Science and Technology Committee
Oral evidence: Digital Government, HC 1455
Tuesday 27 November 2018
Ordered by the House of Commons to be published on 27 November 2018.
Members present: Norman Lamb (Chair); Bill Grant; Darren Jones; Stephen Metcalfe; Damien Moore; Martin Whitfield.
Questions 1 - 80
Witnesses
I: Tom Loosemore, Partner, Public Digital, and former Deputy Director, Government Digital Service (2011–2015); and Dafydd Vaughan, Cedyrn Ltd.
Witnesses: Tom Loosemore and Dafydd Vaughan.
Q1 Chair: Welcome, both of you. It is very good to see you.
Darren, do you want to mention your interests?
Darren Jones: I need to declare my interests in the register, including the fact that a couple of years ago I acted for the Post Office in its contract for Verify. I take no responsibility for the success of that project; I say merely that the contract was of the highest quality.
Q2 Chair: You have got that on the record.
Would you like briefly to introduce yourselves? We will then get on to the questioning.
Tom Loosemore: I am a partner in a consultancy called Public Digital. We work with Governments around the world to help them to digitise. For the purposes of this session, I was one of the founders of the Government Digital Service, of which I was deputy director between 2011 and 2015.
Dafydd Vaughan: I run Cedyrn Ltd. For the purpose of this session, I was a software developer, technical architect and technical lead at the Government Digital Service between 2011 and 2016.
Q3 Chair: Will you start by defining what benefits were attached at the very start of the Government Digital Service? What was it seeking to achieve?
Tom Loosemore: Its genesis was a report that Martha Lane Fox wrote in the summer of 2010 for Francis Maude, who was then the Cabinet Office Minister. I helped to write that report. It was deliberately a very large stone thrown into what I would describe as a fetid pond. It said that not only were we being absolutely left behind by people’s expectations of what internet-era services should be able to deliver, but the Government were spending billions of pounds very poorly, in a way the rest of the world had moved on from in how you design and deliver services.
Q4 Chair: There was a real sense at that point that we were behind the curve internationally.
Tom Loosemore: We were massively behind the curve of what the private sector, and even some of the third sector, was able to deliver, not just in the simplicity, elegance and empathy of services, but in cost and the ability to design services that started with the needs of users—of citizens.
Q5 Chair: At that point, were all Governments behind the curve of what had happened in the private sector?
Tom Loosemore: Estonia had done some very interesting things. The UK was languishing at No. 8 in the UN e-government rankings. Frankly, if we look at other Governments as the bar, we find that at the time they all faced similar challenges. Martha felt very strongly that sitting on our laurels and saying, “We are fine compared with other Governments,” was simply unacceptable.
I should be clear. It is about a way of working—a notion of working in small, multidisciplinary teams—and about bringing the technology and design skills back into government so that it is able to deliver in the same way as Google, Amazon, Airbnb and the best start-ups deliver, with multidisciplinary, iterative, incremental, constant improvement and feedback loops, embracing open source as a way forward and communicating openly.
Q6 Chair: Are you suggesting that those principles were not at the heart of government?
Tom Loosemore: I have to say that they could not have been further from the heart of government. Senior officials had taken a view through the 1990s and 2000s that technology was a bit scary and that we should outsource it to nice men in suits, who would spend our money and take the risk away from us. Obviously, it is expensive, scary and complicated, and government is special, but it is not big—it is not that complicated, technologically. The bill went up to billions and billions and billions. From my perspective, the quality of the services that were delivered was simply lamentable—I use that word advisedly—compared with the amount of money being spent on them.
What was interesting about 2010 and 2011, which was relatively unique in my experience, was an alignment politically between all parties that the way in which things were done was unacceptable. The previous Administration—the likes of Tom Watson, who had been a Minister in the Cabinet Office—were very vocal on the issue. Just after leaving power, Tom Watson blogged a mea culpa, saying, “We should have done the things that Francis Maude is now going to do.”
Q7 Chair: What was the ambition at the start? What were you seeking to achieve?
Tom Loosemore: We wanted to reorient how government delivered services from being centred around the convenience of government and suppliers to being centred around the needs of citizens. We also wanted to change the delivery model from being one of outsourcing to IT—Google does not have an IT department—to being one of much closer ownership by civil servants of the iterative, multidisciplinary design of services, embracing policy, operations and technology.
GDS was an attempt to catalyse a radical movement of reform across the whole public sector in the UK. It was deliberately a very bold and disruptive act, to stimulate and catalyse a movement all over government, not just to be a delivery arm in the middle.
Q8 Chair: Dafydd, do you want to add anything?
Dafydd Vaughan: The main thing to add is that, prior to 2010, if I had wanted to work on public sector IT or building services for the public, I would almost certainly have had to consider joining one of the large outsourced IT providers. I was not interested in doing that, and many people I have worked with over the last couple of years were not interested in doing it, either. They were more likely to go and work for Google, Facebook or one of the big, interesting technology firms. One of the things that I hope we achieved in part at GDS was to make government an interesting place to come and work, and to encourage people to join government and to build interesting technology.
Q9 Chair: What were the big challenges that you faced in those initial phases? You smile. Try to be succinct. Could you describe it to us?
Tom Loosemore: One challenge was tissue rejection of an approach that was centred around the needs of citizens, not the convenience of government. That goes against the grain of civil service culture very fundamentally.
Another was acceptance that multidisciplinary teams—teams from different policy areas, from technology and from operations—should and could sit and work together in the same place.
Procurement was designed almost deliberately to allow only very large businesses in.
Recruitment needed fundamental reform. The process of recruitment almost deliberately pushed away people like Dafydd whom we were desperately trying to bring in—even if he wanted to come and work for us.
Some of the cyber-security standards were terrifying in their inappropriateness, even in 2010. They actively stopped us doing the right thing from a cyber perspective.
I could go on and on. All I would say is that it was not unusual compared with any other Government I have ever worked with, and not that unusual compared with many large private sector organisations. That does not mean it was acceptable.
Dafydd Vaughan: It is worth saying that it is not unusual to see this outside government as well. It happens quite often in the private sector. It is not an unusual set of attitudes to come across.
Q10 Darren Jones: In the initial coalition Government, you had David Cameron as Prime Minister and Francis Maude in the Cabinet Office supporting this project. When we talk about digital transformation in businesses, we say that we need someone at board level—someone in the lines of business—to drive it through. You had that. How do we fix this in the future? If you have the Prime Minister and the Cabinet Office, what more is there that we can do?
Tom Loosemore: You do not have that now, do you?
Darren Jones: No, but it did not work when you had it.
Q11 Chair: Are you saying that these were massive challenges, but you made massive progress?
Tom Loosemore: I am very proud of the progress that the teams in the Cabinet Office and GDS made over the five years I was there. I am inordinately proud of the work they did.
Q12 Chair: Do you think you shifted the culture significantly?
Tom Loosemore: I think we made a very fundamental shift in expectation for thousands of civil servants, particularly at mid-levels, about what good looks like, in how you work and the quality of services. I think we made a fundamental improvement to things like the front door to government, in the simplicity and clarity of gov.uk.
There are dozens of services out there now, including some very big and important ones, that are simple, clearer, faster and supported by the right technologies. There is now a procurement digital marketplace with thousands of SMEs on it. There are now proper open standards—
Q13 Chair: Yet the share of the market serving or contracting with government is still very much dominated by the big players.
Tom Loosemore: My point would be that we made a very good start, but there is an enormous way to go. My worry is that our progress and momentum are not as fast as those of other Governments. Other Governments, such as Denmark, are overtaking us and being much bolder. What those Governments have is digital government. Denmark sees digital government as a fundamental Government priority, right from the top.
Q14 Chair: Is it therefore your argument that we have lost momentum?
Tom Loosemore: Yes, I think so. One of the reasons for that is the lack of political priority. I know that we have Brexit, so that is understandable. But in the five years in which I was in government, you could rely not just on Francis Maude but on a general understanding that this was a Government priority and that change was going to be normal, not something to be avoided. You also had an expectation that, were an Opposition to get into power, they would continue these reforms. That was very much the norm.
Chair: That is the consensus you were talking about.
Q15 Martin Whitfield: May I pursue that point slightly? You say that the UK Government, like a lot of Governments and a lot of industry, found it a challenge to get this started. You have a route map that has been well established in lots of places. You say that you had a very good start and had started to shift the blockage that always exists. How many more years would you have needed to have got there?
Tom Loosemore: The first thing to say is that the leadership that you need at the start of that disruptive process is not the leadership that you need in the middle. We were very aware of that. As you can probably tell, we were quite disruptive, in a positive way. We knew when to leave. I think that 2015, after the election—we did not want to leave before it—was the right time for us, as a leadership team, to hand over to the likes of Stephen Foreshew-Cain, who took the view that the role of the centre in GDS was not to disrupt but to support Departments and agencies trying to do the right thing. That is what I am worried has stalled.
Q16 Chair: You think it was the right judgment, but it has not really happened as it should have.
Tom Loosemore: If you are in a Department and are trying to do the right thing—if you are trying to work in a user-centred, agile and iterative way, procuring the right technology—the grain is still far too against you, compared with how it should be. It is better than it was—there are routes through that were not there before—but you still need support from the centre to help you navigate—
Q17 Chair: You are saying that that support is not really coming.
Tom Loosemore: I am not sure that that support is as active.
Q18 Chair: How would you define the challenges now? In a sense, what do you think we should be saying to government about how we can overcome them?
Tom Loosemore: There is a huge way to go in changing the grain of how things are normally done. The centre has a role around applying IT spend controls on the billions of pounds—we do not even count it any more—that are being spent on technology, mostly still with very large suppliers. That function of the centre, which is to challenge constructively how Departments and agencies are spending their technology money, was quietly—we did not make a big noise about it—the most important lever that GDS had to encourage Departments to do the right thing. That is not a Machiavellian “encourage them to do the right thing”—many Departments wanted to do the right thing.
That role has gone. When describing to me what GDS’s IT spend control function is now, someone said, “They just turn up. We treat them just like birdwatchers.” That is terrifying from my perspective. I worry about the momentum towards using open standards, smaller companies and commodity technology, breaking contracts up into small chunks and owning the knowledge of service design and technology in the civil service, so that you know what you are buying. I worry that that ship could be turning the wrong way again, having started to turn the right way.
Q19 Chair: I am slightly confused. You said that at the start you needed the disruptors, and then you needed a change of approach. It sounds to me as though you still need the disruptors, in a way.
Tom Loosemore: I still think you need a strong centre. It does not need to be as disruptive as we were. I think that that strong centre has weakened. You need a centre that is able to encourage Departments to do the right thing in a very constructive and collaborative way. Stephen Foreshew-Cain, who was there for only a year, very sadly, in 2015 and 2016, changed the mantra of GDS to be facing towards teams and Departments. He would say, “We have got your back. We in the centre have got your back in Departments. If you are facing pressure to do the wrong thing in your Department or agency, we in GDS will come and help you. We will use our central powers and controls.”
Q20 Chair: But that is not happening.
Tom Loosemore: That is not happening. If you talk to teams in Departments, it is not happening.
Dafydd Vaughan: One of the difficult problems that a lot of Departments are facing is with legacy technology. A lot of technology is very old and is very difficult and expensive to change, as it is locked into big contracts. One of the things GDS provided support on was trying to break that up and dealing with the legacy problem. The centre is not there to support the Departments in the same way as it was then.
Q21 Chair: We have heard that we should be setting clear standards, which every Department should comply with, for moving on from the legacy systems. Would you support that as being necessary?
Dafydd Vaughan: Yes.
Tom Loosemore: Government has thousands of standards. What matters is who enforces them. That is the question to ask: who is enforcing them? Francis Maude would get the chief executives of global technology businesses into his office and show them the door, saying, “You are not taking our billions off us any more. Enough—the game is over.” That alignment and political support from the centre, with Jeremy Heywood right behind it, is now absent.
Q22 Bill Grant: I sense and welcome your enthusiasm for the change that you drove in the first half of this decade. It was an attitude shift, as well as a culture shift. During those years, as a selling point to improve digital services in government, did you make a link between the service improvement that is required and productivity in the economy? Was there ever a link between that, as a selling point, and the country being better?
Tom Loosemore: Completely. Partly, the selling point was that we are great at internet-era technology in this business. I have spent the last 25 years in it. We should be leading the world, as an element of our soft power, in getting other Governments to follow our lead in using this way of thinking and working, but that would mean encouraging lots of SMEs and opening the door. We did that as effectively as we could—very boldly, in procurement reform, with the digital marketplace. G-Cloud 10, which is the latest release of that reform, has only just made it out of the door. You may want to ask questions about that and about why the digital marketplace is still struggling. It should be the absolute centrepiece of reform of procurement in government, with pride of place.
One of the really interesting elements is the Treasury and the business case process with it. It is the same with all Governments. You have a Treasury that likes to spend CAPEX on big projects. It is easy. This world is not about CAPEX. It is about teams—investing in teams that continuously improve and develop. One of the problems with going with CAPEX is that your business case is a five-year or 10-year fiction. I have read too many of them that were just made-up nonsense, with false certainty all over the place.
There is the opportunity to make life simpler for citizens and for businesses and to drive productivity gains. You can open a business in Estonia in 14 minutes. How long does it still take to do that here today? By the way, the reason it takes so long is not technology, but that we have lots of different Departments that will not talk to one another. We cannot agree what a unique identifier for a business is in government. Different Departments—HMRC, BEIS and the ONS—are still fighting about it.
Q23 Chair: That is why you need the central leadership.
Tom Loosemore: That is why you need that strong leadership.
Q24 Damien Moore: What is the primary aim of digitising government? Are there any Departments that are not suitable for digitisation?
Tom Loosemore: From my perspective, the word “digital” means adopting the ways of working of the internet era. It is a wave that is hitting every single sector. If you do not keep up, you get left behind. There are no bits of government at all that can ignore adapting what they do, right down to their policies, to the internet era. I would look at places like the Foreign Office. On the surface, the Foreign Office does not do much service delivery. I think that it has done a great job of adapting to the internet era, by getting ambassadors to blog and so on.
As a Government, we are still at the foothills of adopting internet-era styles of delivery. You need the humility to start by accepting that you do not understand what citizens actually need and want. Then you iterate your way humbly towards getting the policy outcome you want, rather than pretending you know what the answer is up front.
That is a fundamental change. It requires people like you to ask different questions in Committee, rather than demand answers from people up front. I would not pretend that we are anything other than still at the start of that.
It is a difficult journey. I work with enough large private sector clients to know that, although I am disappointed with the pace of progress over the last few years, the UK Government are still way ahead of many large private sector global businesses.
Q25 Damien Moore: Do you agree, Dafydd?
Dafydd Vaughan: Absolutely.
Q26 Damien Moore: It does not make for as interesting a conversation when that happens.
Would you describe gov.uk as user-friendly and accessible? You have mentioned the needs of citizens several times. The question goes back to that, because they are the end users of this product. It is for them. They are paying the bills for this service to be delivered.
Tom Loosemore: Gov.uk is symbolically and practically a fundamental shift in how government thinks about delivering for the needs of citizens. Previously, you had 2,000 websites. You needed to know the structure of government before you could find what you needed. Eighty per cent. of people start with Google. If you search on Google for, “What do I do when someone dies?” good luck to you, frankly—good luck to you to find what you need, and good luck to you to understand what you need to do as a citizen. The language was opaque, even if you could find it.
What gov.uk did was say, “There is a single front door. There are standards in place”—let us hope that they are still upheld—"to make sure that the language is simple, clear and fast, so that you can find out what you need to do.” We, as a Government, definitely led the world in that. Lots of Governments all over the world are copying that approach. They are saying, “Let’s make sure that the front door of our Government is as simple, clear, fast and unambiguous as possible. Let’s accept the fact that people who are great at writing policy may not be the people who write our website. That is a specialist skill, around using simple, clear language that people can read on a screen when they are short of time.”
I still think that, in many ways, gov.uk has a lot to improve. If you are a journalist, it is still hard work, but I would argue that, before gov.uk, all the emphasis of comms teams and Governments was on feeding the needs of journalists. I am an ex-journalist. Bluntly, journalists are not the priority for gov.uk. Citizens in need—citizens in desperation, who may struggle to read or to comprehend—are. That is the priority.
It is still world leading. It is copied all over the world, even today, by other Governments.
Dafydd Vaughan: If you look at the way in which gov.uk was built and how the content is managed, you find that the teams spend a lot of effort on user research, talking to citizens, looking at the data for how people use the service and, from there, improving the content and making changes to meet users’ needs. They spend a lot of time focused on users and what they need to do.
Tom is right. There are some areas where there needs to be improvement. In some cases, where content is controlled by the Department comms teams, rather than centrally, you get confusion, because you end up with additional or duplicate content being published. In general, there is an enormous amount of user focus in the primary citizen-focused areas of gov.uk.
Tom Loosemore: I remember talking to the chief executive of a very large and well-known US software services business, who said very proudly, “We release software every day.” That is good. Gov.uk releases software 10 times a day. Previously, it was once every six months. That is the change—being able to release software 10 times a day when you need it. I refer not just to content changes, but to actual software changes.
Damien Moore: On the issue of whether it is user-friendly and accessible, if you were brought up in the digital age, I am sure that it is perfectly fine and you will be able to navigate around it. You can overcome a lot of problems yourself by surfing between different pages, which will solve the problem. A lot of the people in this country, particularly my constituents, were not brought up in the digital age.
Bill Grant: That is me.
Damien Moore: He is not my constituent, but—
Bill Grant: From the same era.
Q27 Damien Moore: It is about overcoming some of the problems and getting them to engage with it. It is not a case of their not wanting to do that. Sometimes there are some unique problems that are not solved by a “frequently asked questions” page. They might need to phone up, but the challenge faced with the phone call might be almost as great as the one faced on the pages. There should be a point where you are going not from option 1 to God knows what, but from not being able to do something to speaking to someone relatively quickly. If you cannot do that, you will put people off using the service.
Tom Loosemore: Thank you. You have just given us the argument that we had so many times with Departments. This comes down to Departments, agencies and, in some cases, Ministers wilfully misunderstanding what we meant by digital by default. Digital by default did not mean online only. It did not mean hiding other channels and making them rubbish, so that people were forced to use the online channel. Personally, I think that that is lamentable. As a civil servant, I could not agree to that.
Q28 Chair: But you felt that that was happening.
Tom Loosemore: We had huge arguments with Departments that wanted to hide the phone number. They said, “Please can you hide the phone number, so that no one finds it.” What are you talking about? All you will do is drive pain, suffering and failure demand elsewhere. It will not save any money, because people will find a way of popping up elsewhere in the system.
The whole point here is that, collectively and holistically, services are simpler, clearer, faster and more empathetic. You want to make sure that the online services are so simple, clear and fast that those who can self-serve prefer to do so. That should save money, in theory—and it does—but you need to invest that money to make sure that those who cannot self-serve, either some of the time or all the time, have suitably empathetic services, sometimes face to face and sometimes by telephony.
One of my most upsetting moments in government was discovering that a DWP service was only online. That is indefensible.
Q29 Chair: Was that one of the Departments that was resistant to the idea of having access by phone?
Tom Loosemore: I cannot remember any specific arguments with the DWP about the phone call, but it had a service that it was impossible to use other than online. It made people go online. Forcing people who are in the benefits system to go online by ministerial edict strikes me as ill advised.
Q30 Stephen Metcalfe: Will you clarify that? It was a ministerial decision.
Tom Loosemore: I think that, possibly, the Minister had got slightly over-enthusiastic about the internet. That is good. It is not a bad thing. Ministers get enthusiastic about things. The role of senior civil servants is to advise the Minister, “Yes, we will encourage and build the very best online service that we can, but we have to be very careful about exclusion and universality and to make sure that there are other channels. It would be very ill advised to force us to have an online-only system.”
To be clear, there are bits of HMRC that are online only for businesses. I am less worried about businesses.
Q31 Chair: But you are dealing with a potentially vulnerable group here.
Tom Loosemore: You simply cannot afford to do that. It does not save you any money, because the pain will pop up elsewhere in the system.
Q32 Chair: When that emerged, were you able to sort it?
Tom Loosemore: Eventually.
Chair: After a battle.
Q33 Damien Moore: You had to alter the strategies for different Departments. What challenges did you face? Were there any particular Departments that you felt were challenging of you in trying to carry this out?
Tom Loosemore: This is a hard thing to do. Departments do very different things. Our strategy in 2012, once we had established ourselves and shown that a different way of delivery was possible, was to work with the big delivery Departments to identify 25 services on which we could work with them and that we could help them to be radical in transforming. Those big seven or eight Departments are obvious. They include HMRC, the DWP, the DVLA—for motoring services—the Ministry of Justice and the Home Office. They were the Departments where we put most of the energy. They had different teams within them that were more or less keen to come on the journey with us.
This is where IT spend control was really useful. With a team such as the Home Office, which in 2013 and 2014 was really keen to change radically, we were able to say, “Fine. We trust you. We will come and see you every three months. Get on with it. How else can we help?” Other Departments, which I will not name, that were rather wedded to their £1 billion-a-year contracts with their IT vendors—I am sure that golf days will have been part of it—were much harder to change. We kept them on a much shorter leash of IT spend control and had to be rather more forceful.
Q34 Chair: You saw enormous variation across government in the willingness to engage.
Tom Loosemore: It is down to the personalities of the leadership. That changed over time. One of the things that we did in 2012, which was very unpopular with perm secs, was demand that all perm secs put on their board a digital transformation leader, whom we would help them to recruit. You need that alignment at board level for this to happen. Once some of those people got into place in some of the bigger Departments, things changed.
Q35 Chair: Did they all do that?
Tom Loosemore: Yes, eventually.
Q36 Darren Jones: Are they still there?
Tom Loosemore: Some of them are, but not all of them. To be honest, they should not be there in 2018. You should not have a chief digital officer at this point. It should just be normal. In 2018, you should not need to force a permanent sec to have someone on their board. I would be disappointed if they had them.
Last week there was a brilliant blog post by Jonathan Slater, from the Department for Education. I could not have written it better myself. Look at Clare Moriarty, the perm sec at DEFRA. I applaud every word that she writes about ways of working as a perm sec.
Q37 Chair: Do you still think that there are variable degrees of embracing of this among permanent secretaries today?
Tom Loosemore: There is a view, which is very persistent at the perm sec level and which, for me, is terrible: their Department is sovereign and their job is to protect the sovereign interests of their Department. Their Departments are not legally sovereign, but that is by the by. That view is far too prevalent. The likes of Denmark and Estonia have managed to jump over that departmentalism.
Q38 Chair: The silo mentality.
Tom Loosemore: Yes. It is deeply cultural. Ministers are rarely there long enough to challenge or to change it.
Q39 Chair: See you in a moment.
Tom Loosemore: They are not. Just look at the maths. Look at the Institute for Government report on how long a Minister lasts.
Dafydd Vaughan: One of the problems is that, to deliver a good, easy-to-use, successful service for people, you have to cross departmental boundaries. As we heard earlier, if you want to deliver starting a business quickly, you have to go through Companies House and HMRC. If you want to deliver quite a few services, you have to cross departmental boundaries. When you have senior leadership whose role is to protect that organisation, it is very difficult to cross those boundaries.
Q40 Chair: We still have not overcome the barriers that make setting up a company very slow.
Tom Loosemore: We do not have the desire in the centre to break those boundaries. That is the fundamental problem. There is no desire in the centre to force those boundaries to be broken. There is no leadership from the centre.
Q41 Damien Moore: Last but not least, to what extent do you think that the problems encountered with universal credit were due to GDS?
Tom Loosemore: I went into universal credit myself in 2013 to help to reset it. I was on the review team that forced it to stop, having spent three years and £450-odd million on delivering, frankly, not very much.
Some of the most impressive and honourable civil servants I have ever met worked, and still work, in the DWP. They are some of my most valued and liked colleagues. I am proud of the work with them to reset that programme and to bring together policy, delivery and operations so that they were in one location, rather than three different cities. I am proud of helping to empower civil servants in the DWP to force on Ministers policy changes that were previously unacceptable and to bring control of technology back into the DWP, so that now you can see that the challenges on universal credit are due not to technology not working but to other reasons that may or may not be related to policy.
From GDS’s perspective, I am very proud of that. I wish that we could have stayed longer. Many of the people on my team from GDS did go and work for DWP; they just swapped over. I am very grateful to them for doing that. It put me in hospital. That is how toxic it was. I was not the only person to be put in hospital by universal credit, as I am sure you well know. I am proud of the work that the team did in that culture.
Expecting a Department that in 2010 had outsourced its knowledge about technology almost completely—there was one person left who knew what was in the systems, and he was nearing retirement—to move away from the mindset of outsourcing everything and to go agile may have been an ask too far in 2010.
Relations got difficult. Looking back, I wish that ministerial relations—not with me personally, to be honest—had not become so difficult as well.
Q42 Chair: Is that sorted now, or being sorted? Are the problems still persistent?
Tom Loosemore: I am sure that you know the problems with universal credit. I am very confident that those problems are not due to one of the best teams that I have ever seen work—weirdly—which is the universal credit full service team, based just around the corner from here. In many ways, they are exemplary. If they were not exemplary, people would find reasons to blame them for problems that are actually policy based. They cannot blame the system any more. It is the policy.
Q43 Martin Whitfield: May I pursue that point, to give you an opportunity to say where GDS, the policy and the strategy come in? With universal credit, there was a problem that if you, as an EU citizen in this country, filled in the online stuff and everything was marvellous, but you did not get any payments, eventually, when you had to resort to telephoning, the people who were looking at the screen in government were seeing a screen that was different from the screen that the individual claimant had filled in. It turned out that if someone in government had panned down three or four pages, they would have got to the same screen that was being looked at outside. On one level, that is a software problem. On one level, it is a policy problem. Where would you say that the responsibility for that to be solved rested—with GDS, the DWP or the individual group?
Tom Loosemore: We were empowered.
Q44 Martin Whitfield: Talk me through how.
Tom Loosemore: I would argue that it is about having a single team that is accountable for the design and the delivery of the service. You have that in DWP. GDS was always there only to help. We were parachuted in. I could not have been prouder or more delighted to help those brilliant people in DWP, who are still there.
Accountability for the example that you give is with the SRO of universal credit, Neil Couling. It is for him to make sure that the design of the service and the policy are aligned, so that the Minister gets the outcome that she or he wishes and citizens are not excluded.
Welfare is a hard problem. Globally, it is a hard problem. The intent behind it—to simplify benefits—is admirable. Let us not forget what the benefits system was like before. I will be honest. When I went into DWP, I was pleasantly delighted and shocked by just how committed and hard working the people there were. Their hearts are in the right place.
Q45 Martin Whitfield: In actual fact, that multidisciplinary team, empowered to make the changes it needed to make as the various feedback loops were coming in, would have worked. It should and will work, if it is empowered.
Tom Loosemore: It will work. The policy still has fundamental problems that are potentially catastrophic, such as the waiting period. I can go into detail about what I think the problems with the policy are.
Q46 Martin Whitfield: I was thinking more about the tech.
Tom Loosemore: There was a period in 2013 when the Minister would come every week and sit with the team, which had done user research on the service. The team would say, “Minister, we have a problem with your policy,” and he would fix it—there and then, at a weekly cadence. He would be brave enough to do that, although he did not have much choice at the time. He would say, “Okay, I can see the evidence. I am going to change my policy now.” That close alignment—that iteration and feedback loop with reality, without layers of senior civil servants writing nonsense papers defending their position—
Q47 Chair: Was that Lord Freud at the time?
Tom Loosemore: It was Lord Freud and Iain Duncan Smith, actually. With Lord Freud, in particular, we were there every week, making changes and being involved.
Q48 Chair: Are you saying that we then lost that?
Tom Loosemore: I think that it gets harder. If the Minister has been there for only a few months, it makes life difficult. They have to learn to trust that and to trust the team. There are some other big policy issues that are beyond the remit of the team to fix. The team in Caxton House cannot fix the Treasury-induced week-long waiting time.
My advice to DWP Ministers has been and will remain to get close to the team, to go and see them a lot and to listen to them. If Ministers can make changes that the team recommends, they should trust them, because they are close to the citizens and the operational reality—much closer than Ministers are, and much closer than some of the people in their policy teams who are writing papers for them will be. I am sounding pejorative around policy people, but there are still thousands of people out there in the DWP writing papers that may or may not be based on reality.
Q49 Darren Jones: I think that we all agree that we want to get to the position where government is delivering services that meet the needs of citizens, and that the technology is inherently a part of that. We need to get to a position in this inquiry where we reach conclusions and make recommendations to the Government about what they do next to try to achieve that outcome.
Today we have talked a bit about the leadership issues. I do not think that my question was quite answered. We could say to the Government, “We want the Prime Minister and the Cabinet Office to provide the equivalent of board-level leadership at Government level and to get people in every Department leading on it,” although perhaps that did not work before. My question is: you have been elevated to the House of Lords, we have put you in the Cabinet Office, and you have the ear of the Prime Minister. What would you do?
Tom Loosemore: The first thing I would do is change the relationship with the Treasury—we did not succeed in doing that in GDS—around how business cases are written and how the Treasury sees its role in approving funding. Bluntly, I would move from a CAPEX model to an OPEX model, where the Treasury funds teams to deliver outcomes that Ministers want, rather than big CAPEX spend, as if we were building a motorway. You are not building motorways—you are building services.
That is the first point. Secondly, I would champion non-stop some of the brilliant work that is going on all over government at mid-levels in civil service teams. I could name dozens of teams. They are not always protected in the way I would hope.
Q50 Darren Jones: Protected from whom?
Tom Loosemore: Protected from people who have grown within a civil service where the grain was different from what it needs to be and whose sense of control and power rests on an old way of working, be it a procurement-based mindset, a policy-based mindset or the kind of nonsense that involves saying, “We don’t sit together as teams here, because we do not do that.”
I would champion thousands of mid-level civil servants who are trying to do the right thing. I would reinstitute IT spend control tomorrow, with the Treasury. I could go on.
Dafydd Vaughan: I think that it is important to spend time with the Department to help it to deal with the legacy technology and the contract problem. There are some good examples from the work that we did with the DVLA. It had a big outsourced IT contract that went from 10 years to 15 years to 25 years. In 2015 and 2016, we brought that back in-house and delivered a bunch of really good digital services that let the DVLA move much more quickly than it had moved before and that were much cheaper than they had been before. We brought skills in-house to allow the DVLA to make those changes quickly. I would look to do something similar in other Departments.
Q51 Darren Jones: You raise the issue of in-house skills versus public procurement. I cannot speak for Margot James, but I think that she is suggesting that there is a midway house between a £1 billion outsource to company X and using private sector providers to help. We did an inquiry on decision-making algorithms and biases. We focused on healthcare, for obvious reasons. I asked the Minister, “Did we really need to give all the data to people at DeepMind? Could we have employed data scientists in the health service to do it ourselves, and then retained the value?” She took the view that we do not have the capacity to do that in the health service, which is why we need to work with private providers. Do you agree? Do you see that happening across Government Departments? Is that the right model? What are your views?
Tom Loosemore: How can you buy something if you do not understand it yourself? At a fundamental level, you need to have the requisite specialist skills. We have thousands of lawyers in government, for good reasons. They are not the only people who do law. We pay them a lot of money as well, which is interesting.
I think that the model should be the same for any technology. You need to have people who understand it deeply in order that you can buy the right things. Buy commodity from commodity providers; do not buy commodity from big system integrators.
If you need to commission something very special, buy it from the right kind of business. If you need to keep it really close and it has not been commoditised yet, build it internally.
When I say that we lost the technology skills, I am talking not about writing basic code—government should not really be doing that, possibly—but about understanding deeply what the architecture should be and how you buy the right thing from the right suppliers. Bluntly, the people reinventing the world around us are not going to big system integrators to build the things that they do. They are going to cloud providers and to talented small businesses, some of them in the UK. That is the skillset that we need in-house. We need the right blend. It is not binary.
I am astonished that the NHS does not invest in data science. I think that that is terrifying.
Darren Jones: I thought so, too.
Dafydd Vaughan: Government will never be able to do everything itself. Government probably should not do everything itself. However, if you are going to go out and get private sector organisations to provide services or products to you, you need to know what you are buying and to understand those products. You therefore need to have at least a base level of technical skill within government and the public sector to make informed choices and decisions about the right way of buying things, the right types of contracts and the right technologies to build services. That is fundamental to building good services in government.
Q52 Darren Jones: My last question is about the attractiveness of coming to work for GDS or Departments on digital transformation projects. We heard that there was a recommendation that a chief data officer be recruited in 2017 but that the role had not yet been filled. I take the point that Tom made earlier—that perhaps we do not need a CDO, because everyone should be doing it—but are we struggling to recruit the talented people that we need? If so, why?
Tom Loosemore: You did not hire a chief data officer because the centre decided that it did not want to be powerful. Basically, the perm secs vetoed it. That is the position I have come to.
I have a worry that at the centre more people are leaving than are joining. Recently, there was an answer to a written question from one of your parliamentary colleagues about turnover of staff at GDS. You might want to have a look at that. The numbers tell a story that worries me.
Q53 Darren Jones: What could we do differently?
In order to take advantage of that, you have to build the right cultures around it. You have to allow teams to get on and deliver things. You have to give them the ability to make mistakes and to learn from and fix them. You have to give them the ability to be open and to talk about what they are doing. You can build a bit of excitement around what is going on.
One thing that is quite important and that we did quite early on in GDS’s life is to be open and to talk about what is going on. We talked at events regularly and wrote blog posts explaining what we were doing and exposing some of the interesting problems. That worked as a recruitment tool on its own and encouraged other people to join and to get involved in what we were doing. Continuing that—ensuring a culture of openness that gives people the space to get on and deliver and the ability to build good things—is really important.
Q54 Darren Jones: It is not just about full-time employees on our books in the public sector. I often find Westminster quite stale compared with the real world. I find huge amounts of energy among those wanting to solve societal problems outside, compared with what I find inside. Clearly, that does not apply to my colleagues, but, from a government perspective, all the interesting innovation seems to be happening outside.
I support the Government’s position on their GovTech Catalyst. We are delivering money to try to support that relationship. Looking at international comparisons, do we find that entrepreneurial companies want to be able to work with government to help us to solve these problems, or is it too difficult? Are they going elsewhere, as they think that we do not really get it or want to do it? Are we missing out on that innovation input?
Tom Loosemore: Only Estonia has cracked how you really engage SMEs. Estonia is small—everyone knows each other. They are usually married to their sister. That is not pejorative; it is true.
There is a long way to go. If you set up a start-up in the UK, it is still really hard to sell to the Government. I have to say that GovTech Catalyst is very unproven as a model, when you look internationally. There are nice headlines for the Minister, with nice, big numbers of money, but I would say, “Go and sell it to HMRC, mate. Let’s see whether they buy it.” I am waiting. Culturally and technically, there is a big gap between the desire and reality.
Having said that, I was absolutely delighted to discover that the DWP has used a small start-up to power its job search engine. It can and does happen, but there is a very long way to go.
Beware ministerial announcement chaff: ask how many of these businesses have actually managed to sell something to government. That, rather than the headline, is the only question that matters.
Internationally, it is a very unproven model. There have been quite a few failures out there. Not enough work has been done to bridge the gap on culture and mindset and to say, “You are moving really fast. You are breaking things. You are not doing some of the things that you need to do to be universally accessible within government, let alone to be able to navigate procurement rules that are insane in some cases and necessary in others.”
That gap is huge. All the effort should be going into bridging it. That is not very sexy work. Everyone wants to go to launches of shiny, new, innovative blockchain nonsense, rather than to say, “How do we fix procurement? How do we make sure that you understand the security requirements of government? How do we make sure that you adopt open standards that government insists on your adopting?” That is hard yards.
Dafydd Vaughan: Buying from and using small companies in government is still too difficult. We have made big steps forward in services like G-Cloud and the digital outcomes and specialists frameworks on the digital marketplace, but it is still difficult to use small companies to deliver government services or even to experiment and try things out. There is still a lot more work to do there.
Tom Loosemore: I have a big concern about GovTech—a blog post will come out in a couple of days. It needs two things. You need to inquire what is meant. Does it mean a technology that government is going to buy to power its services—the job search power engine? Does it mean that we are privatising and allowing people to cream off easy and cheap-to-service customers and paying them the mean, leaving the likes of the NHS to pick up the others? With sexy GovTech stuff, you need to inquire very hard about what is meant.
Q55 Stephen Metcalfe: We have covered a lot of the ground that I wanted to cover on the successes, failures and challenges. Correct me if I am wrong, but it sounds to me as though the culture is more of a challenge than the technology. You have talked about legacy issues. If you were to set out a road map for reimaging and reinventing that culture, where would you start?
Tom Loosemore: You start with a GDS. Governments around the world are all setting up GDSs. With pretty much all our clients, the first thing we do is help them to set up a government digital service. It is about what that Department does and the role that it plays in supporting and challenging the status quo in Departments and agencies. I think that there has been a wilful, deliberate misdirection of the centre in supporting and challenging Departments. I would correct that very quickly, to make it a strong centre.
Q56 Chair: You say “wilful”. Who is responsible for this wilful misdirection? We need to be clear here.
Tom Loosemore: It is on the public record that the chief executive of the civil service wishes departmental sovereignty to be maintained. The buck stops there.
Q57 Stephen Metcalfe: That is the challenge.
Tom Loosemore: That is the challenge.
Q58 Stephen Metcalfe: Effectively, that is siloing of Government Departments.
Tom Loosemore: Let us be clear. We are not going to wave a magic wand and change accountability models, but you need to have a centre that says, “We are going to challenge that sovereignty, in the interests of the citizen, when you want to start a business quickly or when a Department is clearly not following the standards and has signed another £400 million contract over 10 years.” For me, saying that we are going to let that go, because we do not want a fight, is unacceptable as a role for the centre. You need a strong centre.
It is a big supertanker to shift. There is some good support. There are service manuals, and some great work is being done by teams in GDS to support Departments to do the right thing. There is a bit of carrot in the service standard. There is no stick at all. I am afraid that you need a bit of stick.
Q59 Stephen Metcalfe: Does a bigger stick come with a more senior Minister to take responsibility for these things?
Tom Loosemore: It would be lovely to have a Minister who was there for more than nine months and who saw their role, as Francis Maude did, as being to achieve fundamental reform of the machinery of government, whatever its political ends.
Q60 Chair: You had the same person there for five years.
Tom Loosemore: That is super helpful. It is not going to happen again in a hurry. At the end of the day, you need a civil service leadership that can provide the continuity and help Ministers to understand—even if they are there for only 18 months—that this is an important long-term mission and change. The role of senior officials is to advise Ministers of that. If we get another Francis Maude, of whatever colour or shape, I will be delighted, but you cannot rely on it.
Q61 Stephen Metcalfe: We have talked a lot about Estonia, but you said that it is a very small place. Is it unrealistic for us to be able to model ourselves on Estonia by scaling up that system?
Tom Loosemore: No. You can joke about Estonia being small, but it is still 1.3 million people. It is not nothing. They have their challenges.
What is interesting about Estonia is the institutional architecture—the shape of government. It is more different than they sometimes realise, because they reinvented the shape of government when the Russians left. They had a President who is a coder. They have an institutional architecture that has horizontal layers. There is a law in Estonia that, if one Department asks you, as a citizen, for information that another Department already has, you can say no. Constitutionally, it forces Departments to put the citizen at the centre of things. It has data sharing baked into the accountability model and the institutional model.
We are in the 19th century in our institutional model, as are most Governments. I would love to see that shift. If that were to shift in the UK Government—if you had a different accountability shape, with horizontal layers of government—I would be the first person back.
Q62 Stephen Metcalfe: Is that ever going to happen?
Tom Loosemore: Yes, because the force of the internet will make it happen. It is just a question of whether we are first or whether we are a laggard. It will happen.
Q63 Stephen Metcalfe: It will happen.
Tom Loosemore: It will happen.
Q64 Stephen Metcalfe: You are confident.
Tom Loosemore: Places like Denmark, Singapore and some of our clients in South America, who are keen to leapfrog, are all looking at horizontal layers of government. How do we have a Minister for personal data? How do we have a Minister for payments, across everything, not just a Department of X,Y and Z horizontally? If you look at the construction of every single internet-era business, you find that they are built on platforms like that, with services built on top. The UK Government will inevitably have to adopt that. It may take the rest of my blooming lifetime—I hope not—but it will happen.
Q65 Bill Grant: We have heard a number of things this morning. We have heard that government cannot do everything itself, which we accept. We have heard about the barriers to securing the services of SMEs, which seem to be individuals, culture, attitude and so on. How vibrant, efficient and reliable is the marketplace for SMEs that could service the needs of government? What is that marketplace like?
Tom Loosemore: It is brilliant.
Q66 Bill Grant: Answer complete.
Tom Loosemore: It is funny how you do not realise that you do not have problems until you go to other countries that have problems. Having worked globally with countries where they do not have a vibrant technology community, or even a civic technology community, of SMEs, you realise what we are blessed with in the UK. We have fantastic companies here.
Q67 Bill Grant: We have low-hanging fruit, in the form of SMEs that can provide vibrant, reliable and well-run services to government.
Tom Loosemore: Yes, provided that government and those SMEs do the hard yards to understand that you cannot act as if you are the next Facebook, move fast and break things, serve only a subset of the population and build things in ways that are a bit fast and loose and may fall over occasionally. You cannot do that if you are doing government services. They are slightly different.
We have plenty of companies that are already offering services to government, and many thousands more that are on the digital marketplace. I am confident that the ecosystem is not the problem.
Dafydd Vaughan: There is still a tendency towards aggregating things up to a point where it is quite difficult for an SME to provide a service. Some of the services that government is buying and that are published on the digital marketplace, for example, are still quite big. They are aimed at large suppliers, to provide their—[Interruption.]
Chair: Just keep talking.
Dafydd Vaughan: In order to make good use of some of the SMEs, we need to start to break services down a bit more. That will require a bit of a culture change within government. It will require more skills in government, to manage contracts a bit better or to manage suppliers better, for example—but there are definitely a lot of SMEs out there that can provide services for government, as long as we structure those services and contracts well.
Tom Loosemore: There is a reality here. This is not talked about very often. When you had big suppliers in government, they got fat and lazy on billion-pound contracts, and you got their C-team. In terms of quality, you got teams that, frankly, any start-up would eat for breakfast in 10 minutes. As a big supplier in the UK, you do not give your best team to government—you give it to the banks and the people who pay more. You can get away with giving lower-quality teams to government. SMEs are much more exciting.
Q68 Bill Grant: I have a final, risky supplementary. Could it be that government is uncertain about what it requires and has an inability to write the proper specification? Is that a risk?
Tom Loosemore: If you think that you can write the answer in a specification and get the right answer, that is the problem. It is the point that I made about CAPEX versus OPEX. Far too many bits of government still think that they can design the service up front in a contract and their big suppliers will be more than happy to deliver it to them. When that does not work, they will charge you change control fees through the nose.
That was the business model—you charged for change. You pretended to have certainty up front. There are still too many bits of government that would like to think that we can write a specification and get the answer that we want, rather than take the harder path, which is to start small and to iterate, based on reality, with a team. We need to accept the fact that we do not know the answer up front and are going to iterate our way towards it.
Dafydd Vaughan: It is exactly as Tom has just said. You need to focus on user needs and on delivering small chunks of a service regularly, on an iterative basis, rather than specifying everything up front. That requires government to have technical skills or understanding in-house in order to specify those things to begin with.
Q69 Martin Whitfield: I will continue the flow on this, because a lot of areas have been covered.
You have raised a number of times the significant part that leadership plays. You have also raised the problem that elected leaders come and go. The real change needs to be at the leadership level within the civil service—this move across. In my previous question, I suggested that there was a road map for big organisations. Perhaps there is not a road map in government, but it would be one answer. Irrespective of who advised on policy following manifestos and elections, you would still have that digital government—that gov.uk—that worked for the people, rather than the other way around.
Tom Loosemore: I am cautiously very hopeful about the medium term, because I know of dozens of mid to senior-level civil servants who have experienced this way of working and embrace it fully. They do not always get the freedom to adopt it as they would like, but they ain’t going anywhere. They are only going up, because they are quality people. In 10 years, I hope to be able to give you not just two perm secs who I think behave admirably, but 10.
The incentives in place on the perm secs are the thing I would look at. You are absolutely right—the incentives in place on the senior leadership layer are key. They are sometimes incentivised to be sovereign and to treat their Department as sovereign. It will take quite a challenge—from No. 10, bluntly—to change that.
Dafydd Vaughan: There are many very good people in senior roles within government—the senior civil service or just below senior civil service level—who know how to do this stuff well. Given the space and ability to progress, in a number of years we should have a good-quality senior civil service who understand how to do digital technology well.
Q70 Martin Whitfield: The bunker mentality that not just this Committee but many committees across government see will be forced to change simply because of the internet over time, but it would be better to do that sooner and lead rather than later and follow.
Tom Loosemore: I would love that to happen. I would love to see an ambitious, open-hearted reform led by, collectively, a perm sec layer that said, “We are going to work differently. We are going to have some horizontal layers that we all want to embrace,” rather than rejecting it. For that, you need a strong centre. I am not sure that the civil service needs a new chief executive. Remember: the perm secs don’t report to him.
Q71 Martin Whitfield: With this change and the road map that is developing with these areas, how will we ensure that true digital government is accessible to absolutely everyone in the United Kingdom?
Tom Loosemore: It comes down to the teams understanding why that is important and believing in it. I do not have a second’s doubt. I have never come across a team that did not believe that its core job was to make services accessible and inclusive. Sometimes, the incentives put on them were misapplied, but civil servants get up in the morning to make sure that no one gets left out. An empowered team will do everything it can to make that happen. I do not have any worries about that.
I do have worries at the SRO level, sometimes—that it is still seen as IT projects and the delivery of IT rather than delivering an outcome to everybody universally served. The issue is at SRO level, not the teams.
Q72 Martin Whitfield: Those teams need to have someone who has their back.
Tom Loosemore: You said it.
Q73 Chair: You have talked about the modes of alignment between policy and implementation. We have also heard that policy is now within DCMS.
Tom Loosemore: Data policy is.
Q74 Chair: Right. Are you comfortable with data policy being in DCMS and responsibility for GDS being in the Cabinet Office, or do you think there is a case for aligning the two?
Tom Loosemore: Deckchairs is one. There is a case for a strong centre, and the Cabinet Office is the least bad. Interestingly, other Governments have finance ministries, not treasuries, and finance ministries are usually where you have that strong centre. A finance ministry not doing macro stuff but doing delivery would be ideal for me. It was, maybe, an odd move from my personal perspective to see data moved to DCMS. It was the personal passion of the Minister at the time. I am not sure it is the personal passion of the present Minister. There are some great people in DCMS doing work on that. There is a big private sector element to data policy in the UK, although why is it not in BEIS? It comes down, fundamentally, to whether HMRC is going to wait for the morning and be worried about a missive from DCMS on data policy.
Q75 Chair: Exactly. That is what I was hinting at.
You have talked a lot about the potential to improve the experience of citizens in their dealings with government, but, presumably, there is also a significant potential gain in more efficient public service delivery. What is the potential for gain in saving scarce resources in more effective delivery of services?
Tom Loosemore: We did some work in 2011-12 in writing a digital efficiency report, which attempted in top-down and bottom-up ways to estimate exactly the size of the prize. Depending on the complexity of the service, you should be looking at a 10 times difference between a cost of face-to-face—the cost of telephony being 10 times cheaper and online self-service being 10 times cheaper as well. That is not to say that there are not some services where you should have significant face-to-face. Even in 2012, we were talking £2 billion with the numbers we had then. If you ran that analysis now, it would be far more.
Q76 Chair: So the savings are, potentially, huge and, along with our responsibilities to citizens to improve their experience, the problems that you have described—the reassertion of the sovereignty of individual Departments and moving back to a more silo approach—have a massive cost attached to them.
Tom Loosemore: Yes. One of the arguments around the cost savings is just on IT spend alone. Why do we have hundreds of different ways of taking payment in government? You need, as Estonia has, a single payments platform. We have a great team in GDS doing gov.uk/pay, which is a central payments platform, but many Departments are completely refusing to use it. That is unacceptable from a cost perspective.
Q77 Chair: Through this apparent inertia, we are wasting vast amounts of public money.
Tom Loosemore: Yes. I cannot remember the business case for gov.uk/pay, but you are talking hundreds of millions/billions of pounds over time. Lots of money.
Q78 Chair: Given all these developments and the extent to which you consider it has gone off course since the early days when there was a real momentum, how far do you consider we are off course now? How significant is the fall-off and the need to regain the momentum?
Tom Loosemore: The reason I am here being as candid as I am and losing friends is that it is fixable. There are still so many great people, both in GDS and more importantly elsewhere in government, trying to do the right thing, that it is very far from being a lost cause. I worry when I work with other Governments and see the pace at which they are moving, but I do not see the same pace here.
Q79 Chair: So there is an urgency about getting this sorted.
Tom Loosemore: A real urgency. I have recently been to Argentina, a country with many challenges, but there is an amazing team in Argentina. They are ready to leapfrog what we have been able to deliver already. I could list at least another 10 countries where the same position exists.
Q80 Chair: So Argentina moves ahead of the UK. Thank you for that encouragement.
Tom Loosemore: Again, the leadership have alignment with Macri. They have a ministry of modernisation, which is horizontal. They have a crisis, which helps. The crisis of austerity helped change things. They can leapfrog, and we have not used Brexit as a means to leapfrog, I am afraid.
Chair: Thank you all very much, indeed. I appreciate it.