Skip to main content
EVIDENCE

[Recorded by Electronic Apparatus]

Tuesday, March 11, 1997

.1533

[Translation]

The Chairman (Mr. Michel Guimond (Beauport - Montmorency - Orléans, BQ)): Order! Good afternoon everyone.

The Public Accounts committee is meeting today, Tuesday, March 11 at 3:30 p.m. to deal with two items on the agenda.

I first wanted to say that your Sub-Committee on Agenda and Procedure met on March 6 and that we are now in a position to submit a report to you to be adopted by all members.

Secondly, pursuant to Standing Order 108(3)(d), we will consider Chapter 24 of the November 1996 Report of the Auditor General entitled: "Systems Under Development - Getting Results".

We will have to ask our witnesses to be a little patient because we have a few administrative details to attend to.

So I will start by reading the ninth report of the Sub-Committee on Agenda and Procedure, which entailed the co-operation of Mr. Grose and Mr. Silye as well as myself as chairman.

The Sub-Committee on Agenda and Procedure of the Standing Committee on Public Accounts met on Thursday, March 6, 1997 and agreed to report the following:

That is the first point. To make the Public Accounts Committee as efficient as possible, instead of holding a briefing session for each of the chapters in the Auditor General's Report, one in camera and one on the analysis of the relevant chapter, your Sub-Committee on Agenda and Procedure suggests holding more regular hearings on the chapters of the Auditor General's Report in order to review as much as possible and adopt reports that we will then present to Parliament.

.1535

That does not stop us from holding preparatory briefing sessions on new subjects, but we do realize that over the past few years, this committee has changed its practice. This will enable us to study more chapters of the Auditor General's Report.

Secondly, we prepared a schedule for the months of March and April 1997. Here's what we suggest.

Wednesday, March 12, 3:30 p.m.: In camera meeting on Draft Report on Chapter 14 of the September 1996 Report of the Auditor General (Service Quality)

Tuesday, March 18, 3:30 p.m.: Meeting on Chapter 37 of the November 1996 Report of the Auditor General (Enforcing the Income Tax Act for Large Corporations) with the Auditor General and National Revenue

Wednesday, March 19, 3:30 p.m.: In camera meeting on Draft Report on Chapter 26 of the November 1996 Report of the Auditor General (Canada Infrastructure Works Program - Lessons Learned)

Then, we will adjourn business of the House of Commons for two weeks to work in our ridings, and will then resume the committee's business on Tuesday, April 8 for a special meeting. At Mr. Silye's request, we will invite the Finance Minister and the Auditor General for a meeting on accounting policies of the Government of Canada and specifically on the accounting treatment of the payments to the Maritime provinces in 1996 respecting the harmonized sales tax and the accounting treatment of the creation of the Canada Foundation for Innovation. We have not yet received confirmation that the Minister of Finance will attend, but our clerk will do his utmost to get it.

On Wednesday, April 9, we will hold an in camera session on the draft report on Chapter 23 of the Report of the Auditor General, which deals with materiel management in the federal government.

On Tuesday, April 15, we will have a meeting with the Auditor General on the Main Estimates 1997-1998 for the Office of the Auditor General. As you know, it is up to the Public Accounts Committee to review the Main Estimates for the Auditor General.

On Wednesday, April 16, we will hold an in camera meeting on the Draft Report on Chapter 17 of the September 1996 Report of the Auditor General (Human Resources Development Canada - Canada Pension: Disability).

On Tuesday, April 22, we will have a meeting on Chapter 30 of the November Report (Correctional Service Canada - Reintegration of Offenders), with the Auditor General and Correctional Services Canada.

On Wednesday, April 23, we will hold an in camera meeting on draft reports to the House. The chapters have not yet been finalized.

On Tuesday, April 29, we will have an in camera preview of the April 1997 Report of the Auditor General to be tabled that day in the House.

Finally, on Wednesday, April 30, we will have a meeting with the Auditor General on the April 1997 Report of the Auditor General to be tabled in the House on April 29.

There you have the recommendations from the Sub-Committee on Agenda and Procedure for the months of March and April.

For the third and last point of our report, dear colleagues, I would like to remind you that every year the annual conference of the Canadian Council of Public Accounts committees is held in a different location.

.1540

As you know, last year's conference was held in September in Victoria, British Columbia. We have received the invitation for the next conference, which is to be held in September 1997 in Edmonton. I therefore ask you, colleagues, that the Chair of the committee be authorized to seek the authority and the necessary funds to permit four members of the Standing Committee on Public Accounts and two staff persons of the Committee to travel to Edmonton, Alberta, to attend the annual conference of the Canadian Council of Public Accounts committees from September 14 to 16, 1997.

I would just like to make a clarification before someone asks me the question: if Parliament is dissolved, if this 35th Parliament is no longer sitting, the Canadian government delegation will not be able to attend since the adjournment marks the dissolution of House committees. That means our committee will be dissolved and no one will be delegated by the Canadian government.

I am now ready to hear your questions. If there aren't any questions or comments, perhaps someone could move the adoption of our report and we can begin hearing our witnesses.

Mr. Grose so moves.

The ninth report of the Sub-Committee on Agenda and Procedure is adopted unanimously [See Minutes of Proceedings]

The Chairman: We will now move on to our consideration of Chapter 24 of the November 1996 Report of the Auditor General entitled: "Systems Under Development - Getting Results". We will begin hearing our witnesses by starting from the left.

Mr. Harder, we would ask you to introduce yourself and to introduce those who are accompanying you today.

Mr. Peter Harder (Secretary of the Treasury Board and Comptroller General of Canada): Thank you, Mr. Chairman. I would first like to confirm that I plan to be brief, because I would like you to have as much time as possible to ask me as many questions as possible.

Today I will speak mostly about the improvements Treasury Board has made to the management of information technology projects in the context of the Auditor General's recent report that was tabled in the House of Commons.

But I would first like to introduce to you a person with a great deal of authority, who has just joined us as new Chief Information Officer. I will then give you an overview of what Treasury Board has done to rethink the role of government in the evolving area of information technology.

[English]

So before I begin let me introduce the committee to Paul Rummell. Paul is Canada's new chief information officer and comes to us as part of an executive exchange program. Paul was, until this point, a partner with Ernst & Young - I dare say the same firm as the Auditor General comes from - in Vancouver before bringing his expertise to Ottawa.

The role of chief information officer is primarily strategic. Mr. Rummell will help create the vision for the effective use of technology in government in improving the delivery of government programs. We can expect him to make a significant difference in the way government reaps the benefits of the new technologies while at the same time ensuring investment in technology is consistent with strategic planning.

It is clear we need to make certain that ministers, deputy ministers, senior policy people, as well as senior IT executives understand information technology and its vision for the future and how that vision stretches across the entire horizon of how we govern. It's a big task. Mr. Rummell is already recognized as a visionary and a leader in the industry, and I am pleased that he has joined us for this period of three years for the interchange with the private sector.

.1545

Complex information systems are required to support programs such as taxation, scientific research, and statistics. There's no question that substantial improvements in government productivity, service quality, and the cost of administration can be achieved by implementing modern technology.

I would point to our insistence on a modern comptrollership to manage government resources as a prime example of using information technology wisely. We are well aware that without the best use of information technology it would be an impossible task, and that's why we've asked a panel of experts - private sector experts as well as public sector experts - to advise us on this new initiative.

You are aware that the Auditor General notes the responsibilities for information technology are unclear. Let me suggest to you that the responsibilities are indeed clear. Departments are responsible, and the AG strongly supports this. Treasury Board Secretariat does not prefer a hands-off approach. We're taking action and we're getting results - and there is more action to be taken.

Although we are not asking departments to produce action plans, we are requiring them to set out and achieve specific targets. As you are aware, the innovative use of information technology is not without peril. Mistakes are costly in the world of high technology and mistakes have been made since the dawn of the computer age.

The Standish Group International, a leading private sector analyst of information technology, in a study of 8,340 projects in the public and private sectors in the United States found that a staggering 31% of all IT projects are cancelled, 53% of completed projects cost an average of 189% of the original cost estimates and included only an average of 42% of the original functionality, and only 9% of the projects in large organizations were completed on time and in budget.

I mention this not as an excuse but to assure everybody that the concerns and problems of the management of large IT projects are endemic to both the public and the private sector.

In April 1996, the Treasury Board ministers approved the Secretariat's action plan entitled ``An Enhanced Framework for the Management of Information Technology Projects''. It was an action plan based on four principles.

First, information technology projects are to be aligned with and support business directions and priorities. Second, clear project accountabilities are to be established. Third, government information technology project managers are to be developed and are to work within a government-wide discipline. Four, project management decisions are to be based on risk management.

For each of these principles a number of best practices, adopted from organizations that are successfully implementing information technology projects, are described.

Finally, a plan of activities was described to implement these improvements within government. The Treasury Board directed departments to begin implementing these practices in the management of their new projects and to the extent possible to introduce them into the existing projects they had under way.

To lead this implementation the Treasury Board approved funding to extend the project management office until March 1998. The necessary changes are not quick fixes but are significant changes in departments' project management regimes. While departments have begun to make changes, full implementation of changes of this scope will take time, perhaps three to five years in large departments with some of the very large IT projects.

The secretariat's project management office is putting a number of initiatives in place to speed the transition to modern management of information technology. This includes the professional development of information technology project managers and developing appropriate courses to continue their education in a field where today's bright new ideas will become part of tomorrow's desk top.

As you know, the Treasury Board Secretariat tables a fall performance report, and for the next report we are committed to bringing forward an information technology update to report on results.

We are encouraging the adoption of formal, comprehensive risk assessment and risk management approaches. The development of a strategy and a plan for establishing continuous project management improvement models in departments is under way, as is the implementation of revision to government project management policies to address specific concerns of large information technology projects.

.1550

We'll be adopting an improved estimation and performance tracking and measurement process, and we will establish a high-level project managers' guide describing the government's IT policy requirements. Departments are accountable for implementing the government's policy management framework and for ensuring that all of its project managers adhere to it. Departments must ensure that contracts for projects that are outsourced to the private sector are carried out in accordance with this framework, and departments are also accountable for reviewing the status of their projects and for taking any necessary corrective action to ensure the successful completion of their projects.

We at the board will monitor progress in departments' project management performance through established business planning processes and via the departmental project submissions for approval to the Treasury Board. The secretariat will continue to work with departments to identify project management best practices, methodologies, and tools for project managers. The secretariat will also provide a role in the sharing of experiences and lessons learned by project managers.

We believe that getting government right means identifying problems and moving as swiftly as possible to correct them. An example is the looming worldwide year 2000 computer date problem. With many computer systems designed only to recognize the last two digits of the field of 2000, as the computers move to incorporating the year 2000 we have a potential problem to wreak havoc and confusion with government business. My son will get the old age pension and my mother will be eligible for the child benefit.

The federal government recognized the problem in 1994, and since the spring of 1996 an interdepartmental working group of IT managers has been meeting to improve methodologies and best practices. I personally made certain that awareness and accountabilities for this problem have penetrated beyond the IT community to the deputy minister level itself. The federal government recognized the problem in 1994, and we are expecting, in the context of the review that's presently under way, asking the new CIO to give prominence to focusing on the year 2000 issue. We'll be using private sector advice and practice for that purpose as well.

Thank you.

[Translation]

The Chairman: Thank you, Mr. Harder. I had asked you to introduce yourself as well as those who are with you, but I think we misunderstood each other. Because of the language barrier, you did not understand me and you started your presentation. It doesn't matter. Usually, the Auditor General opens the meeting. I wanted to be a little different by first asking the witnesses to introduce themselves in order to break the monotony of our meetings, but I see I made yet another mistake. I am trying to be too democratic.

Mr. Harder: That makes up for the last time when I did not have the opportunity to make any comments.

The Chairman: Fine. I understand the tone you want to give our meeting. I think it is be a good idea to define our respective turfs. That will be appreciated during questions later on and especially during your answers.

Mr. Desautels, please introduce the people who are with you before making your presentation.

Mr. L. Denis Desautels (Auditor General of Canada): Mr. Chairman, I am joined byMr. Doug Timmins and Mr. Eric Anttila, two people who shared the responsibility for the chapter we are dealing with today. I will now proceed with my presentation, if I may.

The Chairman: Go ahead.

Mr. Desautels: Thank you for this opportunity to appear before this committee to discuss our November 1996 chapter on systems under development. In contrast to the financial and administrative systems we reported on in October 1995, we wanted, this year, to look at service delivery systems that represented large dollar outlays. At National Defence, Transport Canada, and Public Works and Government Services Canada, we reviewed four systems under development representing some $3 billion in public investments.

The overall objective of the audit was to review the management of these projects and assess their risks of being late, over-budget and short on planned functionality.

.1555

Our audit confirmed our findings of last year and Chapter 24 highlighted six key issues to getting results.

My remarks, today, will focus on two aspects of systems under development in government: the accountability framework and specific shortcomings in departments.

From the perspective of Parliament and of government, the question of accountability or the question of the roles and responsibilities of the major players in these large IT projects still remains unclear or appears to be misunderstood. Let me take a few moments to outline the position of my Office on the accountability of the department and of the Treasury Board Secretariat with respect to systems under development.

[English]

In June 1996 the Treasury Board directed all deputy ministers to apply a newly approved enhanced framework for the management of IT projects, as we've just heard from Mr. Harder. The framework addresses most IT project shortcomings that our audit identified and establishes an accountability framework for such projects.

Departments hold the primary responsibility for running large IT projects and getting the results expected. This position is generally well received by everybody, and I strongly support it.

In practice, however, as we have seen in our audit, this departmental responsibility is not always exercised fully. First, once the project has been initiated, senior management often fails to be actively supportive of the project under way. Second, in some cases senior management intervenes very strongly, but only once the project is in serious difficulty. I'll come back to this later.

The role of the Treasury Board Secretariat in interacting with the departments with respect to their systems under development is not clear to some people. It is our impression that the Treasury Board Secretariat has preferred in the past a hands-off attitude, or at least a less directive role, while others, including me, are of the opinion that it needs to take more responsibility in overseeing systems under development from a central point of view.

The Treasury Board approves the funding of all major IT projects. Through its secretariat, it needs to have some involvement in the management of IT projects and to play a more active role in the implementation of the framework than in the past. Although departments have the primary responsibility to implement new systems, the Treasury Board Secretariat is in a position of being able to carry out a certain level of monitoring of projects and to ask questions as required to ensure that the intended results are achieved and to protect the common interest.

As the enhanced framework document expresses:

[Translation]

As you know, Mr. Chairman, major information technology projects are often complex, unique, discrete and normally one-time efforts. For this reason, we would expect that the Treasury Board Secretariat would monitor the departmental implementation of the framework by ensuring that departmental action plans are developed and implemented and that new projects are approved only after the enhanced framework has been fully implemented.

Turning now to departmental shortcomings, I would like to draw the attention of the committee to three major deficiencies in large information technology projects: senior management participation, proper definition of requirements and appropriate management information for decision making.

A recurring observation in our audits of large technology projects is the crucial role played by senior management in the successful completion of these projects. Although the project leader and the project sponsor are usually senior executives, at the ADM level, we found that they can be deficient in their role as they already carry a full load of other departmental responsibilities.

.1600

Often, they are not actively involved in a significant manner in the project or they get involved late in the process when the project is almost beyond repair.

In our audits, we observed that, without the active support and involvement of the project's leader and project's sponsor, project managers do not always have the authority or the resources to handle corporate and external issues. The active participation of senior executives in support of their project manager helps to obtain the co-operation of other organizational units, to overcome problems of organizational rank or to meet contractors on even terms.

For project leaders and project sponsors to be effective, they must understand the complexities of managing large projects and believe in the importance of information technology projects to the organization. In large projects, of $100 million or more, the stakes are so high that we recommend that a project leader at the assistant deputy minister level take charge and be fully involved. The project sponsor should also work in close support throughout the development process to ensure that the project delivers on the expected results and that priorities are established and understood. The CAATS project illustrates what can happen when important results like safety and system performance receive inadequate attention.

[English]

We found that weaknesses in defining requirements properly are the cause of many problems in projects. In large, complex, and multi-year projects, the risk is often increased by the need to define the requirements in fixed-priced contracts.

At National Defence on the CFSSU project, several years were spent defining and refining the requirements for an upgrade to a supply support system, yet the contractor spent the first nine months of the project confirming with the department his understanding of the requirements. In the CAATS project at Transport Canada, several key requirements and design issues were still under discussion six years after the signing of the contract.

Because requirements are incomplete or misunderstood, delays in contract amendments add to the cost and schedule of the project. In all projects, but especially in long-term, fixed-priced contracts, the proper definition of the requirements becomes an essential first step. These requirements must be defined for realistic time horizons.

To identify the risks of a project under way and to help in decision-making, project managers need to have appropriate management information on the true state of the project and on its degree of advancement. The information needs to be timely and complete, and there are many possible indicators of progress for systems under development. Most methods recommended by the industry start by comparing planned with actual costs and time per task.

From these comparisons, trends are derived that warn management of deviations from plan and indicate when corrective action is required. In large, complex projects, where software development plays an important role, these measurements of cost and time become critical elements for decision-making. The larger the project the more crucial this information becomes to determine the true state of project tasks and its impact on other tasks and on the project as a whole.

In response to this need, the Treasury Board Secretariat has been requiring all departments to use a performance management system for all major capital projects since 1992 and for all other projects since 1994. The system to be used is one accepted in 1993 as a national standard by both government and the industry for systems under development.

[Translation]

In our audits, we observed that project managers are slow to implement the required management system, often do not have timely management information at their disposal or that their information is incomplete. In general, time and cost performance figures are obtained only from the contractor and do not reflect the Crown effort.

On the CAATS project at Transport Canada, it was only after rebaselining the project in 1995 that performance measures were requested in the contract in an effort to keep better control of the project. Unfortunately, the figures obtained were not as meaningful as could have been expected and were mostly ignored as a tool to manage the project.

.1605

Chapter 12 of 1995 and Chapter 24 of 1996 focused on managing the risks and getting results through effective project management. Nevertheless, in carrying out these audits, we also identified many difficulties associated with procurement and contracting for large information technology projects. Although we have not yet audited these areas, we may do so in the future and, in the meantime, we would encourage government to carry out its own examination of these areas.

Mr. Chairman, the committee may want to ask the Treasury Board Secretariat, first, with respect to its role and responsibilities vis-à-vis the departments, how it intends to ensure that departments produce action plans to implement the Enhanced Framework and that these action plans are followed through, and then follow a gating process where the further release of public funds is tied to progress being made on the project, and secondly, with respect to project shortcomings, how it intends to ensure that departments actively support their projects at the most senior levels, then formally agree with their developers that project requirements have been properly defined over reasonable time horizons, and finally, that accurate performance management information is used to monitor projects.

Thank you, Mr. Chairman. We would now be pleased to answer any questions from the committee.

The Chairman: I am sure Mr. Harder will also be pleased to answer questions if colleagues want to ask him any.

That said, we will now move on to the representative from Transport Canada, Mr. Gauvin, Acting Deputy Minister. Mr. Gauvin, welcome to the Standing Committee on Public Accounts.

Mr. Paul J. Gauvin (Acting Deputy Minister, Transport Canada): I am accompanied by Mr. Ron Jackson.

The Chairman: You are accompanied by Mr. Jackson, who is Assistant Deputy Minister for Safety and Security.

Mr. Gauvin: That is correct.

The Chairman: Do you have a presentation?

Mr. Gauvin: Yes, Mr. Chairman.

We are pleased to be here today, Mr. Chairman, regarding the references to Transport Canada in Chapter 24 of the Auditor General's November 1996 Report.

Unfortunately, the Deputy Minister, Margaret Bloodworth, is out of the country, and could not be present. I am Transport's Senior Assistant Deputy Minister and I am acting on her behalf.

In respect of the amount of time available today, I have just a few brief comments to make, following which we will be ready to take the committee's questions.

[English]

In the Auditor General's report, two major Transport Canada systems were referenced, the integrated departmental financial and materiel management system, known as IDFS, and the Canadian automated air traffic control system, known as CAATS.

In terms of IDFS, the November report highlighted that the system had been implemented and that a cluster group of several other departments had now been formed around the IDFS solution.

This reference is a follow-up to an earlier one in the Auditor General's October 1995 report, which said that of the four major government systems under development that the Auditor General had reviewed, only IDFS was being managed in a way that dealt effectively with the risks associated with the project. We certainly take great pride in that observation.

Furthermore, the AG noted that based on their own research and that of the private sector, the likelihood of large, multi-year systems development initiatives being completed on time, within budget, and with the desired functionality in the public and private sectors was extremely low. In fact, IDFS implementation began in July 1994 and was completed on time within its $60 million budget in March 1996, notwithstanding considerable transition and change within the department.

IDFS is the first integrated finance and materiel management system in the federal government. It has several interrelated modules, such as accounts payable and purchasing, and was implemented to approximately 4,000 departmental users across Canada.

As the single authoritative corporate system, it replaced many separate, overlapping and out-of-date departmental systems, significantly reduced paper burden, and facilitated re-engineering and streamlining of the business processes.

.1610

IDFS interfaces with other government systems, and many of its functional specifications are serving as models for other government departments. In a January 1995 report related to its government shared systems initiative, the Treasury Board Secretariat observed that IDFS's implementation process should be regarded as a model for other government departments.

So IDFS has been an important initiative for Transport Canada and we are gratified that this success has been recognized by both the Auditor General and the Treasury Board Secretariat.

In terms of the CAATS initiative, obviously any discussion of it should include the participation of Public Works and Government Services Canada, the government's contracting authority that oversaw the contractual and legal elements of the CAATS system.

CAATS was certainly one of the most complex and demanding projects ever undertaken by Transport Canada, and we acknowledge there were difficulties encountered at various stages in the project.

It is important to appreciate, however, that at the time the CAATS project was initiated, the government recognized the need for Canada to have a modern and efficient air traffic control system in order to be able to meet the forecast demands of air traffic into the 21st century.

Although the audit refers only to the role of the department in managing the contract, CAATS was and is being developed in partnership with a private sector contractor and a service provider.

When it became apparent the CAATS project was encountering serious problems, Transport took decisive action to protect its investment, including taking measures to ensure timely and effective delivery by the contractor. As the result of tremendous effort by many departmental officials, in full collaboration with the contractor, by late 1996 Transport Canada had succeeded in putting the CAATS contract in good standing.

CAATS was subsequently taken over in November 1996 by NAV CANADA as part of the successful commercialization of the air navigation system, and NAV CANADA has been managing the project since that time.

I would like to add, however, that as the government safety regulator under the Aeronautics Act, the Minister of Transport has the regulatory authority to prevent the use of the system if adequate assurance of the system's safety - based on established safety criteria - for both the industry and the travelling public is not provided.

In summary, Mr. Chairman, while we agree with many of the conclusions reached in this audit, we also share the view of others, including the Secretary of the Treasury Board, Mr. Harder, who has openly stated that where large systems projects are involved, both the public and the private sectors have shared responsibilities for success, and each party needs to do more to develop and improve project management skills.

Both parties must also put as much ingenuity and commitment into finishing projects as they do in beginning them. Not only in government, but also in the private sector, an inordinate amount of time is often spent planning, marketing, and getting the initiative off the ground, but then the project sometimes seems to become less visible until problems are encountered.

More effective and responsible collaboration between the government and the private sector on large information technology projects is obviously essential. A key element of this is risk identification, assessment, and management, along with vigilant monitoring of schedules and deliverables.

Chapter 24 makes a number of recommendations in support of the important work of the public accounts committee and the Treasury Board Secretariat in this area, which we fully support, based on our own experiences, for example, in the area of professional development and training for project leaders, sponsors and managers, as well as sponsorship, ownership, and accountability by senior management.

I trust these brief remarks have been helpful, Mr. Chairman, in clarifying the department's views, and we would now be ready to answer any questions the committee may have.

[Translation]

The Chairman: Thank you.

We will start with a 10-minute round for the questions. I now give the floor to Mr. de Savoye.

Mr. Pierre de Savoye (Portneuf, BQ): As you know, I am somewhat of an expert in these areas because for a number of years, I dealt with businesses around the world who were experiencing this type of problem and I tried to give them some advice.

For the benefit of those listening and members of the committee, I would like to review the nature of the problem. This is not the first time projects are behind schedule, are over budget, do not do what they should and, when they are delivered, may no longer meet the requirements.

.1615

Think about it! If you build based on expectations that were valid five years ago, it is quite possible that five years later, those expectations have changed. The development process must always be based on a proven development methodology. There are standards. IEEE, among others, has had a variety of standards for over 10 years now.

Mr. Harder, you said that mistakes in the high technology sector were costly, and God knows how many mistakes were made. Indeed, your systems are not the first and will not be the last, but you do not have to make the same mistakes as those who came before you.

You outline four principles. In the third, you talk about being able to work within a government-wide discipline. I find those terms somewhat vague. I would have preferred something a little more specific and perhaps a direct reference to the use of engineering systems methods or other recognized standards.

In your fourth principle, you say that project management decisions are to be based on risk management, but I'm sure you will agree with me that the greatest risk is incompetence. I am not deprecating individuals, but I am referring to the fact that, too often - and you admit this yourself - mandates are given to people who will be there only part-time for projects that will cost nearly a billion dollars. You cannot manage those types of projects from a distance or part-time. Mr. Rummell, I sympathize with you, because the contract you have is huge.

My preamble may be a little long, but I think it's important to set things out. I would also like to point out that if you do not take the time to clearly define the requirements from the outset, namely how many stories the house is to have, how many rooms you want in it and what they will be used for, it is a little too late to do so once the concrete is poured.

There is an old saying in the computer industry, and I will say it in English for the benefit of my colleagues:

[English]

``The customer knows what he or she wants at delivery time, when he or she finds out it's not it.''

[Translation]

But that is not necessary, and all the engineering principles I referred to also apply to architecture. If you have three-dimensional plans that explain what the building is to look like, you can immediately refine your expectations: no, you do not want a sloped roof; no, you would prefer to have the door a little off centre; etc. You cannot wait until the concrete is poured to realize things are not right.

There are two other problems as well. There is the problem of "while we are at it"

[English]

- the ``while we're at it'' syndrome -

[Translation]

whereby a user adds requests as the project is progressing. If there is no tight control over changes, you run the risk of significant cost overruns.

Then there is the other syndrome, the "you know what I mean" syndrome,

[English]

- the ``know what I mean?'' syndrome. Every time someone says, ``know what I mean?'', they don't.

[Translation]

You will only notice when the system is delivered or during the advanced testing stage, when the concrete is already poured, and that will be very costly.

Treasury Board has produced a document called "An Enhanced Framework for the Management of Information Technology Projects". I must say it has some good solutions. But nowhere - at least as far as I can see - does it focus on systems development methodologies, which, in my view, are critical to the successful development of systems because they provide a definition of expectations, control points, quality assurance, control of changes and tests to ensure that objectives are met at each stage before proceeding to the next and forging ahead.

I would like to hear your views on that particular point.

.1620

[English]

Mr. Harder: Thank you very much. In many ways you've touched on many elements of the recipe with your remarks.

Let me respond in a couple of ways. Firstly, those of us who have been near IT projects are scarred with the view that any project that cannot be conceived and delivered within 18 months - and some would say 12 - ought not to be undertaken. Otherwise you get into some of the syndromes to which you referred, such as ``while we're at it'' and ``you know what I mean''. So the more precision and the more rigour we can attach in narrowing the timeframes....

In some of the large projects we're going to have to do this in chewable chunks, and what we have in place in our framework is the capacity to do just that. That is to say, every time a department comes forward to Treasury Board for a submission, they must attest to the fact that their project is being managed in accordance with the framework and that the gating processes are in place so we can have the appropriate off-ramps and the discipline necessary to make the real choices as a project proceeds.

There is work we must do in the area of defining well and having contract administration procurement policies that in themselves don't mitigate against well-managed IT programs. What do I mean by that? Too often the leadership of IT programs spend too much time defining fault or potential fault in fear of litigation, rather than on solutions in the application of the project itself. That's certainly one of the areas I've asked Mr. Rummell to focus on.

At the end of the day we need to build a series of instruments, along the lines you've raised and that are in the framework, to ensure that all of these mechanisms, from training and project management skills to appropriate checkpoints as the project unfolds to ensuring gating is in place so that ministers and the board have all of the information before them....

And increasingly we are providing information to Parliament, through either the performance reporting in the fall or the large IT projects that are identified in part IIIs on a regular basis, and in questioning in standing committees and in other fora, to ensure that the senior management of a department is aware that managing their IT well is an aspect of their personal performance.

Those are some of the elements that come to my mind to avoid some of the problems you've addressed. But perhaps on the mechanisms themselves, if you don't mind, I'll ask Mr. Rummell to answer. I know he has given this some preliminary thought, having been in government now a week and one day. I'm sure he has views.

Mr. Paul E. Rummell (Chief Information Officer, Treasury Board Secretariat): Thank you for the opportunity to be here today. I appreciate the opportunity to comment on these issues and I certainly appreciate the points and the endorsement by the Auditor General of some very important ingredients in making large projects a success.

As to the member's comments, I certainly am a very strong proponent of methodologies. We've used the Software Engineering Institute's capability maturity model as part of our enhanced project management framework and we will adapt leading practices, risk assessment techniques, and project methodologies for the implementation of our projects in government.

Mr. Secretary Harder provided some statistics on projects that had occurred in industry. I'm not willing to accept those fatality rates for our projects in government. We're going to be taking the steps, through the chief information officer's office, to ensure that government has one of the best records of delivering the large systems we need to run our government as we move into the next millennium.

.1625

[Translation]

Mr. Pierre de Savoye: I appreciate that. Is my time really up?

The Chairman: Yes. You can resume in the second round. Your preamble was too long and we listened to your speech. I am sorry you did not have anytime to ask your question.

Mr. Pierre de Savoye: But I asked my question.

The Chairman: Yes, but you do not have enough time to ask another. You'll have the opportunity to do so later.

[English]

Mr. McClelland.

Mr. Ian McClelland (Edmonton Southwest, Ref.): Thank you very much for your presentation, which I found most interesting, particularly because I'm involved in a much smaller project and nothing seems to be coming together as it should. I understand that what we're talking about, which is massive, really requires a different kind of thought process to come to it.

I had scribbled down a couple of questions before the Auditor General's comments, and one was addressed to you, Mr. Harder. Why no action plan? What's the difference between an action plan and a target? Is the difference perhaps accountability?

The other question I'd like to pose to the Auditor General is if a project, particularly a massive project initiated by government, has more than one objective, and one of the objectives is defined by efficiency and information technology and the other is defined by political necessity through the outsourcing of different geographic areas or different businesses, how do you go about squaring that circle? How can we get a good deal when we're not necessarily trying to get a good deal because the business venture has dimensions that make accountability particularly difficult?

I'd like to pose one general question. I have read that we in Canada have a problem inasmuch as we blame people who are responsible for outputs for outcomes and we don't make a distinction between those who are making political decisions for outcomes and those who are responsible for doing the day-to-day job. If these two responsibilities are confused, it makes it impossible for either, and it makes it doubly impossible to assign accountability.

Would it be a good idea then that on senior levels of accountability for specific projects people are hired on a contract basis for a period that would transcend a normal political term?

There are three questions, and I would ask the Auditor General, you, Mr. Harder, and perhaps Transport to respond. I'm sure Transport would have comments with regard to the political dimension of sourcing a very complicated information technology project. Thank you.

Mr. Desautels: Mr. Chairman, let me go first. I believe the second question was addressed to me. How does a large project like this, wherein there's a mix of objectives, ones to meet certain technical requirements and certain delivery requirements and others that are, I suppose, what you might call macroeconomic objectives or socio-political objectives...?

I suspect that in a number of these projects there are socio-political objectives, such as encouraging regional development and so on. One of the projects we looked at in this last audit, the so-called CCCS project, I think, was in fact carried out essentially outside of Ottawa, I believe, and I'm told CAATS is the same situation. One of those was, according to our assessment, fairly successful. The CCCS project, when we last looked at it, was performing reasonably well, but the CAATS project ran into significant difficulties, which we've described.

So the lesson, I think, to learn from this is that when we look at these projects and when we audit these projects, we have to focus on the delivery of the technical requirements that are called for, and the other objectives that may accompany the project, such as socio-political objectives, in my view should not diminish any of the expectations to deliver the technical requirements on time, within budget, and with the functionality that's called for. In the two examples I've given, one of them seems to be working and the other one seems to have a problem.

.1630

Mr. Ian McClelland: May I interrupt. What I have learned in my three and a half years in Ottawa is that when push comes to shove and the political considerations are brought to the table, the bureaucrats will not shove back because of whatever reason. We all know what the reasons are, but how can you separate the technical responsibility from the non-technical responsibilities and say, well.... I just don't know how you could run a business that way, when you say you have to get the job done on time, but, by the way, you have to do it and fulfil these social objectives. If I haven't made myself clear, I apologize.

Mr. Desautels: As far as I'm concerned, in all of these projects you have the Crown on one side and you have the contractor on the other side, and the contractor has agreed to deliver certain systems within a certain timeframe for a certain price, and whatever other conditions the contractor may have accepted for delivering that system, he still has accepted responsibility for delivery of the system. So the bottom line is that there's a valid contract between the Crown and the contractor and it has to be respected, period.

Mr. Harder: Let me pick up on a couple of aspects of your question. You asked why there was no action plan. I'd like to suggest that the enhanced management framework itself is the action plan and that within this context every time a department comes in for funds for a particular project, we want to ensure - and they must attest for our purposes - that they have set and will achieve specific targets, and have achieved them as they come back for part two or three of the funding. That includes the kind of gating procedures that we talked about, and the rest.

So I believe we've come a long way from the kind of management of large IT projects. Just as we've learned from each other, there's been enough experience on which to base some of these procedures. And that enhanced management framework itself has been endorsed by ministers of the Treasury Board. So it's not officials either in the IT community or the deputy community; it's at the highest level of ministerial collective endorsement. From our Westminster model of accountability, there is no distinction between a minister's accountability...it all flows through a minister. It is through the responsibility and accountability of a minister that a deputy, and through the deputy the organization of a department, delivers on its commitments. Part of the way of achieving that is through performance reporting to Parliament and the kinds of projects that have been under way to improve performance reporting, the move to the outlook document in the spring and performance reporting in the fall, and the commitment by the government, as I indicated today, to report on the performance of large systems is really important.

The last point I would make is that the enhanced framework has been endorsed by the AG. I don't think this is an embarrassment; I think it's good. So I think the systems are in place and the accountability structures are clear.

The Chairman: Monsieur Hubbard for ten minutes.

Mr. Charles Hubbard (Miramichi, Lib.): First of all, for Mr. Harder, the Auditor General makes certain recommendations in terms of his observations, and I seem to comprehend that there is a difference of opinion even at the table today. But are you in Treasury Board willing to accept, when this committee writes its report, that they report on the recommendations that the Auditor General has made in terms of major projects that Treasury Board is funding? Are you willing to accept in Treasury Board the Auditor General's concerns and his recommendations?

.1635

Mr. Harder: Let me respond. I think the Auditor General's report is quite helpful. If there is a difference of view, I think it's at the margin. What is an action plan? The action plan is in the enhanced framework, I would argue, and the results and the systems that are in place to ensure prudent management of the project are key.

Let me also make this point, which I think is an important one. In appointing the chief information officer for the Government of Canada in the Treasury Board, it's the first time the government has done that. It's an acknowledgement by government that we need a clear focus within the board for the government-wide leadership that is necessary in the IT area to respond to some of the very real concerns the Auditor General's report has raised.

My concern is that we do not have a hands-off attitude, but neither are we the ones who at the end of the day will have to manage the department's accountabilities. The department's accountabilities are clear. So I want to strengthen the capacity of the centre to deliver on its role for government-wide project management. I think the Auditor General in his comments is supportive of that.

Mr. Charles Hubbard: As for the Auditor General and our friend here, Mr. Harder, you probably might have an expert if you could hire him as part of your staff to look at the systems. Pierre, here....

Some hon. members: Oh, oh!

Mr. Charles Hubbard: To the Auditor General, in terms of the contracts as they were written, were they clear, precise? Were there problems in terms of...? I would imagine that Transport sent their needs to PWGS. Somebody wrote up a contract. The contract must have had certain timeframes and progress steps to be made with certain payments to be made as the contract progressed.

Were there certain problems in terms of how those contracts...? Take the old cartoon. There was the way the consumer wanted something and how he explained it to an architect in verbal terms. There was what the architect thought he wanted. Finally, there was the third phase, which was what the product was. Was there a problem in terms of these contracts in terms of what somebody else thought they wanted and how the contractor perceived what their needs were? Was there a problem in that area?

Mr. Desautels: Mr. Chairman, with your permission, I'll ask Mr. Anttila to answer the question.

Mr. Eric Anttila (Principal, Audit Operations, Office of the Auditor General): Very often there were misunderstandings about what was required. We've cited several examples - there was the CFSSU project, the CAATS project, and so forth - where these major contracts were let while there were still major misunderstandings or a lack of clarity about requirements.

Mr. Charles Hubbard: We also heard, Mr. Chairman, that generally with information technology there are continuous changes about every 24 to 30 months. The CAATS project for example has been on line for more than ten years. With the number of companies we have in Canada that are capable of doing that kind of work.... It's my understanding there's a company in.... Can you tell me who received this contract. Who was the general contractor for CAATS?

Mr. Ron Jackson (Assistant Deputy Minister, Safety and Security, Department of Transport): The contract was let in 1989, and there was a full tendering process with a number of competitors bidding. This was a very large air traffic control automation project, as you can appreciate. The contract was awarded to Hughes Aircraft, located in Richmond, British Columbia. They had a number of partners that were part of the consortium that won the contract.

Mr. Charles Hubbard: In terms of letting that contract, I know Treasury Board was involved, but there was also Public Works. In the mind concept of the people who let the contract, was that company capable of doing the work or were they a growing company that was still groping with an industry that...?

.1640

I have heard concerns in terms of the relationships between Transport Canada and that company. In fact, some other companies that didn't get the contract were concerned about certain people working for that company who had formerly worked with Transport Canada. I think in the IT industry there had been concerns as to whether or not Hughes had the capability of doing that contract when it was let.

Mr. Jackson: I will give a bit of context, if I may. At the time of the contract award in 1989 - it was before my time - the U.S. had launched a major modernization of its air traffic control system a couple of years prior to the contract being awarded in Canada. As you may be aware, the successful contractor here was not the same contractor as the one that was doing the U.S. modernization program at that time.

The U.S., in the intervening period, shortly after the awarding of our contract, ceased its project because the contractor was in difficulty, the project was failing to meet milestones, and so on. So they abandoned their similar project to our CAATS project.

It was our view certainly that in the assessment of the tender, bid, proposal that was put in, Hughes and its partners were competent to do the job.

It was a high risk. There's no question about that, because this was a requirement that had not been satisfied anywhere else before. We were breaking ground. At the time it was awarded, we were watching the U.S. situation carefully, but when they abandoned their project, we were in the game alone at that point with, as I say, a requirement that was complex.

To address the point you made about the definition of the requirement, again, the contract called for us, the government, to specify the requirement in a fairly high-level way. It was up to the contractor to develop the technical concepts to meet that requirement. That was the way the contract was worded, which was somewhat problematic because the technical solutions developed by the contractor had to be accepted by the government, the customer.

There were differences of opinion, you're quite right, between the air traffic control experts in the Department of Transport and the Hughes people who were developing the technical solution to meet the fairly high-level requirements specified in the contract.

Mr. Charles Hubbard: Overall, one of you mentioned private groups being involved as well. In terms of the total cost of the contract, there were private.... Or did the government bear the full cost and the full risk of this CAATS contract with you?

Mr. Jackson: For the time the government was the customer, which was until November 1, 1996, it was paying. There were no partners in the business as far as this particular contract was concerned, in terms of being the customer.

As of November 1, as you know, the CAATS system was part of the air navigation system that was sold to NAV CANADA.

Mr. Charles Hubbard: Another group, another party, was mentioned. It was the air navigation people. I think someone mentioned that today, did they not? Was the government the only one involved with this contract?

Mr. Jackson: Yes.

Mr. Charles Hubbard: And in terms of the success of the contract, if it were and is totally successful, who will hold the patents and so forth? If it can be sold to another country, for example, who is holding the -

Mr. Jackson: That is an arrangement between Hughes and the new customer, NAV CANADA.

Mr. Charles Hubbard: Before that, did the Government of Canada hold those rights?

Mr. Jackson: There were licensing and intellectual property considerations as part of the contract.

Mr. Charles Hubbard: Thanks, Mr. Chairman.

[Translation]

The Chairman: Thank you. Before moving on to the five-minute round, Mr. Harder, I would like to make a comment that will put you in a good mood and to ask a question that will put you in a bad mood.

Here is the comment that will make you laugh. I am sure you know that there is a recycling and maximum resource use program in effect on Parliament Hill, especially with regard to paper. We all know that forests are a precious resource and not inexhaustible.

.1645

Taxpayers are fed up with paying for abuse and waste. As an example, at the bottom of most correspondence and official letters from the Office of the Auditor General, you see the words "reduce, reuse and recycle". I imagine you had some meetings before we arrived.

However, Mr. Harder, I am certainly not impressed because you did not practice what you preach. Look at the thickness of your brief; you gave us a 45-page document, and if you look at page 12 of the French version, it only has eight words:

Now for my mean question. But before I ask it, Mr. Harder, I would like to refer back to the first time you appeared before our committee, on November 5th, and make an apology. I realized I had made a joke unwittingly when I said, and I quote the words I used on November 5th:

I realized I made a darn good joke; it was a joke. I don't know if I will be able to sit on the Standing Committee on Public Accounts long enough, since my new leader may give me other responsibilities, but my new pet hobby horse is to try to understand the role of the Treasury Board Secretariat in relation to other departments.

When a deputy minister appears before us - since November the 5th, we have seen the deputy ministers for Human Resources Development, National Defence and many others - I always ask them whether the Treasury Board Secretariat, perhaps not you as an individual but rather you as an entity, is truly in a position to tell us that is how things should happen. I'm still asking myself questions of the usefulness of a Treasury Board Secretariat since the departments do whatever they want to do anyway.

Now coming back to my question. The Auditor General concludes his presentation of paragraph 23 with some questions. He states:

There seems to be an incredible misunderstanding or else I don't get it. I must not be getting it, yet again. What is your role and your responsibilities vis-à-vis departments?

[English]

Mr. Harder: I'll try to be very specific in my responses to the questions to which the Auditor General referred.

.1650

I would say that the gating process is in place; that the enhanced framework does provide for the attestation we would expect departments to provide to the Treasury Board before we approve the release of funds; that we actively support the Auditor General's comments with respect to projects gaining the support at the most senior levels of departments, and I endorse that entirely; and that project requirements be defined properly. In fact, I believe part of defining them properly is to put them into chewable chunks so that they can be delivered and managed appropriately and so that we don't get into these extended and failed project deliveries.

I should add that the projects we're reviewing all precede the enhanced framework. No projects that are presently in difficulty proceeded after the enhanced framework, and the kind of performance information we believe is necessary is part of the framework.

What's the role of the Treasury Board Secretariat and what's the role of a department deputy? Let me give you as an example the large immigration technology program that I was the deputy of at the time it was conceived and a contract let.

I wanted to ensure that the project was being managed consistently with the framework that was provided to me by the Treasury Board, but I was absolutely certain it was my accountability and that of my senior management team to deliver the program on time and in budget. If at some point I was to get into trouble, I was certain I would be able to go to the Treasury Board and, if appropriate, seek their help; or perhaps they would come to me and say they weren't pleased with the next phase of project approval and would require additional information; or perhaps even use the gating that's available to terminate the project. But it is the accountability of deputies to ensure their projects are being well managed and the framework is being applied to the specific case.

There is a requirement for the Treasury Board to have the capacity to fulfil its role, in terms of understanding how projects are delivered, what attestations are appropriate, and when we should take the off-ramp of the gating process. But this is a complex area where we do need good advice.

The accountabilities are clear. The role of the Treasury Board is to provide the framework, and in the face of a particular case it is to ensure the department has come forward with appropriate information that can meet the attestation process and, where necessary, encourage perhaps the use of the gate. But it is the department that is accountable.

When I mentioned earlier that it is helpful when other standing committees speak to technology issues, I was trying to reinforce the need not just for me to ensure senior managers focus on IT, but that all of us involved in service delivery, as MPs are, in the face of a particular department, whether it's Human Resources Development Canada or Transport Canada, raise the IT issues that are relevant to those departments. In the process we need to ensure there is the appropriate information exchange and the appropriate awareness by both officials and parliamentarians on the crucial role IT plays in enhancing the service delivery to Canadians of the services they expect.

[Translation]

The Chairman: Mr. de Savoye, five minutes.

Mr. Pierre de Savoye: During the first round, we understood that the problems you have are not unique and that solutions have existed for 10 to 15 years now, that they have been refined in the interim and that now you have decided to implement them. But between this desire and the implementation, we can't let these problems be solved by chance. At recommendation number 8, the Auditor General indicates that:

.1655

I would like to add that if the Treasury Board Secretariat does not assume this responsibility, I'm willing to wager that the implementation will take much longer than it would otherwise and perhaps even take too long. On the other hand, centralizing these contracts with you would provide an expertise that you could then share with others and that would give you some standards for comparison which would let you see where things are more effective, and where they are not as effective, and then you could provide advice to improve the performance of those who are falling behind. Actually, to put things simply, there are units of measure in the development of systems that we call function points. They allow us to calculate the cost and the time in the same way a builder would calculate his cost per square foot for a given building. This is nothing new, because it already exists. But if it is not managed in a centralized fashion, as the Auditor General suggests, all the potential of these tools will not be applied as it should be. And, as the chairman said so appropriately, after all it is the taxpayers, money.

What is it that you intend to do? Do you intend to adopt the recommendation of the Auditor General or let each one assume their responsibilities as they see fit?

[English]

Mr. Harder: Thank you very much. I believe the Auditor General has described in his report the appropriate accountability structures. We're discussing what the role is within those accountability structures for the Treasury Board Secretariat. I'm trying to describe for you as best I can what the tools or the levers are that the board has to exercise its role. It is not just printing a framework, distributing it, and saying we hope it works. It is to ensure that every time money is requested there is an assessment done by the secretariat with the department concerned to attest to whether or not the management of that project merits continued funding on the basis of all the indicators one would apply.

Mr. Pierre de Savoye: Excuse me.

[Translation]

You have just spoken of indicators; and that is the heart of my question. If these indicators are not standardized in a centralized fashion, there is no standard for comparison. If you impose harmonized standards, you can immediately have points for comparison. These indicators, do you centralize them and then let anybody use them as they see fit, or do you centralize them and then impose specific controls?

[English]

Mr. Harder: I'm going to ask Mr. Rummell to speak to this. This is one of the issues he clearly has as one of his priorities, which is that we have in the framework what some of the indicators would be. There are clearly points of functionality and other aspects that are part of a contract's indicators, which the departments are to establish in their project design. But to the extent that we can at the centre of government establish greater clarity to what the indicators might be, that's certainly one of the roles of the chief information officer.

Perhaps you'd like to add to that.

Mr. Rummell: There have been a number of risk indicators or risk assessment techniques that have been applied to projects over the years. Those risk assessment techniques have really been standardized to the SEI capability maturity model. So we have a standardized set of techniques now for managing risk in projects and to judge the amount of risk in a project.

Computer projects, information technology projects, and high-technology projects like CAATS inherently have risk. The important thing is that we will apply measures on a uniform basis for all projects - whether they be large, very large, or medium-sized projects - in order to make sure those risks are effectively managed. We have the capability and the methodologies, as you referred to in your other remarks, to manage those risks effectively.

In terms of the Treasury Board, I am here to help provide some additional focus in these areas. That's one of my prime responsibilities, and I look forward to helping the government in that regard.

[Translation]

Mr. Pierre de Savoye: Mr. Auditor General, are you satisfied with this issue given the concern rather that you expressed?

.1700

Mr. Desautels: As I said, I believe that as such the model which has been developed to be used as a tool to assist in approving projects which are presented by the various departments is very well designed and may play the role that you hope it will play. It can be used as a central tool in order to allow Treasury Board to ensure that projects which are submitted will be managed according to the criteria set forward in the model. At the outset, the model can be used for the purposes that you have in mind.

It is not in this regard that there is a perception of conflicts between ourselves and officials from Treasury Board Secretariat. I think that we basically agree on the way that projects should be managed within departments. What we have proposed as a tool, what we have included in our audit report and what the government intends to do appear satisfactory. There may be some disagreement, and it is rather more philosophical than anything else, as to how far we should go with centralizing the management of Treasury Board for large projects.

Like Mr. Harder, generally, I feel that in terms of regular, ordinary departmental operations, we should treat the departments as competent entities who can provide the services for which they are responsible. Therefore, in taking this position, like Mr. Harder, I would say that departments are fully responsible for delivering their programs. There is, of course, a presumption of competency on the part of departments in the proper management of their operations.

Now in areas such as the new technologies, especially large computer projects, an exception can be made to the general rule. I think that for most departments, these are major challenges that they do not always have the proper ability to face. The risks for the State are very high given the amounts involved; these days we talk of projects worth $300 million or $500 million or more. Therefore, the risks to the State are very high. So we would say, for sectors such as informatics, Treasury Board should perhaps adopt an approach that will be somewhat different from the rule that I stated earlier.

In our view, Treasury Board should ensure that its model contains action plans which would be provided at the outset, when a request for approval for a new project is submitted. We must ensure that this model, called an enhanced framework, be applied even before any moneys are provided. We would also expect that the Treasury Board Secretariat obtain and analyze on a regular basis risk evaluations throughout the life of the project. We would expect that the Secretariat would regularly review progress reports on large projects and advise the ministers of Treasury Board where certain projects present an unacceptable risk.

Lastly, we would expect that the Secretariat would ensure that there is a process which would be put in place for departments and which would ensure that the moneys are not released until the departments have respected all the requirements for each step in a satisfactory fashion.

.1705

If departments have not implemented such a system, Treasury Board must assume this responsibility which otherwise would have been properly assumed by a government department.

In summary, all what I am saying is that we must treat departments as relatively autonomous entities, but in certain sectors such as informatics, where the risk is higher or where expertise within a given department is perhaps insufficient, we would expect that Treasury Board act properly and implement mechanisms to ensure that the level of risk that the State is exposed to is acceptable. Thank you.

[English]

The Chairman: Mr. McClelland.

Mr. Ian McClelland: I'd like to address my comments to Mr. Harder.

As the department that really has the Pathfinder responsibility, if I'm correct, introducing information technology to the public service, how do you go about managing human nature, the resistance to change, particularly in a public service that has not had a lot of new entrants in recent years? In my experience, resistance to change, particularly to information technology, does in some cases - or at least in my case - have some bearing on age. The younger people are, the less resistant they are to change.

How do you manage this? Do you have a proactive plan to try to address this in the public service, if it is a problem? It may not be a problem.

Mr. Harder: Before I get there, let me add that I have no problem with the way the Auditor General has described the role Treasury Board should play when - and I wouldn't say when things go wrong - the orange light is on. I think that's the situation your comments describe.

Your point about resistance to change is very appropriate, I think. It's clear that to transform service delivery and to truly re-engineer work processes there are two essential ingredients. One is technology, which itself is very much changing the definition of work itself, because it provides for the capacity to do things so much differently from before. We're not talking about it mechanizing existing ways of doing our work. Actually, it gives opportunity to do different things.

The second part of it is, how do you deal with your human resource issues that truly deliver on IT?

One of the reasons I was attracted to Mr. Rummell - if I could share with you some of the discussions we had before he joined Treasury Board - was that he has a firm conviction that IT has everything to do with human resource management. It's not a technology issue, in the main.

I think there are a number of things we are doing and more things we need to do. Treasury Board Secretariat is implementing a professional development program with respect to the necessary competencies to manage IT projects. But it's clear that the issue isn't just managing IT projects; it's using IT projects to transform organizations.

We have some evidence of quite a bit of success, whether it's the CADEX project at Revenue Canada or the way in which in only two or three years Human Resources Development Canada today has more kiosks across the country that act as the principal element of engagement with.... For example, 95% of the UI applications are dealt with through that mechanism. That's a transformation of the organization. That's 5,000 fewer people, etc.

But there are some real challenges. You've identified some. One is the demographic structure of the public service. We have not recruited. In particular, departments have strategically used youth summer employment...but now it's the cycle. These students, these people, who are with us for four months at a time, sometimes through internship programs with universities, are quite computer literate and through their contributions have significantly changed the way we do things.

.1710

This is not unique to our jurisdiction, I should add. International observers of IT in government, Don Tapscott for one, point to Canada as being at the leading edge of IT delivery. What I read into this is that others have a long way to go. One of the reasons the position of the chief information officer was established was to give centrality to the question you posed.

Mr. Rummell's third-highest priority is the HR aspect of the IT community itself, but the deputies I speak to are increasingly saying the change is not just in the HR management but also in the broader program re-engineering that requires multi-skilling, a different way of doing, in a sense, our human resources. That's very much part of what we need to look at.

Mr. Ian McClelland: Thank you.

[Translation]

The Chairman: Two of our colleagues have not yet had a chance to ask questions. I will therefore give the floor to Mr. Shepherd and Mr. Grose. We will try to conclude our deliberations before the bells which will ring at 5:15 p.m. for a 5:30 p.m. vote. Otherwise, we will come back later.

Mr. Shepherd.

[English]

Mr. Alex Shepherd (Durham, Lib.): Thank you, Mr. Chairman.

I would like to quote from a couple of the presentations that were made, especially by the Auditor General. He was talking about the CAATS project of Transport Canada, where several key requirements and design issues were still under discussion six years after the signing of the contract.

Mr. Harder, in your presentation you talked about the implementation to changes in the department itself. You stated:

What the Auditor General is telling us, I guess, is that time is money, and we've failed to meet certain deadlines in the past.

If I interpret you correctly, your reaction to that has been that it's going to take upwards of five years to implement these changes and control the process. Do I understand you correctly?

Mr. Harder: No, I think it's a little more nuanced than that. CAATS and a few of the other programs the Auditor General has commented on pre-dated the enhanced framework. The problems they encountered are manifestations of the problems both governments and private sectors had managing large IT projects.

The enhanced framework, I believe, addresses many, if not all, of the concerns that have been identified in examining and trying to improve management of large IT projects. I was saying that the implementation of that enhanced management framework will take some time. For example, some of the professional training one could associate with growing the project management capacity throughout the organization of government may take more than a year, but the systems in place with respect to attestation and gating are in place already.

A framework is just that. It's designed to ensure we have a broader way in which to manage IT. It will take some years for it to be fully implemented in terms of having in place all of the programs and skills necessary to give greater assurance.

Again, I want to repeat that we have not had a major IT problem with any project that has been subject to the enhanced framework for managing large IT projects. We're dealing with the problems that are legacy system problems.

Mr. Alex Shepherd: Presumably what you're saying, though, because it's going to take so long to implement this review process, is that there must obviously be certain aspects of the IT projects that are escaping your review of them in this manner. There's quite a difference in time dimension between one year and half a decade, which are the parameters you've set in here.

Mr. Harder: Not a project that comes to Treasury Board ministers for funding approval hasn't gone through the kind of assessment that the enhanced management framework assumes and that we're assuring.

.1715

There are other aspects of the framework, though, that will take some time to be fully operational throughout the Government of Canada, such as the kind of professional development we must undertake and the kinds of recruitment strategies we must have to meet what one member recognized as an HR issue as well. But the fundamental management is in place, certainly in respect of the funding issues and whether or not we have the gating that's necessary and whether or not we are prepared to call the shot to terminate a project when necessary.

Mr. Alex Shepherd: Something that always mystifies me is when we talk about this year 2000 project. Getting back to timeframes, here it is the middle of 1997 and people are saying these computers are all going to go on the fritz in the year 2000, which is in two and a half years. Where are we on resolving this?

Mr. Harder: I'll ask Mr. Rummell to respond, because he, most recently in the private sector, has had some acquaintance with the year 2000.

Mr. Rummell: I've had the good fortune of doing a lot of work with different companies and spending quite a bit of time talking about this, trying to sensitize the public to this on television in western Canada. In fact I've done some national television on this.

The year 2000 problem is a universal problem. It's not only a problem we have here in Canada in the Canadian government; it's faced by the major corporations and all businesses across Canada. We have been addressing this problem for several years in the government.

At this point many of the departments have done a very fine job in addressing this problem. We just keep escalating in advertising the importance of updating their systems to make sure this will not cause us a problem. We in the Canadian government have a very important role to play and we are playing a role to take a leadership position in solving the year 2000 problems.

I'd also like to thank the Auditor General for bringing this issue forward to the deputy ministers and sending out a letter about his concerns regarding this problem. I thank them for doing that, because that's going to help us solve the problem sooner rather than later so we can get on with the development of other systems that are probably more beneficial to the Canadian public.

The Chairman: Colleagues, the bells will start only at 5:30. Maybe we'll have a couple of minutes to come back if you want.

Mr. Grose.

Mr. Ivan Grose (Oshawa, Lib.): Thank you, Mr. Chairman.

Incidentally, I'd like to thank you, Mr. de Savoye, and I hope you continue with your preambles, because they're of great assistance to me in understanding just exactly what we're doing. I loved your analogy last meeting, when you said it's like building a house of concrete and after you get the walls up you decide to put the windows in. Somehow that rang a bell.

I should have asked this question first, but it's the luck of the draw. Mr. Harder, you gave us some odds on systems. You sounded like an odds-maker. Back home we call that a bookie, but Mr. Guimond says you're on a much higher level than that. You gave us some odds on delivery of a system. Looking at those odds, I wonder why we're doing this at all, because according to your odds, the chance is not good that we will get the system on time, if at all, and if we do get it, it will not be in the format we want it in.

As part of that question, but possibly my second question, are we that unique that we have to design our own systems? Could we not wait a bit until someone else who's in somewhat the same business as us has designed a system and gone broke in the process? Then we could buy or license the system from them.

It was mentioned about the air traffic control system that the United States was designing a new one of their own and apparently gave up. Maybe they know something we didn't. We've apparently decided to soldier on at additional cost.

So my question is, do we have to be at the forefront of this?

Mr. Harder: Thank you very much.

The document to which I was referring was by, as I indicated, the Standish Group. It's somewhat historical. The data itself was a study that stimulated both private sector and governments to ask, ``How do we mitigate these risks? This is crazy.'' It was a long list of things such as narrowing the time scope, chewable chunks, gating all of these sorts of practices. I don't think the answer is ``Don't use technology, or adapt technology'', but ``Boy, we'd sure better be more prudent in thinking that technology will somehow whimsically fall into place without being managed''.

.1720

Your question is a very good one, and that is, where on the cutting edge of technology's application should government prudently sit? Your observation that too often...or that certainly there have been cases where we probably made an inappropriate assessment of that risk; that we should have waited for either something off the shelf or actually admitted to ourselves that the best available optimization might not meet all of the functionality we would like, but was a more prudent risk in terms of our contracting....

I think there have been enough failures for us to take that into account in some of the procedures that are in place. That study was certainly key in changing the way private and public sectors have managed technology.

Mr. Ivan Grose: Thank you. Does the Auditor General have anything to say about that? I didn't hear a direct comment about whether or not we should be in this at all. You simply commented on the success or failure of the systems.

Mr. Desautels: Well, Mr. Chairman, we've been on the record before as saying that it is very important for the federal government to reap the benefits of technology.

If we are to be able to deliver to taxpayers the services they expect, and still do that within reasonable cost constraints, I think we have to exploit technology smartly. That might involve certain risks, but I think those risks can be mitigated by proper management systems. We have to gauge those risks, but I think we have to move forward and try to reap, as best we can, the benefits that technology offers to a large national government like Canada's.

Mr. Ivan Grose: Thank you very much.

Thank you, Mr. Chairman.

[Translation]

The Chairman: Mr. de Savoye would like to have another turn. Since we have a department representative before us, we will try a little exercise. Mr. Gauvin, is the enhanced framework for the management of information technology projects which was issued by the Treasury Board Secretariat on May 28, 1996, currently in force at Transport Canada?

Mr. Gauvin: Yes, Mr. Chairman.

[English]

I should say that Transport is considerably different from what it was in the past. We were a very large department and we were in very large systems, and CAATS was the biggest one of all. Transport is considerably smaller now. We're down to about 4,000 employees, and a lot of our operations have been commercialised or privatized, including the air traffic system, and CAATS was transferred with it. But from the experience we have had, we certainly would be a lot more careful and prudent in the future.

I'd like to reply to the honourable member who just had a few comments in terms of why can't we take systems that are already there and have been proven. There is a little bit more complication in the government in applying any system, because even a straight financial system or material management system - you can't really go out, buy it off the shelf, implement it, and expect that it's going to work in government.

In government we do things differently. As an example, the department doesn't purchase; it has to go through Public Works or Government Services. In government the department doesn't issue its own cheques. Again, we have to go through Government Services.

What we do is we take a system that has been proven worldwide and is accepted off the shelf. Then we have to customize it, and that's where a lot of problems occur, because we have to interact and interface with other systems. We take something that's working, and by the time we're finished we've complicated it so much that it creates a lot of problems for us. So part of this is also the system we work within.

.1725

[Translation]

The Chairman: Mr. Gauvin, I will not ask you my question in English because I just cannot speak in English to another francophone. Is the enhanced framework for management currently in force at Transport Canada?

Mr. Gauvin: Yes.

The Chairman: Could you undertake to present the committee with the action plan that you have adopted for the implementation and the follow up that you will be carrying out?

Mr. Gauvin: Yes. Rest assured that when we launch a new and large system the new process is in force.

The Chairman: You confirm that it is in force and we are taking note of the fact that you will provide the documents to the clerk of the committee.

Mr. Harder, at the present time, in how many departments is this enhanced framework being used?

Mr. Harder: In all departments.

The Chairman: It is therefore being applied in all departments. Do all departments that have projects involving new technologies send you a copy of the action plan for the enhanced framework implementation?

Mr. Harder: Each project in each department must be managed in conformity with the framework.

The Chairman: Mr. de Savoye.

Mr. Pierre de Savoye: Mr. Chairman, I would not have made this comment if this meeting were not being broadcast to the general public, but it is.

One of my colleagues mentioned resistance to change and indicated that this may have something to do with age. That is a myth. I will give you the example of a business where a young person is very familiar with a particular software; in fact, he is the expert, the guru. The software will have to be changed. Who, do you think, will resist the most? Those other individuals who are not very familiar with the software? Not at all. It is this one person because he is the one who has the most to lose; he will lose his status. Resistance to change often stems from the balance that the person establishes between what he stands to gain or stands to lose, and what type of effort is required. Therefore, to settle the problem, the company sends the individual to the United States for two weeks to follow a training course. It is a nice trip and the person comes home very happy. He is still the expert.

Resistance to change has nothing to do with age, but rather with the way we perceive the fairness of what is happening to us.

Thank you, Mr. Chairman.

The Chairman: I will ask Mr. Desautels to make his concluding remarks.

[English]

Mr. Desautels: I'll be quite brief, Mr. Chairman. I think this was a very useful hearing. I learned a few things today myself as well.

In closing, let me just say that these are early days, but I'm quite supportive of the new enhanced framework and I'm looking forward to its successful implementation. I would like to wishMr. Rummell the best of fortune in ensuring that this is applied properly, government-wide. I can assure the committee that, as with everything else, we will do further work in that area in due course, and we'll be able to report back to you on how it's working.

Let me just raise a second point in that while we now have the new framework put in place and applied to new projects, there is still quite a substantial amount at stake through other projects that have started before the new enhanced framework was put in place. I'm told there is about $5 billion worth of work not covered by the new framework. We would encourage the Treasury Board Secretariat to pay particular attention to that, even though it falls outside of the framework.

Finally, we too are quite interested in the year 2000 issue. We are doing some work in that area now, and we will be reporting back to this committee on the state of preparedness of the government in this area sometime next fall.

[Translation]

Thank you, Mr. Chairman.

The Chairman: Thank you very much, gentlemen, for your co-operation.

The Standing Committee on Public Accounts will now adjourn till Wednesday, March 12, 1997, at 3:30 p.m.. Thank you.

Return to Committee Home Page

;