud proof them by designing them to be inherenty ess vunerabe to fraud. Compex rues of entitement can increase the risks of fraud. Fraudsters can expoit the situation in two ways. The rues may be difficut to poice effectivey, requiring officias to consut voumes of guidance in their everyday work. Where customers are often uncertain of their obigations, it is easier for fraudsters to misrepresent their circumstances and if discovered caim that it was a genuine error The NHS Counter Fraud Service iaises with poicy eads across the Department of Heath and Government in an ongoing initiative to fraud proof the Department s future poicies. The aim is to ensure that, as far as possibe, new and existing NHS schemes are protected from fraud from the outset. Figure 18 shows steps taken to fraud proof the Department of Heath s Heathy Start scheme. The scheme offers free mik, fruit and vegetabes to mothers on eigibe benefits and beneficiaries are sent vouchers to exchange for these goods at approved retaiers. Retaiers then send the vouchers to the Department of Heath for reimbursement. Figure 18 The steps taken to fraud proof the Heathy Start Scheme Cear Rues Anaysis of the Heathy Start reguations to identify fraud risks. Introduction of appication forms for beneficiaries and retaiers to join the scheme. Strict terms and conditions with counter fraud information paced on a iterature. Prevention Physica security methods paced on the Heathy Start voucher anti-photocopying paper, watermarks. Accountabiity Unique information incuded on the Heathy Start voucher Nationa Insurance Number, unique barcodes and sequentia numbering. Appropriate decarations paced on forms to warn of the consequences of committing fraud. Monitoring Detection methods for Department of Heath subcontracted administrators a new payment machine that scans vouchers sent by retaiers for reimbursement which identifies copied vouchers and prevents dupicate payments. Warning etters automaticay sent to beneficiary and retaier if attempt at dupicate payment occurs. Sanctions Remova of retaiers from the scheme if convicted of fraud (in conjunction with a court ruing) Sufficient weight shoud be given to expert advice on the risks of fraud in new programmes and effective counter fraud measures shoud be integrated into the design. Where innovative schemes are being proposed, it is good practice to piot these to identify any further risks of externa fraud. Eary consutation with interna audit and counter fraud speciaists can hep to identify the risks, and to obtain advice on how these can be minimised, at key stages during design and impementation of new programmes. An evauation process is hepfu in determining whether eary risk assessments have been effective in countering fraud risks during deveopment, pioting and initia impementation.

26 26 Good practice in tacking externa fraud Deterring and preventing externa fraud Strengthening interna contros and checks 2.13 It is important that the effectiveness of contros is continuay reviewed. Contros which have traditionay worked we in countering fraud may no onger be effective where fraudsters have aunched determined attacks. Detected cases of fraud may show that fraudsters are using new methods to circumvent contros indicating that these need to be strengthened. Interna Audit s work may aso identify system weaknesses which coud ead to fraud In 2003, the NHS Counter Fraud Service set up a Fraud Prevention Unit to identify specific areas of weakness in systems, recommend poicy changes where appropriate and issue instructions and guidance to make them ess vunerabe to fraud (Figure 19). A recent fraud prevention review invoved the Unit examining the processes within trusts for identifying chargeabe overseas visitors. The findings and recommendations from the fraud prevention review have been provided to the Department of Heath to hep with deveoping further reguations and guidance. The atest fraud prevention instruction issued by the Unit arose from an investigation into an aegation of fraud that a Primary Care Trust was being fasey invoiced for services. The investigation ed to the introduction of standard caim forms, which incude counter fraud decarations. Figure 19 The methodoogy used by the NHS Counter Fraud Service Fraud Prevention Unit Action by the Fraud Prevention Team 1 Identify possibe areas of high risk of fraud in the NHS arising from system weaknesses. Sources wi incude the staff within the Counter Fraud Service incuding the Centra Inteigence Unit, Risk Measurement Team, Quaity Team and feedback from Loca Counter Fraud Speciaists. 2 Research the areas of system weakness and estabish the amounts at risk. 3 Organise a programme of visits to seected NHS Trusts to gain experience of how the systems work in practice. 4 Issue questionnaires to NHS Trusts to assist in identifying the scae of any system weaknesses. 5 Coate the findings and recommendations from the research, reviews and questionnaires. 6 Issue instructions, best practice guidance and/or recommend poicy change taking into account the amounts at risk. 7 Request compiance statements from NHS Trusts to provide assurance that good practice is being appied. 8 If appropriate, recommend that Loca Counter Fraud Speciaists carry out proactive fraud prevention work into the areas of high risk identified. 9 Measure the financia impact of the introduction of any new procedures. 10 Report findings, recommendations and action taken to Counter Fraud and Security Management Service poicy, operationa, and inteigence services as appropriate.

27 Good practice in tacking externa fraud Deterring and preventing externa fraud Strengthening interna contros can aso hep prevent or reduce crimina attacks. For exampe, the most serious VAT fraud that HM Revenue & Customs currenty face is VAT missing trader fraud a Europe-wide systematic attack on the VAT system orchestrated by organised crime groups. As part of its strategy, introduced in September 2000, to tacke this fraud, HM Revenue & Customs enhanced their VAT registration checking procedures to make it easier for it to detect and prevent bogus registrations. Each VAT appication form is assessed according to various risk criteria to identify potentia fraudsters. Further checks are carried out on suspect appications and those that remain suspect are visited. The number of suspect registration appications refused increased from around 3,500 in to around 6,000 in New egisation may be required to improve contros. Figure 20 iustrates how improved egisation has heped to reduce fraud in HM Revenue & Customs. Figure 20 Shows how HM Revenue and Customs has improved fraud contros by introducing new egisation Missing Trader Intra-Community fraud is a systematic crimina attack on the VAT system that is recognised as a serious probem throughout the European Union. The Government s strategy for tacking it was strengthened in response to a rapid and arge rise in attempted fraud in This targets the peope, goods and money fows associated with the fraud using measures incuding: egisation to make traders jointy and severay iabe for VAT debts and to change the VAT accounting practice on domestic business-to-business suppies (a reverse charge ) for the most commony traded goods which invove missing trade fraud; 2.17 Deveopments in technoogy can provide opportunities to strengthen contros in a cost effective manner to reduce the eve of externa fraud. Figure 21 iustrates how new technoogies are used to prevent fraud. Figure 21 Exampes of how new technoogy is being used to prevent and detect fraud in the private sector The Association for Payment Cearing Services (APACS) has reported that the success of chip and PIN has meant that faceto-face credit card fraud has continued to drop, faing, in 2006, a further 47 per cent on 2005 eves. The chip cards prevent counterfeiting of cards by hoding data securey preventing copying or ateration. The use of a four digit number PIN number aso heps reduce the use of stoen cards. The insurance industry uses Voice Risk Anaysis to speed up the processing of caims and to identify possibe higher risk caims. The technoogy detects minute changes in a caer s voice caused by stress eves which give signs as to when the caer may be ying about their circumstances. It then assigns a risk profie to that caer. This prompts the ca hander to probe more deepy into the information provided by the caer. There has been some criticism about the use of voice risk anaysis for the purpose of tacking fraud with it being described as a ie detector that coud deter genuine customers from caiming benefits. The Department for Work and Pensions is triaing the technoogy within 7 oca authority sites on Housing Benefit and Counci Tax Benefit caims and a Jobcentre Pus ca-centre. Expenditure on Voice Risk Anaysis by the Department in was 700,000 of which 460,000 was paid directy to oca authorities pioting the technoogy. In May 2008 a further 1.5 miion was made avaiabe for 15 oca authorities trias. The Department wi evauate the technoogy in August 2008 and has commissioned socia research into the customer experience and whether or not it deters genuine customers from caiming benefits. Foowing this evauation, the Department wi assess whether the technoogy shoud be tested more widey. operationa activity, incuding rigorous checks on suspect appications for VAT registration, and the use of dedicated teams to carry out in-depth verification of suspect VAT repayment caims; and crimina investigation and prosecution of those invoved in the fraud, and the recovery of the proceeds of their crimes. As a resut of the strategy, HM Revenue & Customs estimate that attempted fraud has faen by up to 1.5 biion in , to between 2.25 biion and 3.25 biion.

28 28 Good practice in tacking externa fraud Xxxxxxxxxxx Part 3 Detecting and investigating externa fraud and imposing sanctions 3.1 To show that departments and agencies are serious about tacking externa fraud, they need to detect cases of fraud against them; investigate them where appropriate and impose sanctions which are proportionate to the crime. This wi hep to deter potentia fraudsters in the future by showing that crime does not pay, especiay if the outcomes of cases are we pubicised. Departments aso need to consider whether the frauds detected show new threats are emerging, or are on a arger scae than originay thought. From this work, departments and agencies wi need to consider whether their strategic approach needs updating. They wi aso need to assess whether any frauds revea systemic weaknesses which need to be tacked (Figure 22). In ooking at this Part of the guide, ask yoursef whether your organisation: has a we pubicised teephone hotine, emai and freepost address to which the pubic can report cases of suspected fraud; uses techniques proactivey to detect cases of suspected fraud such as in-depth investigative work into hotspot areas, data matching exercises, data mining and neura networks as appropriate; assesses whether suspect cases of fraud need to be investigated further such as through the use of scoring systems; assesses whether the number of investigations is proportionate to the potentia sums ost from fraud; tracks the progress of individua investigations; has sufficient investigative staff with the essentia technica knowedge and experience; reviews independenty the way fraud investigations have been conducted; imposes appropriate sanctions on fraudsters such as fines, or other penaties, or refers appropriate cases for crimina prosecution; seeks to recover the amounts ost from fraud; evauates the effectiveness of sanctions; works with others to tacke fraud.

29 Good practice in tacking externa fraud Detecting and investigating externa fraud and imposing sanctions 29 Figure 22 The actions to dea with individua cases of fraud Detect cases using various methods Work with others to tacking fraud Investigate cases of fraud Actions to dea with individua fraud cases Evauate the effectiveness of sanctions Impose sanctions incuding: Fines and other penaties Crimina prosecution Recovery of assets Detecting fraud Effective work to detect and investigate fraud must ie at the heart of any professiona, integrated approach. Such work is not ony about eventuay appying appropriate sanctions, it has to be about earning as much as possibe about the nature of the probem, about deterring fraud where this is possibe and about identifying the poicy or systems weakness which has aowed it to occur. Dermid McCausand, Managing Director, NHS Counter Fraud Service 3.2 Frauds may be detected in a number of different ways. Referras may come from staff who have carried out checks on transactions and suspect a fraud. Members of the pubic may contact departments about their suspicions. Departments may aso use a range of techniques and technoogies to identify suspicious cases for further investigation. They may aso carry out specia pro-active exercises to detect fraud in high risk areas. Fraud investigators may deveop their own inteigence by foowing eads on existing cases where there may be inks to other frauds. This section focuses on the use of hotines and computer software techniques. Hotines 3.3 Hotines can be a cost effective way of obtaining from staff and the pubic detais of possibe cases of externa fraud which can be assessed and investigated further. Figure 23 outines the arrangements in pace in some departments. Good practices incude: setting up a singe freephone teephone number, with aternative means of contacting the department incuding an emai and freepost addresses; advertising the teephone number and contact detais on the department s website, in eafets and posters, advertisements during anti fraud campaigns; giving undertakings on confidentiaity; indicating the information that is usefu in a referra, incuding the types of frauds that the department are particuary interested in hearing of and how the department wi dea with the information provided.

30 30 Good practice in tacking externa fraud Detecting and investigating externa fraud and imposing sanctions 3.4 It is aso good practice to record information received onto a standard form. This can hep in prompting the person making the referra into providing as much reevant information as possibe. An eectronic version of the form can be incuded on a website, which can be competed and submitted anonymousy onine. The person may want to know what action may be taken and feedback on what has happened. Whie it is possibe to give genera information on how referras are handed, it may we not be possibe to give specific detais on individua referras where this woud breach confidentiaity requirements. 3.5 Hotines shoud be evauated at reguar intervas, for exampe, anaysing the number and type of referras received, what has happened in each case, and overa resuts. Figure 23 How some departments are using hotines The Department for Work and Pensions has a dedicated teephone hotine ( ) and website ( uk/benefit-thieves) to support the pubic in reporting suspicions of benefit fraud. In , the hotine received 250,000 cas from the pubic. The NHS Counter Fraud Service has a confidentia NHS Fraud and Corruption Reporting Line ( ) which NHS staff and the pubic can ring to report suspicions of fraud. Members of the pubic can anonymousy report a vehice that appears to be unicensed and used/kept on the pubic highway to The Driver and Vehice Licensing Agency by using the nationa untaxed vehices teephone hotine on freephone ( ), by competing an onine form or by writing to a DVLA oca office. The Agency receives around 10,000 reports a month from the pubic. In November 2005, HM Revenue & Customs set up a confidentia Tax Evasion Hotine ( ) aowing members of the pubic to report suspicions on the evasion of income tax, corporation tax, capita gains tax, inheritance tax, VAT or Nationa Insurance contributions. The hotine takes cas Monday to Friday 8am to 8pm, Saturday and Sunday 8am to 4pm. Peope can aso report detais of their suspicions through the Department s website by competing an onine form. In March 2006 the Department commissioned a major advertising campaign, using a combination of TV, press and radio to encourage members of the pubic to use the hotine. The hotine received around 120,000 cas in The number of cas has averaged 7,000 a month since then. The use of computer techniques to detect fraud 3.6 A range of techniques using computer software and technoogies can be used to detect cases of fraud. These incude techniques such as data matching, data mining and neura networks. Smaer departments and agencies may be abe to draw on the experience and essons of others in the use of these techniques. 3.7 Data matching invoves computerised scanning of data hed in different data fies either within the same organisation or in different organisations. It can be used by management for a range of purposes incuding detecting potentia fraud. With increasing computer power, data matching across fies is possibe on a very arge scae. 3.8 To hep focus resources on the matches which indicate possibe fraud, data matching software: highights the highest priority matches; aows users to fiter ony those matches that meet investigators criteria for investigation; expains the importance of each match type and protocos for sharing information between matched bodies. 3.9 The Audit Commission s Nationa Fraud Initiative is the country s argest data matching exercise in reation to fraud (Figure 24). The Nationa Fraud Initiative wi be expanded in future years as a resut of a new statutory power under Part 2A to the Audit Commission Act 1998 as amended by the Serious Crime Act The Act enabes bodies other than those that have a mandatory requirement to provide data for the Nationa Fraud Initiative to vounteer to participate by providing data to the Commission. This means that centra government departments and agencies can participate in future data matching exercises. Figure 25 to 27 shows how the Department for Work and Pensions, the Driver and Vehice Licensing Agency and HM Revenue & Customs use data matching to detect evasion. 10 Serious Crime Act

31 Good practice in tacking externa fraud Detecting and investigating externa fraud and imposing sanctions 31 Figure 24 The Audit Commission s Nationa Fraud Initiative The Nationa Fraud Initiative is the country s argest pubic sector anti-fraud exercise. It is a computer based system which is run every two years and matches information such as housing benefit caims, pensions and socia housing records from oca councis, the NHS, poice authorities, oca probation boards and fire and rescue authorities across Engand. The matching process enabes pubic bodies to share and compare information through a secure website and identify those taking services or money that they are not entited to. Typica exampes incude counci tenants renting (and subetting) counci property in two different authorities, frauduent caims for housing benefit, pensions being caimed for deceased peope, frauduent use of bue badge parking passes and dupicate trade creditor payments. A record 140 miion in fraud and overpayment was detected by the Nationa Fraud Initiative in The Nationa Fraud Initiative has now detected around 450 miion in fraud and overpayments since it started in Figure 25 Shows how the Department for Work and Pensions is using data matching The Department for Work and Pensions has a dedicated Database and Matching Service to identify possibe fraud and error. It matches data: across benefit systems; between other Government departments and Department for Work and Pensions data; for other Government departments; for Loca Authorities on Housing and Counci Tax Benefits; to tacke interna fraud. The Department for Work and Pensions is currenty modernising its Database and Matching Service to deiver a more timey and streamined service to its customers. The objectives are to: modernise the extraction of data from data systems to provide eectronic transfer, cutting down on cerica and manua interventions; operate a fraud and error referra system from a modern IT patform which aows new risk anaysis techniques to be used; provide an integrated service centred around the customer rather than the product; provide more frequent and timey referras. The programme aims to provide reguar and timeier feeds into the data matching system to prevent frauds from being in the system for ong periods of time. Fraudsters are therefore more ikey to be caught sooner, and overpayments are expected to reduce as a resut. Figure 26 Shows how the Driver and Vehice Licensing Agency uses data matching to detect vehice excise duty evasion Automatic Number Pate Recognition (ANPR) cameras, some fixed and some mounted in mobie units, coect images of stationary and moving vehices. Computer software searches for images of number pates and records the registration mark. The recorded mark is then compared with detais from the Agency s vehice database. If a match is found, the image and event detais are stored as evidence of the offence. The images of vehice registration marks that are not matched to the unicensed vehice data bank are discarded. Figure 27 Exampe of how HM Revenue & Customs has used data matching Using data matching, HM Revenue & Customs has identified peope who may have received income from property but have not discosed it. Foowing on from an advertising campaign, HM Revenue & Customs is sending etters to taxpayers for whom it hods information suggesting that they may have received income from etting property. The etters contain cear guidance, aong with a standard form that can be used to return detais of income received. Where no repy is received, HM Revenue & Customs considers what further action needs to be taken which coud incude forma enquiry or, in exceptiona cases, crimina investigation.

32 32 Good practice in tacking externa fraud Detecting and investigating externa fraud and imposing sanctions 3.12 The Information Commissioner has issued guidance on his website isting the eight principes put in pace by the Data Protection Act 1998 which ensure that information is handed propery. These are that data must be: fairy and awfuy processed; processed for imited purposes; adequate, reevant and not excessive; accurate; not kept for onger than is necessary; processed in ine with the individua s rights; 3.10 The Serious Crime Act 2007 enabes a pubic authority to discose information as a member of a specified anti-fraud organisation for the purposes of preventing fraud or a particuar kind of fraud. The information may be of any kind and may be discosed to the specified anti-fraud organisation, any members of it or any other person to whom discosure is permitted. As part of the reguatory impact assessment for the new provision, four pubic sector organisations provided in tota 1,636 records to match against CIFAS database. CIFAS is a Fraud Prevention Service with 270 member organisations spread across banking, credit cards, asset finance, retai credit mai order, insurance and other sectors. Neary one third of the records matched demonstrating that many of those who commit fraud against one organisation aso commit fraud against others Data matching between different bodies is faciitated greaty by common data descriptors but is possibe ony if there is appropriate authority for data to be transferred or shared between these bodies. This authority may derive from a statutory basis for demanding, or discosing, the data or both. Uncertainty regarding powers to share data may sometimes have hindered the use of data matching. Data matching has aso raised concerns about the possibe infringement of individua rights to privacy. Concerns about individua privacy are the subject of the Data Protection Act and the Human Rights Act secure; not transferred to countries without adequate protection. 11 The Commissioner has aso produced guidance on impementing these principes in the documents on Compiance advice: Data sharing between different oca authority departments; Framework code of practice for sharing persona information; Sharing persona information in the pubic sector: A new approach and Sharing persona information: Our approach Schedue 8 of the Human Rights Act 1998 gives rise to questions about the extent to which data matching compies with the provisions regarding persona rights to privacy. There are exceptions to these provisions where: necessary in a democratic society in the interests of nationa security, pubic safety or the economic we being of the country, for the prevention of disorder of crime, for the protection of heath or moras, or for the protection of the rights and freedoms of others Eight principes of the Data Protection Act 1998, Information Commissioner s Office 12 Human Rights Act 1998 Chapter 42

33 Good practice in tacking externa fraud Detecting and investigating externa fraud and imposing sanctions Data mining is the process of seecting, exporing and modeing arge amounts of data to revea previousy unknown patterns, behaviours, trends or reationships which may hep to identify cases of fraud. Because of the arge amount of data that need to be anaysed, speciaist computer software is used which usuay contain a range of data mining toos (Figure 28). A number of software companies have deveoped such products. Data mining can be a powerfu way of interrogating data and reveaing anomaies that woud not be reveaed by other techniques. However, to enabe it to function most effectivey, staff need to be trained in the use of the software, and to gain experience in seecting the most appropriate toos to scrutinize the data and in foowing up anomaies to detect cases of fraud. Figure 28 How some organisations have used data mining The BBC uses data mining software toos to match detais of icensabe paces with externa commerciayavaiabe data to identify specific paces or segments of the popuation for targeted enforcement activity. The faciity has been used to target maishots, posters and teephone chasing on specific groups, such as students, and offers opportunities to focus enforcement activity on paces or segments of the popuation where the ikeihood of evasion is greatest. The NHS Counter Fraud Service uses data mining and anaysis software to examine pharmaceutica and denta data. The software is capabe of advanced data anaysis that estabishes data profies and highights anomaies. These can indicate potentia fraud for further investigation. For exampe, the software highights any individua dentist caiming for unusuay high eves of intricate or expensive work. These cases are identified for potentia investigation of whether the dentist is making frauduent caims Neura networks are computer based mutiprocessing systems which are designed to connect data from mutipe sources to identify structures and patterns and exceptions to an identified structure or pattern. The abiity of neura networks to identify patterns of activity and exceptions to a pattern that may be associated with fraud, gives organisations an abiity to focus their detective efforts on these exceptions One of the probems of using these techniques more widey in the pubic sector is that the data may not be hed in a way that ends itsef to such anaysis. The move towards providing services onine may change this and aow rea time anaysis of transactions through Departments websites using some of these techniques. Investigating cases of fraud 3.17 Where fraud has occurred, the Department shoud consider: stopping the fraud at the eariest opportunity and ook at whether weak contros have been expoited which need to be tightened up; whether to refer the case for crimina prosecution or impose a penaty; coecting any arrears and any penaties to ensure that the economics of the crime are undermined and to deter others Some departments have criteria or scoring systems to determine those that shoud be investigated with a view to prosecution with the remainder subject to other forms of sanction. An inteigence and risk based approach to assigning cases for investigation is used by the Department for Work and Pensions (Figure 29) and HM Revenue & Customs has set out criteria for cases submitted for crimina investigation when serious frauds are suspected (Figure 30). Unike the arger Departments, smaer departments and agencies may not have trained fraud investigators in which case they wi need to invove the Poice in deaing with detected cases of externa fraud. They aso need to have pans in pace so that they know the steps to take if cases are uncovered.

34 34 Good practice in tacking externa fraud Detecting and investigating externa fraud and imposing sanctions 3.19 Departments aso need to ook at whether the tota number of investigations is commensurate with the potentia sums ost from fraud. The costs of investigating cases can be resource intensive. Assessing the financia return achieved on the overa caseoad, and different categories of case wi indicate the ikey benefits of undertaking more investigations or a different mix. Figure 29 The Department for Work and Pensions has a dedicated inteigence function to evauate cases of potentia fraud and decide which to investigate The Department for Work and Pensions has a network of 10 Operationa Inteigence Units ocated across Engand, Scotand and Waes. When referras are received for investigation the Fraud Referra and Intervention Management System (FRAIMS) appies a rues engine that determines routes for investigation and provides a Gateway through which cases can be channeed. This carifies whether cases shoud be investigated via the Fraud Investigation Service or Customer Compiance route (see Figure 14). For fraud investigation cases, the Operationa Inteigence Units then gather and evauate inteigence on cases to assist fraud investigators, incuding exchanging information with other agencies where appropriate Tracking the progress of fraud investigations aows managers to assess the overa workoad (such as whether investigations are concentrated on the main types of fraud set out in the Department s strategy); identify probem areas such as where progress is sower than woud be expected; understand the cost impications of investigations and the effects on panning future resource usage or the consequences of increasing or decreasing resource eves (Figure 31). Figure 31 Information to hep managers track fraud investigations Usefu information can incude: number of investigations carried out; vaue of fraud oss identified; duration of individua investigations; costs of individua investigations; methods of investigations used; outcome of investigations (prosecutions; fines etc). Figure 30 HM Revenue & Customs uses criteria for deciding which cases to refer for crimina investigation HM Revenue & Customs has a pubished poicy setting out the types of behaviour it wi usuay consider for crimina investigation. These incude factors such as whether the fraud is an attack on the system, the type of person carrying out the fraud and previous behaviour. The objective of the criteria is to ensure that resources can be targeted where they have the greatest effect on compiance. Whenever a case is referred for possibe crimina investigation HM Revenue & Customs evauate it against these criteria, as we as considering the strength of avaiabe evidence and ikeihood of a successfu investigation Where departments investigate frauds they wi need to consider whether there are sufficient staff with the right technica and investigative knowedge and experience. The Counter Fraud Professiona Accreditation Board was set up in 2001 as a professiona body for counter fraud speciaists. It sets professiona conduct guideines for its members who have successfuy competed the speciaist counter fraud training and university courses. Figure 32 sets out the deveopment of a professiona and ethica approach to countering fraud by the NHS Counter Fraud Service.

35 Good practice in tacking externa fraud Detecting and investigating externa fraud and imposing sanctions 35 Figure 32 The NHS Counter Fraud Service approach to counter fraud work In 1999 the NHS Counter Fraud and Security Management Service devised a professionaism strategy, which set out the competencies required for counter fraud speciaists and estabished a series of practica skis-based training courses that coectivey formed a Foundation Leve. It is mandatory for everyone in the NHS who counters fraud to receive this training and by 2008 over 800 peope in the NHS have successfuy competed it. The Department for Work and Pensions, HM Revenue and Customs, oca authorities, Identity and Passport Service, the Border and Immigration Service, Chid Support Agency, Abbey Nationa and Capita Ltd. Have a adopted the Foundation Leve and the quaification of Accredited Counter Fraud Speciaist. By 2008, over 10,000 hed the quaification. The Foundation Leve has been accredited by a number of universities speciaising in crimina justice. The Certificate of Higher Education in Counter Fraud and Crimina Justice Studies (Advanced Leve) is the first academic quaification specificay for the counter fraud fied and hundreds of students have proceeded from the Foundation Leve to study for the Advanced Leve. There is aso a BSc. (Hons) in Counter Fraud and Crimina Justice Studies. Since 2007 there is an NHS-deveoped MSc Counter Fraud and Corruption Studies specificay designed to meet the needs of graduates and professionas speciaising in countering fraud and corruption. It seeks to equip students with a critica knowedge base foowing the introduction of the Fraud Act 2006, the Government s Fraud Review and the changes in corruption egisation Other arge departments provide a range of training for fraud investigators. For exampe, HM Revenue & Customs has deveoped procedures, provided training on these for a investigators and impemented an assurance process to ensure that investigators adhere to them. New investigation recruits are given 14 weeks of training in basic investigation techniques which can be foowed by more speciaist courses such as the handing of informants. The investigators are aso supported in their work by speciaists such as forensic accountants and computer anaysts In the Department for Work and Pensions, a programme of training is provided for investigative and inteigence officers and managers as we as those from oca authorities. Investigators in the Fraud Investigation Service receive Professionaism in Security (PINS) training which is accredited by Portsmouth University. Additiona guidance is provided in the Fraud Procedures and Instructions Manua which is reguary updated Investigations into fraud shoud be consistent with the aims of the crimina justice system to reduce crime and the fear of crime and to dispense justice fairy and efficienty, promoting confidence in the rue of aw. 13 Fraud investigations need to be of high quaity. Independenty reviewing the way in which fraud investigations have been carried out can hep to ensure that appropriate standards and ega requirements have been foowed. The findings can highight areas where improvement is needed. The reviews can be undertaken by: independent interna teams, with expertise in fraud investigation, to review the conduct and quaity of fraud investigations; Surveiance Commissioner ( surveiancecommissioners.gov.uk/) for use of covert investigative methods; the appointment of an externa assessor. 13 Aims and Objectives; Crimina Justice System

36 36 Good practice in tacking externa fraud Detecting and investigating externa fraud and imposing sanctions Figure 33 sets out how the Department for Work and Pensions ensure professiona standards are maintained through its emphasis on standards assurance. HM Revenue & Customs approach to assuring standards is set out in Figure 34. Figure 33 How the Department for Work and Pensions reviews investigations The Department operates a number of systems to ensure they are compiant with the crimina justice system: investigations and prosecutions are run separatey; authority to gather inteigence under the Socia Security Act 2001 is ony avaiabe to a imited number of authorised officers who must have undergone reevant training and hod an appropriate icence; a Quaity Assurance Framework has been impemented within the Fraud Investigation Service which incudes checks on key eements of the inteigence and investigation processes to ensure these compy with instructions. The Department for Work and Pensions set up the Fraud Investigation Service in 2006 to investigate cases ikey to resut in a crimina sanction. The Service has deveoped a Strategy and Panning Support Team structure to improve the standards of fraud investigations and ensure that Learning and Deveopment products meet the changing needs of investigation staff. The Surveiance Commissioner reports annuay on the use of covert methods in investigations and compiance with the Reguation of Investigatory Powers Act 2000, Data Protection Act 1998 and Human Rights Act The resuts of these assurance checks are used to identify areas for further deveopment and scope for improvement. The Products and Services Management Division (P&SMD)is responsibe for ensuring that the Fraud Procedures and Instructions Manua compies with egisative and operationa requirements. This ensures investigators, when engaged in fraud investigations and authorised surveiance, are abe to gather sufficient, reevant and reiabe evidence, in a egay compiant manner for use in any prosecution or sanction. This guidance was deveoped in iaison with oca authority partners and made avaiabe to them. The Department used Home Office guideines as a basis for their written procedures on the use of surveiance. Figure 34 The arrangements used by HM Revenue & Customs to review investigation processes A separate team in HM Revenue & Customs Crimina Justice and Enforcement Standards is dedicated to maintaining professiona standards in crimina investigation. This team is independent of the Crimina Investigation directorate. Crimina investigation is aso subject to review by the Department s Interna Audit directorate. Externa oversight is provided by the HM Inspectors of Constabuary and the Scottish and Northern Ireand inspectors. Compaints about investigations can be made to the Poice Compaints Commission, the Ombudsman and the Adjudicator. Imposing sanctions Effective sanctions can do more than simpy punish the individua: they have a wider deterrent impact and shoud aso provide an incentive to encourage the non-compiant to return to compiance. Visibe sanctions that work can reinforce the perception among compiant taxpayers that the system is fair. Simon Norris, Deputy Director, Review of Powers Team, HM Revenue and Customs 3.25 Where investigations find evidence of fraud, departments wi usuay seek to impose some form of sanction. The purpose is to deter others from carrying out simiar types of fraud against the organisation; recover the money defrauded and punish the fraudster by imposing a penaty, such as a fine, or confiscating an asset, or by prosecuting them criminay in the courts. Some departments have pubished their approach to deter potentia fraudsters and ensure that a consistent approach is taken. The NHS Counter Fraud Service imposes parae sanctions to increase the deterrent effect (Figure 35). Departments need to consider whether the eve and range of sanctions imposed on fraudsters provide a sufficient deterrent.

37 Good practice in tacking externa fraud Detecting and investigating externa fraud and imposing sanctions 37 Figure 35 Exampes of departments approaches to imposing sanctions Where the Department for Work and Pensions has sufficient evidence of fraud to take a case to court they wi give a forma caution or impose an administrative penaty; or prosecute those invoved. Cautions are usuay aimed at the ess serious benefit frauds and those where the overpayment is under 2,000. The offender must admit to the offence, and the amount overpaid must be paid back. (This appies to Engand and Waes. There are sighty different arrangements in Scotand). An administrative penaty is imposed where the vaue of the fraud is under 2,000. The person must repay the amount overpaid and an administrative penaty of 30 per cent of the amount of the overpayment. Whether cases are prosecuted depends on a number of factors. The vaue of the fraud shoud be over 2,000, the strength of the evidence shoud suggest that conviction is ikey and prosecution shoud be in the pubic interest. Magistrates or the Crown Court decide on the penaty to be imposed if the person is convicted. The NHS Counter Fraud Service seeks to combine the appication of discipinary, civi and crimina sanctions where fraud is found to dismiss an empoyee or suspend or de-register a professiona; to obtain civi aw orders to freeze assets and recover funds and to seek a crimina prosecution. The approach is set out in Appying the Appropriate Sanctions Consistenty, pubished in Apri 2003 and updated in December Copies can be obtained from HM Revenue & Customs operates a progressive regime of sanctions against fraud which is aimed at deterring and penaising fraud and evasion through significant financia penaties. Whist the deterrent threat of prosecution remains, and is appied in the most serious cases, the majority of evaders can be hit where it hurts most through financia penaties of up to 100 per cent of the sums being evaded. This approach seeks to match the eve of the penaty to the amount of the tax understated, the behaviour giving rise to the understatement of tax and the extent of co-operation by the taxpayer. Fines and other penaties 3.26 Fines and other penaties imposed on those committing fraud need to be recovered to ensure that they act as a deterrent. Fines and other penaties imposed on those committing fraud need to be recovered to ensure that they act as a deterrent. In 1999 egisation provided for a new civi penaty charge to be imposed where a person wrongy fais to pay any amount in respect of NHS charges or obtains goods or services to which they are not entited. If the fine is not paid by the due date then a surcharge of 50 per cent of the penaty can be made up to a maximum of It is important to monitor progress in recovering the fines and penaties invoved, incuding the enforcement of fines imposed by the courts for convicted fraudsters. Athough in such cases it is not the departments that coect the fines, they shoud consider working with the Ministry of Justice to determine the eve of fines actuay paid. The extent of re-offending may indicate whether sanctions provide a sufficient deterrent. The Department for Work and Pensions has previousy estimated that 9 per cent of those frauduenty caiming Income Support and Jobseeker s Aowance had been caught before. To hep combat this recidivism, the Department introduced egisation in 2002 to aow benefit to be stopped for 13 weeks if individuas were convicted of committing benefit fraud twice within the space of three years. The Department extended the inking period for benefit fraud offences from three to five years in the Wefare Reform Act The NHS Counter Fraud Service may take crimina proceedings against patients for serious cases of repeated or persistent fraud. This crimina offence is designed to compement the civi penaty charge and attracts a fine, on conviction of up to 2, Departments need to be aware of Artice 6 (1) of the Human Rights Act 1998, which provides for rights to a fair tria, which says, In determination of his civi rights and obigations or of any crimina charge against him, everyone is entited to a fair and pubic hearing within a reasonabe time by an independent and impartia tribuna estabished by aw. As a fina recourse, someone who has been subject to a fine or other sanction shoud be aowed to dispute the sanction in either the magistrates or the county court.

38 38 Good practice in tacking externa fraud Detecting and investigating externa fraud and imposing sanctions Crimina prosecution 3.29 The Fraud Act incudes three casses of fraud: Fraud by fase representation; Fraud by faiing to discose information; Fraud by abuse of position. In a three casses of fraud, the Act requires that for an offence to have occurred, the person must have acted dishonesty, and that they had to have acted with the intent of making a gain for themseves or anyone ese, or inficting a oss (or risk of a oss) on another. There are aso offences of fraud specific to particuar departments activities Preparing cases to the state of proof required for a crimina prosecution can take a ong time and invove significant resources. Decisions on whether to refer cases for prosecution may depend on whether: there is sufficient evidence to obtain a conviction; the case invoves a systematic attack on the department s systems and has ed to substantia amounts of money being ost; there is a history of re-offending; professionas such as awyers and accountants are invoved in the fraud; prosecution wi increase the deterrent effect These factors need to be baanced against the time and cost of bringing a case to court, and the avaiabiity of other forms of sanction which may be more appropriate. Some departments have aid down the circumstances in which they wi refer cases for prosecution to ensure they take a consistent approach in each case (Figure 36). Departments wi need to consider whether the number of prosecutions is commensurate with the potentia sums at stake in ost revenue, provide a sufficient deterrent and meet the wider pubic interest in prosecuting significant dishonest conduct. Figure 36 HM Revenue & Customs normay refer for prosecution the most serious cases of VAT evasion In Apri 2005 the Revenue and Customs Prosecutions Office was set up, under the superintendence of the Attorney Genera, as an independent prosecutor to prosecute cases referred to it by HM Revenue & Customs and the Serious Organised Crime Agency within Engand and Waes. In Scotand, cases are reported for consideration of prosecution to the Crown Office and Procurator Fisca Service. Prosecution is an important part of HM Revenue & Customs armoury to tacke VAT fraud. It refers the most serious cases of evasion for prosecution. Cases are referred where a strong deterrent message is required and the use of other options wi not achieve this. However, HM Revenue & Customs may investigate any case of suspected dishonest evasion of VAT with a view to referring the case for prosecution. Cases are referred for prosecution where HM Revenue & Customs pubished criteria for crimina investigation may appy. The Revenue and Customs Prosecutions Office wi consider whether the evidence is sufficient, whether prosecution is in the pubic interest and who shoud be charged and with what offence(s). It foows the Code for Crown Prosecutors, issued by the Director of Pubic Prosecutions that sets out the genera principes Crown Prosecutors shoud foow when they make decisions on cases. Revenue and Customs Prosecutions awyers can aso provide guidance to the crimina investigators of HM Revenue & Customs and Serious Organised Crime Agency during the investigation and prosecution process. 14 Fraud Act

39 Good practice in tacking externa fraud Detecting and investigating externa fraud and imposing sanctions 39 The recovery of money defrauded Asset recovery prevents crimina proceeds being reinvested in other forms of crime. By reducing the rewards of crime, it begins to affect the baance of risk and reward, and the prospect of osing profits may deter some from crime. Fundamentay it serves justice, in that nobody shoud be aowed to continue to profit from crime. The ong term purpose of recovering the proceeds of crime is to reduce harm. Asset Recovery Action Pan, Home Office, May The means of recovering assets whether in or outside of the UK may be achieved through the crimina process or through the civi courts. Where it is decided to embark on a prosecution then crimina remedies shoud take priority over civi. Any decision to proceed with a civi action shoud be taken in consutation with the prosecutor. As part of an investigation, departments may ook into the financia affairs of the suspected fraudster to see whether evidence can be provided to the court on the extent of the benefit obtained by the defendant, and to make a confiscation order. Before the suspected person or persons become aware that an investigation is taking pace, action may be needed to secure misappropriated funds by a crimina restraint order or seeking a civi injunction Where an organisation seeks to recover stoen monies through the civi courts it wi have to prove on a baance of probabiities, that it has cause of action against the defendant. Further the caimant wi have to prove the amount taken. If successfu the court wi then make an order against the defendant requiring him or her to compensate the caimant together with an award of costs in most cases. Lega costs can be high. Organisations wi need to consider: the amounts stoen and which therefore coud be recoverabe; the prospects of winning the case; the vaue of assets hed by the suspected fraudster; the ikey ega costs; whether it wi be possibe to pursue a civi action whist a crimina investigation is underway Organisations may be abe to recover stoen monies as part of crimina proceedings where the prosecutor secures a compensation order in favour of the victim as part of the offender s sentence. Further information about civi and crimina actions to recover assets can be found on the website of the Serious Organised Crime Agency The powers to deprive criminas of their assets increased substantiay as a resut of the Proceeds of Crime Act The egisation brought together and strengthened in one Act the drug and non-drug confiscation egisation previousy contained in the Drug Trafficking Act 1994 and Part VI of the Crimina Justice Act Under the Act, any offence that generates proceeds is capabe of attracting crimina confiscation, and the avaiabiity of restraint (court freezing of property pending confiscation) is brought forward from the point at which a person is about to be charged to any time from the start of a crimina investigation. The Act estabished the Assets Recovery Agency to investigate and secure crimina assets. Law enforcement agencies coud pass confiscation cases to the Agency for it to appy for restraint, to undertake the confiscation hearing or to enforce confiscation cases. In addition, the Act created a new civi recovery scheme, empowering the Agency to sue in the High Court to recover property derived from crime without the need for anybody to have been convicted of an offence. There is aso an option to tax the proceeds of crime where there are reasonabe grounds to suspect a person s income, profit or gain was derived from crime.

40 40 Good practice in tacking externa fraud Detecting and investigating externa fraud and imposing sanctions 3.36 The Serious Crime Act 2007 aboished the Assets Recovery Agency and transferred its civi recovery and taxation powers to the Serious Organised Crime Agency which now undertakes civi recovery and tax investigations in Engand, Waes and Northern Ireand. The Act aso conferred on the Revenue and Customs Prosecutions Office, the Crown Prosecution Service, the Serious Fraud Office and Pubic Prosecutor Northern Ireand the power to bring civi recovery proceedings in the same way as the Assets Recovery Agency Whist other agencies are deveoping their civi recovery capabiities, the Serious Organised Crime Agency is providing support by taking on cases referred to it, where the use of civi recovery and tax powers woud be in the pubic interest. Where a aw enforcement agency or prosecution authority has a crimina case which it has been unabe to prosecute successfuy it can refer the case to the Serious Organised Crime Agency. For a case to be considered for adoption by the Agency recoverabe property must have been identified and have an estimated vaue of at east 10,000; the recoverabe property must incude property other than cash or negotiabe instruments (athough cash is recoverabe if it is in addition to other property); and there must be evidence of crimina conduct that is supported to the civi standard of proof, that is, on the baance of probabiities. Figure 37 outines actions taken by the Department for Work and Pensions to recover assets and Figure 38 provides an exampe of Revenue and Customs Prosecutions Office appication of crimina confiscation powers The Home Office has recenty consuted on proposas to increase dramaticay the quantity of crimina assets recovered. Its proposas are set out in the Assets Recovery Action Pan of May Figure 37 Action to recover assets and seek crimina compensation The Department for Work and Pensions has emphasised the punitive, deterrent and disruptive effects of the recovery of assets deemed to be the proceeds of crime and are making use of the new powers of confiscation for offences committed under the Proceeds of Crime Act In , the Department obtained 136 confiscation orders, 33 compensation orders and 85 vountary payments as a direct resut of the Proceeds of Crime powers. These totaed 7.55 miion with a further 54,266 paid in costs to the Department for the confiscation proceedings. Figure 38 Revenue and Customs Prosecutions Office use of crimina confiscation powers to recover the proceeds of crime In the Revenue and Customs Prosecutions Office restrained miion in assets to prevent them being dissipated and obtained 445 confiscation orders worth 55.7 miion. The Revenue and Customs Prosecutions Office coected confiscation receipts of 24.2 miion at a cost of 2.2 miion. Revenue and Customs Prosecutions Office prosecuted the owner of an accountancy practice in Torquay athough he was not a quaified accountant. He took on the affairs of a weathy oca famiy who had recenty paid substantia amounts in Capita Gains Tax on the sae of their business. He submitted a frauduent caim for the repayment of the Capita Gains Tax on his cients behaf and forged their signature on a payment authority. He then stoe the refund of 300,000 and used it to pay for an extravagant ifestye. He peaded guity to forgery and was sentenced to 30 months imprisonment and a confiscation order of 370,000 with five years in defaut.

41 Good practice in tacking externa fraud Detecting and investigating externa fraud and imposing sanctions 41 Evauating the effectiveness of sanctions 3.39 Evauating the effectiveness of sanctions is not straightforward, mainy because of the difficuties in assessing the deterrent effect. In broad terms, the deterrent effect of sanctions wi be refected in whether the amount of fraud has reduced, athough it is hard to disentange the effects of sanctions from other action to reduce fraud as we as wider economic effects. Trends in the indicators given beow in Figure 39 can hep to determine whether the eve of activity may be having a desirabe effect. The Department for Work and Pension has carried out a fu review of its sanctions regime, which has been in use since It concuded that the regime was fit for purpose, but that more coud be done to increase caimants awareness of the sanctions imposed on benefit fraudsters. Figure 39 Exampes of indicators number of frauds identified; number of identified frauds with no sanction imposed; number of cases where re-offending has occurred; Working with others in tacking fraud 3.40 Individuas and businesses may be committing frauds against more than one government department or agency. Joint working enabes departments to identify common threats and poo their knowedge and expertise to investigate fraudsters. Other benefits of working together to tacke fraud are: good practice can be shared across departments; information can be exchanged more efficienty; skis, informa systems and cuture are deveoped across participating departments; a more consistent approach from the different departments can be deveoped; the consistency of information provided by customers to different departments can be tested; trust and understanding can be buit across departments. Figure 40 Exampes of working with others to tacke fraud number of forma cautions given; number of penaty charges imposed; amount raised by imposition of penaty charges; number of cases recommended for crimina prosecution; number of convictions achieved; amount of fraud oss and amount recovered; amount of confiscation order and amount recovered; amount of assets seized from the fraudsters. HM Revenue & Customs has worked in partnership with the Gangmasters Licensing Authority to identify businesses that shoud be registered for tax. The Gangmasters Licensing Authority reguates businesses that suppy abour for agricuture, horticuture, shefish gathering and food processing and packaging, to protect vunerabe workers from expoitation. During the Authority and HM Revenue & Customs discovered that at east 40 businesses that appied for a icence shoud aso have been registered for VAT. As a resut of this cooperation, those businesses have paid an extra 2 miion in VAT, in addition to nationa insurance contributions and tax for their 6,000 workers, and corporation tax. HM Revenue and Customs is making of use of suspicious activity reports to detect fraud. Under the Money Laundering Reguations organisations are required to make suspicious activity reports to the Serious Organised Crime Agency where they know or suspect that a transaction invoves money aundering. In the period Apri 2004 to the end of March 2007 HM revenue & Customs competed 7,150 investigations invoving over 27 miion in tax or around 3,800 in each case.

42 42 Good practice in tacking externa fraud Detecting and investigating externa fraud and imposing sanctions 3.41 Joint working arrangements can be set up to share information and data, discuss issues of common interest and carry out research (Figure 40). Such arrangements can be covered by a Memorandum of Understanding with the other organisations setting out for exampe the arrangements for sharing data and carrying out matching and profiing with their data. The sharing of data may be faciitated through data warehouses accessibe to the organisations invoved. The data warehouse can incude data from each organisation and from externa sources such as nationa insurance numbers, driving icences, passport hoders and eectora ros. In a piot project, HM Revenue & Customs is using speciaist computer software to anayse various interna and externa information to hep identify potentia ghosts (peope who work in the hidden economy and pay no tax on their earning) and moonighters (peope who pay tax on certain earnings but fai to decare other sources of income) Joint working may aso incude co-operation on fraud investigations. This enabes departments to identify and investigate cases of common interest, avoiding dupication of effort. The Wefare Reform Act 2007 enabes oca authorities to investigate suspected frauds on benefits administered by the Department for Work and Pensions. The Department aso cooperates with oca authorities to identify and jointy investigate cases. There are a number of factors that need to be considered for effective joint working in fraud investigations (Figure 41). Figure 41 Good practice in joint working to investigate externa frauds Aim to estabish with the other organisations invoved in joint fraud investigations: common criteria for seecting which cases to investigate; the time scaes for investigations; agreed procedures for deaing with cases by either the civi route or crimina prosecution; the different powers of the departments to investigate and the sanctions that can be used.

43 Good practice in tacking externa fraud Xxxxxxxxxxx 43

44 44 Good practice in tacking externa fraud Xxxxxxxxxxx Design and Production by NAO Marketing & Communications Team DG Ref: 8383RA Printed by Maxim Nationa Audit Office 2008