COMMITTEE OF PUBLIC ACCOUNTS debate -
Thursday, 8 Apr 2004

Department of Social and Family Affairs — Value for Money Report on Evaluation of Control Activity.

Mr. J. Purcell (An tÁrd Reachtaire Cuntas agus Ciste) called and examined.

Mr. J. Hynes (Secretary General, Department of Social and Family Affairs) called and examined.

Chairman

Witnesses should be aware that they do not enjoy absolute privilege and should be apprised as follows. Members' and witnesses' attention is drawn to the fact that as and from 2 August 1998, section 10 of the Committees of the Houses of the Oireachtas (Compellability, Privileges and Immunities of Witnesses) Act 1997 grants certain rights to persons identified in the course of the committee's proceedings. These rights include the right to give evidence; the right to produce or send documents to the committee; the right to appear before the committee, either in person or through a representative; the right to make a written and oral submission; the right to request the committee to direct the attendance of witnesses and the production of documents, and the right to cross-examine witnesses. For the most part, these rights may only be exercised with the consent of the committee. Persons invited to appear before the committee are made aware of these rights and any person identified in the course of proceedings who is not present may have to be made aware of them and provided with a transcript of the relevant part of the committee's proceedings if the committee considers it appropriate in the interests of justice.

Notwithstanding this provision in legislation, I remind members of the long-standing parliamentary practice to the effect that they should not comment on, criticise or make charges against a person outside the House, or an official, either by name or in such a way as to make him or her identifiable. They are also reminded of the provision in Standing Order 156 that the committee should refrain from inquiring into the merits of a policy or policies of the Government, or a Minister of the Government, or the merits of the objectives of such policy or policies.

I welcome Mr. Hynes, Secretary General, Department of Social and Family Affairs, and invite him to introduce his officials.

Mr. John Hynes

I take the opportunity to welcome the Comptroller and Auditor General's report.

Chairman

Will Mr. Hynes first introduce his officials?

Mr. Hynes

I am accompanied by Ms Bernadette Lacey, director general, social welfare services; Mr. Eoin O'Broin, regional director with responsibility for control activity in the Department; Ms Anne Tynan, until recently the accountant in the Department, and Ms Fiona Ward, assistant principal, control division.

Chairman

Who is present from the Department of Finance?

Mr. Dermot Quigley

I am from the public expenditure division which deals with the Vote for the Department of Social and Family Affairs.

Mr. John Purcell

Social welfare fraud and how to counter it have been persistent matters of concern to this committee over the years. This is understandable when one considers that between €10 billion and €11 billion is paid out by the Department annually in benefit and assistance payments. If only 1% of this figure constitutes fraudulent payments, it amounts to approximately €100 million. It is important, therefore, that the Department has systems in place to minimise the risk of fraud and overpayments generally. With this in mind, I undertook an examination of control activity in the Department, particularly the extent to which that activity is informed by an assessment of the risk attaching to the individual types of welfare payment and the effectiveness of the detection measures employed. I also reviewed how the Department evaluated the performance of its control activity.

The report shows that the Department takes its control obligations seriously. It spends an estimated €18 million alone on dedicated control measures, by which I mean resources applied over and above the normal claim checking that takes place at the first point of contact between the claimant and the Department. Over the years the Department has gradually developed its approach to fraud detection and some of its initiatives, particularly in the data matching line, have yielded substantial savings. Our work has established that what is needed now is a more strategic approach to tackling social welfare fraud and overpayments. The Department also recognised this and adopted a new control strategy last May.

As part of that strategy, the Department has built on some earlier work which identified possible gaps in its payment control procedures and is moving to a risk assessment model, culminating in a risk rating for each scheme. It is intended that this will inform the targeting of control reviews in order that, ultimately, groups of clients receiving payments will be subject to a degree of review appropriate to their assessed risk of overpayment. Within this strategy, we found there was scope for better application of the information gleaned from reviews on the various causes of overpayments.

At the time of the examination the Department used two main measures of performance — how many reviews it had conducted and the savings which arose as a result of those reviews. By using multipliers based on the average amount of time before the person recommences claiming, the Department calculates the notional savings achieved as a result of its control activity. The figure so calculated for 2002 was €283 million but this estimate of savings gives no indication of how effective control activity is in the first place. Internationally, it has been recognised that the appropriate indicator of effectiveness is the movement in the levels of welfare overpayments occurring. To introduce such an indicator, it is necessary to establish an opening baseline measure of the scale of overpayments in schemes. With my encouragement, the Department is moving in this direction.

During the examination and in consultation with my office the Department carried out a special review of randomly selected unemployment and one-parent family claims and payments with a view to arriving at initial estimates of the scale of the overpayment problem in those schemes. The preliminary results of this exercise indicate that about 28% of one-parent family cases and 16% of unemployment assistance cases required revision of the level of payment. These results should be treated with caution as the methodology used needs a fair degree of refinement before definitive conclusions can be reached. The Department is improving the methodology and broadening the scope of the special reviews to cover payments under other welfare schemes. Once the methodology and scope have been fully developed the Department should be in a position to monitor the effectiveness of its prevention and control activities on the level of overpayments occurring.

My examination also looked at the performance of the Department in recovering overpayments. Overall, by value, nearly one third of overpayments are recovered but only 18% in those cases where the overpayment is attributable to fraud. The Department is in the initial stages of developing a centralised overpayments and debt management system which in the long run should help to boost recovery rates. The report gives details of the Department's prosecuting activity and some information on the number of persons who repeatedly defraud the system.

Mr. Hynes

I welcome the Comptroller and Auditor General's report on control activity in the Department which is responsible for delivering in the current year €11.3 billion in income supports to approximately 1.5 million people. Control of fraud, abuse and error is an essential part of the work of the Department to ensure social welfare funds only go to those entitled to them.

The evaluation by the Comptroller and Auditor General was undertaken at an opportune time for the Department. It came as it was undertaking a review of control activity and preparing a revised control strategy. A key objective of the control strategy is to ensure we pay the right person the right money at the right time. A four pronged approach has been adopted by the Department to meet this objective: prevention of fraud and error at initial claim stage, early detection through effective review of claims in payment, measures to deter fraud and abuse and the pursuit and recovery of overpayments.

One of the key recommendations in the report is that the Department should introduce a formal risk management approach in its control and review activity. The Department has always had a wide ranging control programme, including controls at the point of claim, a programme of investigation and reviews of entitlement and pursuit of overpayments. However, while its programme has hitherto focused attention on areas where experience has shown that there is greater exposure, this has not been done in a systematic way or by using a risk management approach as recommended by the Comptroller and Auditor General.

Systematic risk analysis will in the future be a key element of control strategy. Scheme managers will be required to identify the areas of high risk in the schemes for which they are responsible and take appropriate action to address them in a systematic way. We have begun this process with nine major expenditure areas included in the first phase. The Department is also undertaking a number of specific control initiatives in the ten regions, the outcomes of which will be evaluated with a view to expanding successful initiatives to other areas. These initiatives will also feed into the process of refocusing control activity to those areas and types of cases which present the highest risk of fraud and error.

The benefit of a risk management approach will be that review activity can be targeted more effectively. The report also recommends that this approach be combined with the more effective use of the information already held in existing databases. There is already extensive matching of data with that held by agencies such as the Revenue Commissioners under powers given in legislation and the quality and timeliness of this information exchange are improving. More integration of systems within the Department will also improve the ability to cross-reference information from different sources for control purposes.

The review points out that the Department has no reliable estimate of the level of fraud and error occurring. In the context of the review, a survey was undertaken of the unemployment and lone parents schemes for this purpose. The results indicated a significant proportion of cases where people were being paid incorrect amounts. In the bulk of these cases, this was due to increases in the earnings of the people concerned and the fact that the cases had not been recently reviewed. A project has been put in place, using information on previous earnings, to review similar cases. Arrangements will be made for more frequent reviews of such cases in the future.

The Department will include surveys of the levels of fraud and error as an integral part of the control strategy. During 2004 the Department plans to undertake three such surveys, the first of which is to commence shortly. The outcomes of these surveys will feed into the development of the Department's control approach in the future and be used to identify the types of claims which should be prioritised for review purposes.

The report highlights the need for an effective system of identifying and recovering overpayments and using information from the analysis of overpayments to identify where procedures need to be changed or where higher levels of review activity may be required. The Department is developing a new overpayment and debt management system which will provide better information on overpayments and overpayment recovery and better linkages between different schemes in that regard. In the short term the Department has initiated a project at regional level to analyse overpayments which will feed into a claims risk profiling process.

As regards deterrents, another major component of control strategy, the prosecution of cases where fraud has occurred is an important aspect. During 2003 the Department issued revised prosecution guidelines which have had the effect of increasing the numbers of cases being submitted for prosecution. In 2003, 405 cases were referred to the Chief State Solicitor's office with 88 being forwarded by the end of February this year. To complement prosecution policy, the Department is examining the issue, also raised in the report, regarding the possible imposition of administrative sanctions as an alternative to prosecution in appropriate cases.

The Comptroller and Auditor General's report has provided a valuable commentary on the Department's control effort and an input into the Department's ongoing efforts to tackle fraud, abuse and error. It has already had a significant impact on the direction the Department will take in this area. The essential elements in the strategic approach to control which the report recommends and on which the Department is following up are regular surveys to establish baseline levels of fraud and error in different schemes; a better analysis of the causes and emerging patterns of fraud and irregular payments; a risk management approach to direct the Department's review activities more effectively; better use of information and IT systems for the identification of control priorities; more comprehensive systems of debt recording and management and greater analysis of the underlying factors giving rise to overpayments; a wider array of sanctions for fraudulent claiming; and a wider and improved set of performance indicators to measure the ongoing effectiveness of our control systems.

Any system which involves the transfer of funds of the magnitude involved in social welfare payments is likely to be subject to abuse. The majority only get their legal entitlements but we must ensure the extent to which they do not is minimised. Implementation of the findings of the report and the development of a new control strategy will help to keep our control systems up to date and effective. This will have implications for our management systems, IT systems, staff training programmes and the level of resources we apply to control and their deployment. I acknowledge the value of the report to the Department and assure the committee that the findings of the review will be taken on board and given effect in the context of moving forward the programme of modernisation of the social welfare system.

Chairman

Can we publish the report?

Mr. Hynes

Yes.

Chairman

The report is important. The Department operates 34 social welfare payment schemes with a total fund of €11 billion and 1.5 million recipients. The Comptroller and Auditor General stated that even a 1% miscalculation would cost €100 million. That shows the scale with which we are dealing. The Department does not have a benchmark of its control activity in terms of savings. It is difficult to know how the figure of €283 million was calculated given that the Department does not seem to have a method of calculating such figures.

Mr. Hynes

The figure of €283 million is an estimate of the savings which arise from the stopping of payments in particular cases. It involves projecting how long the person whose payment has been stopped would have stayed in the system if his or her payment had not been stopped. To that extent, it is not an actual accounting figure which appears in the accounts. However, it is useful to us in the sense that it gives an indication, year on year, of how we are doing on the control front. It indicates if we are doing better or worse than previous years. It is calculated by applying a multiplier to the value of the benefit a person is receiving when his or her payment is stopped. The multiplier varies with the type of payment in question. We use certain information which came to us from the United Kingdom where a similar exercise was undertaken some years ago. We have also undertaken some verification work by looking at samples of cases of people whose payment was stopped and how long they stayed out of the system before coming back in again. We have tried to verify in a number of ways the multiplier used.

The report questions the accuracy of the multiplier and the basis on which the figures are derived. We are looking at other ways of evaluating or measuring the effectiveness of control activity. However, we have found it useful as an indicator of how we are doing year on year, particularly in terms of our control activity. We set targets at the beginning of each year for each scheme in terms of the number of reviews and the level of savings we expect them to achieve. In that sense, it is useful as a management tool.

Chairman

The approach used extensively in America is data-matching which identifies investigations by certain characteristics. Has the Department examined this scientific system which has produced a significant yield?

Mr. Hynes

The report refers to the importance of data-matching which we have done in the past and hope to do more in the future. We use information on earnings from the Revenue Commissioners and match it against our databases to highlight cases where people in the Revenue system appear to be in receipt of social welfare payments also. This is a good indicator of the level of fraud. We use it to feed into our review activity. I agree that such an approach is important. As IT systems get better and more integrated, we should have more opportunities to use such an approach.

Chairman

Does the Secretary General agree that many surveys can be evasive in getting the figures? A scientific method of data-matching, particularly in cases involving huge amounts of money, may not achieve the necessary level of savings.

Mr. Hynes

That is one way of putting it. It is only one indicator. We hope to develop better and more factually based indicators in the future from the surveys we will carry out of fraud and error. We will examine the materials coming from the risk assessment process, recommended in the report, to see where schemes are more open to risk and where there are greater levels of risk and try to target our review activity. We will also examine our review activity and the percentage of reviews which yield savings as an indicator of the effectiveness of our activity. We will use a wider range of indicators in the future.

Chairman

I am delighted to hear it.

Deputy Dennehy

I welcome back Mr. Hynes and would like to put this issue in context. We must recognise the huge volume of work the Department is doing. Some 1.5 million people benefit from a fund of €10.2 billion. It is difficult to get one's head around such figures but it is important to highlight the number of recipients because what could grab the headlines is the fact that there were 39,000 cases of overpayment during 2002 with a total value of €29 million. Of these, 48% were fraudulent. This meant that approximately 19,000 fraud cases emerged from 1.5 million claimants. Am I correct in assuming this set of figures?

Mr. Hynes

Yes, the figure of 48% is correct.

Deputy Dennehy

Therefore, we are talking about 19,000 cases that were fraudulent.

Mr. Hynes

That is correct.

Deputy Dennehy

When one considers that the Department has a 5% error rate, it shows what we are dealing with. There is always a fear that if the system is not right, there will be room for fraud but we need to look beyond this. I might almost start at the end. There is an argument — probably a general consensus — against imprisoning people for fraud, unless the case is an extreme one. I noted that Mr. Hynes was examining a wider range of sanctions where fraud was involved. Will he outline what he has in mind and what is intended?

Mr. Hynes

The report mentions the issue, whether it would be possible, short of taking prosecutions, to apply other sanctions. Examples were given of what happens in the United Kingdom where people may receive a formal caution, rather than being taken to court, or where a fixed penalty can be applied in certain cases by the relevant Department. In some cases the overpayment amount which the person has against him or her can be increased, or a percentage increase can be applied to it, and he or she is required to pay back a higher amount. Therefore, there are a number of possibilities. As there are legal issues in our system which are different from those in the United Kingdom, we have to examine the legal implications but we certainly need to address this area.

Deputy Dennehy

In the past one of the encouragements to keep claiming was that when one went to court a fine of only £20 or £40 would be imposed. Five or six years ago there was a judgment to the effect that the amount outstanding had to be repaid. Is that still binding? Does the amount claimed have to be repaid before other penalties can be imposed by the court?

Mr. Hynes

When we take a case to court, we are not really talking about the overpayment but about a criminal offence. In that sense, there is a separation. Frequently, in practice, the person concerned is required to pay back the overpayment before the court decides on the appropriate sanction. As I understand it, the Probation of Offenders Act cannot be applied until the overpayment has been repaid. The Department would prefer to have the fraud dealt with as a separate issue. The recovery of the overpayment is something we set about doing separately.

Deputy Dennehy

Looking at it from our perspective and examining prevention measures purely, it was certainly an encouragement in the old days to chance claiming if one would get away with a fine which was one-fiftieth of the overpayment concerned. I thought, however, a judgment had been made that such overpayments had to be repaid, which would be important.

The Chairman mentioned the question of data-matching but I was shocked to discover that other Departments did not do this; Revenue is an example that has been quoted here. At one time it certainly did not seem to be in touch with the Companies Registration Office and there was no cross-referencing. When did the Department begin cross-referencing with the Revenue Commissioners?

Mr. Hynes

That goes back a long time but I will have to ascertain the exact date. The fact that the Revenue Commissioners are responsible for collecting PRSI contributions, for example, means there is a close working link between them and the Department. For that reason information is exchanged. In 1997 we initiated the "financement of earnings" project with the Revenue Commissioners, whereby on a regular basis we are informed by Revenue when people take up employment. When we obtain this information, we match it against our unemployment records to see if there is an overlap — that is, if it appears, according to Revenue's information, that a person is back at work and still claiming in our system. We follow up all such cases. This has been happening routinely since 1997.

Deputy Dennehy

Throughout his report the Comptroller and Auditor General pointed to the fact that such initiatives had not been taken. He found that people were able to register six, seven or eight companies. One builder in Dublin had registered about ten companies and nobody knew about it, although he was building all over the city. Such a situation would be revealed if an individual had the same social security number, rather than using initials or a combination of his or her name in English and as Gaeilge. In this way fraud could be tied down but I presume the Department has been using such methods for a long time.

Mr. Hynes

Yes. We are now trying to have the PPSN or public service number as it is now called — it used to be known as the RSI number — used more widely across the public service. One of the benefits is that information can be matched, subject to having the legal basis on which to do so. It offers the opportunity to match information more effectively. We also have an information exchange with the education sector concerning those involved in third level education which enables us to match that data against the data in our own systems. There is huge potential for the better control of payments by using such data-matching techniques.

Deputy Dennehy

I note that client error accounted for 25% of overpayments. I am worried, however, that, given legislative changes and alterations in the number of stamp contributions required, there will be room for genuine error. Are we getting the message across plainly enough to applicants concerning these changes? I know the Department is issuing a range of pamphlets, which I have received and which are both excellent and of great help. Constituents come to me about such issues but if they do not and choose to take their own route, are we getting the message across? It appears not, given that there was a 25% error rate in 39,000 cases. Does Mr. Hynes have a view on this?

Mr. Hynes

There is no doubt that the system is complex. I accept that from time to time it is difficult for people to find their way through the various rules and regulations. We have a strategic policy programme in place to try to improve the level of information available within the system. When people make a claim at our local offices, they are told what the rules and regulations are concerning whatever payment they are claiming but I suppose we can never get it totally right in that regard. We can never provide enough information but are conscious of the need to explain the rules in order that people are fully aware of the conditions attached to whatever payments for which they might apply.

Deputy Dennehy

While I do not want to contradict myself, I pay tribute to Mr. Hynes and his staff for the level of information provided. It concerns me, however, that one quarter of overpayments are down to client error. For years we had a standard question which we put to Revenue at a time when there were all sorts of allegations about what was going on. That question was: do you have any requirements by way of legislation? Mr. Hynes mentioned that whatever the Department would do would have to be legally based but are there any changes in legislation he would like to see which would help in controlling the system?

Mr. Hynes

We mentioned the issue of administrative penalties as an alternative to prosecution, an area at which we will be looking. If there are proposals to be made, we will come forward with them, subject to what the legal requirements are. We are continuously looking at the situation regarding the exchange of information and from time to time have discussions with the Data Commissioner about what can be done. Perhaps that is an area in which there is scope for the further exchange of information and greater integration of systems.

Deputy Dennehy

From time to time we make recommendations as happened in the DIRT inquiry. However, we cannot enter the area of policy. If there are areas where you want to see changes, we would like to hear about them.

I want to refer to the breakdown of overpayments included in the report, specifically on pages 20 and 23. While I know one can do anything one likes with percentages and figures, the proportion of the total value of overpayments is 41% for unemployment payments and 24% for old age pension payments. A figure of 73% is included in the report. According to paragraph 234 on page 23, which deals with fraud, unemployment assistance, unemployment benefit and single parent payments accounted for over 80% of the amount recovered. Will Mr. Hynes talk us through these figures giving the amount overpaid on old age pension and unemployment payments?

Mr. Hynes

In the case of unemployment payments, there is a difference between long-term and short-term schemes. In the case of short-term schemes, there is a lot of activity, with people moving in and out. We carry out a lot of investigations of unemployment claims. For that reason the scale of overpayment identification is higher because of the level of review activity. The scope for fraud is possibly higher because, as pointed out in paragraph 234, the main cause was concurrent working and claiming unemployment payments and not revealing means when applying for unemployment assistance. That is why the scale of overpayment is higher is this area than in others.

In the case of old age pension payments, as the Deputy knows, when a pensioner dies, there is an obligation on his or her personal representative to inform the Department of the assets of the person concerned. There is a lot of activity in recovering overpayments. This explains the high level of recovery in this area.

Deputy Dennehy

That is a little complicated and probably distorts the figures. Between 1998 and 2000 the average value of overpayments was €999 while the average weekly amount overpaid was €55. In the case of lone parent and separated person schemes, the average amount detected was €12,000. In the case of the unmarried lone parent scheme, the average amount detected was €11,000. I accept the difference between the long and short-term schemes. Does the Department prioritise checking based on the amount overpaid or the number of cases involved? Obviously, the matter has to be prioritised in some way.

Mr. Hynes

We try to identify the schemes in which fraud is most likely to occur. The approach recommended in the Comptroller and Auditor General's report is that we should do this in a more systematic way and that we should identify the specific risks to which we are open. We would have said that the short-term unemployment and lone parent schemes involved people of working age who had the opportunity to work. In the case of the lone parent scheme, a person is entitled to work and earn up to a specific amount and still qualify for payments. Obviously, this is an area on which we need to focus. There is an opportunity to engage in fraud and a lot of movement in and out of schemes. As a general rule, we try to target the short-term schemes as in the longer-term schemes, there is not the same scope for overpayment, for example, in the case of those in receipts of pensions. In the case of the old age pension scheme we look at the situation afterwards and then recover overpayments. Therefore, the level of activity during the payment period is reduced.

Deputy Dennehy

I am trying to look at the significance of the breakdown provided. Some of the schemes mentioned are administered by community welfare officers who are separate from the Department and I presume report to the health boards. Arguments have been made about the rent subsidy scheme which is administered and monitored by community welfare officers. Is there a difficulty for Mr. Hynes in the fact that another Department administers this scheme? What proportion of the total budget of €10.2 billion is involved? Is the amount big or small?

Mr. Hynes

It involves a sum of €600 million to €700 million. I do not have the exact figure.

Deputy Dennehy

Would it be less than one tenth — approximately 8%?

Mr. Hynes

Yes. As the Deputy says, the scheme is administered by the health boards under the general direction and control of the Minister. That is how it is phrased in the legislation. We have responsibility for ensuring the legislation is applied. There is a lot of contact between the health boards and the Department on the level of control activity and so on. The fact that a lot of supplementary welfare payments are made while people are waiting on other social welfare payments means there is a lot of contact between community welfare officers and the Department. To that extent, we are involved in the level of control activity in ensuring it is adequate and given a high priority by community welfare officers as well as by the Department.

Deputy Dennehy

There was an argument — the debate is ongoing — that the rent subsidy scheme should come within the local authority regime. Has there been any debate on bringing CWOs under the aegis of Mr. Hynes's Department? I am not being mischievous and believe the rent subsidy scheme is being administered well. However, I have argued against it being subsumed into the local authorities.

Mr. Hynes

That issue has cropped up a number of times over the years. Community welfare officers operate schemes closely linked to social welfare schemes in that much of their work involves dealing with people awaiting social welfare payments. They also deal with rent payments, in respect of which there is a linkage with local authorities. They straddle an area with links to a number of agencies.

On the question of where community welfare officers should be located, this matter has always been on the agenda to some extent. Operationally, we try to keep as close a link as possible with the work they do. On whom they report to, this is a secondary issue as far as the administration is concerned. A value for money report compiled a number of years ago on the supplementary welfare allowance scheme raised such issues. While on the agenda, they are not being actively pursued.

Deputy Dennehy

I believe there should be a consistency of approach when dealing with clients. Accountability is also very important. Much of what we are dealing with is history, from which we are learning.

I am pleased Mr. Hynes has welcomed the report which coincides with the internal review. In responding to the risks posed, the committee would like to help him in any way possible. The argument is made from time to time, for example, that social welfare fraud is treated more seriously than white collar crime. Does Mr. Hynes agree that times have changed in that respect?

Mr. Hynes

I would not like to comment. I can only speak on the question of social welfare fraud which the Department has always treated seriously and will continue to do so. I would prefer not to comment on other types of fraud.

Deputy Dennehy

Is Mr. Hynes happy with his response to the risks as outlined? Is he taking other steps to control and monitor the system?

Mr. Hynes

The risk management approach recommended in the report offers us opportunities to be more focused on what we do and target schemes better. In the past, when circumstances changed due to the arrival of another child, for example, a review of the whole claim was triggered but this may not be the most effective way to deal with the matter. We should probably adopt a more focused approach to identify the types of claims which offer most scope for possible fraud. Where we are working with finite resources, we must ensure we use them in the best way possible. That is the main priority.

On data-matching and exchange, there is a recent project to computerise the General Register Office to allow a better flow of information on births, marriages and deaths. If we get information more quickly, we can take whatever measures are necessary to change payments, which is very important. The development of IT systems and more efficient technology will——

Deputy Dennehy

The Southern Health Board has being fully computerised, an issue on which we had a long debate with medical representatives recently. It was discovered that payments were being made in respect of people who were dead. It was a genuine error. There are examples where IT systems have not been used. Such systems have massive potential. Some argue that Big Brother is watching but in a range of issues, particularly deaths, there could be a trigger mechanism. Births are of greater significance. Is Mr. Hynes happy with the level of interaction and is there a need for expansion of IT systems?

Mr. Hynes

There is always a need for better IT systems. The huge benefit is that such systems enable us to provide a better service for customers. They also give us the opportunity to exert better control — an example is the GRO project. We can now pay child benefit much more quickly and at the same time have the information available to ensure the correct person receives payment. Good IT systems will enable us to combine a better service with a much improved control environment — the pay-off. There are a number of ongoing projects to improve our own systems. The REACH project launched recently will enable us to provide a much more integrated service for the public and exert much better control, using the PPS number as the key.

Deputy Dennehy

As a committee which seeks value for money, we should be concerned about the 19,000 fraud cases. However, I admire the work being done in the Department and welcome the fact that the automatic right to free travel and various other services is being notified, in respect of which I compliment the Department. I would like to see IT systems being used extensively. Perhaps Mr. Hynes will let us know if there are difficulties. I also thank the Comptroller and Auditor General for his excellent report.

Chairman

I have a question about the level of fraud which stands at 18% and the level of recovery in probate cases which stands at 68%. It is indicated in the report at page 41 that 12% of individuals who commit fraud reoffend at a later date. What is Mr. Hynes's comment in this regard?

Mr. Hynes

Obviously, it is a matter of concern. We will consider the issue when trying to direct or control activity. People who previously defrauded the system——

Chairman

Table 3.11 indicates that 11,804 persons were detected committing a second offence and that 2,232 committed a third offence.

Mr. Hynes

That is correct. This happens mostly in the case of persons moving in and out of short-term schemes. They include people who move from job to job. It may be the case that there will be an overpayment on a number of occasions.

Chairman

If a person reapplies for a welfare benefit, is it possible to recoup an unlawful payment?

Mr. Hynes

Absolutely.

Chairman

Can it be withheld from a future benefit?

Mr. Hynes

Yes. One of the ways we recover an overpayment is by making deductions from future entitlements. In fact, that is the main way we recover overpayments. We also follow up on those who do not come back into the system and try to recover overpayments from them. However, when people come back into the system, we note the overpayment against them and deduct from their ongoing payment, subject to a code of practice.

Chairman

If a person did not have the cash and was in real need of benefit, it would be difficult to recover the overpayment. Would it be deemed a bad debt?

Mr. Hynes

Under the code of practice, we do not reduce a payment below the level of the supplementary welfare allowance to which the person would be entitled. There is a limit to the level of overpayment recovery one can have. That is a baseline.

Chairman

It appears that the probate procedure is the best method of recovery of overpayment of pensions.

Mr. Hynes

That is right. It is a very effective method of recovery because the law requires us to be informed of the assets of people who have died. We can institute recovery action on foot of this.

Deputy Connaughton

I welcome Mr. Hynes and his management team. The role of the Committee of Public Accounts is, strictly speaking, to examine the value for money aspect of spending. Nevertheless, I mention the following matter. In his initial statement Mr. Hynes referred to the important principle of trying to pay the right person the right money at the right time. That is reasonable and fair. Over the years I have found the Department always aspires to this ideal, even if it does not always achieve it. Compassion and understanding in making payments are also important. For a person who must depend on a social welfare payment the clouds can be very low. On the other hand, the system must incorporate caution regarding those who wilfully break the law. This committee is interested in that aspect of the social welfare system.

There are 34 social welfare schemes. Is there a different computer system for each of those 34 schemes? Do the Department's computers speak to each other?

Mr. Hynes

We have a number of large computer systems. Our short-term schemes such as unemployment and disability payments operate on an integrated short-term system, ISTS. Pensions are operated on an old system. Apart from being able to make payments, it does not have the facilities of the short-term benefits system. That is one of its disadvantages.

Deputy Connaughton

Why does it not have those facilities?

Mr. Hynes

We are in the process of introducing a new system, the service delivery model. We have introduced it into the child benefit system where it has enabled us to have much better linkages with the General Register Office. It is a much better system for staff and the public. We are extending it into the pensions area.

Computer systems are very expensive and it takes a long time to change them. When we put in the pensions system years ago, we did not think it would be in use for as long as it has been. To change a system investment is needed and people must be available within the Department who are capable of managing a transition to a new system. It is not simply a matter of bringing in a new IT system. Work systems and organisational arrangements must also be changed. It is both difficult and complex.

Deputy Connaughton

Every Thursday, irrespective of what Department comes to this committee, we hear of trouble with IT. The reasons given are usually similar to those Mr. Hynes has outlined. Many of the greatest mix-ups in Departments have arisen because of IT difficulties. Do the computers in the Department of Social and Family Affairs speak to the computers in the Office of the Revenue Commissioners?

Mr. Hynes

We have a central records system which holds the insurance records of everyone on the social welfare system. It is the central repository of information. I am not sure if it is fair to say they are linked but we receive information from the Revenue computers which can be matched with information on our systems.

Deputy Connaughton

That is very good for the people to know. The question is often asked as to whether there is a throughflow of information between the computers in the Department of Social and Family Affairs and those in the Office of the Revenue Commissioners. Can the Revenue Commissioners gain access to the file of a social welfare recipient?

Mr. Hynes

That is subject to the requirements of data protection legislation. We have access to information from Revenue which is necessary for the purposes of controlling social welfare payments. That is the general principle. We can get information on the earnings of people also paying PRSI and on people taking up employment.

Deputy Connaughton

What about attachment orders and such matters?

Mr. Hynes

We do not operate attachment orders.

Deputy Connaughton

Is there a system of attachment orders against earnings within the social welfare code?

Mr. Hynes

No, not at present.

Deputy Connaughton

Is that something the Department would like to see?

Mr. Hynes

It is a possibility. There are considerable legal issues surrounding the introduction of such a system. It is not in place at present. If it was to be introduced, we would have to consider the legal implications very seriously.

Deputy Connaughton

Does Mr. Hynes consider the question of ID cards to be as important as many seem to think? I know the Department provides for some degree of central registration but I assume it is not all-inclusive.

Mr. Hynes

There is a card which we use for certain purposes. A variety of cards are in operation throughout the public service for different schemes and purposes.

Deputy Connaughton

There is a plethora of them.

Mr. Hynes

One could envisage having one card to cover everything. It would not have to be an ID or national identity card.

Deputy Connaughton

What would Mr. Hynes suggest that it might be?

Mr. Hynes

It would be a public service card, for the purposes of accessing public services. The question of a national ID card, is a different area.

Deputy Connaughton

Would Mr. Hynes like to see that type of system being adopted?

Mr. Hynes

I suppose it would be an advantage if we had a public service-wide card framework in place which would enable people to gain access to different services using the card.

Deputy Connaughton

The traffic flow in the opposite direction is that the Department would then be able to find out who they were and where they were?

Mr. Hynes

The only value of the card is to enable people to get access to a service. The information on the person would be held in a centre of information, a repository of some kind. Different Departments have records. We have a central records system; other Departments hold information on their clients. Whether there could be one information repository for all the information is another issue. Under the Reach project, there are proposals that people would be able to deposit information that they wished to deposit with the central public service broker and that information could be used to access a variety of different services through the broker. It would be up to people themselves to decide——

Deputy Connaughton

It would be voluntary?

Mr. Hynes

It would be a voluntary thing. The Department's main interest is in providing opportunities for people to get their entitlements quickly and efficiently and to get them from different Departments quickly without having to fill in application forms for different purposes.

Deputy Connaughton

Although not specifically in the witness's Department, one must acknowledge that because of all the moving parts in social welfare, it is a huge logistical matter to try to do what the Department does. In fairness, I think it works very well most of the time, except that there will always be black spots.

The witness referred to my colleague, Deputy Dennehy, a few minutes ago. I remember being in another forum where the medical people very specifically told us that they genuinely do not know either the names or the addresses of at least 20% of the babies born here. They do not have the information on a simple register. This matter was raised in the context of the vaccination processes. If that is the case, are we not a million miles away from having a computer-based system for information on who is entitled to the children's allowance, for instance, and the cut-off date for the allowance, all that kind of information?

Mr. Hynes

We have made considerable progress on the General Register Office, the register of births, marriages and deaths. The information coming through that office is comprehensive as far as the Department is concerned. We have access to that information and we have very good information on——

Deputy Connaughton

Obviously there must be huge gaps. The medical people say that because of people relocating and with the movement of families, they find it very difficult to pinpoint where they are and when a child was born. That obviously does or does not feed into the Department's system.

Mr. Hynes

One of the difficulties is keeping up to date the addresses of people who move. We spend many resources in cleaning up, if you like, the data on our systems and making sure that it is up to date. If somebody applies for a payment from a different address, we then check whether that is their new address and we update our records accordingly. We never have a perfect system but our system is much better than it used to be. We have arrangements in place to update our information as new information becomes available.

Every child born has a public service number given at birth. They have a record from that date. To that extent, we have a much better system of information about people from birth. Obviously when people change address the information must be updated and that is a challenge we must face. The Department is pretty good at doing that and when people make claims, we link back to their record and if there is a change to the information we already have, we check it out and see whether the record needs to be amended and that is done if necessary.

Deputy Connaughton

To return to the 39,000 cases of overpayment in 2002, of which I understand 48% would be in the area of fraud, when does an overpayment turn out to be a fraud? How is that signalled?

Mr. Hynes

The official has to determine whether the overpayment was incurred by deliberate fraud, whether the person knowingly withheld information from the Department or was working while claiming. The officer must decide whether that is to be treated as a fraud overpayment or a non-fraud overpayment. Fraud means deliberately giving false or misleading information or failing to disclose information deliberately on a claim form or at an interview with an inspector. It is reasonably clear-cut. Obviously in some cases, people incur overpayments without deliberately intending to defraud the system.

Deputy Connaughton

I understand the value of those 39,000 overpayments was €29 million.

Mr. Hynes

That is correct for 2002.

Deputy Connaughton

My figure would suggest that it costs €18 million to collect or identify the overpayments. Is that a correct figure?

Mr. Hynes

The sum of €18 million is the cost of the Department's investigation staff. They carry out many other activities apart from the recovery of overpayments. They carry out reviews of entitlements.

Deputy Connaughton

What was the cost incurred in identifying the €29 million overpayment? Did it cost €18 million? I will deal with the figure of €283 million in a moment. On a superficial examination, I reckon that there was a net benefit to the State of €11 million, that is, if it cost €18 million to identify the fraud and the overpayments.

Mr. Hynes

The sum of €18 million is the cost of the staff that investigate and work in the control areas of the Department. Their job is to examine and detect fraud where it occurs but they also have a prevention role in that they investigate claims. They also have an awareness-raising role. It is not just related to the detection of fraud after it happens. Most of our effort goes into making sure that fraud does not happen in the first place.

Deputy Connaughton

If there are 39,000 cases of overpayment in 2002, was that a random sample? If the Department had the resources, which it never will have, to examine all the payments, I assume the figure would be far higher than 39,000?

Mr. Hynes

The figure of 39,000 is the number of overpayments that were detected.

Deputy Connaughton

Is it reasonable to assume that there were far more than 39,000 overpayment or fraud cases? This goes back to the baseline proposals of the Comptroller and Auditor General. Does Mr. Hynes' long experience of the Department and other ministries around the world suggest there is a system through which to link the figures to what the ultimate total overpayments might have been? While we pride ourselves on the ability of our security services to apprehend drugs smugglers, one can be sure that for every person who is caught four or five get through. This situation might be mirrored in terms of the 39,000 social welfare cases. Does Mr. Hynes or his team consider they might have discovered that had they been able to go through them line by line?

Mr. Hynes

I will not speculate. It is very difficult to estimate fraud levels in social welfare or any other area. Any estimate one came up with would be purely speculative. The surveys carried out by the Comptroller and Auditor General and the Department in the course of the review indicated the level of incorrect payments. The incorrect payments were not attributed to fraud.

Deputy Connaughton

Were they overpayments?

Mr. Hynes

They were overpayments which were made for one reason or another.

Deputy Connaughton

Were they as high as 25% in some cases?

Mr. Hynes

There was no indication that the overpayments were due to fraud in those cases. overpayments might have been made in respect of schemes where people combining earnings with payments changed jobs or experienced income increases. A person's circumstances can change very quickly. The level of overpayment might have been quite small and resulted where a person took up a few days extra work resulting in a change in his or her circumstances. It would be totally wrong to attribute such overpayments to fraud.

Deputy Connaughton

If Mr. Hynes were to adopt the Comptroller and Auditor General's baseline on evaluations, would he be in a position someday to tell the committee how many cases of overpayment were the result of fraud? Is that the direction in which the Department is heading?

Mr. Hynes

The idea of the survey is to obtain better information about the level of incorrect payments in different schemes to enable us to target our activity at schemes in which the level of incorrect payment is highest. It will be very difficult to identify fraud from a survey as we have seen from exercises carried out in other countries.

Deputy Connaughton

I direct Mr. Hynes's attention to the €283 million which the Department has stated has been saved. If find it very difficult to get my head around this one. I have read the Comptroller and Auditor General's report and listened carefully to Mr. Hynes. I assume the Department uses some mechanism to compare one year to another. By comparing numbers, the Department could assess whether it is becoming more successful. If the €283 million which was saved is related to the 39,000 cases to which I referred, it appears the sum is entirely notional. For all I know, it may have been saved. One hopes it was. However, I am not convinced the figure is not, for example, €83 million or €183 million.

Mr. Hynes

The €283 million is a different type of figure. It is established by considering the following. If someone's claim is stopped we assume, if it is an unemployment claim, that the person will remain off the system leading to a saving over 32 weeks. This is the number of weeks we assume a person would have remained on the system if the claim had not been stopped. That is how the overall savings figure is derived. It is an estimate of the value of the activity. In the case of other payments, we have other multipliers, as we call them, which we use to work out the amount of savings resulting.

Deputy Connaughton

Is there any evidence that hard line crime is involved in social welfare fraud? Are there any crime gangs involved or is the system being manipulated in any way? Given the big, bad world in which we live, one would not be surprised at what one might hear. Does Mr. Hynes have any evidence that this might be a feature of the social welfare system at a very low level?

Mr. Hynes

There is no evidence of that at the moment. We have come across cases of identity fraud particularly in recent years where it was suspected that an organised effort was made to establish false identity for the purpose of claiming entitlements. We have put a great deal of effort into ensuring that staff members responsible for allocating personal public service numbers are aware of the types of data and false information which can be provided. We try to keep as up to date as we can on that front in co-operation and consultation with the Garda.

Deputy Connaugton

Does the Department have special investigators?

Mr. Hynes

We have a special unit centrally which is specifically examining the issue of identity fraud. It works in close co-operation with the Garda to ensure that as far as possible we are up to date. Ultimately, this is a matter of concern to the Garda also. We try to ensure our staff are fully briefed of the possibilities in this area.

Deputy Connaughton

I thank Mr. Hynes. Most public representatives will agree that his Department has provided them with a fine level of service on a personal basis down the years. We hope that will continue to be the case.

Deputy Boyle

I am conscious that there will be a division at 1 p.m. in the Dáil. As that is likely to entail a series of votes, I will be as quick as I can. From 1998 to 2002, the average overpayment was €779. Early in the report is a reference to the fines levied on foot of fraud, but I cannot tell to what the figure €455 refers. As a proportion of €779, a sum of €455 would be quite high. It would be well over 50% of the original sum.

Mr. Hynes

I ask the Deputy to repeat the question.

Deputy Boyle

The figures are in different categories and there may not be a direct relationship between them. The average overpayment, which includes fraud——

Chairman

What page is this on?

Deputy Boyle

The table is figure 2.6 on page 22. The average overpayment over five years was €779. In the summary of the findings, which refers to the fines levied specifically on fraud cases, it is stated that the average fine was €455. I cannot see a correlation between what the €455 refers to, the number of fraud cases and the amount. As a proportion of the average overpayment, €455 seems quite high.

Mr. Hynes

The second figure refers to the fines imposed where cases were prosecuted. Not every case of overpayment results in prosecution.

Deputy Boyle

We do not have a corresponding figure. Is that the difficulty? I am trying to find out the extent to which the sum of €455 relates to the amount on which the prosecution was based. What proportion is it? Does it relate to the fines being levied by the Revenue Commissioners, for instance, for the non-payment of income tax and the like? Perhaps that information can be supplied subsequently.

Mr. Hynes

I will certainly try to do so.

Mr. Purcell

It is stated on page 39 of the report — if I am picking up the point correctly — that the average fine imposed between 1998 and 2002 was €371 and €455 in 2002. Also shown are the results of court cases over that period. In 609 cases penalties were imposed. Of these, fines were imposed in 315 cases. There were also 23 cases where fines and suspended sentences were imposed. The amount of the fine arises only as regards that cadre of people who have been prosecuted for fraud. There is no real correlation between the figures mentioned in Table 2.6 and those in——

Deputy Boyle

They were identified frauds on overpayments relating to higher amounts, I presume.

Mr. Purcell

Yes.

Mr. Hynes

In general, we would prosecute——

Deputy Boyle

Is there any measurement as regards underpayments or is that treated as a time-specified circumstance within the Department?

Mr. Hynes

It is an issue. We would be concerned that people get their full entitlements. The phrase we used earlier was that we had tried to pay the right amount to people at the right time. We try to ensure, through our information services and so on, that people are aware of their entitlements. Staff try to ensure people are made aware of other entitlements they might have when they claim social security.

Deputy Boyle

How is it recorded in the Department's systems? Is it the time delay between a valid application and the time it is received?

Mr. Hynes

No, normally people are paid from the date of entitlement. Even if there are delays in the lodging of claims, they will still be paid from the dates of entitlement.

Deputy Boyle

If there is a time delay, is that the only payment that is backdated? Is there not a reverse consideration in terms of the fine and the fraud? Is there a payment in respect of the value of the money and the time delay?

Mr. Hynes

In certain cases, for example, where a person claims a pension and there has been a long delay in payment, which is the fault of the Department, it is possible that compensation will be paid.

Deputy Boyle

My final series of questions relates to information technology. When the Department makes policy decisions as regards qualifying criteria in respect of entitlements — as was the case last November when a series of decisions were made — does it also involve the rewriting of computer programmes that reflect how people are entitled or otherwise?

Mr. Hynes

It depends on the nature of the case but in many instances the IT systems need to be altered to cater for the changes.

Deputy Boyle

Would it have involved significant expense to alter the computer programmes to cater for the changed entitlement qualifications?

Mr. Hynes

It would not have been significant. It depends on the size or nature of the changes being made. In many instances this can readily be done and does not involve significant cost.

Deputy Boyle

On the reverse side — the decision to change one of those last night — what was the impact in terms of the potential cost of rewriting programmes?

Mr. Hynes

It will have little impact on our computer systems.

Deputy Boyle

In view of that decision, where does the principle of "only one social welfare payment" now stand within the social welfare system?

Mr. Hynes

I cannot comment on that. That is definitely a policy matter.

Deputy B. O’Keeffe

I have a number of questions but in view of the time I am willing to postpone them. As the Secretary General will come before the committee again on a number of other issues, I am quite willing to allow the Chairman to conclude the meeting.

Chairman

I thank Mr. Hynes and his team for coming today. Does the Comptroller and Auditor General have any comments?

Mr. Purcell

There are a couple of items arising from earlier questioning, one of which relates to the multiplier. As the report states, I have my doubts about its validity as an indicator. It has a value if used in consistent terms from one year to the next but if any indicator is being used in that line, it might be better to take the actual numbers rather than the amount using a multiplier.

Deputy Connaughton talked about trying to establish fraud levels through surveys. I agree with the Accounting Officer that this would be difficult to do. I recollect in the distant past that the Department commissioned PricewaterhouseCoopers to carry out some kind of survey which was not altogether satisfactory.

The other matter related to correspondence as regards supplementary welfare allowance.

Chairman

Is it agreed that we should dispose of the report? Agreed.

Mr. Purcell

Supplementary welfare allowance is an area in which, anecdotally at least, there is a high risk of overpayment. I note the Department has earmarked three areas for baseline survey work. It might consider supplementary welfare allowance as one of the areas in which that work could be carried out as it might prove worthwhile. It would be instructive in the light of general concern among members of the committee, and beyond, about the operation of the scheme. By its nature, a necessary level of discretion is involved. However, a control problem is created commensurate with that level of discretion.

The Department was supportive in this work. I hope we can continue to work with it in developing a risk approach to its control activity.

Chairman

I concur with what the Comptroller and Auditor General has said. The level of co-operation has been evident from the start of this meeting, which the committee much appreciates. The risk assessment and methodology on how this may be benchmarked are welcome as is the total co-operation of the Department vis-à-vis the report. This is good news. I congratulate the Department on the outstanding work it is doing. Its level of commitment to international best practice must be recognised. I look forward to its representatives coming before the committee again when we can look at the new methodology for effectively indicating base levels.

I thank everybody for attending. A summit is agreed for Thursday, 22 April. At our next meeting we will consider the following: Vote 1 — President's Establishment; Vote 6 — Office of the Minister for Finance; and Vote 7 — Superannuation and Retired Allowances.

I would appreciate if Mr. Hynes supplied me with a note on the level of payment on pre-1953 pensions.

The witness withdrew.

The committee adjourned at 1.10 p.m. until11 a.m. on Thursday, 22 April 2004.