ACKNOWLEDGEMENTS

Merv Harrower, Director, Prevention, Compliance and Enforcement

Glen Tadsen, Manager, Prevention, PCE

Mel Weisgerber, Manager, Enforcement, PCE

In addition, without the enthusiastic support and assistance of all the American states which were contacted, this report would not have been possible. Special thanks are owing to Elaine Minkler, President, United Council on Welfare Fraud (UCOWF), Betty Apt, Atlanta Region, Office of the Inspector General, and the directorship and members of the UCOWF organization. Their help and encouragement was invaluable!

      C. A. (Tina) MacDonald                                                                       Sheila Blair

TABLE OF CONTENTS

TABLE OF CONTENTS

EXECUTIVE SUMMARY

INTRODUCTION 1

BACKGROUND 1

OBJECTIVES FOR PROJECT 1

METHODOLOGY 2

STRUCTURE OF REPORTS 3

EARLY FRAUD DETECTION AND PREVENTION 5

HISTORY OF INITIATIVE 5

OVERALL DESCRIPTION OF PROCESS 8

The Intake Interview 9

The Referral Process 9

The Feedback Loop 10

CHARACTERISTICS OF SUCCESSFUL EDP PROGRAMS 11

AMERICAN EXPERIENCE WITH PROGRAM 13

Data Matching 14

Administrative Disqualification 15

MAJOR BENEFITS ACHIEVED FROM PROGRAM 17

Cost-Benefit of Early Detection Programs 17

OTHER BENEFITS OF EDP PROGRAMS 18

MAJOR DIFFICULTIES EXPERIENCED 19

RELEVANCE FOR BRITISH COLUMBIA 20

DESIGNING AN EARLY FRAUD DETECTION AND PREVENTION PROGRAM 23

THE EARLY FRAUD PREVENTION PROGRAM 23

REFERRAL CRITERIA 24

RESULTS OF A FINDING OF ATTEMPTED FRAUD 27

STAFF TRAINING 28

HIRING ELIGIBILITY INVESTIGATORS 29

REPORTING STRUCTURES 31

CLIENT RIGHTS 33

TIMING OF INVESTIGATION PROCESS 34

LINKAGE ACROSS OFFICES 36

IMPLEMENTING AN EARLY FRAUD DETECTION AND PREVENTION PROGRAM 39

THE AMERICAN EXPERIENCE 39

SELECTING A PILOT SITE 40

DEVELOPING AN IMPLEMENTATION PLAN 41

DEVELOPING A COMMUNICATIONS STRATEGY 42

DEVELOPING AN EVALUATION/MONITORING PLAN 43

IMPLEMENTATION COSTS 44

CONCLUSION 47

HIGHLIGHTS OF NEW PROGRAM 47

SUMMARY 47

APPENDIX 1 - SUMMARY OF RECOMMENDATIONS 49
 
 

EXECUTIVE SUMMARY

The goal of Early Fraud Detection and Prevention programs in the delivery of Income Security is to detect and prevent fraud during the application/intake process so that assistance is denied before benefits are issued. This is in contrast to traditional fraud investigation programs where active assistance cases are investigated upon suspicion or a complaint of existing fraud.

There are currently 27 American states who have implemented these programs, and a recent study by the federal Office of the Inspector General has identified a weighted average cost-benefit ratios across all of these states of $8.43 saved for every dollar spent on implementation.

This report explores ways in which this American experience could be implemented in British Columbia. This implementation would have several major impacts:

1. Significant cost-benefits will be achieved. If British Columbia achieved even average returns, estimated at $8.43 saved per dollar expended in American states and counties which have implemented the program, this would provide a significant savings.

2. The implementation of an early fraud detection and prevention program will be seen publicly as having a major impact on improving program integrity.

3. The implementation of this program will clearly demonstrate government concern for responsible stewardship of public funds and improved accountability for expenditures.

4. Demonstrated concern for client rights will allay any possible criticism about inappropriate and reckless invasion of privacy.

5. A carefully designed and implemented program, with well-trained professional staff, would ensure that individual applicants for benefits were treated fairly and consistently through the application of a pre-designed set of referral criteria which would identify high-risk cases.

6. Staff, particularly in the intake area, will feel in greater control of the benefit approval process, and will have a greater ability to ensure benefits are granted appropriately in questionable cases.

Two reports have been developed as part of this project. The main report contains overview material, highlights of the most effective approaches to the various components of early prevention and detection, a proposed fraud detection and prevention delivery model for British Columbia, implementation strategies, and conclusions and recommendations.

The State Directory accompanying this report provides state-by-state descriptions, for all states which indicated that they had implemented, or were planning implementation, of an early prevention and detection strategy, and which sent details of the state program.

Given the Province's ever-increasing income support caseload with the attendant stresses on caseworkers, investigative staff, and all other income support staff, any initiative which provides some tools to help manage and control this caseload is worthy of consideration. A well designed and implemented front-end fraud and error detection and prevention program would meet a number of the goals of the PCE Division, as well as those of the Department as a whole.
 
 
 
 

INTRODUCTION
 
 
 
 

BACKGROUND

The establishment of the Prevention, Compliance and Enforcement Division within the Ministry of Social Services marked the beginning of a new approach to fraud and error control within British Columbia. The Division's mission statement is clear:

A major component of the recently prepared Strategic Plan for the Division is the development of a comprehensive prevention program, which must be integrated into every aspect of the Income Security program. It is recognized that there is no Canadian jurisdiction which has developed such a comprehensive prevention strategy although most American states have implemented prevention initiatives. This report uses this American experience to move towards the design of such a program for British Columbia.
 
 

OBJECTIVES FOR PROJECT

The objectives for the Early Fraud Detection and Prevention Project were defined as follows:

1. Conduct a review of welfare fraud prevention and early detection programs or initiatives in the United States.

2. Produce a report containing the following:

3. A summary or description of each state's prevention, deterrence or early detection program where one exists (approximately 25 states.

4. Descriptions of supporting policies, procedures, and computer systems.

5. Evaluation results, if available.

6. A contact person for each state identified.

7. An identification and analysis of:

8. those programs or systems which are the most innovative or developmentally significant and productive.

9. those programs, systems or innovations which would be compatible for introduction as improvements to BC's system.

10. Development of a model for selected programs or innovations identified under 2(e).

This project will provide an important first step in assisting the Prevention, Compliance and Enforcement division to establish a comprehensive front-end detection system for implementation within British Columbia.
 
 

METHODOLOGY

Every American state was contacted by telephone and interviewed to determine the status of their approach to early prevention and detection. Those states which had implemented prevention programs were identified, and a comprehensive follow-up request for information was forwarded to the state.

All states were extremely helpful in providing both verbal and written information describing their programs. Some states also provided detailed manuals, and training materials which were in use in their programs, as well as whatever evaluative material was available for the prevention initiatives. Follow-up phone calls assisted in clarifying information.

Much of the material collected was duplicative, but some of the more important and well-presented materials will be provided directly to the PCE Division.
 
 

STRUCTURE OF REPORTS

Two reports have been developed as part of this project. The main report contains overview material, highlights of the most effective approaches to the various components of early prevention and detection, a proposed fraud detection and prevention delivery model for British Columbia, implementation strategies, and conclusions and recommendations.

The State Directory accompanying this report provides state-by-state descriptions, for all states which indicated that they had implemented, or were planning implementation, of an early prevention and detection strategy, and which sent details of the state program.
 
 
 
 

HISTORY OF INITIATIVE

The goal of early detection and prevention in the delivery of Income Support programs is to detect and prevent fraud during the application/intake process so that assistance is denied before benefits are issued. This is in contrast to traditional fraud investigation programs where active assistance cases are investigated upon suspicion or a complaint of existing fraud.

The Early Detection and Prevention (EDP) movement began in 1981/82, when Orange County, California, District Attorney's Office piloted an EDP initiative for both Food Stamps and the Aid to Families with Dependent Children programs. During that year, 12 Fraud Investigators completed 1596 investigations, finding fraud in 802 or nearly 50% of the cases referred. With the very positive cost avoidance shown during the pilot study, Early Fraud Detection/Prevention was implemented County-wide on June 1, 1982.

Within 18 months, the program had achieved an $11.6 million tax cost avoidance for Federal, State and County governments.

Since its inception in 1981, the Orange County program has:

successfully generated immediate (before benefit issue) investigative referrals,

provided for timely completion of intake investigations, and

encouraged and facilitated prompt feedback from investigative staff to Intake Workers.

A federally-funded study of the program, in 1984, found it to be highly cost effective, with savings of $33 for every dollar spent. Referrals for investigation from intake have increased from 100 per year prior to the EDP program implementation to 917 per month in 1992/93.

For fiscal year 1992/93, Orange County's EDP program recorded these results:

1) AFDC/FOOD STAMP PROGRAMS

Intake Cases % Cases

Referrals Denied Denied Cost-Avoidance

10,192 6,854 67% AFDC/Food Stamp

Combination Cases $70,805,760

Food Stamp Cases $122,454

Total $70,928,214

2) GENERAL RELIEF PROGRAM (100% County Funded)

Intake Cases % Cases

Referrals Denied Denied Cost -Avoidance

5,987 5,398 90% $2,074,616

California counties now receive 100% Federal and State funding for the administrative costs of Early Fraud Detection/Prevention Programs, if State approved. The EDP Program is supplemented with a strong criminal Fraud Investigation Unit to prosecute proven cases of fraud, with the consequential deterrent effect of criminal prosecution. An integral part of the overall fraud detection, prevention, investigation and criminal prosecution program is an efficient mechanism for civil collection of any overpayments established which are deemed not appropriate for criminal prosecution, and a high profile information campaign informing the community of the aggressive Fraud Detection/Prevention Program.

In 1988, the federal government passed legislation under the Family Support Act, requiring States to establish pre-eligibility fraud detection measures in their AFDC programs, and recommending the Orange County program as a model. As the Federal Government funds 100% of the cost of the Food Stamp Program and about 50% of the AFDC Program, they have a critical interest in proven cost avoidance strategies such as Early Fraud Detection and Prevention Programs.

These initiatives have become widely recognized and supported throughout the American States as a critical component of any cost control program. In fact, the Minnesota Fraud Prevention Investigation Program was honoured as a "best methods" example of reinventing government in 1993 by the U.S. Department of Health and Human Services.
 
 

A recent survey of American States by the United Council on Welfare Fraud found that 22 of the 30 States surveyed had implemented a front-end prevention process. Those States responding to a request for information for purposes of this study confirmed the effectiveness and importance of the EDP Programs, especially in light of possible reduced federal government cost-shared funding for assistance programs.
 
 

OVERALL DESCRIPTION OF PROCESS

The fraud prevention programs implemented in the various American states are known variously as FRED (Fraud Early Detection), EDP (Early Detection and Prevention), Fraud Prevention Investigation, Fraud Detection Investigations, FEV (Front-End Verification), and Front-End Investigation. But, whatever their titles, most were patterned after the Orange County experience and the goal of the programs is the same: the selective identification at the intake stage of suspicious cases for investigation, thus permitting the identification of misinformation that otherwise could not be detected, even by front-end data matching programs, and preventing benefit payments being made in error.

Most American cases referred for investigation from the point of application are characterized by unsubstantiated deprivation (parent not in the home), household composition and residence concerns. These concerns are not easily dealt with by Eligibility Workers, as they require investigative field work or contact with sources not available to Eligibility Workers. Routine quality control audits, which are principally a paper review process, are also not likely to determine that there is a problem with these cases, once benefits have commenced. Only public or other types of complaints might bring such fraudulent activities to light. In Canada, because of the lack of data matching systems available to American Intake/Eligibility Workers, undeclared income is likely to be a much larger part of the investigative program.

Even when violations are reported on active cases and post-investigation is successful in halting or reducing benefits, the cost of investigation and possible criminal prosecution is much higher than the cost of front-end investigation and prevention of benefits being inappropriately paid. And, the record of social service agencies in recovering overpaid funds is dismal. EDP Programs, by emphasizing prevention, place a very broad interpretation on what is included as "fraud", including unintentional misrepresentations of facts by clients (client errors). For purposes of front-end prevention, some inconsistency in the application for benefits, or a previous record of program abuse, causes the Eligibility Worker to make a referral. If the subsequent investigation finds misrepresentation, the case is normally rejected as ineligible. In a small minority of cases, prosecution may be sought. However, the key to success for the EDP Programs is the cost-avoidance - any cost benefit is fully realized. Savings identified from post-fraud investigation are more speculative as only a small portion of client overpayments are ever recovered.

Major components of a successful front-end investigation process are illustrated on the next page, and include the intake interview, the referral process to an investigator, the actual investigation, and the feedback process to the worker.

The Intake Interview

During the normal intake interview, the Financial Assistance Worker is responsible for doing normal eligibility checks to ensure that the applicant is eligible for benefits. However, if an early fraud detection and prevention program is in place, the Financial Assistance Worker is also on the alert to determine whether the applicant displays any of the criteria that would justify a referral for investigation. Some criteria, such as forged or altered identification documents, can be fairly obvious without additional probing on the part of the worker. Others, such as expenses consistently exceeding declared income in the previous months, may require careful interviewing on the part of the worker. Generally, intensive training of the Financial Assistance Workers in recognizing indicators of applicants who are at high-risk of committing fraudulent behaviour is required. Although some states allow workers to refer a case for pre-eligibility investigation for any reason that "doesn't feel right" to the worker, most states have a specific list of criteria. The Financial Assistance Worker must demonstrate that one or more of the criteria exist before referral can occur.

Thus, in those states which have implemented an early fraud detection and prevention initiative, the Intake Worker has gained an additional option. Without this initiative, the worker can make only two decisions: based on the information presented during the intake interview and the Intake Worker's limited ability to verify or validate that information, the applicant is either eligible to receive benefits, or the applicant is not eligible. With pre-eligibility investigation in place, the worker has a third option, that where eligibility is not clear, or where the applicant seems to be eligible but there is some question of the accuracy or reliability of the information presented, to refer the case for further investigation to assist in the determination of eligibility.

The Referral Process

If an applicant demonstrates one or more of the referral criteria, and the Intake Worker is unable to resolve the issue, a referral is made to a Pre-Eligibility Investigator. Although a few states allow Intake Workers to make a direct referral to an Eligibility

Investigator, the majority of states involve the Intake Worker's supervisor in the referral process. The client is informed that they may be contacted for additional information and the interview is terminated. Most states try to ensure a rapid turn-around of information and attempt to ensure that benefits are issued promptly to the client. However, in most cases benefits are not issued until the investigator's report is returned to the worker, if it is within established timeframes regarding the state's "standard of promptness".

The Pre-Eligibility Investigation

Many states separate the investigative process into two components: a "quick-and-dirty" review to see whether there are grounds for questioning the accuracy and reliability of applicant information, and a more in-depth investigation, which may involve a home visit with the applicant(s), request for and review of substantiating information, as well as contact with landlords, bank managers, employers, neighbours, etc. At the preliminary investigation stage, the investigator has available linkages to existing data bases through the many and varied data linkages which are available to most American states. It is in the later stages of investigation that the more traditional investigative techniques come into play.

The Feedback Loop

Most states set a "standard of promptness" requiring feedback on information accuracy from the Pre-Eligibility Investigator to the Intake Worker (or, more probably, the Intake Worker's supervisor) within 5 days. This seems to be the best balance between ensuring that the investigator has sufficient time to complete the investigation while the applicant does not suffer unduly from hardship while awaiting benefit issue. However, staffing restrictions have meant that some states have been forced to accept a longer time period between referral and feedback. Since few states can use front-end investigation as a rationale for delaying benefits, this means that the client is issued interim benefits (reducing the effectiveness of the program), which may not be recoverable if a finding of inaccuracy is returned.

It is important to differentiate between the role of the Intake Worker and the role of the Pre-Eligibility Investigator. The investigator's role is NOT to determine eligibility -- that is the role of the Intake Worker. The Pre-Eligibility Investigator is responsible for confirming or denying the accuracy of the information provided by the applicant in applying for benefits. Then, based on that determination of accuracy, the Intake Worker determines eligibility and the level of benefits.
 
 

If it is determined that the client willingly or accidently misrepresented information, the client is contacted and the situation is presented to them. This results in either a decision of ineligibility on the part of the Intake Worker, or a decision of eligibility with either reduced or full benefits. For those applicants who are determined to have intentionally defrauded the program, states are required to use an administrative disqualification process for the Food Stamps program, and many states also use this process in the Aids to Families with Dependent Children (AFDC) Program. In other words, not only does the applicant have their benefits denied, but they are disqualified from receiving benefits for a period of time depending on the number of offenses. If the client is determined to be ineligible for benefits after investigation, the start of this disqualification period is delayed until the individual becomes eligible for benefits again.
 
 

CHARACTERISTICS OF SUCCESSFUL EDP PROGRAMS

Based on the survey of American initiatives, the most effective EDP programs display these common characteristics:

1) A Social Services agency philosophy that welfare fraud and its attendant costs to government programs must be reduced, and that the most cost-beneficial method of so doing is through an aggressive front-end detection and prevention program, backed by an effective fraud prosecution capability for serious fraud found on ongoing cases. This philosophy must be communicated throughout the Social Services organization widely and frequently.

2) A reporting structure for EDP Program Investigators staff independent of local Income Security program supervisors. These programs are usually managed from a separate departmental agency, i.e., Office of Inspector General. In the case of Orange County, California, the Early Detection Program is administered by the District Attorney's Office, under a Plan of Co-operation signed with the Social Services Agency.

3) Close relationships between EDP investigators and intake or Eligibility Workers. To this end, the investigators are most often physically located with intake staff to foster a common bond, co-operation and communication between the staff who perform these two critical functions. This arrangement is considered key to the success of EDP Programs by facilitating;
 
 
 
 

effective, frequent face to face communication between the Investigator and Intake Worker,

timely referrals of suspected abuse,

quick turnaround time and feedback to the Intake Worker, and

ready access to Intake Worker files for the investigator.
 
 

4) A clear expectation that Intake Workers will fulfil their responsibility, as established in procedures and policy, to conduct eligibility checks to the best of their ability before referring cases for investigation.

5) A smooth and efficient referral and feedback process, with a minimum of red tape, along with reasonable workloads for Investigators. This allows quick turnaround of investigation requests, preventing a situation where benefits must be issued in questionable cases because of investigation delays. Minnesota permits direct referrals and investigation results feedback between Intake Worker and the Investigator, with the supervisor reviewing the results after completion of the case, to reduce turnaround time.

6) Intensive and on going training in all aspects of the program for both Investigators and Intake Workers. The Intake Workers should be trained particularly on the mechanics of the program, i.e., applying referral criteria to ensure quality referrals. The importance of training cannot be overstated.

7) An effective awareness campaign at program implementation. As well as ensuring the taxpayers of a high level of stewardship in the delivery of public assistance programs, a public campaign will alert clients and potential clients about the program, thus eliminating any "element of surprise" during the intake process.
 
 
 
 
 
 
 
 
 
 
 
 

AMERICAN EXPERIENCE WITH PROGRAM

As mentioned earlier, for the purposes of this study, each American state was contacted to determine whether an early fraud detection and prevention program is in place for any of their Income Security programs, and, if so, they were requested to provide a program operation description, criteria for referrals, supporting policies and procedures, staff training materials or processes, and evaluation results if available. As well, some states provided information on other fraud control initiatives they have in place. The State Directory accompanying this report summarizes the information provided by each of the states which sent material on EDP programs. The following states have implemented, in all or most counties, an early prevention and detection program. Not all of these states provided materials and not all of the states which did provide materials on their fraud programs have implemented EDP programs. However, the following states do have early detection and prevention programs in place.

Alaska

Arkansas

Arizona

California

Connecticut

Florida

Georgia

Iowa

Kansas

Kentucky

Massachusetts

Minnesota

Mississippi

Nevada

New Jersey

New York

Ohio

Pennsylvania

South Dakota

Tennessee

Texas

Utah

Vermont

Virginia

Washington

West Virginia

Wisconsin

In addition, 4 states (Colorado, Delaware, Illinois, and New Hampshire) are currently at the stage of piloting the early fraud detection and prevention program. Maine is planning to pilot but has not yet begun the pilot. The remainder of the states place their dependence on the IEVS system and a comprehensive post-fraud program. An interesting case is represented by Missouri, which piloted an early detection and prevention program in several locations but discontinued the program due to the poor quality of referrals from the Intake/Eligibility Workers. The state could not obtain funding for the intensive training that would have been required to improve the referral process, which is key to the success of any EDP program.

Two other programs used throughout the United States (which, together with early detection and prevention initiatives form a comprehensive fraud prevention system) are Data Matching and Administrative Disqualification.

Data Matching

Data-Matching of client records is much more widespread in the United States than in Canada, and in fact, some data-matching has long been mandated by the Federal Government. The most comprehensive data-matching is done through the federally mandated Income and Eligibility Verification System (IEVS), established by Congress under the 1984 Deficit Reduction Act to reduce errors in determining eligibility and benefit levels in the Food Stamp, Aid to Families with Dependent Children (AFDC) and Medicaid programs. Program regulations require state welfare agencies to compare income reported by program applicants and recipients with income from several data sources:

1) Internal Revenue Service (IRS) data on interest, dividends, and other types of unearned income;

2) Social Security Administration data for Retirement, Survivors, and Disability Insurance benefits, Supplemental Security Insurance benefits, and annual earnings; and

3) State quarterly wage reports and unemployment insurance benefits.

States are required to follow up within 45 days on at least 80 percent of all IEVS information received on applicants and information targeted for review on applicants. Since the Federal legislation mandating computer matches was passed in 1984, individual states have made various degrees of progress towards implementation of the full range of data-matching. As well, some States are conducting matches in addition to those mandated and some States are requesting access to additional Federal data banks such as Veterans Administration benefits, Federal employee salaries and pensions, U.S. Savings Bond holdings, and State death records maintained by the Social Security Administration.

There are two aspects to IEVS - the applicant system and the recipient system. For the applicant system, client identifying information, i.e. name, Social Security Number, sex, Date of Birth, etc. is submitted via computer by Eligibility or Intake Workers once an application for benefits is completed. Information from the data sources listed above is usually available within 3 - 5 working days for those states with this system, and can be used to confirm benefits or deny eligibility as appropriate. However, counties cannot unduly delay initial assistance while awaiting match results. As well, prior to submitting client identifying information for data matching, counties must inform clients in writing that the match will be conducted.

The recipient portion of IEVS provides information from the matching of active clients with a variety of other systems at various times of the year, depending on the match type. Some information sources are available through on-line access systems. Various income threshold levels are established to ensure that only information that could impact eligibility or benefit levels is sent to the workers for follow-up.

The applicant system (up-front matching) has been shown to improve the morale of Eligibility Workers by helping them establish client eligibility with more assurance. It also allows Eligibility Workers to more easily meet case disposition timelines and documentation requirements. The computerized data bases can be a way to save applicants' time by relieving the clients of the need to locate the required documentation themselves.

The IEVS process supplements the early prevention initiatives, but it is the Eligibility Workers who receive the data-matching information, or who have on-line access to the data banks, and evaluate the match results, rather than Pre-Eligibility Investigators. For instance, if discrepancies in client-provided information are found through up-front data matching, the Eligibility Worker considers the information in determining client eligibility rather than referring those cases for investigation. Unreported income at the intake level becomes a relatively small problem for those states with comprehensive on-line or data-matching systems.

The cost savings reported for the American States' Early Fraud Detection and Prevention Programs are in addition to savings related to data-matching initiatives.

Administrative Disqualification

There has been an Administrative Disqualification process in the Food Stamp Program, which is administered by the Food and Nutrition Administration, Department of Agriculture, since 1977. Under the Food Stamp Act, all States are responsible for investigating cases of alleged intentional program violation and ensuring that cases are acted upon either through Administrative Disqualification Hearings or through the criminal court system.

Individuals found to have committed Intentional Program Violations are notified in writing of the specifics of the overpayment, and for cases where criminal prosecution is not to be undertaken, the client can choose to request an Administrative Disqualification Hearing (ADH) or to sign a waiver of right to a hearing. By signing a waiver, the client signals agreement with the finding of Intentional Program Violation and any resulting overpayments.

Hearings are held at the state level, in some cases by telephone conference, and afford clients the opportunity to explain why they believe they did not commit a violation. Some States find that providing an "evidence packet" which outlines the evidence supporting the violation in advance of the ADH is quite successful in convincing clients to sign the hearing waiver at that time.

On the basis of a signed hearing waiver, an Administrative Disqualification Hearing which confirms the violation, or a successfully prosecuted criminal case, the overpayment is established and the client becomes ineligible for further benefits as follows:

For a period of six months for a first violation

For a period of one year for a second violation

Permanently for a third violation

If the disqualified individual is not eligible for program benefits at the time the disqualification period is to begin, the period is postponed until the client applies again and is determined to be eligible for benefits. The disqualified individual's household is responsible for repayment of the overissuance which results in the disqualification, regardless of the household's eligibility for program benefits. States are required to report, to the federal Food and Nutrition Administration, information on each individual disqualified for Food Stamp eligibility.

In 1991, federal legislation was passed allowing States to extend the Administrative Disqualification process to the Aids to Families with Dependent Children program. A national registry is maintained for access by state welfare agencies.

In most EDP Programs, Pre-Eligibility Investigators can refer cases of confirmed Intentional Program Violation to the Administrative Disqualification process. For applicants who have been investigated, but whose benefits were not denied or reduced, normally records are not retained. Thus, the investigation does not go "on their record". The client will not, therefore, be referred due to prior history. However, in Texas, for example, Investigators are also able to refer cases where there has been a second or third early fraud detection and prevention investigation on the same applicant, even if all the cases have been proved unfounded.

Administrative Disqualification does result in some cost avoidance attributable to the benefits not paid for the period of disqualification, however, more important is the deterrent effect - clients who know they may be disqualified for future benefits are considered to be less likely to break the rules when applying for or receiving benefits.

Although these initiatives are very important and effective processes for combatting Income Security program abuse, Early Fraud Detection and Prevention is a stand-alone process which is even more effective when operating in conjunction with these other initiatives, but which has proven to be very cost-beneficial on its own.
 
 

MAJOR BENEFITS ACHIEVED FROM PROGRAM

The Office of Inspector General, Federal Department of Health and Human Services, surveyed 39 local Social Services offices across the nation having active front-end fraud detection and prevention programs. In a November, 1994, report, the Office of Inspector General (OIG), stated that the EDP Programs in the 39 offices had resulted in savings of over $41 million in program funds in 1992. The cost and savings information provided by the 39 selected offices showed a weighted average savings of about $8.43 for every $1.00 spent operating the program. It was noted that savings resulting from pre-eligibility fraud investigation are in addition to savings that might otherwise result from Eligibility Worker decisions, and investigative and prosecution activities of regular post-fraud investigation units. As well, preventing improper payments before they occur prevents the need for costly prosecution processes, including documentation, overpayment calculation, and recovery activity resulting in additional cost-avoidance.

Cost-Benefit of Early Detection Programs

The chart on the following page, taken from the OIG Report, shows the breakdown in savings per dollar of EDP Program operating costs of the 39 offices which reported savings. The most commonly found discrepancies were reported to be:

Applicants underreport their income so expenses exceed income, ensuring eligibility for benefits

Applicants incorrectly report household composition. They may show that a parent is absent when in fact the parent is residing in the household with their children. Or they may report children in the home who are not actually residing there.

Applicants may report incorrect addresses, or

Applicants may underreport their property or assets.

Other Benefits of EDP Programs

Front-end prevention efforts yield many benefits in addition to the very attractive cost-benefits shown by the programs:

Improves and facilitates increased teamwork between Eligibility or Intake Workers and the Investigators, in contrast to the sometimes adversarial relationship experienced between workers and the traditional post-fraud Investigators

Results in higher morale of Eligibility Workers who feel some sense of control over the process and who feel they have an important role to play in preventing client fraud and error

By preventing fraud and the resulting overpayments, EDP Programs yield true cost-avoidance and eliminate the need for (often ineffective) overpayment recovery and collection procedures

By preventing fraud, effective EDP Programs reduce the need for costly criminal prosecution efforts and the attendant criminal record thereafter attached to the prosecuted client. Clients with criminal records frequently find it difficult to secure employment and in some cases, may lose a job they had held prior to the prosecution. This may result in increased dependency on the Income Security system.

A change from the traditional reactive (post-fraud investigation) approach to a preventative approach can not only uncover a high proportion of attempted fraud and unintentional error during the application process, but can also identify problems in program design and implementation which facilitates fraud. This can be done through analysis and assessment of those cases and circumstances found to be at highest risk for abuse, and

A public awareness campaign accompanying implementation of EDP initiatives may increase the "deterrent effect", leading to voluntary compliance with eligibility requirements by increasing the perception in the community that it is difficult to abuse the system.

The long-term effect of an EDP program is that the Social Services agency will be recognized as more professional and accountable by both the taxpayer it represents and the clientele it serves.

The Office of Inspector General, on the basis of the 1994 study made a recommendation that the Federal Administration for Children and Families should actively promote the establishment of pre-eligibility fraud investigative units in those States without such units.
 
 

MAJOR DIFFICULTIES EXPERIENCED

Most states noted that inadequate staff training would almost assuredly result in failure of the EDP program, as evidenced by Missouri's experience in cancelling their EDP program. This stresses that the importance of staff training, for both Intake/Eligibility and Investigative staff, is critical for the following reasons:

To combat a lack of support for the program by workers who perform the intake role, and who likely have been trained to be empathetic and considerate of client's stated needs, without the balancing concern for ensuring eligibility

Training in understanding and interpreting referral criteria can help ensure good quality and quantity of referrals for investigation

training which stresses the very important role Intake Workers must play in increasing program integrity and preserving the Income Security programs for the truly needy can greatly improve their commitment to reducing program abuse

Investigators should also have relevant training programs available to allow for upgrading of investigative skills as well as skills relating to their co-operative and supportive role in the teamwork environment

Some States reported insufficient staffing as a major problem for the success of their EDP Program. With insufficient investigative capability, it becomes impossible to complete the cases within established timeframes (usually five working days) and may result in benefits being issued prior to the completion of the investigation report, leading to overpayments and defeating, to some extent, the purpose of pre-eligibility investigation programs. California's 1994 Strategic Plan (see overview in the State Directory) recommends one pre-Eligibility Investigator for every 300 applicants.

The timely use and follow-through with investigation results by Income Security Intake Workers or other staff was noted as a problem area by Florida, as well as the high rate of referrals in lower socio-economic areas and the attendant intensive field work.

Minnesota noted these concerns with their very successful Fraud Prevention Investigation Program:

Establishing appropriate investigative caseload levels;

Inadequate staffing in large areas with concentrated public assistance caseloads;

Whether FPI will be successful in counties with smaller caseloads; and

The impact of Electronic Benefits Transfer on the investigative process.
 
 

RELEVANCE FOR BRITISH COLUMBIA

Early detection and prevention of fraud and error has been identified as a primary focus for the Prevention, Compliance and Investigation Branch of British Columbia Social Services.

Given the Province's ever-increasing income support caseload with the attendant stresses on caseworkers, investigative staff, and all other income support staff, any initiative which provides some tools to help manage and control this caseload is worthy of consideration. A well designed and implemented front-end fraud detection and prevention program would meet a number of the goals of the PCE Division, as well as those of the Department as a whole:

1. It would recognize that Eligibility Workers are the cornerstone of any effort to prevent fraudulent applications for benefits. Due to very heavy workloads, the expectation that eligibility decisions must be made and cheques issued within short time lines, the lack of training in recognizing and verifying potentially fraudulent situations, and the lack of back-up support to assist the workers when eligibility concerns are beyond their ability to resolve, may leave the Eligibility Workers feeling that there is little they can do to combat potentially fraudulent situations. A front-end prevention program puts the Eligibility Investigator in partnership with Income Support staff, particularly the Financial Assistance Worker, and by providing support, advice and quick turnaround on requests for investigations, can give the beleaguered worker some sense of control over questionable situations on which benefits might be otherwise be issued without further examination;

2. It would, by detecting and preventing cases with questionable circumstances being approved at intake, reduce the number of cases on which overpayments must be later established and recovery procedures undertaken, reduce the number of cases which the post-fraud Investigators must handle once the abuse comes to light later on, and demonstrate that the true measure of effectiveness of a fraud control program is the absence of fraud;

3. It would improve and maintain public confidence in the Income Security Program through a well publicized awareness campaign, which stresses the benefits of an early detection and prevention program to both clients and taxpayers, while ensuring a high degree of respect for the rights and freedoms of the clientele; and

4) It would demonstrate British Columbia's commitment to provide a high quality service to clients in the most cost-effective manner.

Recommendation 1: Based on the experience of the majority of American States, it is recommended that British Columbia develop and implement an early fraud detection and prevention system.
 
 

THE EARLY FRAUD PREVENTION PROGRAM

An early fraud detection and prevention program could be implemented in British Columbia relatively easily since many of the underlying requisites are already in place, including the appropriate mandates (particularly in the PCE program), training and background for various types of investigative staff, and appropriate reporting structures.

The most major change to current processes would be in the intake role. Establishment of the Eligibility Investigator function provides another option for the Financial Assistance Worker to use in determining eligibility. It is expected that these two roles would complement each other. The establishment of the role of Eligibility Investigator does not replace, and in fact, emphasizes, the role of the Financial Assistance Worker in verifying and documenting all eligibility factors that are, by policy, their responsibility. If, after completing this determination to the best of their ability, there are still questionable factors identified in the referral criteria, then the Eligibility Investigator takes over to verify information which is difficult or impossible for an Financial Assistance Worker to confirm. States report that the most common techniques used by the Eligibility Investigators include:

1. Personal visits to the applicant's home.

2. Matching financial and personal information provided by applicants with various computer databases.

3. Collateral interviews with relatives, friends, neighbours, landlords, employers and others.

4. Assisting Financial Assistance Workers in the interviewing process.

5. Surveillance of an applicant's home.
 
 
 
 

All of these techniques are difficult, if not impossible, for an Financial Assistance Worker to undertake personally.

REFERRAL CRITERIA

Each state has developed a set of pre-defined criteria to identify which applicants are at high risk of committing fraud and program abuse. It is critical that these criteria are clearly defined to ensure that client rights are protected and that any referral for pre-eligibility investigation respects and recognises these rights.

The criteria that are defined for referral purposes are very similar across the various states, although some sets are more complete than others. The following major topic areas are included in almost all of the referral criteria lists (these have been adapted somewhat to put them into the Canadian context):

VI. Identification -

VII. Documents appear altered, invalid, or forged

VIII. Applicant has no documentation

IX. Applicant is unable to obtain replacement documentation

X. Applicant cannot demonstrate that children are in school

XI. Residency -

XII. Address does not exist

XIII. Address may be a mail drop

XIV. No telephone reported

XV. Mailing address different from street address

XVI. Highly mobile - rarely stays in one location more than three months

XVII. Household Composition -

XVIII. Applicant does not have appropriate documentation identifying children

XIX. Prior case history indicates different household members

XX. Case previously terminated - whereabouts unknown

XXI. Case previously terminated - failure to provide essential information

XXII. Applicant reports children temporarily absent from the home

XXIII. School age children not attending school in the district where family lives

XXIV. No legal separation, divorce, nor court order for child support exists

XXV. Changes in household circumstances which appear to have occurred or to have been unreported solely in order to allow the household to gain eligibility

XXVI. Resources/Assets -

XXVII. No bank accounts reported

XXVIII. Recent unexplained deposits made to bank accounts

XXIX. Income -

XXX. Households where paid expenses are consistently higher than earned income

XXXI. Evasive or unclear explanations about how expenses are being met

XXXII. Reports household being supported by family/friends

XXXIII. Periods where no income is reported but bills are up to date/expenses continue to be paid

XXXIV. Employment verification documentation appears to be invalid, altered, or forged

XXXV. Household where a wage earner has lost job and UI is not reported

XXXVI. Children in home of Relative -

XXXVII. Documentation in respect of children for whom assistance is being requested appears to be invalid, altered or forged

XXXVIII. Frequent change of caretaker for children

XXXIX. Deprivation (situations in which the applicant provides contradictory information relative to the whereabouts of the absent parent)

XL. Prior History -

XLI. Has been convicted of fraud against the department

XLII. Previous record of frequent overpayments on file

XLIII. Previous history of failure to report changes

These criteria have been selected, from and combined as an example of a criteria list from three of the American states which provided details of referral criteria: Texas, Minnesota and Ohio.

Recommendation 2: It is recommended that British Columbia start their development of a set of referral criteria by adapting a combination of the criteria developed by Texas, Minnesota and Ohio ("best state" model).

High Risk Profiling is the statistical analysis of available data used to identify clients with a higher than average risk of incorrect payments. This methodology maximizes the chance of detecting fraud and incorrect payments while minimizing intrusion on the privacy of clients. The process of high-risk profiling is nearly as important as the criteria list itself.

Starting from a base of criteria from other jurisdictions; a process involving front-line staff, investigators, and supervisors; and preparation of a high- risk profile using data relevant to British Columbia should be undertaken. British Columbia data which would be relevant to the development of high- risk profiling would include:

I. Analysis of "presenting" factors from actual fraud case files.

II. Statistical analysis of fraud data with respect to major reasons for fraud investigation and criminal prosecution.

III. Review of file audit information completed by eligibility officers.

IV. Results of any random case reviews.

Then, with the involvement of a variety of staff, and based on the referral criteria derived from the "best-states" model, this British Columbia specific information could be used to refine a "high-risk" profile for applicants who should be referred for further pre-eligibility investigation. This development of referral criteria provides an ideal topic around which to construct joint workshops involving both front-line and investigative staff. These workshops would be designed to achieve several objectives:

1. Finalize the development of an agreed-to set of referral criteria

2. "Sensitize" front-line staff to the existence and seriousness of the "fraud problem"

3. Emphasize team-building between the front-line workers and the Pre-Eligibility Investigators

4. Develop a cooperative approach to solving the "fraud problem"

5. Develop a mutual appreciation of the respective roles of the Intake staff and the Eligibility Investigator staff.

This process of consultation could be built into either a fraud awareness seminar, a staff training process or into the implementation plan for establishing a British Columbia Early Detection and Prevention program.
 
 
 
 

Recommendation 3: It is recommended that British Columbia use a staff consultation model in finalizing the referral criteria for the British Columbia Early Detection and Prevention Program.

RESULTS OF A FINDING OF ATTEMPTED FRAUD

In the United States, the early fraud detection and prevention programs are invariably linked with an administrative disqualification process, described on page 15. In Canada, since the Canada Assistance Plan (CAP) defines need as the only determinant of receiving benefits, administrative disqualification has not been an option. For the majority of clients who were found ineligible after investigation, they would have additional income or assets, and thus, would remain ineligible regardless of the existence of an administrative disqualification process. But, even when an applicant was clearly demonstrated through investigation as having attempted to abuse the Income Security program, they could still reapply when they became eligible without any penalty.

However, with the effective demise of CAP as part of the last federal budget, this is no longer a barrier. The federal government has requested assistance from the provinces in developing a set of national standards to replace the Canada Assistance Plan. The timing could not be better to develop an administrative disqualification program in Canada.

Recommendation 4: It is recommended that, with the demise of the Canada Assistance Plan, British Columbia consider the implementation of an administrative disqualification process for clients who are found to have willfully misrepresented their circumstances in order to obtain program benefits.

It is critical that Financial Assistance Workers respond quickly to information being returned by the Eligibility Investigator in order not to unnecessarily delay the issuance of benefits for those clients for whom the information submitted at intake was found to be valid, accurate and reliable. Similarly, clients who have been found to have submitted inaccurate or fraudulent information, and who are clearly ineligible, should be dealt with quickly in order to ensure inappropriate benefits are not issued, or where temporary benefits have been issued, to close the case. If administrative disqualification becomes an option, a process comparable to the U.S. disqualification hearing would need to be established in order to protect client rights. Without an administrative disqualification system in place, applicants continue to have access to the system of appeal panels currently operating throughout the province.

Recommendation 5: It is recommended that the role of the present Appeal Panels be expanded to include the hearing of appeals of applicants whose benefits were reduced or who were found ineligible as a result of an eligibility investigation.
 
 

STAFF TRAINING

Most U.S. states emphasize training for all staff involved in EDP as a major factor responsible for program success. In the words of one American program administrator, "Without training, workers either want to refer everything, or they won't refer anything."

For front-line Intake/Eligibility Workers, the training programs developed by several states tend to focus on:

The complementary roles of the Intake Worker and the Pre-Eligibility Investigator

Use and understanding of the referral criteria, with the use of many examples

Interviewing techniques, including the interpretation of body language

Recognition of forged and altered documents

Program policy and processes, including the use of forms

Dealing with difficult clients

The training provided to the Pre-Eligibility Investigator tends to focus on:

The complementary roles of the Intake Worker and the Pre-Eligibility Investigator

Appropriate investigative techniques

Client rights

Sources of data, including training in the use of any available on-line systems

Program policy and processes, including the use of forms

It is important that some of this training occur in joint training sessions in order to develop a team orientation, a common understanding of the problems, and to assist in the development of cooperative and productive working relationships between the two sets of staff. Supervisors also need to be integrally involved in this training process, since they are often the obvious personal link between the Intake/Eligibility Workers and the Pre-Eligibility Investigators and are the first line of appeal for disgruntled clients.
 
 

Recommendation 6: It is recommended that the 28-day On-The-Job Training Program be expanded to include at least one day on the Early Fraud Detection and Prevention initiative, focusing on the use of the referral criteria, interviewing techniques to ascertain whether an applicant meets those criteria, and dealing with clients who are ineligible as a result of an eligibility investigation.

Recommendation 7: It is recommended that all current Financial Assistance Workers, supervisors and investigative staff receive full training on the early fraud detection and prevention program during the implementation phase.
 
 

HIRING ELIGIBILITY INVESTIGATORS

The American states contacted took two different approaches to the establishment of the Eligibility Investigator function. The majority of states hired Eligibility Investigators from a police service background, providing additional training on program policies and procedures. These states felt that the success of their program was partially due to the strong investigative background of the staff involved. A few states, on the other hand, hired staff from a program delivery/intake background and provided training in investigation techniques. These states felt just as strongly that their program success was due to the involvement of well-trained staff who had a great deal in common with program delivery staff, thus, strengthening the critical relationship between the Eligibility Investigators and the Financial Assistance Workers.

British Columbia has recently established the role of Assistant Investigator. These positions are drawn from the ranks of the front-line workers. This role would seem to be a good fit with the requirements of the Pre-Eligibility Investigator position, therefore retaining those staff with detailed criminal investigation expertise, the Ministry Investigators, for more serious cases where prosecution is a real option. Therefore:

Recommendation 8: It is recommended that the role of the Assistant Investigator be expanded to include the conducting of eligibility investigations based on a referral from an Financial Assistance Worker.

Different states have taken a variety of approaches to staffing the Pre-Eligibility Investigator positions. The three main approaches are:

Hiring full-time, specialized staff for the positions

Rotating existing investigative staff through the positions

Distributing the responsibility for completing the pre-eligibility investigation function among all investigative staff positions

Texas has evaluated the impact of each of these approaches, and has established that the last option, with all investigative staff spending a portion of their time on pre-eligibility investigation, is the most effective. Not only does it prevent "burn-out", but it allows for the establishment of long-term working relationships between specific front-line staff (both Financial Assistance Workers and ongoing financial staff) and the complement of investigators. Thus:

Recommendation 9: It is recommended that all Assistant Investigators spend some portion of their time on eligibility investigations, rather than hiring specialized staff, in order to maximize the working relationship between the Assistant Investigators and the Financial Assistance Workers.

Determining the appropriate level of staffing is a function of many factors, including the number/proportion of cases which are referred from intake. In turn, this is a function, not only of the potential level of fraud being attempted by the new applicants, but also of the referral criteria selected and the skills of the Financial Assistance/Eligibility Workers in identifying individuals who meet those criteria. California is the only state which has attempted to develop a caseload standard for Pre-Eligibility Investigators, selecting one investigator for every 300 new applicants as their standard (see California Strategic Plan in the State Directory report). The most critical factor is that the timelines which have been established for feedback to the Financial Assistance Worker must be met. This may take some experimentation in any given site.

Most states who provided information on this issue informally have indicated that they aim for approximately a 50% "turn-down" rate. That is, of the cases referred, approximately half are determined to be fraudulent. A lower "turn-down" rate is not seen to be a cost-beneficial use of the Pre-Eligibility Workers' time and indicates that the Intake Workers are applying the referral criteria too loosely. A very high "turn-down" may indicate that Intake Workers are not referring enough cases and some clients may be obtaining benefits inappropriately. This could indicate that the referral criteria are too rigid or limited, or that the Intake/Eligibility Workers are not supportive of the program.
 
 

REPORTING STRUCTURES

For most States that provided information on EDP Programs, it was noted that they prefer and recommend that pre-eligibility fraud investigation units be organizationally independent of the income security benefits payment sections of the Social Services Departments. In some cases, Investigative staff report independently of the entire Social Services organization, i.e., to the District Attorney's Office, under a "Plan of Cooperation" between the two departments as is the case in California. The pre-eligibility investigation programs most generally have the same reporting structure as the existing post-fraud investigation units, and are usually part of the same "Welfare Fraud" office.

In the Office of Inspector General's 1994 Report on AFDC Pre-Eligibility Investigative Units, it was found that 62% of the pre-eligibility fraud investigative units surveyed reported to supervisors that were independent of a benefit payment section (which generally determines eligibility for AFDC, Food Stamps and Medicaid assistance). The remaining 38% of pre-eligibility units reported to supervisors that were organizationally a part of a benefit payments section.

The argument for organizational independence focuses on the difference in missions of benefit payment staff (Eligibility Workers) and Investigators, wherein Eligibility Workers are concerned with processing applications and payments quickly, their goal being to get assistance into the hands of needy people. Because of this orientation, Eligibility Workers may be inclined to deem applicants as eligible even when there is a doubt about eligibility. Conversely, the goal of Investigators is in preventing and detecting fraud, and in providing benefits only to those who can prove they qualify. This difference in missions may create a conflict of interest and unnecessary tension between the two types of workers. Some Eligibility Workers may object to certain investigative techniques, and some Investigators reported to the OIG that they had been asked to "back off" on cases where the supervisor did not agree with the information-gathering methods, such as contact with an applicant's neighbours. There seems to be wide-spread agreement that the process works best and is most effective if the investigative units report independent of the benefit payment sections.

There also seems to be agreement on the benefits of physically locating Pre-Eligibility Investigators near benefit payment Eligibility Workers. Advantages of close physical proximity are cited as:

Facilitating frequent and effective communication between Investigators and Eligibility Workers, including advice to workers on questionable cases,

Assuring timely referrals of cases from Eligibility Workers to Investigator,

Facilitating timely feedback on cases, and

Accessibility to Eligibility Worker's case files.

The most important tool for ensuring an effective program and for enhancing and maintaining morale of all staff is a high degree of communication and interaction between program staff at all levels.

The reporting structure presently in place for BC Social Service's Prevention, Compliance and Enforcement Division, relates quite closely to the reporting structure recommended for pre-eligibility investigation units; that is, fraud investigation program staff report to (PCE) Supervisors who are independent of the district offices which issue benefits. Although the PCE Supervisors report, ultimately to the same Regional Director as do income support supervisors, the Regional Directors are assumed to be "neutral" in terms of representing all aspects of the organization fairly and with equal concern for what may seem to be differing goals. Although the Regional Directors clearly have the responsibility for delivering this program, the PCE division also has overall responsibility for program mandate and success. Thus:

Recommendation 10: It is recommended that the PCE Division undertake the role of monitoring the Early Fraud Detection and Prevention Program to ensure that the program is fulfilling expectations.
 
 
 
 

CLIENT RIGHTS

States with pre-eligibility investigation programs consider it important to balance the need to ensure eligibility with client rights to expeditious service, privacy and confidentiality.

Benefits are not withheld beyond the 30 day (Food Stamps) and 45 day (AFDC) mandated period allowed for eligibility determination and issue, and in most cases, the actual timeframes are much shorter - from 48 hours to five days. Most Counties have an emergency provision they can utilize to ensure immediate needs are met while circumstances are verified.

Orange County, whose program is widely used as a model for EDP Programs, has encased Special Provisions relating to client rights within their Early Detection/Prevention Program:

1. Referrals will not be made before the applicant(s) has signed the application for aid.

2. No intimidation, based on fraud allegations, shall occur, whether by referral or threat of referral, prior to completion of the application.

3. Intake fraud referrals and investigations shall not delay the timely receipt of aid

4. No uniformed Investigators shall be employed.

5. The County Welfare Department will abide by the confidentiality requirements and protection in the California Right to Financial Privacy Act.

In the State of Washington, as a result of concessions gained through court challenges to the EDP program early in its existence, Fraud Early Detection Investigators are now required to give clients a warning before entering a dwelling, to the effect that the client does not have to talk to the Investigator or provide access into the home. The warning also states that benefits will not be denied because client refuses entry into the home or refuses to speak to the Investigator. This requirement is considered restrictive and counter-productive by Investigation staff and the State legislature is presently reconsidering the issue.

Most States also have an appeal hearing process, either as part of the Administrative Disqualification Hearings for Intentional Program Violation system, or as a separate appeal process. This provides applicants who feel they have been unfairly dealt with by the pre-eligibility process, an opportunity to present their case to a neutral third party.

In all cases, it is considered crucial that client rights be protected while ensuring ineligible clients do not receive benefits.

Recommendation 11: It is recommended that a clear statement of client rights be developed and incorporated into the investigators' and Financial Assistance Workers' manuals and training guides to provide a clear framework for the operation of the early fraud detection and prevention program.

In British Columbia's case, as noted earlier, it may also be desirable to extend the role of already existing appeal panels to allow the hearing of pre-eligibility investigation-related cases.
 
 

TIMING OF INVESTIGATION PROCESS

In order to successfully prevent benefit payments from being paid in error, fast turnaround times for investigation requests are a most critical component of EDP Programs. American Federal regulations stipulate that payment must be made within 30 days to eligible clients on Food Stamp Program cases and within 45 days on Aid to Families with Dependent Children (AFDC) cases.

From the information provided by States on their EDP Programs, it appears that the most commonly established time frame for investigations is five working days. This time frame permits benefits to be issued where the client is deemed eligible well within the mandated time lines. The Office of Inspector General reported that about 51% of the 39 offices included in their 1994 report on AFDC pre-eligibility investigation units, required that investigations be completed within five working days after referral. Only about five percent of the sample offices allowed more than 30 days for conducting a pre-eligibility fraud investigation.

This chart provides a breakdown of investigation timelines for the 39 surveyed offices:
 
 

TURNAROUND TIMES	NUMBER OF LOCAL OFFICES
Within 5 days	20
6 - 10 days	4
11 - 15 days	5
16 - 20 days	0
21 - 30 days	8
More than 30 days	2
TOTAL OFFICES REPORTING	39

 

Some States have mandated other time-lines and requirements which must be followed, and which may impact the work of Pre-Eligibility Investigators. For example, clients in Wisconsin must be notified, verbally or in writing, of intention to conduct a home visit within 10 days.

In Connecticut, Investigators use a technique called the Ten Day Letter, which is left at the applicant's home if the applicant is not home when the Investigator attempts a home visit. The letter details information required and a date by which the Investigator must be contacted. If no response is received within ten days, the case is closed and benefits are denied.

In Iowa, immediately upon receipt of a referral, the Investigator sends a letter informing the applicant that an investigation is underway and the nature of the concern. The letter also informs the client that refusal to cooperate will result in the application being denied.

Texas Fraud Prevention Investigators must provide notification to Intake Workers within 24 hours on those cases not accepted for investigation.

In the case of British Columbia's Programs for Independence, there is an expectation or standard that benefit cheques are issued very quickly after applicants are deemed eligible, usually within the same day. This admirable standard of promptness makes it difficult to delay issue on suspect cases in order that an Investigator has time to check into the situation and report back to the Eligibility Worker. It may be possible to utilize the program's Temporary Hardship Assistance provisions to provide assistance in cases where need is of an emergency nature, in order to provide the Investigator a reasonably sufficient time to validate or refute the applicant's stated circumstances.

Recommendation 12: It is recommended that British Columbia explore the possibility of relaxing the standard of same day/next day service for cheque production for cases where a referral for pre-eligibility investigation is underway.
 
 

Recommendation 13: It is recommended that only Temporary Hardship Assistance be provided in cases which are being referred for pre-eligibility investigation and only where need can be clearly established.
 
 

Recommendation 14: It is recommended that Pre-Eligibility Investigators in British Columbia be given a five day turnaround time standard to respond to an Financial Assistance Worker's referral.
 
 

LINKAGE ACROSS OFFICES

In order to maximize the effectiveness of an Early Fraud Detection and Prevention Program, it is important that there be in place a province-wide information system which can be used by Eligibility Workers and for the Pre-Eligibility Investigators to record information on each case where an investigation has resulted in denial of or reduced benefits. This type of capability is most often built into the existing eligibility and case management system in the American programs. Quick access to a data bank of clients who have unsuccessfully attempted to obtain benefits using incorrect information will prompt Eligibility Workers to carefully scrutinize another application by the same person. Such cases may be considered at high risk for another fraudulent attempt and most States include in their referral criteria all applicants who have been previously investigated, and have had benefits denied as a result of that investigation, at the pre-eligibility or post-eligibility stage. Shortly after Orange County implemented their Early Detection/Prevention Program, the administration rechecked a sample of 350 cases on which assistance had been denied as a result of the pre-eligibility investigation. On almost 20% of those cases, it was found the applicant had reapplied and been granted assistance - in most instances through a different District Office than the original, leaving some concern about their eligibility. The County thereafter developed a system of flagging the Master Index and the file of any applicant and recipient referred for investigation so all workers in the County had ready access to such information.

British Columbia Social Services presently has an "alert" capability on the client index system. This system could also be utilized to record pre-eligibility investigation cases in which benefits were denied or reduced. Thus, some systems development work would be required to accommodate this new "alert" category, upon implementation of an early detection initiative.

Recommendation 15: It is recommended that a new category of alert be added to the current Income Security computer system to identify clients who have been found ineligible as a result of a pre-eligibility investigation.
 
 

THE AMERICAN EXPERIENCE

It has been 14 years since the first experimental early fraud detection and prevention program was implemented in Orange County, California in 1981. Since that time, well over half of the American states have implemented some variant of this program. The federal government continues to encourage states to undertake this proven method of reducing state and federal costs for Income Security programs.

Without exception, when American states or counties embarked on the implementation of an early fraud detection and prevention program it was done via a piloting process. Even in states which "borrowed" another state's plan and design, along with program and training material, implementation still occurred slowly, with office-by-office implementation after an initial pilot period of, in most instances, at least one year.

There is no Canadian jurisdiction which can be used as a model for British Columbia. Given the differences between the American Food Stamps and Aids to Families with Dependent Children programs and the British Columbia G.A.I.N. program, and given that British Columbia cannot rely on an extensive computer matching network, and does not have an administrative disqualification program, a pilot approach is even more critical.

Implementing an early fraud detection and prevention program within British Columbia through a pilot process in several carefully selected sites would have the following advantages:

1. Staff from the pilot sites could take a lead role in the development and adaptation of the referral criteria, giving them a "stake" in making the criteria effective. It would even be possible to use different criteria in different sites to test out the differences in referral patterns that could be attributed to the different criteria in order to determine the most effective overall set of criteria.
 
 

2. Policy, procedures and forms can be developed, changed and improved during the pilot period.

3. Improved estimates of staffing requirements can be obtained during the pilot period.

4. Improved estimates of cost-savings attributable to the program could be obtained, making it easier to justify program expansion.

5. Selection of pilot sites in different types of locations would allow some insight into whether the overall level of attempted fraud at intake is comparable regardless of the characteristics of the site.

6. Communication to all staff of the success of the pilot projects would develop an increased awareness of fraud issues, as well as establishing "live" sites for staff not involved in the pilot(s) to visit prior to implementation in other offices.

Recommendation 16: It is recommended that British Columbia implement an early fraud detection and prevention program by using a pilot project approach.
 
 

SELECTING A PILOT SITE

Depending on the assumptions made about client fraud, the results from a pilot early fraud detection and prevention project could be highly variable depending on the location within the province where the pilot was implemented. If one assumes that clients who are most likely to attempt fraud on intake tend to gravitate to certain inner city locations within British Columbia, then one might expect a higher cost-benefit from a pilot located in the Vancouver inner city than from a pilot located in Chilliwack. If on the other hand, one assumes that fraud is a result of opportunity rather than location, there should not be much difference in cost-benefit between the two sites. Thus:

Recommendation 17: In order to ensure that the results of the pilot test of the early fraud detection and prevention program are not questioned on the basis of location, two or three sites should be selected for simultaneous pilot implementation, in varying urban/rural configurations.
 
 

Types of urban/rural configurations which could be considered would include:

Metropolitan inner city (eg. Vancouver)

Small urban/suburban (eg. practically anywhere in the lower mainland, such as Langley, Surrey, etc. or south Vancouver Island)

Urban with rural fringe (eg. Kamloops, Nanaimo)

Isolated small urban with rural fringe (eg. Prince George)

Alaska, which borders on British Columbia, is the American state most comparable to northern British Columbia. Alaska has had a very effective early fraud detection and prevention program in place in the greater Anchorage area for several years. Although they would like to expand the program to Fairbanks and the Kenai Peninsula offices, they also recognize that since the concentration of income security cases is so sparse in much of the state, and the individuals are so well known in their place of residence (because of the very small community size), that it would not be an effective use of resources to attempt to cover the entire state with this program. British Columbia may want to test out this assumption by ensuring that one of the pilot sites is in the north of the province where caseloads are small and dispersed.
 
 

DEVELOPING AN IMPLEMENTATION PLAN

Although it is beyond the scope of this report to develop a complete program design and implementation plan for piloting of an early fraud detection and prevention program in three sites spread across British Columbia, it is important to note the major components of that plan. These components of a detailed program design phase would include development of:

1. Full program description and rationale

2. Policy, procedures, and forms for supervisors, Financial Assistance Workers and Pre-Eligibility Investigators

3. Referral criteria

4. Information access and investigation guidelines for investigators

5. Description of client rights

6. Clarification of roles, authorities, expectation and reporting relationships

Major components of an implementation plan would include development of:

1. Project Management strategy and selection of Steering Committee and project manager

2. Systems Development plan

3. Staff Consultation strategy

4. Staff Training plan and accompanying materials

5. Strategy for and selection of pilot sites

6. Staff reallocation plan or hiring schedule

Timeframes for developing the appropriate materials and implementation plans would be approximately 6 months from the date of approval to proceed to the date of implementation in the first pilot site. Therefore, with rapid approval of the recommendations in this report, it would be possible to see a pilot site in operation by October/November, 1995.
 
 

DEVELOPING A COMMUNICATIONS STRATEGY

A comprehensive communications strategy would be necessary to support the implementation, even on a pilot basis, of an early fraud detection and prevention program. Stakeholders who would have an interest in the development of this type of program would include:

Government of British Columbia

Staff and managers

Taxpayers/the general public

Local police and other law enforcement departments

Current clients and new applicants
 
 

Programs and jurisdictions providing information to investigators

Media

Provincial and local client Advocacy Groups

Local aboriginal communitiesA proactive strategy is always preferable to a reactive strategy. Thus, early consultation with departmental staff, local police and law enforcement officials, and officials from programs and jurisdictions who would be asked to provide information to investigators, followed by later involvement of other provincial and local stakeholders would be important to ensure that the program implementation had the strongest possible support in the local community. Thus:

Recommendation 18: It is recommended that a strong, proactive communication strategy accompany the implementation and pilot testing of the early fraud detection and prevention program to ensure that the program intent and operation is clearly understood by all stakeholders.
 
 

DEVELOPING AN EVALUATION/MONITORING PLAN

Given that the long-run intent is to justify the establishment of an early fraud detection and prevention program throughout the province of British Columbia, evaluation and monitoring must be considered before implementation on a pilot basis begins. In particular, the development of an evaluative framework must be completed very early in the overall implementation planning process to allow for baseline data to be collected before implementation in the pilot sites begins. Data definitions and appropriate monitoring instruments will also be required before the pilot is actually "up and running". It may be useful to also identify several "control sites" where baseline data is also collected to provide a basis of comparison for the pilot sites.

Key evaluative themes would include:

1. Impact and cost-benefit of the early fraud detection and prevention program, including number of new and reopened cases approved by the Financial Assistance Worker, cases denied benefits by the Financial Assistance Worker, cases referred for pre-eligibility investigation, cases denied benefits and cases receiving reduced benefits as a result of the pre-eligibility investigation.

2. Effectiveness of the procedures and referral protocols, forms and data collection instruments.

3. Additional information sources required by investigators to improve effectiveness of the program.

4. Changes to staff relationships and morale as a result of the program. Do Financial Assistance Workers feel more in control of the caseload with the additional options for decision-making? Is there a smoother, stronger working relationship between the investigation staff and the Financial Assistance Workers and other front-line staff?

5. Is there any noticeable deterrence effect as the pilot proceeds with fewer new applicants even attempting to provide inaccurate and/or fraudulent information on intake?

6. Public, client and staff reaction to the operation of the pilot project.

The intent would be to make any necessary "mid-course corrections" at the earliest possible point in the implementation process to maximize the returns from the early fraud detection and prevention program. The information provided by this evaluation would enable senior management to make informed decisions on the overall validity of the project. Major decisions which would be made at the end of the pilot period (at least one year) would be to cancel the pilot(s), to continue the pilot(s) as originally designed for some additional period of time, to make changes to the design prior to continuation, to expand the pilot(s) to other major sites, and/or to expand the pilot to province-wide implementation.

Monthly collection of data from the monitoring instruments in both pilot and control sites could be reviewed on an on-going basis to identify at the earliest possible time whether preliminary estimates of savings were accurate. This data could be compared to detailed monthly implementation data from several American states to determine whether the British Columbia experience is following the expected patterns.

Recommendation 19: It is recommended that an evaluation/monitoring plan be established before the implementation of the early fraud detection and prevention program in the pilot sites to enable baseline data to be collected before implementation occurs.
 
 

IMPLEMENTATION COSTS

British Columbia local offices usually have a caseload of approximately 2,000 to 2,500 cases. Given that intakes usually run at about 15% of overall caseload, then an "average" office would have approximately 300 to 375 intakes per month. This is roughly equivalent to the workload "standard" identified by California. Thus, each pilot site selected should start with one full-time equivalent Pre-Eligibility Investigator assigned to the site. Given the earlier recommendation regarding establishing this as a partial function for all Assistant Investigators, this "full-time equivalent" may represent several individuals.

This level of staffing may require adjustment as the pilot continues and the referral patterns become more apparent. However, the initial "investment" in staff in order to run pilots in three sites across British Columbia would be 3 man-years, which could possibly be seconded or reallocated from existing Assistant Investigator positions.

Other implementation costs are identified in the Prevention, Compliance, and Enforcement Division Strategic Plan: Quality Service with Integrity, February, 1995. These costs include an additional program manager within the PCE branch, and approximately $50,000 in contract resources for program design, implementation, and evaluation assistance.

Hidden costs of the implementation would be the staff time required for staff consultation and training on-site in the pilot offices.
 
 

HIGHLIGHTS OF NEW PROGRAM

The implementation of an early fraud detection and prevention program in the British Columbia Income Security intake process will have several major impacts:

1. Significant cost-benefits will be achieved. If British Columbia achieved even average returns, estimated at $8.43 saved per dollar expended in American states and counties which have implemented the program, this would provide a significant savings.

2. The implementation of an early fraud detection and prevention program will be seen publicly as having a major impact on improving program integrity.

3. The implementation of this program will clearly demonstrate government concern for responsible stewardship of public funds and improved accountability for expenditures.

4. Demonstrated concern for client rights will allay any possible criticism about inappropriate and reckless invasion of privacy.

5. A carefully designed and implemented program, with well-trained professional staff, would ensure that individual applicants for benefits were treated fairly and consistently through the application of a pre-designed set of referral criteria which would identify high-risk cases.

6. Staff, particularly in the intake area, will feel in greater control of the benefit approval process, and will have a greater ability to ensure benefits are granted appropriately in questionable cases.
 
 

SUMMARY

Taking the lead in Canada in the development of an Early Fraud Detection and Prevention initiative would be an exciting challenge for the British Columbia delivery system, effectively addressing a crucial problem that costs millions of dollars every year. As the only province in Canada which has clearly stated its intention to address prevention and deterrence rather than after-the-fact detection and prosecution, implementation in a pilot setting of an early fraud detection and prevention initiative is a progressive first step.

Recommendation 1: Based on the experience of the majority of American States, it is recommended that British Columbia develop and implement an early fraud detection and prevention system.

Recommendation 2: It is recommended that British Columbia start their development of a set of referral criteria by adapting a combination of the criteria developed by Texas, Minnesota and Ohio ("best state" model).

Recommendation 3: It is recommended that British Columbia use a staff consultation model in finalizing the referral criteria for the British Columbia Early Detection and Prevention Program.

Recommendation 4: It is recommended that, with the demise of the Canada Assistance Plan, British Columbia consider the implementation of an administrative disqualification process for clients who are found to have willfully misrepresented their circumstances in order to obtain program benefits.

Recommendation 5: It is recommended that the role of the present Appeal Panels be expanded to include the hearing of appeals of applicants whose benefits were reduced or who were found ineligible as a result of an eligibility investigation.

Recommendation 6: It is recommended that the 28-day On-The-Job Training Program be expanded to include at least one day on the Early Fraud Detection and Prevention initiative, focusing on the use of the referral criteria, interviewing techniques to ascertain whether an applicant meets those criteria, and dealing with clients who are ineligible as a result of an eligibility investigation.

Recommendation 7: It is recommended that all current Financial Assistance Workers, supervisors and investigative staff receive full training on the early fraud detection and prevention program during the implementation phase.

Recommendation 8: It is recommended that the role of the Assistant Investigator be expanded to include the conducting of eligibility investigations based on a referral from an Financial Assistance Worker.

Recommendation 9: It is recommended that all Assistant Investigators spend some portion of their time on eligibility investigations, rather than hiring specialized staff, in order to maximize the working relationship between the Assistant Investigators and the Financial Assistance Workers.

Recommendation 10: It is recommended that the PCE Division undertake the role of monitoring the Early Fraud Detection and Prevention Program to ensure that the program is fulfilling expectations.

Recommendation 11: It is recommended that a clear statement of client rights be developed and incorporated into the investigators' and Financial Assistance Workers' manuals and training guides to provide a clear framework for the operation of the early fraud detection and prevention program.

Recommendation 12: It is recommended that British Columbia explore the possibility of relaxing the standard of same day/next day service for cheque production for cases where a referral for pre-eligibility investigation is underway.

Recommendation 13: It is recommended that only Temporary Hardship Assistance be provided in cases which are being referred for pre-eligibility investigation and only where need can be clearly established..

Recommendation 14: It is recommended that Pre-Eligibility Investigators in British Columbia be given a five day turnaround time standard to respond to an Financial Assistance Worker's referral.

Recommendation 15: It is recommended that a new category of alert be added to the current Income Security computer system to identify clients who have been found ineligible as a result of a pre-eligibility investigation.

Recommendation 16: It is recommended that British Columbia implement an early fraud detection and prevention program by using a pilot project approach.

Recommendation 17: In order to ensure that the results of the pilot test of the early fraud detection and prevention program are not questioned on the basis of location, two or three sites should be selected for simultaneous pilot implementation, in varying urban/rural configurations.

Recommendation 18: It is recommended that a strong, proactive communication strategy accompany the implementation and pilot testing of the early fraud detection and prevention program to ensure that the program intent and operation is clearly understood by all stakeholders.

Recommendation 19: It is recommended that an evaluation/monitoring plan be established before the implementation of the early fraud detection and prevention program in the pilot sites to enable baseline data to be collected before implementation occurs.