COMMITTEE OF PUBLIC ACCOUNTS debate -
Thursday, 8 May 2003

Vol. 1 No. 17

2001 Annual Report of the Comptroller and Auditor General and Appropriation Accounts.

Vote 5 - Central Statistics Office.

Mr. D. Garvey (Director General, Central Statistics Office) called and examined.

Chairman

Witnesses should be made aware that they do not enjoy absolute privilege and advised 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 introduction of documents, and the right to cross-examine witnesses. While most of these rights may be exercised only with the consent of the committee, persons invited before it are made aware of these rights and any persons identified during the course of proceedings who are not present must 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 the 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 provisions within 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 or objectives of such a policy or policies.

I welcome Mr. Donal Garvey and ask him to introduce his officials.

Mr. Donal Garvey

I am accompanied by Mr. Gerry O'Hanlon, director with responsibility for the census of population and demographic statistics; Mr. Bill Keating, director with responsibility for macro-economic statistics, and Mr. Joe Treacy, director with responsibility for business statistics and general administration in the office.

Chairman

Will the official of the Department of Finance please introduce himself?

Mr. David Hurley

I am a principal officer in the organisation, management and training division. I deal with the administrative budget.

Chairman

Paragraph 3.1of the report of the Comptroller and Auditor General, reads:

Census of Population

Background

The first census of population was taken in 1841 and since 1951 has been repeated every 5 years, with the exception of 1976, when it was cancelled because of budgetary considerations.

In March 1999, the Government decided that a Census should take place on 29 April 2001, and that, as for previous censuses, it would be carried out by the Central Statistics Office (CSO).

Due to the threat posed by Foot and Mouth Disease, the Government, on 27 March 2001, postponed the census, and subsequently rescheduled it for 28 April 2002.

Audit Objectives and Scope

The objectives of the audit were to:

· Ascertain and evaluate the actions taken and the procedures put in place to provide for the taking of the 2001 census of population.

· Examine and evaluate the specific procedures followed in placing contracts for goods and services, in reacting to unforeseen events and in protecting the States interest.

· Ascertain the cost of the project and evaluate the efficacy of the financial controls in operation.

The audit consisted of a review of the structures and procedures employed to manage the project. Files of the CSO and the Department of Finance were examined including Government Decisions and Memoranda, minutes of Consultative Groups, Evaluation Board and Project Board meetings, legal agreements and correspondence. Discussions were held with CSO staff and site visits were made to the Processing Unit at Swords in Dublin.

Audit Findings

Planning of Project

The following key tasks were involved in carrying out the census:

· Planning

· Conducting a pilot study

· Securing professional advice on the technology to be used

· Selecting a supplier to provide computer hardware, software and related printing services

· Raising public awareness

· Recruiting additional staff

· Distributing and collecting census forms

· Processing analysis of completed forms

· Publishing results

The previous census held in 1996 involving 1.1 million households cost €19.5m, according to CSO estimates. The information contained in the 1996 census forms was captured on computer using keyboard data entry. The processing of the forms, including keying, coding and editing, took approximately 480 person years. The Census 2001 workload was expected to increase by about 20% because of a rise, also of that order, in the number of households and because of additional questions to be included in the census form. A rise in wage and field costs was also predicted taking account of general wage inflation and the impact of proposed minimum wage legislation and a tighter labour supply situation. When account was taken of these factors it was estimated that, using 1996 methodology, costs would rise to €31.7m.

At this time, the computer system in use within CSO consisted of a VAX network. The Department of Finance strongly advised CSO that because of EU procurement regulations, future hardware acquisitions should conform to a PC network architecture. Given this constraint, it was decided to invest in new computer hardware and software and avail of advances in image technology, which would enable the data in the census forms to be automatically read rather than having to be manually keyed into the computer. An analysis carried out by the CSO indicated that the cost of this new technology could be offset by reductions in labour and other direct costs of about €8.3m. The new technology would also speed up the processing of the census forms so that all relevant statistical reports would be published by April 2004, a gain of 8 months The final CSO Reports relating to the April 1996 census were released in December 1998; it is expected that those relating to the April 2002 census will be released in April 2004.on the Census 1996 timetable. It was also expected that the new technology could be used, to a certain extent, on ongoing CSO projects and in processing the returns relating to the next census due in 2006.

The National Statistics Board, which comprises members drawn from public, commercial and academic bodies, has as its main function the guidance of the strategic direction of the CSO, including the setting of priorities for the compilation of official statistics. In its Strategy for Statistics 1998-2002, the Board advocated maximum consultation with users and application of modern technology in carrying out the 2001 Census of Population. Acting on this recommendation, a Consultative Group was formed representing Government Departments, local authorities, the Social Partners and the transport, planning, academic and research sectors. In consultation with the Group, CSO conducted a pilot study in September 1999 involving 8,900 households to assess public reaction to possible new questions and amendments to existing questions, and to evaluate revised layouts for the census form and alternative collection procedures. Public comment was also invited through advertising in the national press and on the Internet. Using the results, that Group assisted CSO in drafting the questions to be included in the census form. The final questionnaire was approved by Government on 30 May 2000.

For administrative purposes, CSO divided the country into some 4,000 enumeration areas, each containing approximately 350 households. Each of these areas had been mapped by CSO. Existing records were updated by the inclusion of recent known household developments and individual developments noted by enumerators. As a result, CSO were confident of achieving total coverage in its distribution of census forms to all 1.35 million recorded households. After Census Day, forms from all households were collected by enumerators, with priority being given to establishments where the population was likely to be mobile. Approximately 1.3 million forms were returned, the difference being explained by second homes, unoccupied buildings, spoiled forms etc. Regular progress reports on the project were submitted to and considered by a Senior Management Committee within the CSO.

Acquisition of Computer System and Printing

Appointment of Professional Adviser

In November 1999, the CSO requested Government Contracts Committee (GCC) approval to engage a consultant to provide specialist technical advice in managing the tendering process of engaging a contractor to provide the computerised scanning and recognition system. The timetable involved in delivering the solution had, in the opinion of CSO, become extremely tight due mainly to delays in securing union agreement. In these circumstances, which CSO regarded as being exceptional, permission was sought to dispense with normal competitive tendering procedures and to engage a named consultant. That consultant had assisted the UK Office for National Statistics in its tendering procedures to outsource the processing of its census and would, it was felt, be in a unique position to immediately provide an excellent service. CSO was also of the opinion that because of the uniqueness of the skill-set required, it was unlikely that a competitive process would result in very many candidates. GCC approval was given on 30 November 1999 for the engagement, the cost not to exceed the CSO estimate of €31,743 plus VAT.

CSO decided to expand the original role envisaged for the consultant and to also involve him in the detail of drawing up the legal contract and technical schedules for the procurement process. Following negotiations with the consultant, a cost of €158,717 plus VAT was agreed and the GCC gave its approval on 26 January 2000. CSO believes that the alternative would have involved acquiring substantial legal and technical advice from a number of outside sources at a considerably greater cost and, more importantly, could not have been obtained in sufficient time to allow the tendering process to go ahead within a tight timeframe.

The consultant assisted in streamlining the procurement, negotiation and award phase of the outsourcing contract by designing the tendering documentation, providing a tender evaluation model, acting as a facilitator during evaluation, providing advice during negotiations and designing the main contract. Actual expenditure on the consultancy totalled €177,763 plus VAT. This included additional expenditure necessitated by the re-negotiation of the contract because of the postponement of the census.

Selection of supplier

The tendering process leading to the selection of a contractor for the provision of the new computer system was handled by an Evaluation Team comprising CSO census support management with IT and financial backgrounds.

A Contract Notice was published in the Official Journal of the European Community in January 2000. Following a process that involved an Open Suppliers meeting, and the evaluation of Contract Notice Questionnaires, four of the forty-one companies which had expressed interest in the contract, were issued with a Statement of Service Requirements. Two companies responded and, following detailed evaluation and negotiations, both were invited to submit a Best and Final Offer. While the two companies were rated very closely in non-financial areas, one offer was significantly cheaper. The Evaluation Team recommendation that the lowest tenderer be awarded the contract was endorsed by the Director General and a contract was signed on 24 October 2000 with that company. The contract amounted to €8.2m, including VAT, and was payable in instalments up to and including 31 December 2001.

The evaluation model used by the Evaluation Team was comprehensive and detailed and was sent to the GCC, and the procedures followed by the Team were reviewed for adequacy by a Director of CSO. Site visits were undertaken by Team members to inspect systems in other organisations provided by the two competing companies.

The contract covered an 8-year time span and was designed to give CSO flexibility in relation to carrying out the 2006 census.

Postponement and rescheduling of census from April 2001 to April 2002

Following the postponement and rescheduling of the census, the additional and amended services required for Census 2002 were negotiated by way of a Change Request procedure included in the original contract. Legal advice received indicated that this was permissible and that there was no obligation on CSO to re-open the tendering process. The revised substitute contract was signed in January 2002, and together with the cost of re-printing the census forms, amounted to €10.7m including VAT. Approximately €1.27million of the increase in contract cost relates to re-printing the census forms. The remainder is accounted for by additional work in the areas of system specification, technical design, and additional maintenance.

After the contract had been signed, a Project Board comprising representatives of CSO and the successful bidder was formed to oversee the implementation of the contract. The Board, which is still in existence, met regularly, and kept minutes of its meetings and decisions. It considered progress reports and agreed action to resolve problems encountered at the different stages of delivery of the project. Legal advice to the Evaluation Team and the Project Board was provided by the Attorney Generals Office.

System Performance

The new computer system was signed off for customer testing on 24 April 2001, some two months later than the scheduled date. With the postponement of the census, both CSO and the successful tenderer agreed that the priority should be on the optimal development of the system rather than on meeting the strict deadlines originally set. Correctly and incorrectly completed forms were fed into the system to determine the standard of recognition it carried. Once errors identified in the system had been logged, their source was established and corrective action was taken.

A number of system errors remained unresolved in early 2002. Most of these were considered by CSO to be of a minor technical nature and not to seriously compromise the overall functionality of the system, and were subsequently rectified. Some problems were encountered with scanning and other equipment, resulting in the replacement, under warranty, of all PC motherboards. Also, in June 2002, in a live processing environment, pressure placed on the network due to the volume of batches caused system errors, which, if uncorrected, could have lead to a loss of processing time. The ability of the system to handle such conditions could not have been tested prior to live processing as such a volume of completed forms was not available to CSO. The system was cleared for customer testing on 28 February 2002 and went live on 27 June 2002. The target performance level is 800 batches (12,000 forms) per day. Since the commencement of processing in early June 2002, there has been a steady improvement in the number of batches being processed. At end-June 2002, a level of 500 batches per day had been achieved.

While CSO were at all stages confident that the desired system was technically feasible, it did consider that a risk existed in relation to the project over-running the exacting timetable set. CSO reservations regarding system performance have been notified to the company under the terms of the contract, which entitles CSO to withhold payment in respect of a non-conforming service. In addition, it provides for the retention, by CSO, of €520,593 pending completion of the project to its satisfaction.

Printing of Census Forms

Under the terms of the original contract with the company, CSO were to be supplied with 1.8million census forms at a cost of approximately €1.2m. The company sub-contracted the printing of the forms. However, based on a quality sampling exercise, approximately 55,000 of the forms were deemed to be of unacceptable quality. It was agreed that the forms would be reprinted. However, inspection of those reprinted revealed further deficiencies. On 22 February 2001, CSO agreed to accept the defective forms in return for a reduction in the contract price of €29,204, together with a provision that a further deduction might be made in the event of reasonable additional costs being incurred by CSO as result of the deficiencies.

The situation was then overtaken by the postponement of Census 2001, rendering all the forms useless. They are currently stored by the company pending the outcome of legal proceedings and a decision as to their disposal . While CSO considered the position regarding payment for these 2001 forms, it soon had to make arrangements for the printing of Census 2002 forms. This was effected through the use of a contract Change Request mechanism. On 9 August 2001, the company were contracted to print the Census 2002 forms at a cost of approximately €1.27m. This time, the company, at CSO's request, introduced a Quality Control Procedure in respect of the printing of the forms.

Legal proceedings, not involving CSO, are currently under way in the UK regarding payment owing by the company to sub-contractors in respect of the printing of the 2001 forms.

Under the terms of the revised contract signed with the company on 7 January 2002, CSO retained €190,460, pending settlement of its claim regarding the 2001 forms.

Staffing

Department of Finance sanction was received for the recruitment of an additional 178 headquarters staff to assist in the mapping of enumerator areas, the fitting out of regional offices, the recruitment and payment of field staff, logistical support and the processing of returned census forms. In accordance with previous practice these staff would be recruited as permanent civil servants and following completion of their census duties would be redeployed to other Departments in the Civil Service. Of this number, 110 had been recruited by March 2001.

In the period April-December 2001, 19 were re-deployed to other Government Departments while the remaining 91 were retained pending the commencement of the Census 2002 project. CSO informed my staff that during this time they were engaged in census preparatory work such as, decommissioning all 2001 materials and preparing revised 2002 materials, investigating new housing, standing down and re-recruiting field staff, and developing educational material and the census website. They also undertook work on behalf of other divisions within CSO, primarily relating to the Census of Agriculture, Balance of Payments, the CSO business register and Quarterly National Household Survey mapping. In the normal course of events, it would have been necessary for CSO to carry out this work through outsourcing or through the use of overtime. In addition, some staff continued to be involved in training and additional system testing. From January 2002, CSO recommenced recruitment to achieve the sanctioned level.

The costs incurred in paying staff from the time of the postponement decision until end-2001 is estimated by the CSO to be about €1.2m.

In the fourth quarter of 2000 and in early 2001, 412 supervisory field staff were recruited on short-term contract to supervise the delivery and collection of the 2001 census forms. However, due to the postponement of the census, their contracts were terminated at the end of April 2001. Salary payments to these staff amounted to €2.7m.

An additional 3,974 were on the point of being hired on short-term contract as enumerators but had not taken up duty at the time the census was postponed in March 2001. No money was paid to these staff.

In early 2002, 4,386 field staff were again recruited on short-term contract to supervise and carry out the delivery and collection of the 2002 census forms.

Recruitment of supervisory field staff and enumerators was administered by CSO personnel. Security checks were carried out on all staff, who, on appointment, became Officers of Statistics and were required to sign Official Secrecy undertakings. Training was provided to all grades and a job manual issued to each staff member as detailed in Table 2.

Table 2 - Census Staff in Place

Quarters	1999		2000		2001		2002
	FieldStaff No.	HQStaff No.	FieldStaff No.	HQStaff No.	FieldStaff No.	HQStaff No.	FieldStaff No.	HQStaff No.
Q1	-	42.5	-	52.5	412	141	412	183
Q2	-	45.5	-	63	-	135.5	4,386	226.5
Q3	-	47.5	-	80	-	116.5	-	219.5
Q4	-	45.5	40	113	-	122	-	-

The terms of employment and post-census arrangements relating to all staff were sanctioned by the Department of Finance in line with Government Decisions.

Miscellaneous Costs

Premises

At the request of CSO, the Office of Public Works (OPW) secured premises suitable for housing the Processing Unit and the attendant equipment. The building, comprising 3,462 sq metres at Swords has been rented by Office of Public Works on a 25-year lease at an annual rent of €634,869. The cost of fitting out the building, including electrical and telecommunications wiring was €1,373,857, and €322,513 was incurred on office furniture. The building was occupied from October 2000.

OPW also leased 20 regional offices on short-term leases in late 2000 and early 2001. Office of Public Works carried out security audits on all premises and where necessary, alterations were made to bring them up to an acceptable standard. On postponement of the 2001 census, 18 of the buildings were retained and the leases extended so that they could be used on the 2002 census. The buildings remained unused in the interim period. Two additional premises were leased in 2002 for the 2002 census. The rents paid in 2001 totalled €265,375. Rents arising in 2002 in respect of the 2002 census are projected to amount to €165,066.

Office furniture purchased by CSO in fitting out these premises cost €38,092.

Office equipment

A number of items of equipment were purchased or leased for use in the Processing Unit, in the regional offices and by field staff. These included a telephone system, 24 photocopying machines, 2 fax machines, 400 mobile phones, storage boxes and satchels. In all cases, CSO engaged in a tendering process in acquiring the equipment.

Following the postponement of Census 2001, lease terms on photocopiers obliged payment in full, although the machines were not used during the period of postponement. New tenders were sought in respect of photocopiers used during Census 2002. Mobile phones, which had been distributed to staff, were de-activated and stored at Swords until the commencement of Census 2002.

The cost of office equipment is estimated at €657,724.

Advertising

Fourteen companies responded to a Contract Notice placed in the Official Journal of the European Union on 15 September 2000, in respect of the provision of advertising services for the 2001 census. Following an evaluation process and consultations with senior administrative and field staff, the approval of the GCC was sought to engage one company. That approval was received in November 2000 and, the following month, a contract was signed in respect of recruitment and public awareness advertising campaigns.

Because of the difficult labour market situation pertaining in 2001 it was decided to use radio advertising on a wide scale basis for the recruitment of enumerators. Expenditure of €283,152 was incurred on the enumerator recruitment campaign in December 2000 and January 2001. The cost of the public awareness campaign up to the postponement date was €184,112.

The advertising agency was retained to carry out the awareness campaign in respect of Census 2002. This was effected by way of an amendment to the original contract and was in line with legal advice received by CSO. Payments in respect of this campaign are projected to cost €976,429. As a panel of potential candidates for enumerator positions had been established at the time of the proposed Census 2001, it was not necessary to carry out a recruitment campaign for Census 2002.

Project Cost

The total projected cost (excluding costs incurred by Office of Public Works) of the project is estimated at €43.87m, as shown in Table 3.

Table 3 - Project Cost

Area	Census 2002 €m	Postponement Costs €m	Total Costs €m	1996 Costs* €m
Salaries
HQ	5.65	1.63	7.28	8.51
Field Staff	14.46	2.14	16.60	7.46
Total	20.11	3.77	23.88	15.97
Travel	2.74	0.18	2.92	1.22
Computer	7.10	1.64	8.74	0.52
Office Equipment	1.09	-	1.09	0.22
Premises	0.69	0.36	1.05	0.19
Printing	1.23	1.28	2.51	0.32
Advertising	1.35	0.18	1.53	0.41
Consultancy	0.22	-	0.22	-
Postal & Telecommunication	0.22	0.10	0.32	0.13
Incidental	1.37	0.24	1.61	0.54
Total	36.12	7.75	43.87	19.52

*Includes costs of staff redeployed to other duties following postponement.

The actual amount paid up to 30 June 2002 was €28.98m.

Payment Procedures

All payments relating to the project were authorised by the Census Support Unit prior to processing and payment by the Finance Unit, where they were checked for correctness and proper authorisation. Invoices were tracked from receipt to payment. Capital assets were highlighted for inclusion in the fixed asset register.

Project expenditure under each budgetary heading was also monitored, on a monthly basis, and records were reconciled with Finance Unit figures.

Conclusions

· The overall management of the census by CSO was satisfactory.

· The postponement of the 2001 Census cost €7.75m. The procedures put in place to cope with the implications of the postponement were reasonable.

· CSO internally deployed 91 of the extra staff recruited for the postponed April 2001 census for the period April to December 2001. While the deployment ensured that the State got a measure of value for the pay costs of these staff, the very nature of the circumstances which gave rise to this unexpected resource windfall suggests that it is unlikely that an optimal return was obtained.

· CSO did not initially comprehensively scope the range of services to be provided by the professional adviser ^ a consequence being that there was not competitive tendering for a contract which was ultimately worth €177,763 plus VAT.

· CSO should continue to pursue an appropriate financial concession in respect of system performance and the defective census forms supplied.

Mr. Purcell

This paragraph centres around the fact that the census takes place on a regular basis every five years or so and entails considerable amounts of taxpayer's money. This prompted me to undertake an examination of the financial management of the 2002 census by the CSO. The committee will recall that this census was originally intended to take place in April 2001 but was postponed by the Government because of the foot and mouth disease outbreak. The decision to postpone the census was made just one month before its scheduled date. Therefore, it was inevitable that there would be a cost associated with the postponement. The report records the estimated cost of the postponement, mostly comprising salary costs of staff recruited for the census, computer related costs and, ultimately, the nugatory expenditure incurred in printing 2001 census forms. In all, the cost amounted to an estimated €7.75 million, although this figure does not take into account the benefit the CSO would have obtained by the temporary internal deployment of the surplus staff, including any consequential savings that may have accrued.

The overall cost of the census, including the cost of the postponement, worked out at approximately €44 million, well up on the €19.5 million cost of the 1996 census. The details are shown in Table 3 on page 10. Members will note the significant variations. One of the major contributory factors in the increase was the decision to invest in new computer hardware and software and avail of advances in image technology. Apart from the economies to be realised from reducing the data input effort, the new system was expected to enable the CSO to bring forward the dates of its final census reports by about eight months. I understand it is on schedule to meet this target.

The report concludes that the CSO managed the census in a satisfactory manner. I had a minor quibble about how it had engaged a consultant and his team to provide professional services in connection with the computerised scanning and recognition system. Effectively, it resulted in work to the value of €177,000 plus VAT being placed without competitive tendering. To be fair to the CSO, it obtained Government contracts committee approval for its actions and is convinced it got good value for money from the consultant. Allowing for unavoidable costs arising from the postponement of the census, the report gives positive assurance on the moneys expended.

Chairman

In November 1999 the CSO sought the approval of the Government contracts committee to engage a consultant to provide specialist technical advice in managing the tendering process of engaging a contractor to provide the computerised scanning and recognition system. Due to time constraints it took the exceptional step of engaging a named consultant to do this rather than follow normal competitive tendering procedures. At the time the cost was €31,743, for which Government approval was granted. In January 2000, however, the CSO decided to expand the original role with the result that the final cost of the services provided amounted to €177,000. Why did the CSO change its mind so soon after receiving initial Government contracts committee approval on 30 November 1999? Why did it not follow public service guidelines and put these extra services out to competitive tender?

To use "exceptional circumstances" as justification for dispensing with normal competitive tendering procedures on two occasions over a three month period is unacceptable. These procedures were established to ensure value for money for taxpayers and guarantee commercial and public confidence in the awarding of public contracts. What steps has the CSO taken to ensure it will not circumvent public contract procedures for the 2006 census? Has it applied to the Government contracts committee for dispensation from the procedures on other occasions?

Mr. Garvey

As Mr. Purcell said, the census is an extremely complex and demanding undertaking. The last census involved a considerable leap in terms of the use of technology, a scale not taken on previously. It resulted in a considerable reduction in the number of headquarters staff used to process census information and a considerable gain in the availability of information. We are on schedule to produce all the information on time.

The original intention in relation to the contractor was to seek advice on procurement issues. The technology road we travelled had been travelled by other countries and we were aware it was quite complex. When we engaged the consultant and looked more closely at what had happened in the United Kingdom and other countries, we realised that we needed to get involved in a complex negotiated procedure. At that point we realised we needed complex legal and technical advice. The consultant had been engaged to precisely provide this advice on a huge contract in the United Kingdom. If we had not taken him on, with the expertise he had, we would have had to go looking for both technical and legal advice, which, as far as we could establish, was not available to us. We followed an acceptable procedure. Single tendering is acceptable in certain circumstances. We followed this procedure and spoke to the Government contracts committee and got its approval for the process we followed. I am not sure whether my colleague, Mr. O'Hanlon, wants to add anything. He is the director directly involved in the census.

Mr. Gerry O’Hanlon

The one issue I would add is that the time available to us to deliver the contract was extremely short on this occasion. That was because we had to obtain staff agreement to bring in the new technology. That took longer than we had envisaged. Therefore, we had very little time available to us. One of the features of the census is that it is done on a particular date and one either has one's system available on that date or one does not. That is what made the situation so exceptional. The second issue was that the consultant, in many respects, had a unique competence in the technology, having advised on a similar contract in the United Kingdom. That, effectively, made his expertise, perhaps, unique in the circumstances. Those two circumstances, the time and the uniqueness of his expertise, made the circumstances exceptional.

Deputy B. O’Keeffe

I welcome Mr. Garvey and his team. I do not think there is anything major in the report of the Comptroller and Auditor General. The Central Statistics Office is getting a ticking off here and there but, generally, there is not a problem.

Does Mr. Garvey think the fact that the original census in 1996 cost approximately €19.52 million and that we are now up to €43.87 million represents good value for money? Is there an inordinate increase in the cost involved? Was it warranted? It seems excessive.

Mr. Garvey

There is a reference at the top of page 5 of the report of the Comptroller and Auditor General to estimates of repeating the 1996 exercise exactly in the same way. In 1999, when we went to Government, we had estimated that repeating the 1996 approach would have cost approximately €31.7 million at 1999 prices. Allowing for price changes between 1999 and census date, we roughly estimate the figure at between €35 million and €36 million. Table 3, to which Mr. Purcell referred, shows that, excluding the postponement costs which arose because of the foot and mouth disease crisis and over which we had no control, the new approach using technology delivered the census at roughly the same cost. There was additional expenditure on technology and reduced expenditure on headquarter staffing but the real gain is that the valuable information will be available eight months sooner. There are spatial strategies and all kinds of other policy needs. The CSO will be able to deliver very valuable census information.

Deputy B. O’Keeffe

When is it due?

Mr. Garvey

We have a schedule. The first report is due in the middle of June and the full set of census volumes will appear over the next 12 months. The full set of census volumes is expected to be completed in April next year. Correspondingly, in 1996 the final census volume was available in December rather than April. Principal results are available in the middle of June followed by 13 volumes. These deal with population; age; households; usual residence; migration information, which is always very critical because we are such an open country; industry; occupation; education; a special report on the Traveller community arising from a new question on the census; travel to work; disabilty and carers, which were also two new questions in the census; Irish language; and religion. The volume on religion will be available on 8 April 2004 and the final volume on 22 April 2004.

Deputy B. O’Keeffe

What about the areas of small population in which we might have a particular interest?

Mr. Garvey

Constituency information will be provided in special tables and released on 19 June. It is always of interest to politicians.

Deputy B. O’Keeffe

Let me ask two further questions. One relates to computerisation and IT, the bane of our lives because when anything goes wrong, we are told they are putting in a new system. The CSO installed a new system which was tried and tested following much preparation and soul searching. However, when it was called upon to act, it was not able to respond adequately. It was projected to be able to deal with 800 batches a day but the maximum ever achieved was 500. I see from my notes that "a risk always existed". How would we enter into a contract that cost more than €8.7 million for an IT system when a risk existed that it would not be able to do the job properly?

Mr. Garvey

There are a couple of issues. During processing the number of batches dealt with increased to about 700 a day. It started off rather low because the staff were untrained and their expertise had to be developed. During the exercise we reached a peak of about 700 batches a day, still below the 800 hoped for at one time. In setting out the targets for our activity our public undertaking was that we would process the information within one year of census date. We published this as our target. We tried to beat it but we reached the target. The full information was processed within one year of the census.

This was a complete leap of technology for us and it worked very well. Not every target was met but the throughput, more or less, met the target. The original target of 800 batches was a working one which stretched the boat out and at which we tried to aim. We did better than our target which we published in our statement of strategy by dealing with the processing in one year.

Mr. B. O’Keeffe

As Mr. Garvey is talking about withholding payment, he is obviously dissatisfied with the system.

Mr. Garvey

No.

Mr. O’Hanlon

Perhaps I could be more specific about what was involved. One makes assumptions when one is designing something in advance of delivery regarding both the input of staff and the capacity of the computer system to handle through-flow in an ideal situation. For various reasons, we often have difficulty recruiting staff in time and in ensuring we have enough people to do the job. We also have to ensure they are familiar with new technology. We took certain decisions regarding staff input and those we were taking on from a CSO point of view. Civil servants were working on it.

It was up to us to ensure, from a delivery point of view, that they built a computer system capable of delivering 800 per day if other parameters such as staff involvement demanded it. Indeed, at certain periods during the six months it was in operation the throughput of 800 a day was realised. On average, given the varying competencies of the staff and the circumstances in which it was required to operate, it came out at under 600. We did not view that as a failure on the part of the contractors to deliver their part of the bargain. As Mr. Garvey said, the ideal set in the joint partnership between the CSO and the contractors was to deliver a system which, if it delivered 800 a day would have allowed us process the census by end December of last year. In the event, we finished processing before the end of February. All things considered, we judged that to have been a very satisfactory outcome given that we were building something new.

Deputy B. O’Keeffe

So there was no reason to hold back payment for non-conforming service?

Mr. O’Hanlon

No.

Deputy B. O’Keeffe

What about the retention of moneys pending completion of the project in a satisfactory manner? Was that withheld? Is it still pending?

Mr. O’Hanlon

The payment schedule for the contractors entailed a retention in the event of there being a difficulty with the system whereby we would have some moneys which we could bargain over at the end. The only issue that involved the CSO seeking a refund related to the printing of the forms for the 2001 census. We were not satisfied with them. Following protracted negotiations we received a rebate of £100,000 from a contract of approximately £700,000. That matter was resolved to our mutual agreement some months ago.

Mr. Purcell

As I said earlier, the financial management of the census was well handled by the CSO. There were some minor quibbles but one would not be human if that were not so.

Deputy B. O’Keeffe

Recruitment of 178 staff began in March before postponement of the census. By the time of the postponement, 110 people had been recruited. Some 19 of them were redeployed and 91 were retained within the CSO, and an outline is given saying they were involved in various functions. What type of contract were they offered by March 2001? Is it usual to contract people before the census gets under way or is it normal practice?

Mr. Garvey

Let us look at what is involved in a census. We have to recruit in advance approximately 4,500 members of the public, people who know nothing about statistics, 4,000 of whom are enumerators. We have to divide the country into 4,000 enumeration areas, with pockets of approximately 350 or so households in rural areas and pockets in urban areas which are sometimes a little smaller in areas which are difficult to enumerate. We have to provide them with maps and train them in contacting every person and household in the country in a concentrated period of time. We also have to set up payment schedules to pay them. There is a great deal of preparatory work in advance of the census.

Following the census, we then have to process 1.4 million household forms or information for 3.9 million people. It is not just a matter of contracting in headquarters staff to do the census. There is a great deal of preparation involved such as mapping, training and delivery of forms to 4,000 enumerators around the country. As I said, we also have to pay these people. The work does not all happen at the time of the census but commences months in advance.

Recruiting people at that time was very difficult. We had scheduled to recruit more than the numbers we actually took on by the time the Government decided to cancel the census. We were then left with a dilemma. We knew the census effort would recommence later in the year. We took the decision to retain the people we had rather than risk not getting them again because we had experienced difficulty getting them for the 2001 census. Had it gone ahead we would have been 40 or 50 people below our target. We redeployed some of those people and we turned others to useful work within the office. Mr. Purcell said he wondered whether the resource windfall generated optimum value for the State. I would say it did not in the sense that we were left with a number of people without notice. We did not plan the cancellation of the census and did not plan having more than 100 people had it gone ahead. Nevertheless, we turned them to useful work in the office and progressed some activities that would not otherwise have been progressed. They were usefully engaged and provided value for money. The Comptroller and Auditor General's observation is probably fair.

Deputy B. O’Keeffe

My final question relates to the quality of the printing. It cost €1.3 million to have the census forms printed. Having been provided with the contract, the people involved subcontracted out the work. A proportion of the forms were not of good quality and had to be returned. Even then, the quality of others was not great so you then changed contractors.

One tests the value, efficiency, effectiveness and quality of the work provided by those to whom one awards a contract. Is there anything within those contracts which debars them from subcontracting to another company?

Mr. Garvey

The forms which were considered to be partially deficient were not used. They had been printed for the 2001 census which was later cancelled. As Mr. O'Hanlon said earlier, we recouped €100,000 on them even though we would have been disposing of them. We had to reprint forms with a new census date. There was one overall contract with a supplier of services who was involved in delivering certain things. Perhaps Mr. O'Hanlon will explain the details of that as he knows them better than I do.

Mr. O’Hanlon

There are two requirements of a census form - it must be acceptable to the public who have to fill it in and it must be legible and easy to follow through. That was not the problem we had with our census forms. On this occasion, because of bringing in the new technology and because of using scanning techniques, a sophisticated form was required from the point of view of design and colour - certain colours must drop out and not be recognised or scanned electronically. It was on the latter aspect that we ensured that the production of the forms was part of the overall processing contract for the census. We were advised that it happened in a number of cases where forms were printed independently of the people providing the scanning service and where there were defects in the throughput of the forms due to the printing that it was difficult to determine who had responsibility.

What we were looking for in this contract was assurance that the forms would meet the quality required for scanning. This is relatively new technology, as we found when we went looking, and it was our view that what was provided for 2001 did not fully meet the high quality required. It was that issue we pursued even though we could not prove it because we did not have the 2001 census to run through the machines. Eventually we got a rebate of around 12% or 13% of the cost price which I think was reasonable. We were so concerned about it in regard to 2002 that we persuaded the contractors to change the printer. They changed to a Dublin based printer who, I am glad to say, in the context of the 2002 census provided very satisfactory forms.

Deputy Connaughton

There is a vote. I have a few questions to put but I do not think we will have time for them before the vote.

Chairman

Should we suspend?

Deputy Connaughton

Yes. We have no alternative but to go for the vote or we will have a bigger problem.

Deputy B. O’Keeffe

The last time I did this my Chief Whip was not happy.

Chairman

We will suspend for 15 minutes.

Sitting suspended at 12.45 p.m. and resumed at 1.05 p.m.

Deputy Connaughton

I have a few questions which have not been answered already. I am delighted the officials from the Central Statistics Office are in attendance because I appreciate the importance of the work of the CSO. The census of population is very important for the economic future of the country. It is by listening to Mr. Garvey's contribution earlier that one learns of the complexities of the census.

As the witnesses will be aware, this is a value-for-money committee. Our job is to ensure we get the best possible value for the taxpayer. There are some issues that have not been mentioned. The jump from €19 million in 1996 to €44 million, even with all that has happened, seems huge. Implicit in this I notice in the Comptroller and Auditor Generals report that it appears that part of the cost of that computerisation programme will have some beneficial effect for the next census. I assume it would be fair to point out that by that time there will be a need to upgrade the system. I wonder if it is that wise to believe that there will be a savings there.

As in every other sector in Irish life, the foot and mouth problem certainly caused its own problems for the CSO. On my reading, about €7 million to €8 million was lost, through no fault of the CSO because it could do nothing about it. I imagine that figure would have been a good deal higher if one was to calculate the cost of keeping those people employed and which I accept was the right thing to do in the circumstances. At the same time it was a cost to the State. I imagine that the overall cost would be more like €10 or €12 million if all that was factored in.

Given the commercial importance of the information held by the CSO, is it sold to commercial firms? If not, will some consideration be given to ensuring we get even better value for money?

Historically it was the practice for the enumerators to be chosen from those on the unemployment register. Did the Celtic tiger make it more difficult to find enumerators? What was the profile of the persons recruited when that category of unemployed was no longer there? I personally know of many people who were delighted to get that work. I know the CSO trains them very well to carry out their duties. I hope it will continue with this social aspect to the work.

Mr. Garvey

I will begin and I will ask Mr. O'Hanlon to answer as well because Deputy Connaughton ranged over quite a number of points.

In relation to the value of information, I have spent 35 years with the CSO and I have been lucky that I have always been a totally passionate believer in the value of information to society. As far as I am concerned, we do not do statistics for the sake of doing them; we are there to inform the public - the ordinary people, the media and the political system. I regard statistics as a public good for the most part.

That said, there are some particular uses of the census information where we enter into agreements with people and sell some of it. For example, supermarket chains, banks and people making commercial decisions use census information to decide where it might be best to locate a premises or outlet. We are in agreement on that point. I am not sure how much the total take-up is but it is small in the context of the total cost of a census. The real value of the census is to society and the people.

Deputy Connaughton

I hope Mr. Garvey is not selling it under value because the company with which he is dealing undoubtedly does not sell its services under value.

Mr. Garvey

When I am finished, Mr. O'Hanlon might have some additional information. I know in the scale of the cost of the census it is not a particularly large return to us in that sense. I see the return as something else altogether.

In terms of the staff with which we were left, it is without a doubt true that if somebody told me that in three or four months they would make 90 staff available to me for six months we would plan things in a way that we could get maximum use from them. We were in a situation where the Government decided towards the end of March, not long before the census, to defer it. Like I said earlier, Mr. Purcell's comment is fair, but I believe we got good value, gave the people good training and we were in a stronger position to proceed with the resumed census than we otherwise would have been.

In relation to the cost of the census, table 3 to which Mr. Purcell referred gives a breakdown of a number of costs. Members will see there, for example, that headquarter staff was cheaper even though the cost of labour increased probably by about 100% in the period. That is shown, for example, in the field staff costs, which doubled from €7.5 million to €16.6 million. Those are the payments we make to the 4,500 people we recruit specially for the purpose.

Many things happened in society between 1996 and the new census. The minimum wage was introduced and all those kinds of things had a significant impact on the cost of labour. As I said earlier, we were also in a very tight labour market at the time of recruiting for the 2001 census so there was a real fear that we would not have as many applicants for the temporary posts, and indeed that was the case. I remember being involved in the 1981 and 1986 censuses where there were 20,000 applicants for 3,000 or 3,500 posts. That was not the situation we were in at the time of the last census so we had to make sure the rates we were giving to the field staff were at least as high as the minimum wage. That was a guiding principle for us.

We must bear this in mind when doing the comparison. Mr. Purcell's report includes a reference to repeating the exercise as we had done it in 1996. We estimate it would have cost €31.7 million at 1999 prices. I am convinced that if we had recalculated fully the whole exercise in 2002 prices - I have gone through this with my staff - we would find the census this time cost roughly the same as the old exercise. In other words the cost of technology offset the cost of the staff we would have had to recruit otherwise. However, the big advantage to society is that the information will be available when it is needed, considerably sooner than before. I do not know whether I have missed some of the points.

Chairman

You might like to come back to Deputy Connaughton's point.

Deputy Connaughton

What about the commercial side?

Mr. O’Hanlon

On the commercial value, the only census products that we market in any commercial way are what we call the small area statistics. These are the analyses that one can obtain at the level, for example, of the electoral division of maybe 300 or 400 households. Particularly with new technologies in regard to geographic information systems, one can now get information linked with mapping and other issues like that. There are a number of specialist companies which will package this small area data——

Chairman

The demographic data.

Mr. O’Hanlon

Yes, the demographic data with, for example, information they may obtain from other sources such as where the public facilities are available and link it into a mapping base and they sell that information. We have an arrangement with a number of those, which we term a reseller's agreement. Going on 1996 data, we charge a royalty for every time that data is sold on in such a package. Unfortunately, I cannot give the committee what the value of it was in 1996, but my recollection is somewhere in the order of maybe €500,000.

Deputy Connaughton

As time progresses this information will become even more valuable. As the Central Statistics Office is the only organisation doing work at this level, is it not possible to make more money on it, to put it bluntly?

Mr. O’Hanlon

Certainly, with the advent of the geographic information systems, in other words being able to produce information with maps to back it up and show and display it better, that market certainly has increased. Having said that it is very finite and it is limited to this production of small area data because the opposite trend is taking place in regard to statistics in general. At the national or at the regional level they are seen more as a public good which in fact should be made available as quickly as possible by, for example, using new technology such as the Internet to push the information out there and encourage people to use it. Because, as Mr. Garvey has said already, the potential return even in the most market driven statistical service would still only be a tiny fraction of the cost of statistics, we really see the value of the information itself as the justification of statistics.

Mr. Garvey

I would like to make a point on that. Much public sector thinking over the past ten or 15 years in some countries would have been towards developing a kind of market orientation and selling data and so on, not just in statistical offices but also in other areas of the public service. It is interesting that in statistical offices in some of the countries where the market approach was stated almost religiously they have retrenched quite a bit. Whereas some countries would have set revenue targets in the selling of services, the realisation has come about much more so in recent years that to support democracy, public information and so on, the drive must be away from that and towards putting as much information as possible on a website, making it available to everybody and trying to use statistics and information as a way almost of counteracting disenchantment with participation, whether it be in voting or interest in what is happening in administration and so on. There has been quite a shift in thinking in the past number of years towards getting as much information out there as possible. There always will be scope for providing higher value added information to some key users, but the effort in that, the packaging and so on will be done by commercial suppliers and we will take a royalty where appropriate.

Chairman

I wish to ask Mr. Garvey a few questions about the figure of £1 million in the advertising budget. When one looks at table 3 of the graph, which relates to 1996, it seems the figure is £1.35 million. Has the CSO carried out a study to assess the value for money, effectiveness and benefits of the money that has been spent on television and radio advertising?

Mr. Garvey

I do not know if we have carried out a formal exercise, as such. Why does one conduct a census of population, as opposed to a survey? The reason one holds a census is to achieve perfection, almost, for small areas. A certain amount of the advertising would have been aimed at ensuring that we got good people to carry out the work locally, etc. It was also necessary to ensure that people expected calls from enumerators and were favourably disposed towards giving the information. The real value for money test is the extent to which people co-operated with the census and they did so almost to a household.

Chairman

How confident can the CSO be of the 100% coverage of the population in the 2002 census, bearing in mind changing work patterns, apartment works and new housing developments? A massive amount of development has taken place since 1997. I am aware that there was a huge inaccuracy in a study that was conducted into the level of housing some time ago. How confident can the CSO be that all the new apartment blocks were covered, as there are many of them?

Mr. Garvey

It was a huge job.

Chairman

How confident can the CSO be that we are talking about 100% coverage?

Mr. Garvey

As I said earlier, one of the most important parts of census preparation is to place a map in every enumerator's hand. The map is based on a long-standing relationship with the Ordnance Survey of Ireland, which goes back many years. We are provided with the most up-to-date mapping that is available. The CSO's visual inspection on the ground is used in some circumstances to update our information about new housing. Our people on the ground can check the landscape after the Ordnance Survey's efforts. We have regular meetings with planners in Dublin. We consciously decide to take a hit in terms of publicity in our appointment of enumerators, but it is an investment in quality.

Chairman

Is the €1 million advertising budget well spent?

Mr. Garvey

We appoint enumerators in areas they know. We take a little hit from the public when people say, "You are telling me that the information is confidential, but my neighbour down the road is the enumerator". The appointment of such people guarantees that the person on the ground will know the area in which they are enumerating and will know where all the new houses are. They are, in effect, enumerating their neighbourhood, which is probably the best guarantee. Other countries do not follow such a policy, but instead have a post-back system. Our system involves collection.

Chairman

The enumerators have done a fantastic job - 1,350,000 forms were sent out and 1,300,000 were returned.

Mr. Garvey

A form is dropped into every household where there might be someone present on census night. Contact is made in that way. Some people are away on census night, for example in hotels or outside the country. There is a possibility of legal proceedings in about 12 or 15 cases as a result of non-compliance, but I am not sure if we will progress them to finality. The local enumerators impute the completed forms for the handful of houses that do not co-operate, when people fold their arms and say "No, we are not going to co-operate". There is only a handful of such cases.

Mr. O’Hanlon

The Chairman asked specifically about the effectiveness of the advertising campaign.

Chairman

Yes.

Mr. O’Hanlon

Most advertising people can provide details of hit rates in the various media. The effectiveness of campaigns can be measured from that point of view, as we did in 1996. The main function of the census advertising is to create awareness among the public of the start of the census, as well as its importance. It acts as an aid to the 4,000 people who work as enumerators.

Chairman

Yes.

Mr. O’Hanlon

When they knock on doors and introduce themselves as census enumerators, the effect of the advertising is that the person answering the door will have heard of the census. The effectiveness is not measurable, but the CSO is quite satisfied that the vast majority of people are aware that the census is taking place and know what is required of them by the time the census day arrives. The CSO and the people who work for it in the field offices believe that the task of the enumerators at the doorstep would be much more difficult if there was no advertising awareness campaign to accompany the census.

Chairman

The point I am making is that there has been a massive increase since 1996. If one takes inflation into account, one is talking about an additional €1 million since the 1996 budget. I am making the point that someone is making money on it.

Mr. O’Hanlon

The increase can be explained by reference to two factors. We had to invest part of the advertising budget in looking for people to take census jobs on this occasion.

Chairman

Yes.

Mr. O’Hanlon

That was a first and reflected the tight labour market. As Mr. Garvey indicated, an advertisement in the newspaper or contact with the local unemployment or FÁS offices was sufficient to find six or seven candidates for every position on offer in previous censuses. The labour market had changed dramatically by 2001 and the CSO had to pursue people very actively to find enumerators. It is unfortunate that resistance to censuses or statistical inquiries is increasing among the public in Ireland and most other countries. We judged, therefore, that we needed to put more resources into an awareness campaign regarding the need for the census and the need to co-operate with it. When that is added to inflation and increases in advertising rates, the apparent jump mentioned by the Chair can be explained.

Chairman

Has the CSO disposed of the 1.8 million unused forms, or are they in storage? I refer to the forms that were not used in 2001. Where are they now?

Mr. O’Hanlon

The unused forms have been shredded and disposed of.

Chairman

I note that the CSO pays an annual rent of €634,869 for its property in Swords, on which it has a 25-year lease.

Mr. Garvey

That is an Office of Public Works lease.

Chairman

I believe additional properties were rented. Did the CSO have long-term lease arrangements on such properties? What was the contractual arrangement with the Office of Public Works in that regard? It seems like a lot of money.

Mr. O’Hanlon

As Mr. Garvey has said, the lease is taken on board by the Office of Public Works. It has been used by the CSO, essentially, for the 2002 census. From its own operational point of view, the CSO has put a great deal of work into customising the building for census purposes and would like it to be available for the 2006 census. There will be an interim period of about two years between censuses, essentially, during which it will not be required by the CSO and will be available to the Office of Public Works. On the one hand, the Office of Public Works intends to use the major storage we have and it may also be used as part of its accommodation portfolio if necessary.

Chairman

I know you are not directly responsible for the Office of Public Works, but a building with a 25 year lease arrangement at an annual rent of €634,000 will be vacant until 2006.

Mr. Garvey

It will not be vacant. We have already been in contact with the Office of Public Works and it knows precisely the position in relation to our wind down of staff. My understanding is that the Office of Public Works intends to secure short-term tenants pending our occupation for the next census.

Chairman

I observed in the report that in 2001 a lease was taken on photocopying machines which were not used. How much was paid for that?

Mr. Garvey

I am not sure I have those details.

Chairman

It seemed to be quite a hefty figure.

Mr. O’Hanlon

Before addressing that point, I cannot comment on the leasing policy of the Office of Public Works, but there was extreme difficulty in sourcing a premises for the census in 2001. There was nothing available in the market at the time and we spent a number of months trying to identify a suitable premises. In that context, it may not have been possible——

Chairman

Whereas at the moment the Office of Public Works is selling surplus property. It is amazing that it is selling property while signing a lease which is in excess of €500,000 a year. It seems a contradiction.

Mr. O’Hanlon

I could not comment on the Office of Public Works's overall policy. In 1999 and 2000 when we were attempting to secure a premises, the Office of Public Works had extreme difficulty finding or making available a suitable premises. The situation the Chairman described perhaps obtained more in 1996. The 1996 census was carried out from the office the CSO vacated in Dublin when it moved to Cork in 1994. The previous census was done from Hume House in Ballsbridge as there was spare capacity at the time within the Office of Public Works porfolio. My recollection is that there was no spare capacity available to meet the needs of the 2001 census. It may be different now.

Chairman

Another concern I have is that most items do not include VAT. Given that you are dealing with a trader from outside the State, is there tax harmonisation in terms of value added tax? For example, are you paying VAT on computer equipment? How effective will that investment be in 2006? Deputy Connaughten raised the point that there is a huge investment of €10 million in computer equipment. Did an Irish company tender for the printing of the census forms? Was there a process of selective tender?

Mr. O’Hanlon

The CSO is liable for full VAT in the same way as any other organisation.

Chairman

In terms of the figures quoted, what rate of VAT applies?

Mr. O’Hanlon

It is 21% for most services.

Chairman

When you add VAT to the actual costs a different figure entirely is given.

Mr. O’Hanlon

The figures here are VAT inclusive. Perhaps Mr. Purcell will confirm that.

Mr. Purcell

They are VAT inclusive. The figures represent the full costs.

Chairman

They are inclusive of VAT.

Mr. O’Hanlon

The Chairman's second question refers to the printing contract. As I indicated, we included the printing of the forms as part of the overall contract to process the census. The tender appeared in The EU Journal and full procurement procedures were followed. The contract was won by an international consortium and it included quite specifically the provision that it would be the consortium's business to print the forms. When it came to reprinting, we had to renegotiate the overall contract with the consortium to take account of the postponement of the census. There was a need for the consortium to change its programming routines. Part of that included the printing of the 2002 forms. It was the consortium's decision to source a new printer for the forms due to our dissatisfaction with the 2001 forms. It switched from the UK firm which had printed the UK census forms for 2001 to, as I indicated earlier, a Dublin printer whom I am glad to say was very satisfactory.

Chairman

It is good to think that we have the capability to print the form in this country. It is essential to get the best possible deal and the best value for money, but it is also important to have the work done here. What is the annual contract to maintain the computer technology?

Mr. O’Hanlon

The processing of the census is effectively over and maintenance was included in the overall figure supplied to the committee. I do not have the details. The Chairman asked about the technology implications for 2006. Unfortunately, six years is a very long time in computer obsolescence terms. Many of the work stations detailed in our documentation will be used within our main office and certain other items of equipment which we feel can be used effectively. There are a number of issues to be resolved as equipment such as scanners must have maintenance cover. To ensure that our operation continued to flow we had a four-hour fault resolution maintenance contract on those items. While the machines are probably quite capable of performing the job in 2006, the maintenance charges we would incur between now and then would make using them then uneconomical. We are attempting to sell them or to find uses for them elsewhere within the public service.

Chairman

That is the very point Deputy Connaughton addressed and we are glad to glean that information. The system we have will not be used in 2006 as it would cost so much to hold on to the equipment. That equipment is now for sale.

Mr. O’Hanlon

It is for sale and we are attempting to decide what elements of our system's hardware can be held until 2006.

Mr. Garvey

We are also trying to ascertain which elements of the hardware can be used in other aspects of our office's operation.

Chairman

You might provide the committee with a report on that in the future and on a re-evaluation of your office's equipment going forward.

Deputy Fleming

My question is on the Vote rather than on the census of population. While most of this conversation has addressed the census, the CSO carries out ongoing labour force surveys and household income surveys. I know the CSO regularly conducts surveys on various aspects of business activity. Given that the census of population is an opportunity to get as precise a figure of the population as possible, does the CSO have to revamp its ongoing surveys, which may have been drifting in a certain direction for several years? Will the more accurate information in the census be used correctively in such circumstances?

I am aware the household income surveys carried out by the CSO are based on regions and that it is harder to stand over their accuracy on a county by county basis. Can information be obtained from the census of population which would help the CSO provide more accurate information on a county basis as this would be important for the individual counties?

People in business often tell me they do not respond to questionnaires they receive from the CSO. Approximately what percentage of the forms issued by the CSO to businesses are returned?

Mr. Garvey

To deal with the final question first, one can measure returns and the level of response in a couple of different ways. Very often, what is most important is that one covers a large proportion of total employment in a certain sector. This is not quite the same as the percentage response rate to the number of forms one sends out. The reason for that is there are probably between 50 and 100 enterprises in the country which can shape our economic figures, namely, some of the large multinationals. The delay in publishing figures is an indication of the reluctance of business to give us information. We put a great deal of effort into reminding and telephoning companies and chase the important respondents and large enterprises to the last, that is, until they give us responses.

We are satisfied when we get a reasonable representation among smaller enterprises. In terms of, say, our monthly production indices we get something in the order of 80% to 90% coverage by employment, possibly less in terms of the number of enterprises.

On the Deputy's first question on how we might use the census to benchmark other activities we carry out, the census is a highly important baseline for surveys with which everybody in the country is familiar, for example, the quarterly national household surveys, and the labour market information they provide, as well as the range of special social modules we produce on housing, health and so on. Last week, for instance, we published one on voter participation.

Deputy Fleming

This morning we heard about the most popular names, which are Jack and Sarah.

Mr. Garvey

One of the reasons we can do sample surveys and get good quality results is that we use the census of population as the base for benchmarking our sampling and grossing procedures. This enables us to get fairly good quality sample based estimates even, as the Deputy stated, on regions.

The second question related to whether household and census information could improve estimates of county income. The whole question of county income is very technical. My colleague Bill Keating visited this building within the past month to talk to a few Deputies about the reason certain counties are low down in the order. Perhaps he would like to elaborate on this in simple terms.

Mr. Bill Keating

There are some elements emerging from the census which will probably help us to revamp our estimates at a county level. However, certain aspects of these are not affected by the census, for example, the income we attribute to each county for industry comes essentially from the census of industrial production and the way we do this is to approach every manufacturing enterprise. These surveys will not be affected by information emerging from the census.

On the other hand, our estimates for certain other sectors, including certain parts of the service sector, are effectively calculated by multiplying employment by an average wage. The extent to which the census leads to the estimates of employment at regional level in the quarterly household survey being revamped, as the Deputy stated, or the emergence of a fixed benchmark for employment, particularly in services, will dictate the degree to which the estimates at county level are firmed up or perhaps even changed. As the Deputy will be aware, we make it clear when we publish the county data that it should be regarded more as indicative information than as accurate estimates.

Chairman

I have an observation regarding the CSO and Ordnance Survey Ireland. Recently, when checking the electoral register I found a shocking level of inaccuracy. When mapping areas and the location of houses and so forth, do Ordnance Survey Ireland and the CSO match their information? The information I received from this source was completely inaccurate.

Mr. Garvey

To which source is the Chairman referring?

Chairman

Ordnance Survey Ireland. The electoral register is also an example of inaccuracy. A large number of people who are deceased are still on it. While this may not be the responsibility of the CSO, it is a major problem as there are many errors in it.

Mr. Garvey

The electoral register is the responsibility of the franchise section of the Department of the Environment and Local Government, which updates it. Without wishing to speak on its behalf, I understand the difficulty it faces. I was directly responsible for a number of censuses running back as far as 1979. One of the issues I was always conscious of was the discrepancy between population figures and electoral register figures as it was possible to make comparisons in terms of, for example, people of voting age. When discussing this with my colleagues in the Department of the Environment and Local Government at the time - I am sure the position has not changed in the meantime - they expressed strong reluctance to disenfranchise people. Often people on the register at the time it is compiled are students who move on or others who leave the country before the next overhaul of the register is undertaken. While I should not speak on behalf of the Department, part of the principle behind enabling and facilitating people to cast a vote is not to readily disenfranchise them. If there is a doubt as to where exactly a person is, I suspect one will find his or her name on the register.

Chairman

This issue is important in the context of Deputy Connaughton's point about getting added value from services. I checked one electoral register to ascertain whether it was accurate on demographics and locations. I received information on an area with which I was familiar and when I checked it out I found the register was completely inaccurate. I was amazed by this.

Mr. Garvey

A number of different measures are available from the census of population. Traditionally, the census has always been a de facto census as it records data on the people available on census night.

Chairman

I thank Mr. Garvey and his team for a worthwhile meeting and congratulate the CSO on the outstanding job it did on the census, which was a major logistical undertaking. Many questions were answered today. Is it agreed to dispose of the chapter? Agreed.

The witnesses withddrew.

Sitting suspended at 1.45 p.m. and resumed at 1.50 p.m.

Vote 4: Ordnance Survey Ireland, Chapter 2.1 (Shortcomings in Financial Control).

Mr. R. Kirwan (Director of Ordnance Survey Ireland) called and examined.

Chairman

I welcome Mr. Richard Kirwan, director of Ordnance Survey Ireland. Witnesses should be made 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 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 who are identified in the course of the committee's proceedings. These 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 introduction 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 being invited before the committee are made aware of these rights and any persons identified during the course of proceedings who are not present may have to be made aware of these rights and provided with the transcript of the relevant part of the 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 members 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. Members are also reminded of the provision within Standing Order 156 that the committee shall 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 policies.

I ask Mr. Kirwan to introduce his officials. They are most welcome.

Mr. Richard Kirwan

I am accompanied by Mr. Tom Clarke, head of corporate services and human resources, and Mr. Ray Jordan who is the recently appointed acting financial controller.

Chairman

I thank Mr. Kirwan. Will the official from the Department of Finance introduce himself?

Mr. Hurley

I am from the Department of Finance.

Chairman

Paragraph 2.1 of the report of the Comptroller and Auditor General reads:

Shortcomings in Financial Control

The Appropriation Account for Ordnance Survey Ireland (OSI) for 2001 was submitted by the Accounting Officer for audit on 31 March 2002 ^ the statutory date by which Appropriation Accounts must be submitted.

During the course of audit it soon became apparent that the Account was materially incorrect. In particular, my staff established that the Account did not agree with the banking records of the Paymaster General and another bank account maintained for the purpose of credit card sales. On enquiry, it became clear that full monthly reconciliations between the accounting records and the banking records had not been performed throughout 2001. Such reconciliations are a fundamental control in the financial management of any organisation.

Over and above the failure to carry out bank reconciliations, certain financial transactions were not being correctly processed in the accounting records due to software problems and some human error.

The OSI, on being notified of the position withdrew the Appropriation Account and set about investigating the erroneous figures.

Following completion of the bank reconciliations and the resolution of accounting errors, the necessary adjustments were made and a revised Appropriation Account was prepared and submitted for audit on 4 September 2002.

My audit established that the revised Account properly presents in all material respects the receipts and expenditure of the Vote for 2001.

As I was concerned about the breakdown in financial control in OSI during 2001, I sought information from the Accounting Officer on the circumstances that gave rise to it and on the action being taken to prevent a recurrence.

The Accounting Officer informed me that in previous years OSI payments for supplies and travel had been administered by the Valuation Office, but that these functions had been taken over by OSI in 2001. Moreover, during 2001, OSI finance staff were participating in planning the legal transition of the organisation from a Government Office to a semi state body which involved a move to accrual accounting, and there was also additional work associated with the Euro changeover. While OSI had increased its staffing to allow for this additional work, including putting in place the necessary professional financial expertise, it was apparent that the overall capability to manage this substantial body of work had been less than fully adequate, especially in the early part of 2001. This was the main reason why anomalies and errors in the accounting records were not always detected and addressed.

The Accounting Officer believes that procedures and processes are now generally robust. However, to obtain full assurance in this respect, OSI has appointed external experts to review them, compare them to external requirements and best practice and to make recommendations in relation to any required changes. It is his firm intention to implement any recommended changes by the end of 2002.

He informed me that OSI had also been working with the suppliers of its accounting software to resolve residual issues in system functioning. He stated that the bulk of these issues had now been resolved and that the remainder would be resolved by the end of 2002.

Mr. Purcell

Chapter 2.1 sets out the difficulties encountered in the preparation and audit of the appropriation account for Ordnance Survey Ireland for 2001. The signed account submitted for audit contained material errors and also did not balance with the banking records. On further examination it was established that proper reconciliations between the accounting records and the banking records had not been carried out during the year. To compound the difficulties there was uncertainty about the integrity of the data recorded in the computerised ledger system primarily due to software problems. The most important thing at that stage was to pull out all the stops to enable a sustainable appropriation account to be produced. To its credit the OSI did so and ultimately I was able to meet the statutory date for presentation of the accounts to Dáil Éireann. Nevertheless, the fact remained that accounting controls fell far short of what they should have been during 2001 and into 2002. Clearly this was of concern to me. Members have seen the report where the Accounting Officer sets out the particular circumstances that gave rise to the lapses in the control framework. As a result of the experience and in preparation for getting the accounting system up to speed for legal transition of the OSI from a Government office to a semi-State body, consultants were engaged to examine the system in depth. They confirmed that the accounting system at the time was not reliable as a basis for providing management and financial information. They also drew attention to the need to bring the key control account reconciliations and fixed asset register up to date. The latest report from the consultants records that the necessary corrective action in these areas has since been taken and that the accounting system now provides a solid basis for the preparation of the appropriation accounts for the first two months of 2002 and commercial accounts thereafter. All of these matters will, of course, be checked by my staff as part of the forthcoming audit of the OSI.

Chairman

I thank Mr. Purcell. Does Mr.Kirwan wish to make an opening statement?

Mr. Kirwan

I have furnished the committee with a statement, which I can read if you wish.

Chairman

Is it abbreviated?

Mr. Kirwan

It is a short statement of just one page.

Chairman

As it is quite short, it can beread.

Mr. Kirwan

In the latter part of July 2002 staff of the Comptroller and Auditor General's office commenced their audit of the 2001 appropriation accounts for the OSI. Shortly after the commencement of this work it became apparent that there were certain errors between the statement of the account and the underlying accounting records. These errors were concerned with a failure to adequately carry out bank reconciliations and the incorrect processing of certain financial records. By agreement with the Comptroller and Auditor General the audit was withdrawn and the errors were corrected. Accordingly a new set of accounts was submitted to the Comptroller and Auditor General on 4 September 2002.

As Accounting Officer I deeply regret that errors were made in the compilation of the organisation's accounts. While such errors are inexcusable and should not have occurred I would ask committee members to have regard to certain changes which the finance unit and the OSI as an organisation was undergoing at the time when these errors occurred.

In 2001, the OSI commenced a significant part of the process of establishing its own accounting system which previously had been carried out jointly with the Valuation Office. In that year the OSI for the first time administered its own purchase ledger and this involved taking all invoicing and travel and subsistence accounts processing from the Valuation Office. While this change was in progress the staff of the finance unit were involved in the preparation for our change of status from Civil Service Department to that of State body and this involved a move to accrual accounting. The planning arrangements for the implementation of the euro in that year, particularly having regard to the commercial nature of the organisation, was a further issue which had to be managed. Alongside these changes there was a significant increase in the level of the organisation's ongoing mapping business which put even greater pressure on staff and systems support.

In the advanced planning arrangements to meet the above contingencies the staffing levels in the finance unit were increased to meet the increased workloads. This increase in staffing included the appointment of qualified financial expertise. In retrospect it is now apparent that our overall capability to manage this substantial body of work was less than fully adequate from both a staffing and systems support level and this is to be regretted.

I do, however, wish to reassure the committee that the organisation has learned some valuable lessons from this serious breach of financial procedures and the organisation's financial systems are now more robust as a consequence of this incident. Following the withdrawal of the OSI 2001 appropriations accounts I put arrangements in place to comprehensively complete all bank reconciliations and to resolve other accounting errors. In addition I arranged for the appointment of external financial experts to review the organisation's financial procedures and to make recommendations on achieving best practice. I am glad to report that this review has now been completed and progress on implementation has commenced.

Chairman

Did you sign the 2001 appropriation accounts without checking them?

Mr. Kirwan

I signed the 2001 accounts having been assured that they were in order. I was satisfied at the time with the answers I was given by my professional staff. It was only later that certain matters came to my attention.

Chairman

I have listened attentively and I know you have indicated this massive turnaround. The organisation submitted its appropriation accounts on the very last day that they could be issued, which was on 31 March. The Comptroller and Auditor General found that fundamental financial management controls were not being performed and that the accounts were inadequate, meaning that they had to be withdrawn. Was that not extraordinary?

Mr. Kirwan

Yes, I regret that and make no excuses for it. It should not have happened but it did. We were, as I said, under considerable pressure because of the range of activities that we had to engage in and, inevitably, some things slipped.

Chairman

Involvement in a process of privatisation or obtaining commercial semi-State status does not give any public body the right to treat the 1993 Act with total defiance. I know the pressure is huge but that does not permit derogation.

Mr. Kirwan

I accept that.

Chairman

Does Mr. Kirwan consider bank reconciliation to be a fundamental control in any business, big or small? If so, why were full reconciliations not carried out through 2001?

Mr. Kirwan

I accept that they should be done. I was not aware in 2001 that they had not been done up to that point. I had been assured that the various procedures were in place.

Chairman

What is the cost of the consultants?

Mr. Kirwan

The consultant cost to date is €190,000. That involves a number of things, including an examination of the problems we had. Equally, there were a number of issues on which we had to seek expert advice when adopting semi-State status, such as the accounting policies, which were fundamentally different from the old accounting policies. We had to determine how we would value a database and how we would treat the various sales and receipts. Various things like that had to be done. Also, in moving through the accrual system, there was a requirement for expertise. There were a number of factors that would have had to have been considered in any event.

Chairman

What does Mr. Kirwan expect the final cost of the consultants to be?

Mr. Kirwan

I expect it to be around €200,000. The figure I have given - €190,000 - is almost the final figure.

Chairman

Will the books be in order for Mr. Purcell to carry out the 2002 audit?

Mr. Kirwan

I am satisfied that the two sets of accounts we have to produce are proper. These include the appropriations account up to March, when we were still a Civil Service organisation, and the accruals account, which we produced up to the end of December.

Chairman

Have all the consultants' recommendations been implemented in respect of the work to be carried out?

Mr. Kirwan

All the critical and fundamental ones, such as the need for reconciliations, have been implemented. We had major problems with the accounts system, which Mr. Purcell mentioned. It has been stabilised and is now giving accurate end-of-month statements. There are other recommendations which we will implement this year and we will continue to monitor the quality of our reporting in this respect.

Chairman

Has the Comptroller and Auditor General been provided with a copy of this report on the ongoing changes that have taken place since his audit?

Mr. Kirwan

He has been provided with all the reports from our consultants.

Chairman

Has the OSI resolved the residual difficulties in the functioning of the accounting software?

Mr. Kirwan

Yes, it has. The software is now robust and it is turning over the accounts on a monthly basis, to our satisfaction.

Chairman

The point was made that administrative responsibilities for the OSI payments for supplies and travel were transferred to the OSI in 2001. Did anyone in management foresee that this might cause difficulties for finance staff in 2001?

Mr. Kirwan

Yes. We were fully aware that fundamental change was occurring in the organisation and we had started a process to increase and strengthen the finance function. In early 2002 we were in the process of adding another management accountant to the staff. We were in the process of putting in a full-time, dedicated credit controller and we had also engaged some administrative staff.

Chairman

How many qualified accountants have worked in the finance section of the OSI over the past five years?

Mr. Kirwan

In the past five years, we have always had at least one qualified accountant. That includes the period when we were part of the Valuation Office. A qualified accountant was shared between both offices. Currently, we have two qualified accountants in the organisation.

Deputy Fleming

It is extraordinary what happened. At the end of July the accounts were withdrawn and, to give credit, Mr. Purcell had the new set of accounts within a five week period, by 4 September. Nine days later, by 13 September, he was able to sign off his audit, which is a remarkable feat given what had happened during the course of the year. Maybe he will tell us if this is very unusual or if it happens in some organisation every year. Does he consider the accounts to be in need of special attention next year? Are they error-prone in view of what happened last year?

Mr. Purcell

It is unusual, but not rare. Two years ago, I think, we had similar circumstances in respect of the Vote for the Department of Arts, Heritage, Gaeltacht and the Islands. It was almost a carbon copy of the case under discussion. It stemmed from a lack of appreciation within the organisations of the importance of carrying out reconciliations and the fundamental importance of agreeing to bank records in ensuring that an account did not have any unauthorised transactions. Of course, one is exposed to this danger if one does not reconcile. As the Deputy will appreciate, it is not just a matter of niceties.

What compounded matters here is that we did not get around to doing the audit until July because we thought there would be no problems. The section that deals with the OSI deals with the Office of Public Works and some of the bigger Votes as well and it got those out of the way first. One might say that if we had started our audit in April, we might have found this problem in April but the main point is that an account was presented which was manifestly incorrect. That is of concern.

It is clear from what the Accounting Officer has said and done - I have seen all the reports - that things are being put right. It is important because the OSI is now a semi-State body. The audit committee that was appointed to the OSI asked to meet me and we had a long and fruitful meeting in the latter part of last year. All of these things are indicative of a really different approach to corporate governance and accountability within the organisation. There will be two audits for 2002, one of the appropriation accounts for a two-month period and one of the OSI's commercial accounts for March-December 2002. While I know it is tempting fate to say this, in carrying out the audit for 2002 all of the indications are that things are pretty well in order.

Deputy Fleming

Now that the OSI is a semi-State body, am I right to assume that the Comptroller and Auditor General will not audit it in the long-term? Am I correct in saying that as a commercial semi-State body, the OSI will have an independent firm of auditors and this will move outside our remit from now on?

Mr. Purcell

In 1992 there was a White Paper on what should fall within the remit of the Comptroller and Auditor General, and therefore within the remit of the PAC. The policy decision at that time was that commercial semi-State bodies, in general, should be audited by private sector auditors and reported to the then joint Oireachtas committee on commercial semi-State bodies. Non-commercial semi-State bodies, such as development and training agencies etc., would fall within my remit and that of the committee. This is a general rule. The view was that if we wanted to cut people loose and release the talent they have for the benefit of the State, without having the spectre of the Comptroller and Auditor General or the Committee of Public Accounts hanging over them, theaudit would go outside. However, I carry out, on a statutory basis, audits of certain commercial semi-State bodies. Although the OSI is a commercial operation, because it depends on State support for the service it is providing, the policy decision was taken that the Comptroller and Auditor General should be the auditor. It is the beginning of a new and beautiful relationship between the committee and the OSI.

Deputy Connaughton

I am a new member of this committee and while I have been around for long enough, I am amazed at the various things that can go wrong in all organisations. I am amazed that something so dramatic could have gone wrong. Given the guarantees we have been given, I will not pursue this. I assume everything is in order. It shows the importance and value of the Comptroller and Auditor General and the Committee of Public Accounts. No matter how well organised businesses appear to be, there are always problems of this nature.

I remember being involved in the Dáil discussion when the legislation was changed to allow OSI to become a semi-State body. The product it produces is hugely important to the country. Can I take it that the production of maps, for example, is done on an entirely commercial basis? Does it have to pay for itself? What is to stop the OSI from charging exorbitant prices?

A huge amount of properties around the country have no maps - if one goes to the Land Registry Office there are no maps available. Does the OSI carry out this mapping? For example, there are huge problems over commonages and bogs in the west where no maps are available.

Mr. Kirwan

The Deputy asked about the commercial nature of the organisation. Under the Act, we have a twin remit - we must act in the public interest while also acting in a commercial manner. In acting in the public interest, we are obliged to maintain all mapping for the State. This obviously includes areas that are commercial and non-commercial in sales terms and we receive subvention from Government to do this. We would not be in a position at this time to make full cost recovery as part of our remit is to map in non-commercial areas. Cost recovery in 2001 was 74% as opposed to 54% in 1997. We are improving our cost recovery.

The Deputy also asked about competition and pricing. We have some, but not many, competitors. They are to be found in the lucrative areas where mapping is in demand. We find a downward pressure in prices, not only in Ireland but also in other countries with similar mapping organisations and policies. When one looks at our pricing one will see that we have had very little price increase in recent years, excluding this year. Our pricing for normal mapping has been reduced. For someone seeking mapping for planning permission, one map of their site or house will cost approximately €38. A number of years ago that map would have cost at least €60. We have lowered our prices through the implementation of new technology.

As regards our responsibilities vis-à-vis property, we are responsible for the topographic mapping of the State. We are not responsible for the legal boundaries in the State - this is a matter for the Land Registry. We have a number of programmes for topographic mapping. In urban areas all mapping is up to date and revised on a one-year cycle. In suburban areas, we are more or less up to date and while we operate on a three-year cycle, the intention is to reduce this to a one-year cycle. In rural areas where little building development has taken place - excluding small towns and villages - our mapping has been out of date. However, we now have a programme in place that will ensure the areas are resurveyed and remapped by the end of 2004 or early in 2005. That will ensure maps are up to date. The intention, provided the demand is there, is that we keep them up to date on a five-year cycle.

Deputy Connaughton

Does that cover even the more remote areas, such the mountain and boglands areas?

Mr. Kirwan

It covers all those areas.

Deputy Connaughton

That is good news.

Mr. Kirwan

Obviously those areas will not be as detailed as city areas because there is not as much detail there. We keep our tourist-type mapping, which we completed a number of years ago, up to date on a five-year cycle.

Deputy Fleming

The part of the accounts I wish to look at is the appropriations in aid. The OSI had income of over £10 million, including £6 million in royalty fees, in the year under review, which is great. Will Mr. Kirwan tell us who the OSI's major customers are - local authorities, public bodies and roads authorities - and can he explain how the system works? Does the OSI have contract arrangements with the local authorities to update their maps? Mr. Kirwan mentioned all the urban areas, but I presume he includes the county towns in that. I know from dealing with local authorities that maps can be woefully out of date. Last year we were all looking for maps of new estates and some of the local authorities had maps that were 20 years old. Perhaps that was all they were prepared to give to us - I do not know. Are maps available to all local authorities or do they purchase them from the OSI on an annual basis?

Of the £10 million income, how much was collected in the course of the year and how much was outstanding at year-end? Mr. Kirwan mentioned a credit controller. However, given that the OSI has a monopoly, I would be surprised if it did not use its position of strength to ensure that it did not need a credit controller to chase money six months after the service and invoice are provided.

Mr. Kirwan

By and large, most of our customers are in the public sector. The local authorities are a big and significant customer. We operate a five year agreement with the county managers association, whereby we agree on a pricing structure for the supply and use of our mapping. Other big areas are the utility sector - electrical, gas and telecoms - which has been a growth area in recent times. Government services and Departments such as the Land Registry and the Department of Agriculture and Food are the most notable customers by way of income. However, we find that Government Departments are using mapping more and more, particularly in digital form, for their own information and for supplying information to others. We also have smaller users such as engineers, architects and solicitors.

With regard to royalties, there are a number of ways in which we supply our data services. We sell paper mapping, but for digital mapping, which is a growth area representing some 65% of our income and sales, we licence it to local authorities and Government Departments. They pay us on an annual basis for the use of that which we estimate, through negotiation, depending on volumes and usage. For smaller users, such as architects and engineers, we have another royalty or licensing system whereby, if they wish to buy our digital product and use it continuously in their business, they can do so for a fee. If they wish to buy individual products, we have an individual royalty system. That is a significant part of our income.

Deputy Fleming

Why does the OSI need a credit controller with all these blue-chip customers?

Mr. Kirwan

We need a credit controller because we find it difficult to get that sort of money. It is to bring down the level of debt and ensure that as we move to a semi-State organisation we have regular cashflow.

Deputy Fleming

How much was outstanding at year-end?

Mr. Kirwan

The figure was £2.3 million.

Deputy Fleming

That is three months' worth of business.

Mr. Kirwan

The figure is high, but it can reflect some of the bigger invoices which may have gone out. It is not something we are entirely happy about, which is why we have employed a credit controller to ensure the figure comes down.

Chairman

I look forward to Mr. Kirwan's next accounts and audit reports which will be of interest to the committee. Is it agreed to dispose of the chapter and note the Vote? Agreed.

Can we agree the agenda for next week, which is the 2001 appropriation accounts for the Garda Síochána, including the cost of bank escorts, the Criminal Assets Bureau accounts for 2001 and the Vote of the Minister of Justice, Equality and Law Reform - Vote 19; the Garda Síochána - Vote 20; the prisons - Vote 21; and the Land Registry and registry of deeds - Vote 23? Is that agreed? Agreed.

I thank Mr. Purcell and his team and Mr. Kirwan and his team for a good, frank and worthwhile discussion.

The witnesses withdrew.

The committee adjourned at 2.25 p.m. until11 a.m. on Thursday, 15 May 2003.