Integrated Financial Management System: DPSA Deputy Minister, Treasury, SITA

Meeting report

The Chairperson said the Committee was interested in the progress and clear milestones of the upgrading of the IT system. The Committee was aware the National Treasury (Treasury) was the project leader, SITA was the service provider and DPSA was the policy owner. The Committee was concerned that PERSAL (Personnel and Salary Information System) was outdated and it does make serious financial errors. When you upload HR related matters, it duplicates or ignores it at times. It is believed that the number of ghost workers could be minimised. In terms of HR management, the IT system should assist senior management in monitoring the performance of the Department in meeting their objectives. Another area of concern was how much had been spent on this project so far.

Integrated Financial Management System (IFMS2)
Mr Kenny Govender, DPSA Deputy-Director General: Human Resources Management & Development, Dr Setumo Mohapi, SITA CEO, and Mr Stadi Mngomezulu, Treasury Deputy Director-General: Corporate Services, presented the IFMS2 progress report. The IFMS program, as mandated by Cabinet, is a joint initiative between National Treasury, DPSA and SITA to modernise the management and execution of “back-office” processes in the Public Service (excluding public entities, local government, legislatures, intelligence agencies) in the following domains: Financial Management (FIN); Human Resource Management (HRM); Payroll Management (PAY); Supply Chain Management (SCM); Business Intelligence (BI). There is an IT element to the IFMS but it is mostly a business improvement initiative.

The IFMS objectives for business improvements are:
• Facilitate ongoing financial, supply chain and human resource management improvements and reforms.
• Improve service delivery by standardising and automating Public Service business processes.
• Process and compile data to improve quality information for planning, reporting, better decision making.
• Utilise new features that modern technology presents such as employee self-service.
• Take advantage of global lessons learned in the management of public resources.

• Technology modernisation to replace ageing technology and avoid increasing costs of maintaining obsolete transversal legacy systems.

Discussion
Mr S Motau (DA) complained that the gestation period of the project was too long. The government has a very bad name for not getting anything done. In the process, parliamentarians get blamed for not doing proper oversight.

Mr Motau encouraged the three departments to stress the business improvements of the IFMS. Rapid implementation was essential. He advised the departments that if their ministers are not to be doing what they are expected to do on IFMS, the departments should alert the Committee and the Committee would unlock what needed to be unlocked.

Ms D Van der Walt (DA) encouraged the three departments to provide documentation prior to the meeting so that MPs should do their homework before the meeting. MPs do not come to the meeting, look at the presentation and ask questions. Instead, they need to do their own research and come to the meeting prepared.

She asked Treasury to provide more information on the 98% discount offered by Oracle. The discount was too large and the ordinary person would think that the company was left with nothing and would not understand why the company would bother to tender. Why did the company give 98% discount and what does this deal include? She asked if the system would integrate with the other systems to provide information on which officials were doing business with government. What provision for cost escalation was included in the signed contract?

Mr M Khosa (ANC) noted that the departments would return with a project plan. What made them think they were ready for implementation if they still had to come with a project plan. Is there value for money?

It was reported that IFMS was considered too complex to implement. What are the risks that have been identified from both IFMS1 and IFMS2? He asked if the departments were up to the task of implementation.

 Mr Khosa said that it would be appropriate if the roll out was directed where there are serious challenges. There are serious challenges in local governments. Why start with national government when it is known that there are serious problems with financial management in local government.

Mr M Ntombela (ANC) wondered why so many departments were getting their services from overseas companies. Is there no company in South Africa that can provide the same services as Oracle or is it a question of not providing enough incentives? It was the responsibility of government to develop companies to the extent they are able to match companies such as Oracle. However, government did not seem to be doing this. He wondered why that was the case.

 Mr Ntombela said that it was government’s intention to build capabilities. However, the licences were in the hands of Oracle. So it becomes their decision whether to involve local engineers. When they need engineers, they will use their engineers at an exorbitant amount and not South African engineers. It all depends on them. Why is local talent not being utilised? He named a software specialist company in Pretoria and asked did they not provide enough incentives? What were they running short of?

Mr Ntombela was aware that the pilot sites for the project happened to be new ones. The understanding was that once you discover that this is a pilot site, you obviously have to put some resources there to make it possible for you to achieve your objectives. Why not concentrate on the old pilot sites instead of news ones as a way forward. What is the rationale for this?

DPSA Deputy Minister, Dr Chana Pilane-Majeke, replied about having to pay all that money on licencing. The plea would be that moving forward, one looks at the utilisation of local talent. One needed to come up with a system that can move government away from paying lots of money on licences.

Dr Pilane-Majeke said that licences can hold the government to ransom when it wants to pull out of the contract because half the time, the companies go away with their licences. The government would then have to begin the whole process from scratch. This was what had happened even in Correctional Services. In terms of the IT systems they currently had, they had regressed when compared to what they used to have when Gijima was there. The Gijima program was excellent.

She said that licences were a waste of time because government did not have intellectual ownership of them. Government builds up a system and after some time it finds that the system is gone for whatever reason and the government is left with nothing after paying so much money.

Dr Chana Pilane-Majeke said that there was the need for SITA to assist in the development a sustainable IT system so that government monies are channelled towards IT that is sustainable and that articulate what would make government and public service more viable.

The Chairperson asked the departments to explain the decision to abandon the old pilot sites and go to new ones. What guarantee is there that technical mistakes from the first pilot sites are not repeated? And how much had been spent?

The Chairperson said money was being wasted because while government was dealing with the IFMS system, departments were doing their own thing to improve their IT systems, supply chain and HR systems. She noted that Treasury had appeared before the Standing Committee on Public Accounts (SCOPA) due to a forensic audit on IFMS commissioned by Treasury. What is the status of that report? Is there anything to worry about and how much was spent?

The Chairperson was concerned about the timelines. She requested the departments to revisit the timelines.

The Chairperson said that it was unfortunate that everyone perceived Treasury to be corrupt free. There was no oversight in monitoring and evaluating Treasury. There was no oversight of Treasury in terms of DPSA regulations. It was high time the government paid attention to that especially because of this project. It cannot take more than 15 years. The project has to be finalised in our lifetime.

Mr Ntombela was under the impression that the whole project would amount to about R4 billion. What is going to be the cost of Oracle licences for the next 10 years? And what is going to be the cost of developing and customising the Oracle system to conform to South African standards?

Mr Ntombela asked how ready the state organs were for implementation. Government will have to implement but, in the meantime, how are the state organs getting ready for purpose?

The Chairperson mentioned that government was looking to review the public sector structure organogram. She wondered what this meant for the IFMS project.

Response
Mr Kenny Govender, DPSA, responded that the three departments needed to go back and put together a constructive response to a number of these questions. He agreed about the need to finalise the project as soon as possible. There was the need to revisit it as the three parties responsible for its funding, policy and implementation to see how rapidly the project could be done without compromising implementation.

Mr Govender said a core responsibility for DPSA was to ensure that it builds the skills in the country. DPSA wanted to grow capacity in the country as the moment the project is implemented, it is going to be the system in place for the next 20 to 30 years. So there is need for capacity to maintain it, further develop and enhance it going forward. The Department does not have a choice but to build the skills to ensure the system functions effectively. That work will be done. DPSA will have to give figures and projections on what capacity is needed and the cost of it. He said National Treasury would provide details on the actual cost versus the 98% discount and what was included.

Mr Govender pointed out that there were a number of other challenges around this study. In 2013 there was very slow progress with the implementation of IFMS1. The challenge at that stage was the bespoke approach vs commercial off-the-shelf. Trying to integrate all of them was just not working. SITA had insourced some of the skills to do some of their bespoke work and also outsourced to local companies. As a result, there were serious challenges in trying to get that project off the ground and get the integration working. This led to the commission of research papers – one commissioned by SITA and two by Treasury. Ernst & Young, the auditing company, did some of the work. That research study pointed to a particular direction.  Cabinet was asked to provide direction based on these research papers. Cabinet requested that a tier one organisation be brought on board. There was no tier one supplier in South Africa, and DPSA had to look at it from that perspective. However, that does not mean that DPSA was not going to use local people in the implementation and to support the maintenance going forward.

One of the reasons the Department decided to go with an open tender process was to find a tier one supplier. The Department will come back with more details on why Oracle was chosen and how the country was going to benefit going forward. No matter which way one looked at it, the maintenance and support and work at departmental level will have to be done by local people. This cannot be done by international people. Local people will have to be trained to ensure they are ready to work with it.

Mr Govender said that there was a challenge at the moment in relation to departments “doing their own thing”. In 2005 when the IFMS project started, Cabinet took a decision to place a moratorium on departments procuring any HR system as that would lead to a double cost. Thus departments did not go down that route. Currently, DPSA was trying as much as possible to limit that cost and manage it going forward.

Mr Govender said that the old pilot sites will be used in IFMS2. DPSA and National Treasury had piloted the first one and DPSA and Treasury should be part of this one so that the lessons learned are not lost. They will improve on those lessons learned and the capacities, skills and readiness will be carried through to IFMS2. The advantage was that SITA was on board as well so there are three pilot sites that can be run.

Mr Govender said that R4.3 billion was the full cost of implementation in terms of the licencing, roll out, implementation sites, hosting sites. DPSA, Treasury and SITA had done a cost benefit analysis of maintaining the legacy systems of LOGIS, BAS and PERSAL. Going forward the government would have to maintain the legacy systems, pay licencing, pay hosting and a lot of other work needed to be done along the way. The cost- benefit analysis of maintaining the legacy system, which is not giving the information needed, was compared to moving to a modern system that would provide all the information at one’s fingertips. That cost-benefit analysis indicated that it will cost the government some money to implement the new system but over the period of time, it will break even and end up saving money.

Mr Govender said it was not possible to state how ready they were because they had not started the work in terms of change management and readiness at the departmental level in terms of implementation. There had been engagements on the pilot sites at DPSA and Treasury and probably also at SITA. From the pilot site perspective, in terms of readiness, change management and preparedness had already started.

Mr Govender said that there were two big projects that DPSA was committed to. One was the macro organisation of the state. The President had already indicated that there was the need to relook government at a national level starting from ministries down to departments. DPSA together with DPME and the Presidency are supporting that work. The Minister had also committed that there was the need to revisit departmental structures to ensure optimum utilisation of personnel and structures. The challenge had always for this particular project, in terms of the IFMS1, they had never had dedicated staff to managing this project. Part of the 2013 evaluation indicated the need for a specialised, dedicated, appropriately resourced programme management unit (PMU). That has now been established in Treasury although it was still being capacitated.

Dr Setumo Mohapi, SITA CEO, replied that it was possible that the system will help in enabling better detection of public servants who are doing business with the state. There was already such a system in the Central Supplier Database where every supplier registers. The e-commerce system that tracked the supply chain was connected to the Central Supplier Database. In the IFMS, there will be one module that looks at clean data for HR, one that looks at payments in the finance module and then one that is the supply chain module. There will be other systems that will use the system outside of itself. It will be used to manage the HR systems of government, but there are other systems outside that also use it. So there are integration points that have to be sorted through.

Dr Mohapi said that the most important thing is not just about the integration. It is about building that business intelligence layer where the business questions get asked and answered very quickly. There was the need for that capacity within the ecosystem.

Dr Mohapi agreed that the biggest risk to a project of this size is that you start it wrong if you do not have the right people upfront. You have to design the right organisation, resource it with the right people and right skills. If you do not, you will start making mistakes and you are going to spend a lot of time afterwards fixing those mistakes as opposed to getting things right first time.

Dr Mohapi said that, although not satisfactory to the Committee, the three departments were not yet at the point where they were supposed to have built the centre of excellence. However, on the basis of the Chairperson’s call to revisit timelines, the departments will provide a full project plan for the establishment of the centre of excellence. That will probably give comfort on how capacity was going to be built.

Mr Stadi Ngomezulu, Treasury DDG: Corporate Services, replied that the Treasury audit committee had called for a forensic investigation due to so many allegations in the media, which the Minister concurred with. Forensic investigation is mainly in the domain of audit committees, so they provide oversight. Treasury went through a tender process and Deloitte got the forensic investigation tender. After engagements with the audit committee, it came to light that the scope was too limited. It did not give the audit committee the comfort it needed that the investigation covered all areas. Obviously, there was a misunderstanding with Deloitte. Both parties decided to part ways as there was a clear difference in expectations about the forensic investigation. Treasury was currently using Nexia SAB&T Forensics which will be finalising their audit by the end of July. Treasury will present the audit report to SCOPA but it was more than happy to share the audit report with the Portfolio Committee.

Mr Ngomezulu said that when Treasury had presented to SCOPA the total estimated cost was R4.3 billion. This is the estimate that Treasury believes is achievable. However, in a way, it also addresses the cost of overruns. Treasury has the complete R4.3 billion breakdown and will make it available to the Committee. Treasury will also get back to the Committee with the explanation of the 98% discount to make it clear how a company can make money after giving a 98% discount. He explained that Oracle was making money with the maintenance and support. The contract is made up of two elements. You buy licences once off but on an annual basis you pay maintenance for a period of 20 years. That is how they make money.

The meeting was adjourned.