Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Legal and Constitutional Affairs References Committee
01/11/2019
Impact of changes to service delivery models on the administration and running of Government programs

RIZVI, Mr Abul, Private capacity

[10:00]

Evidence was taken via teleconference—

CHAIR: Welcome, Mr Rizvi, and thank you for taking the time to give evidence today. Information about parliamentary privilege has been provided to you and is available from the secretariat. The committee has received your submission as submission No. 17. Do you wish to make any corrections or additions to your submission?

Mr Rizvi : No, thank you.

CHAIR: Do you have any comments to make on the capacity in which you appear?

Mr Rizvi : I am a former deputy secretary of the department of immigration.

CHAIR: Would you like to make a brief opening statement before we go to questions?

Mr Rizvi : Yes, please. Australia's immigration arrangements are heavily dependent on the efficacy of the ICT platform on which they operate. Continuous improvement of the ICT platform has been, and will continue to be, essential but the crucial question is whether the best way to develop the next phase of the ICT platform on which all visas are processed is by transferring ownership to the private sector. In this regard there are four crucial issues on which I have not, to date, seen an adequate response, and that includes in the Department of Home Affairs' submission.

Firstly, what drove the Department of Home Affairs to choose the privatisation option? Was it the staffing cap, was it the difficulties the Home Affairs leadership is having with staff morale, or was it a reluctance on the part of the Department of Finance to adequately fund the next phase of development despite the significant and growing level of excess revenue from visa application fees compared to resources devoted to visa processing?

Secondly, what arrangements does the Department of Home Affairs have if the private provider, like all private providers before, underestimates the cost of developing such a complex ICT platform? What contingency does the department have if the private provider chooses a cheap form of rebuild that is effectively putting lipstick on a pig, rather than what it may have first promised? Is the department prepared to bail out the private provider with additional funds or allow more time for a proper build?

Thirdly, once visa processing has been transferred to the new ICT platform owned by the private provider, what guarantees has Home Affairs given on the minimum rate of return the private provider will receive and how the private provider will be paid for the regular changes to the ICT platform that government will inevitably demand due to policy changes? Given Home Affairs cannot possibly afford for the private provider to get into financial trouble, will Home Affairs have to agree to whatever visa-application-charge increases the private provider demands? How will Home Affairs deal with any natural decline in case load due to either the economic cycle or policy changes? What if the massive increase in case load that has been assumed is not realised? Will Home Affairs compensate for that? Is there a risk that having to make such compensation payments could affect how Home Affairs advises on immigration policy?

Finally, given the pressure on the private provider to provide strong returns to its shareholders, how will Home Affairs deal with relentless demands on it to allow the private provider to use the ICT platform to increase profits, but which are negative to the national interest? How can we be confident Home Affairs will be able to resist such pressure, especially as the opportunities requested by the private provider may proceed in small, incremental steps?

Without adequate responses to these four sets of questions, I would strongly suggest the committee recommend that the proposed privatisation not proceed or, if it is to proceed, that Home Affairs ensures it has in the contract an adequate get-out-of-jail clause that minimises cost to the taxpayer when the ICT platform has to be brought back into public ownership.

CHAIR: Thank you very much. Perhaps I could kick off with a few questions. You've used the term 'privatisation'. You would have noted that the department is actually claiming that this is not a privatisation. What do you say to the suggestion that this is not a privatisation?

Mr Rizvi : I would acknowledge the department and the government remain in control in terms of the legal aspects through the Migration Act. No-one's arguing that. No-one has. The fact that the government keeps pushing that issue seems bizarre to me. No-one has suggested that that's been changed. What is being changed is that the ownership of the platform on which visas are processed will go from public hands to private hands. To me, that is privatisation, pure and simple.

CHAIR: So you're basing your case on the ownership of the asset?

Mr Rizvi : Correct.

CHAIR: The department manages visa processing through 50 computer systems that have been developed over 30 years. The IT visa application system is over 25 years old. Recognising just how long your experience is in this area of the department of immigration—through various names and structures—can you tell the committee whether the department has a successful record in managing IT projects, in your direct experience?

Mr Rizvi : There have been many instances when the department has managed these IT projects extraordinarily successfully. The transfer to the electronic travel authority in the mid-1990s was one very good example. We were world leading at that time. Indeed, we were criticised by others around the world for being so silly to establish such a system, yet 25 years later the rest of the world has followed suit. So there are examples when the government, the department, has successfully implemented major innovative IT transformation. But those, in my experience, have generally been due to their management by public servants with extensive knowledge of visa requirements, how the existing IT systems work and what is needed. The real risk is bringing in someone from outside who has little knowledge of immigration requirements and little experience of what can go wrong in this space. There was a very major problem with IT system development in the department when we partnered with IBM in around, I think, 2008-09. That led to a major cost blowout and we did not get the kind of outcome that we were looking for.

CHAIR: And your argument is that, if you're not in control of the platform, you lose control of the capability to actually adapt to changed circumstances. Is that the thrust of what you're saying?

Mr Rizvi : I think the private provider will happily change the IT system to whatever the government wants. It is a question of what the government is prepared to pay.

CHAIR: I see. So let me just go through this. You're suggesting that the business case has been deficient. Do we know if there has been a business case put forward?

Mr Rizvi : Not one that is publicly available. I assume there's one that's internally available.

CHAIR: You've suggested that there'll be mounting pressure to change the cost structures of the proposal. What led you to that conclusion?

Mr Rizvi : I think what led me to that conclusion is past experience with private providers rebuilding IT systems in the public sector—the frequency with which these have fallen short of what was required, the cost blowouts and delays. There is no reason to have confidence that we won't have something similar, given the ambitious nature of what is proposed. And, at the end of the day, the private provider will need to have a return on this massive investment, and I think the $35-per-visa application fee is only the start of the process here.

CHAIR: The department is handling, what, nearly 10 million applications a year now; is that right?

Mr Rizvi : I think that's about right, yes. But that includes large numbers of visitor applications, which are mainly automated anyway now.

CHAIR: Yes. So you're suggesting that this arrangement will not solve the backlog problem, given that it seems to be growing quite substantially, and there's the risk of a further blowout in waiting times and a reduction of departmental funding for the visa program. I'm just wondering if you can expand on that. What's the basis of your assertions on that matter?

Mr Rizvi : As I understand it, the visa platform development may take two to three years. We don't know. It could take longer. Once the visa platform has been developed, the visas that will be automated, as I understand it, will generally be very simple visas. That is not where the backlogs are. The bulk of the backlogs that the government faces is, in fact, in other places. They are in spouse visas, asylum seeker visas, citizenship applications and more complex visas. I do not see the transfer or the automation of those visas happening anywhere in the near future. They are a long way away, which means the government will continue to have to deal with the backlogs.

CHAIR: You've indicated that there's no publicly available cost-benefit analysis for this proposal. I'm at a loss to understand what the argument is in favour of this proposal. Has there been a public statement or a public argument as to why we need to actually undertake this privatisation?

Mr Rizvi : I read through, very carefully, the Home Affairs submission and I think it utterly failed to make a business case. It failed to make a business case in terms of how the public interest benefits, and, moreover, it failed to deal with how the extraordinary risks involved will be dealt with. Indeed, it actually just ignored all of those risks.

CHAIR: In my experience with these matters—and I've watched this now for going on for 27 years—the biggest issue is that once the government outsources a capability it's extremely difficult to rebuild it. Do you see any evidence that the Department of Home Affairs is building or maintaining an internally controlled backup system so that, if a new consortia can't hold a system together, the government can actually reassert its capability?

Mr Rizvi : That would be an intelligent fallback option to have in place. Whether the government has invested enough money or has the capacity to build a backup system, as well as this extraordinary new system that they're talking about, I don't know. It just sounds unlikely to me, although, as you've suggested, it would be a sensible thing to do.

CHAIR: Yes, but do you know of any evidence that that capability is being duplicated anywhere within the department?

Mr Rizvi : I'm not aware of any, and there's no mention of it in the Home Affairs submission. If they were developing such a capability, one would assume they would have mentioned it.

CHAIR: If the winning contractor is demonstrated to be underperforming, do you think it's feasible for the Commonwealth to actually terminate the contract?

Mr Rizvi : I can't see, without a backup system, how it could terminate the contract. It's like saying, 'Turn off the visa system for a few years'—it's just implausible.

CHAIR: That raises the issue, does it not, about what the actual real cost is?

Mr Rizvi : Absolutely. The real cost in this, I believe, is ultimately a loss of control in the visa system.

CHAIR: What about the question of the integrity of the visa system? In our previous iterations, Dr Rizvi, when you were on the other side of the table, I pressed you on many occasions—

Mr Rizvi : For many hours.

CHAIR: Yes, for many hours. I know. I remember you enjoyed it immensely. But there were serious problems with fraud; is that right?

Mr Rizvi : Yes.

CHAIR: The department has always had a problem with dealing with fraud, and problems of criminality—

Mr Rizvi : Yes.

CHAIR: particularly in the student visa area. They've had problems in terms of organised crime. What provisions are there, within these arrangements, to ensure the integrity of the visa processing system in terms of the risk liability?

Mr Rizvi : I think the Department of Home Affairs would argue that the responsibility for checking visa integrity will remain with it—and I accept that. I think that's probably correct. The problem I see emerging is if the government allows the private provider to move down what is known as the premium service stream. The premium service stream would involve a stream that is promoted by the private provider as being a bit like a first-class seat on an aeroplane. It's the one where you pay a very large amount of money, for which you get much faster service and, I think—hint, hint, nudge, nudge, wink, wink—a more likely 'yes' than a 'no'. If the premium service stream doesn't deliver a more likely 'yes', how in the world would the private provider ever promote it?

CHAIR: Can I make this suggestion: the protections contained within the tender would be difficult, if not impossible, to implement without causing significant disruption to our visa processing system. Would that be fair conclusion to draw from the evidence that you're putting forward?

Mr Rizvi : I think that would be a very kind way to put it.

CHAIR: Thank you.

Senator STOKER: Thank you very much for appearing before us, Mr Rizvi. Your experience is really helpful. You said something pretty provocative about the premium service stream that you just mentioned. As I understand it, it's designed to give faster processing times—is that right?

Mr Rizvi : I don't know. I don't know whether the premium service stream is an option the government will pursue or allow a private provider to pursue.

Senator STOKER: Right.

Mr Rizvi : I'm merely going on what happened in the United Kingdom.

Senator STOKER: So your suggestion that a premium service stream would increase the prospects of approval is not really much more than speculation in the Australian context at this point in time?

Mr Rizvi : It's certainly the case in the United Kingdom. Just think about it logically, Senator Stoker: how will you promote a service stream which is going to cost you many thousands of dollars extra to apply for if you're not confident you're going to get a yes? If it's an equal chance of getting a no, why would you pay?

Senator STOKER: Because speed has a commercial value and a comfort value for many people. But it strikes me that the answer which you have actually provided in relation to the Australian context is that it's speculative at this point in time. Can I move to a different issue, please—

Mr Rizvi : Well, it's speculative because the government has not yet decided whether to allow such a thing.

Senator STOKER: That's right. Exactly. Do you accept the proposition that the purpose of a contract is to set out the terms of an arrangement between two parties? Someone agrees to provide a particular service and another agrees to pay a certain sum—at its most basic. Do you agree with that?

Mr Rizvi : At its most basic, yes. But contracts are much more complex, as you would know—

Senator STOKER: I understand—we'll get to that! One thing at a time. If the purpose, once we get to more complex contracts, is to set out in some detail the precise scope of the work that is being provided for in that contract, including providing for different permutations and different things that can crop up in the course of its execution—and, similarly, to provide for the cost of the provision of that service or product and any of those permutations—is there a degree of certainty provided by the fact of having a contract in place that you don't have in circumstances where a product is being built internally?

Mr Rizvi : I would disagree with that. In my experience, more often than not, dealing with a private sector contract increases complexity very significantly and it increases uncertainty very significantly. That has been my experience, and it has certainly been my experience where the government has privatised natural monopolies. Wherever the government has privatised natural monopolies it has got itself into deep, deep trouble.

Senator STOKER: Tell me this from your experience: when you have a product being built from within the department, is there a notional contract in place that sets the parameters of the work to be done and how much will be charged for it?

Mr Rizvi : Yes, there is a notional contract in place. I never actually ran the IT side of the department, so I was heavily reliant on what the IT people could deliver. That involved, always, a very extensive process of negotiation in terms of what needed to be delivered and what it would cost. Yes, there was that sort of contract. What there wasn't, was a profit motive.

Senator STOKER: Right. I understand your point about profit motive. Do you understand, or do you accept, the converse? We're looking at the pros and cons of each. We set out the pros and cons—particularly the cons—of a commercial arrangement. Do you accept that when we consider the pros and cons of developing something internally to the department that we don't have a competitive tender process which can help to drive better efficiency and we don't have a contract in place that can help to provide certainty about the costs, such that there can still be cost blowouts and expensive processes that arise when a product is built internally?

Mr Rizvi : There can be.

Senator STOKER: It's not a trick question; it's just acknowledging the differences.

Mr Rizvi : I gave two examples to Senator Carr. One example was in the mid-nineties, when we developed the electronic travel authority—a world-leading development, developed by the department's internal staff. That was highly successful. It was world-beating. Indeed, the rest of the world has copied it. In the other example, when we partnered with IBM in 2008-09 to develop a very big redevelopment of the platform, the cost blow-outs were far greater, the delays were far greater, and what we got was much less than what we expected.

Senator CHANDLER: Mr Rizvi, the first issue that you raised in your submission was around what drove the Department of Home Affairs to move to privatise this ICT platform. You felt that there wasn't necessarily justification provided for that decision. But I note that the RFT documents state—and I'm referencing from Home Affairs' own submission here—why the proposed reforms are required and, indeed, established as five or six intended outcomes of the process that stipulate why we're going out to tender on these services. Do you accept that that does provide some justification of why the department has moved to outsource this platform?

Mr Rizvi : No, I do not. All they are stating is they are seeing an increase in the size of the caseload. Yes, that's happened many times in the last 30 years that we've seen rapid increases in the size of the case load; that's nothing new. Secondly, it's pointing to examples of people trying to defraud the visa system. Yes, we've had that for many, many years; that is not new. They then talked about increases in risks to national security. Yes, that's been going on for a long time as well. What they haven't explained is why a privatised system is better at solving that than an in-house-built system. There is nothing in their submissions that explains that.

Senator CHANDLER: But doesn't that go back to the questions that Senator Stoker was just asking—that there will be costs incurred in establishing a new IT system to deal with this? I think the department is pretty clear in its submission in terms of the age of its current systems and that a new one is required. It's going to—

Mr Rizvi : I can assure you the base systems that Home Affairs are referring to—I was there when they were developed—were intended to be replaced when many iterations of development of the ICT system have taken place. The problem is that those core systems contain massive amounts of old data which is relied upon by the new systems. The new system will not do away with the old systems. What it will do is build on top of them.

Senator CHANDLER: Right. Well, I think we'll have a couple of questions for the department, then, later in the day to get—

CHAIR: When I was Minister for Human Services, this was the same problem—the Cuba system, I think it was called. It is also a problem where governments have not invested in upgrades to the point where you then find that there's a difficulty with the scale of capital investments required. So these so-called quick fixes then become expensive over the longer-term but appear to be cheaper in the short-term. Isn't that the argument that's often presented to the expenditure review committee and the like? But there is a cumulative effect in failing to upgrade systems. You get to a point where obsolete systems are built upon and built upon and repaired. I know, in the Human Services area, we have to retrain people on systems that are effectively 25 years old, and they're not run anywhere else. We have to actually have special programs to operate obsolete systems because the investments have not been made by government over time to replace them when they're due for replacement. It's the same in the science program, where we've built money into our super computer systems. They are due to be replaced at a certain time but the money is not found. Look at what is happening up at ANSTO at the moment.

Senator CHANDLER: I'm not necessarily refuting that point, Senator Carr. I guess where I'm coming from in my line of questioning is to say that we are where we are. We could wind the clock back 30 years and build a system that's designed for now so to speak, but if we've got this issue then are we not better off getting the best possible system in place to correct the—

CHAIR: The same problem will occur.

Senator STOKER: Can I suggest that we focus on the evidence rather than debating one another?

CHAIR: Sorry. Mr Rizvi, will the same problem occur when this system's due to be replaced—whatever the system is that you put in?

Mr Rizvi : What you're pointing to, Senator, is an issue that exists in tax, home affairs, customs—just about every large client service agency has the same issue. It's not as if governments haven't had presentations on this issue in the past, they have. The problem is every time governments have moved down this path the problem of replacing those 25-year-old systems is so incredibly difficult that the providers, or the relevant agencies, have always fallen back to building on top of those systems rather than replacing them. I have seen nothing that suggests to me that any different will take place in this instance.

CHAIR: Sorry to interrupt you.

Senator CHANDLER: No further questions. Thank you.

Senator STOKER: Mr Rizvi, you mentioned towards the end of your questions with my colleagues your concern that such an arrangement would mean that a contract can't be terminated and that there would be problems with the ownership of the platform. Do you remember saying things to that effect?

Mr Rizvi : I remember saying that it would be really, really valuable for this contract, if it proceeds, to have a mechanism in it that enables the ownership of the system to come back into public hands if and when problems occur.

Senator STOKER: Great. That means you and I are on the same page in saying that these are things that can be provided for in the contract—both the arrangements for termination and what circumstances and the arrangements for ownership when the contract comes to an end—

Mr Rizvi : And they come at a massive—

Senator STOKER: If I can finish, Mr Rizvi. Let me finish the question—

Mr Rizvi : I thought you were making a statement.

Senator STOKER: No, I'm asking a question. My question is: these are all things that are able to be negotiated and provided for in a contract aren't they?

Mr Rizvi : At a price, Senator—

Senator STOKER: Thank you.

Mr Rizvi : It is a question of whether you are prepared to accept that price. If you are a government that is prepared to accept any price, then yes. But I have found in my experience that most governments are not prepared to accept any price.

Senator STOKER: Thank you.

CHAIR: Thank you very much, Mr Rizvi. Your information today has been very, very useful to the committee's deliberations. That will conclude our considerations for this time.

Pr oceedings suspended from 10:28 to 10:47