Interview: Tim Kelsey

Ten years ago, says Tim Kelsey, who could have predicted the massive success of Google and Facebook? He’s got a point. In 2001, it seemed likely that online sales would dominate the web’s moneymaking potential – but these two firms have become global phenomena simply by gathering, organising and redistributing other people’s information.

Soon, Kelsey believes, new online business models will emerge to blend users’ personal information with public data and produce big social and economic benefits. “I personally think that certain parts of the public data infrastructure are going to be the things that drive the next generation of Googles and Facebooks”, he says. “We can imagine a world in which there’s a continuous feedback loop from state to citizen, citizen to state.” Given control over their own government records and other relevant data, Kelsey suggests, individuals might – for example – “take their medical record and mash it up with their gym data and other non-NHS data, and from that develop real value in terms of managing their own risk of being ill. That, for me, is the future – and the UK is about to be the world leader in setting that agenda.”

When Tim Kelsey spots a way in which businesses could make profitable use of public data, it’s wise to listen carefully: he’s got form. In 2000, he established Dr Foster: a company that broke new ground by publishing comparative data on the quality of NHS hospitals. It was an instant hit with patients, and steadily won over sceptical medical professionals; in 2006 the NHS spent £12m buying a share in the company, and commissioned Kelsey to produce its flagship health information service NHS Choices.

Now working in the Cabinet Office as director of transparency, Kelsey is behind many of the transparency announcements in this week’s Autumn Statement (see news, p3). “The public data assets that we’re sitting on have enormous capacity to drive enterprise,” he says, applauding the winners of a recent government-backed ‘Appathon’: teams of students produced smartphone ‘apps’ that, for example, can help citizens report local ‘eyesores’, find out hospital A&E waiting times, and compare local schools.

Being clear about transparency
For Kelsey, the potential for public data to catalyse business growth makes a powerful argument for transparency. On top of that, though, there are five other benefits, which he hurtles through at breakneck speed – his default pace. Accountability and choice are in there, of course; but until service users are genuinely able to select between providers across large swathes of the public sector, more important are productivity, outcome improvements and service transformation.

Better productivity and outcomes follow naturally from transparency over operations and performance, says Kelsey: releasing data both throws providers into a gentlemanly but very public competition to raise their game, and hands them the information to understand where and how better results are being achieved. As for service transformation: well, says Kelsey, just look at retail banking. “Their strategy was to give us our data – to make our data transparent to us – and to wrap around that data a series of transactional opportunities,” he says. So while customers used to queue to see a bank clerk, most now access services without delay and do the legwork themselves, reducing the banks’ staff costs. Customers are “delighted, and the banks have managed to reap all the benefits of digital productivity”, says Kelsey. “We need that same kind of transformation, based on transparency of data, in our public services.”

By launching an independent arbiter of hospital care quality in the shape of Dr Foster, Kelsey himself laid some of the groundwork that could that kind of transformation within the medical field. He’s keen to point out that “there’s been a long and honourable tradition of transparency in healthcare, going back to Florence Nightingale – who was, of course, primarily a statistician not a nurse, and published in 1859 something called Notes on Hospitals: the first publication in the world of a comparative league of hospital death rates.” Nonetheless, he acknowledges that many medical staff initially resisted the idea that a business should score their performance. “It wasn’t greeted unanimously with approval by the medical profession,” he says diplomatically. “The profession was nervous because they didn’t know what it meant.”

Yet nowadays, says Kelsey, “it would be almost unimaginable to have a conversation with a doctor of the kind that we were having in 2000, because [transparency] has become a deeply ingrained part of the system.” Like every health secretary since Alan Milburn, who succeeded Frank Dobson in 1999, Kelsey is sure both that the British public require more information about – and control over – the health services they receive, and that data on service quality must form the basis of NHS and patient decision-making. “In 10 years, we’ve gone from: ‘Oh God, that could be potentially dangerous,’ to [transparency] being the fundamental organising principle of healthcare,” he comments. “There’s been a complete transformation in engagement with the power of transparency to drive real improvement.”

Onwards and upwards
Over the last decade, Kelsey says, the UK has moved “ahead of the world on transparency in healthcare, but we’re still in the foothills of what needs to be done”. So while “we do have good transparency around, for example, hospital care, we really don’t have very much at all around primary care.” To support the current set of NHS reforms, he believes, we’ll need to develop “an integrated information infrastructure”.

The people running the NHS are well aware of these gaps, he says, and in the spring the Department of Health will publish an information strategy; meanwhile, next month the government will start publishing data on individual GP practices “for the first time anywhere in the world”. Outside healthcare, Kelsey accepts, there’s been a “relatively variable response” to the challenge of open data, “but I’ve been really impressed by the speed at which departments have embraced with urgency the challenge.” The transparency announcements in last week’s autumn statement, he says, “underline how far people have gone right across the piece. My own sense is that Whitehall is moving quite fast.”

If departments are moving fast, though, they’re starting in quite different places; and the quality of data held on Whitehall, as Kelsey acknowledges, is often significantly lower than that of frontline service providers. When the coalition came to power, its original transparency proposals concentrated on central spending and management data rather than the kind of delivery-end performance info that Kelsey focuses on – and, as CSW discovered while examining departmental business plans for our last Special Report (p13, CSW 5 October), much of this data isn’t robust or comparable enough to be used as the basis for decision-making.

“There are some big issues around departmental lack of comparative capability,” acknowledges Kelsey, slipping into consultant-speak (he’s on secondment from McKinsey). “I think there’s a proper challenge in relation to enabling central government, for its own business management purposes, to develop a much more comparable transparency, and I think that’s understood and noted.” Meanwhile, though, he wants to concentrate on transparency at the sharp end. “Public service delivery has been the focus of the current round of [initiatives], because we need to give our hospitals, our schools and others management information with which they can address themselves to the financial challenge,” he says. “The productivity agenda has been our focus in the current proposals.”

Comparing like with like
That agenda, of course, requires organisations to publish data in a way that enables service users, providers and businesses alike to make meaningful comparisons – but here Kelsey seems to have changed his emphasis since the summer. In July, he told Civil Service Live: “The centre has a fundamental role in setting standards for the collection of data so that it can be comparable. That isn’t contradictory to the notion that there should be flexibility in local presentation or metrics development.” Asked the same question four months on, however, he stresses that “transparency is the opposite of central government telling people what to do”. While he accepts “a role for central government in setting standards so that at least a portion of [the data] is comparable”, he says it would be “entirely wrong-headed for the information marketplace in, for example, Luton healthcare to be dominated by national standards. It will be what the local people who use healthcare in Luton want it to be.” The “real drive” for comparable data won’t come from government, he says, but from service providers who wish to benchmark themselves and learn from the top performers.

In most public services, Kelsey is confident that there are clear forces pushing for greater transparency; but on local authority performance, he sounds less sure. At Civil Service Live, he warned that with the loss of the Audit Commission, “we’ve lost a big swathe of crucial, in my view, indicators relating to the national indicator set”. And while he now studiously avoids commenting on the role and quality of the Audit Commission, he accepts that “there is a priority for open data in relation to providing a much more meaningful information market in local government. Where there is a problem, I think, is that we need more transparency in local government outcomes.” Too often, Kelsey believes, service commissioners award contracts without strong data on the provider’s performance; and when the Audit Commission is no longer around to check service quality, data transparency will become still more important so that commissioners and service users alike can make informed choices.

Risks and solutions
As the open data agenda moves forward, risks will arise; but while Kelsey acknowledges the dangers, he believes that most can be turned into benefits. For example, if public bodies publish data “that doesn’t tell you anything at all, we’d run the risk that the public will become cynical about the value of transparency,” he says.

If the data is simply flawed rather than impenetrable, however, “publication is itself the answer to improving data”, because “by exposing it to daylight you quickly spot the blemishes and inaccuracies”, says Kelsey. “That first horrible moment when you publish data, you have to hold your breath because the first thing you learn is that the quality is not what you thought it was going to be. But we know from our experience in healthcare that that initial moment passes quite quickly” – and service providers, embarrassed by the flaws in their figures, quickly improve data collection and handling (this, he believes, is where many of the departments are currently with their management data).

Another risk is that data can reveal the kind of ‘postcode lotteries’ that so excite the media, Opposition frontbenchers and backbench MPs. But Kelsey is confident that the government is ready to ride this one out: “I’m really impressed by the degree to which this administration is committed to openness, despite the fact that it won’t always deliver a comfortable outcome,” he says. And again, this problem contains the seeds of its own resolution: Kelsey argues that data showing variable performance can be used to tackle black spots and concentrate resources where the community needs them most, often by directly involving local people. When Miami’s city council began plotting complaints about street cleansing on a real-time website map, he says, it was able to create a much more responsive service – “and suddenly the streets got a lot cleaner”. It’s important in such cases to not only “provide the mechanics by which people can participate, but also to help people learn how to participate”, he adds – a springtime white paper will concentrate on participation as well as transparency.

Other sceptics have argued that transparency can create perverse incentives by, for example, encouraging surgeons to turn away patients needing risky operations to protect their performance figures. “Thankfully, the evidence is that we should trust our professionals,” comments Kelsey. “What’s happened in healthcare is that doctors have taken the data, responded to it, and massively improved their outcomes.” Six years ago, he adds, heart surgeons “took the bold decision to publish the data on their individual performance, and now they’re reporting improvements in death rates from core heart surgery procedures of 20 to 50 per cent.”

Money money money
One other major obstacle lies in the path of greater public transparency: the need for some key public organisations to produce revenue from their expensively-gathered data. Asked how the government will balance the push for open data with the income requirements of bodies such as those absorbed into the nascent Public Data Corporation (PDC), Kelsey says that the recently-closed open data consultation revealed “a clear consensus that the government will create more value by releasing data more freely when it can”. The PDC’s component bodies could be doing more “to get more of their core data out into the hands of innovators and entrepreneurs more freely”, he adds.

Last week’s autumn statement announcements, he continues, “speak clearly of the commitment of those trading funds to maximise the economic growth performance of their data, which does mean giving much more of it away either at lower or no cost”. Nonetheless, he adds, “that’s not to say that there won’t be some contexts where data has to be paid for, because industry should bear some of the burden of the costs of data collection.”

Kelsey is clearly setting out a course towards providing more free and subsidised data for business use. Indeed, he even argues that private contractors could be subject to the same level of transparency as public sector service providers. “Broadly, the principle should be that transparency should follow the money, and that people who receive public funds should welcome the opportunity to describe how they’re spending that money,” he says, though he recognises that “there are lots of tactical questions that we need to follow through” (2012’s open data white paper, he says, “will resolve those questions”).

However, he adds, the trading funds won’t be left out of pocket. As well as releasing more data more cheaply, says Kelsey, the PDC will be tasked with “driving synergies through bringing [the PDC agencies] together in a more joined-up way, and enabling them to benefit more from partnership with the private sector to improve investment and their own technology and capacity”. The result is, he believes, that there is “no conflict between making more of the data freely available to entrepreneurs, and at the same time having a coherent business objective of cost recovery in those bodies.”

Finding a way through these complexities will, however, not be an easy task. Kelsey argues that the business department – which oversees the PDC – is sold on the agenda he’s sketching out: “They’re not immune to the idea that economic growth is crucial,” he says. Yet with the open data white paper due next year and the results of the PDC consultation still to be decided, Kelsey still has a lot of work to do inside government.

Given this, while Kelsey’s secondment from McKinsey is due to finish at the end of this year, he hopes to stick around for a while: “I’d expect that to be extended,” he says. Whitehall is moving really fast on transparency: “I’m not having to push departments nearly as hard as I thought I’d have to,” he comments. “I do feel that we’re very rapidly embedding a sustainable cultural change.” Yet there’s much left to do, in a field of government work which has vast potential to improve performance on three crucial fronts: service quality, cost-effectiveness and user experiences. “We’ve done an enormous amount in the last six months, and departments are embracing the agenda,” says Kelsey. “But I’m certainly not going to be leaving it until we’ve got to a natural point of it bedding in.”