One of the challenges faced by governments when transitioning to a digital services delivery model is one of timing. With vast legacy IT estates and contracts in place, deciding on the optimum time to migrate to a new approach is a challenge.
Given the mission-critical nature of public services, is one that must be met head on. These are IT systems whose failure or lack of availability have genuine human consequences. If the IT isn't in place to collect taxes, for example, governments faces a funding crisis that in turn would have a knock-on effect on the ability to deliver other essential services.
That's precisely the difficult position the UK government finds itself in right now. The IT underpinning the operations of Her Majesty's Revenue and Customs (HMRC), the UK tax collection authority, is currently managed under the auspices of the ASPIRE outsourcing contract, with prime contractor Capgemini which counts on the deal for 60% of its UK revenues.
While ASPIRE was once lauded as an exemplar for how to do outsourcing contracst, it's now emblematic of the type of 'big ticket', multi-billion pound, multi-year IT engagement that the UK government wants to see an end to.
ASPIRE formally ends in 2017 with the plan being that it will be replaced by (a) a more digitally-centric approach to service delivery and (b) the involvement of a lot more suppliers hired on shorter contracts with values of less than £100 million. (By the end of its life, ASPIRE will have cost the UK taxpayer £10.4 billion, considerably more than the planned £4.1 billion.)
So far, so sensible. But what's started to alarm legislators in the UK government is that with just over 2 years to go until the end of the contract – and a hugely disruptive election due next May that will bring most government activity to a halt for months – there is at this late stage no formal business case for digital transformation in place from HMRC. In fact, attempts to renegotiate with the existing ASPIRE suppliers have been rumbling on for 2 years, still with no conclusion.
A scathing report from the National Audit Office earlier this year rang alarm bells when it warned that time was running out and that it was concerned by the absence of planning in place. Couple that with what appears to be a dogmatic insistence from the Cabinet Office, which manages IT for the UK government, that there will be no extensions to existing legacy contracts as they wind down and suddenly there's a lot of nervousness around about whether HMRC will be able to guarantee vital tax collection come 2017.
But it works...
There's another factor at play here as well. Despite the fact that there's considerable consensus that the existing ASPIRE contract has cost far too much and that the deal was unnecessarily extended only 3 years into its original lifetime when it still had 7 years to run, it....er, well, it works! It delivers the revenues to the government that enables it to function and provide the services it must do to its citizens.
Or as Margaret Hodge, chairman of the feared House of Commons Public Accounts Committee, puts it:
It may have been stupid ten years ago, but it’s working.
So is it worth the risk of pursuing HMRCs digital aspirations on the current schedule when it seems that the most basic groundwork has yet to be completed?
At a legislative hearing of the PAC this week, HMRC's main adminstrative official, Permanent Secretary Lin Homer was on the defensive:
The approach when we put this procurement out in 2004...was [one of] a huge effort for 10 years, and then not very much. This is about procurement as we go forward, so we have to increase not only our technology skills but our commercial skills. We also need to ensure that we are better at designing what we need.
We have taken a long time to scope and build some things, but we have then been locked into them for quite long periods. Some of that has meant that we have not been able to take advantage of changes in the market, not only in the way that things are done in HMRC but in the way they are done in the market.
We have to have the kind of designing and architecture skills that we do not have. There are opportunities and risks throughout, and I suspect that, as we go forward, as with all risk management, we will not avoid all of those risks, but anticipating and mitigating them will be a big part of the shift.
What's the plan?
OK, but it's nearly 2015 and the deadline is 2017, so where's the business plan that's needed to be signed off before things can progress?
Everyone has known for years when ASPIRE's time was up, so why at the end of 2014, isn't there a published business case and plan on view for where HMRC goes from here? Again, Honer is on the backfoot, spouting high-level generalizations rather than detail:
The pace at which we are planning our digital work is not as fast as some people would like it to be, but we believe it is deliverable.
We have not reached the point of making firm decisions, but we are clear that, in the future, we will not contract for a decade through one prime and withdraw from maintaining the expertise internally to be a strong and active client.
We would want to do that to get better value for money without losing quality or increasing risk and we would want to enable ourselves to move speedily with the changes which we believe will continue to happen at pace in the technology environment.
That's just the blandest missions statement and vague objectives. Where's the detail?
It's left to HMRC's newly appointed director of digital Mark Dearnley to try to fill in the blanks. Poached from Vodafone, Dearnley comes at this with the advantage of course of nothing up till now having occurred on his watch. That said, he's the one who's now had the buck passed to him, so he needs to come up with the goods.
He argues that some progress has been made since the NAO report was published, especially in the area of bringing more design and project work inside to HMRC with 40% by volume of all projects now run by HMRC staff:
These are largely in the digital space, because obviously that is our main growth area. Now, as of right now, that is still only 20% of the spend, because a huge proportion of our spend is on infrastructure and the projects around infrastructure, and they are still with ASPIRE; but that will start to move as we move more infrastructure towards a cloud-based model.
In the past year we have changed the organisation, so that we have got clear accountability, particularly for operations. We are now starting to do this with more contracts, where we have been out to the market. Admittedly, it is not a large number yet, but we are starting to manage things operationally and in delivery more directly ourselves.
One piece of transitional work that we have to do ahead of migration is to redesign our business processes and all our systems, so that we can move from a world where there is one very large broken-down bill every month with Aspire, to a world where we can manage 400 suppliers and a supply chain for those.
Dearnley does point to digital successes within HMRC:
There are all the digital exemplars we are doing with Government Digital Service (GDS). We now have three of those: one that allows you to switch over from getting paper from us to getting digital messaging from us; one that allows you to update your company car tax—it is the first in a series around PAYE; and one that is for business, so that all their tax affairs are in one single account space.
The other big one is that we are digitising all our forms. We have 300 live so far. Today we have moved them from being paper to being a version that you can use on your phone or your tablet. If you go to one today, there is a button that says “Print” at the end, so I am not entirely happy. In a few weeks’ time, we will have the first one that says “Submit”.
We had 1,700 paper forms. By doing the i-forms, we have managed to prove that, from memory, about 300 or 400 were obsolete, but we were still printing them. We will rationalise the whole thing and we will get it down to about 500 or 600 in the end. We hope to have all those forms online by the end of this financial year.
There is a lot to do, but I think pace is really important in this because we are talking about a lot of people’s lives. One of the really important things to do is to give them clarity and certainty as we go through. I don’t think I would sit here and ask for more time.
That's just as well of course as the Cabinet Office and in particular Chief Technology Officer Liam Maxwell have been adamant and highly vocal that there will be no extensions to existing contracts once they reach their termination point.
But here's where things took an interesting turn yesterday as it emerged that those so-called red lines laid down by the Cabinet Office might not be quite as color-fast as they have been presented.
Maxwell yesterday said of the red lines:
All of those are provided on the basis that we do provide exceptions. The confusion that the red lines would be the dogmatic implementation of a diktat is not the way that we have designed it, nor is it the way that we have implemented it. When we've talked to departments and when we致e seen business cases where it痴 sensible to wait for a bit and we can help to implement it in a clearer, more effective way, we have always done that. We are pragmatists at heart.
So this means that despite all previous declarations and despite HMRC having had literally years to put its house in order, it's possible that a contract that is described as over-priced may in fact be extended after all? Maxwell says carefully:
If HMRC said ‘we’ve had a go and the only option is to extend the contract’, if there was a reasonable and strong case to do so, we could extend it...I think if the business case appeared in December and it was an unconvincing piece of work, we would have an idea of whether we need to extend then.”
But at the end of the day, Maxwell cannot directly make a decision about this:
I can formally make a recommendation that it doesn’t proceed. I cannot formally stop it. That is a Treasury thing.
That said, he's the one who'll carry the can:
If it goes wrong, I am accountable.
This just cannot be allowed to go wrong. HMRC's tax collection capabilities cannot be exposed to levels of risk that would impair revenue flow into the government.
If that means ASPIRE being extended, then that will have to happen.
But if that is how this plays out, then serious questions will have to be asked about when the right commercial and technology leadership skills will be in place at scale to cope with some of the most radical (if entirely necessary) reforms of how IT services are procured and delivered in government and when full-on digital transformation can be tackled with confidence.
HMRC's Homer's performance in front of the PAC did little to inspire confidence. When asked about whether HMRC was comfortable with the Cabinet Office's 'red lines', she was non-commital in her response:
It may well be that my successor in 10 years says, “That was really interesting. We have had 10 years of doing it that way and are now swinging back to having big, monolithic contracts.” One isn’t to say, but there are pluses and minuses.
Now I may be overinterpreting here, but that sounds more like a safe bureaucratic answer than an enthusiastic endorsement of a radical new way of operating.
That has to change, as the cultural resistance to change posses an enormous threat to successful delivery of a new way of delivering public services.
At one point during the PAC sessions this week, talk turned to how GDS and the Cabinet Office advise and assist government departments, but can run into barriers. Was this about Universal Credit, demanded one member of the PAC. Maxwell's uncomfortable acknowledgement spoke volumes.
Everything now hinges on Dearnley pulling a business case out of the bag before Christmas, which is the time Maxwell hopes to be able to come to a conclusion about whether HMRC sticks to its original schedule or whether Capgemini gets to keep its own revenues topped up for another year or so.
Maxwell admitted yesterday:
Would it make life easier? It would make life easier to do.
He then posed a second question:
Would it be a good use of public money?