Public Strategy

Stockholm: Public Services Summit

For the last few years, Cisco have run a public services summit in Stockholm for two days in December (since they are based in California, that's one way to suggest that this is not a jolly).  They have been good at getting high-powered speakers and participants - this year from 55 countries.  Last year David Miliband was the senior UK government representative; this year that honour fell to Joe Harley.

I have written a number of notes of both formal sessions and some of the informal conversations I was involved in.  The complete set can be seen grouped on a single page here.  My notes don't cover the full summit agenda, but I will link to some of the presentations when they become available online.

Stockholm: Breaking down silos

The Stockholm summit divided its time almost equally between fairly formal plenary presentations, some of which I have already written about, and smaller and less formal working sessions, where the discussion to presentation ration was much higher.  I went to the one on 'breaking down silos', billed as being about 'the strategies governments need to adopt to break down departmental silos and secure the real benefits of e-transformation'. 

Speakers included

• Joe Harley, whose slides are here,
• Kim Nelson, CIO of the US Environmental Protection Agency - though perhaps more pertinently, she announced at the end of the session that she was leaving government, and from this month is executive director of e-government at Microsoft.
• Gordon Macatee, Deputy Minister [= permanent secretary] of Labour and Citizens’ Services, Government of British Columbia and Chair of the BC Buildings Corporation, Canada
• Andreas Nemec, Chief Executive Officer, Federal Procurement Agency, Austria

I wasn't taking extensive notes as these were more workshop based sessions, but slides from the other speakers are due to be circulated - I will add links here when I get them. 

Joe's presentation was designed to give a UK government as much as a DWP perspective on shared services and drew on Transformational Government material as well as our own.  It was very well received - the discussion in the group over the following 24 hours came back to many of the points made in it

Of the other presentations, I thought Gordon Macatee's was the most interesting. The British Columbia government is tiny by our standards - HR shared services could mean going from an HR team of eight people for a single department to one of thirty for a cluster of departments - but the underlying issues were all very recognisable.  Gordon's department includes a shared Public Service Operations Agency as well as a service delivery initiative which spans the entire provincial government. The consistent theme was the need for shared services to be customer-driven - with two particular hard edged consequences:

• budgets stay with the customer, they don't go to the provider
• customers always have a right to leave the shared service.

Consequences of applying those two principles include:

• defection is a real threat for some services:  departments have clustered for services such as finance and HR, which makes it viable for an unhappy department to leave one cluster and join another
• on a larger scale, a department wanting to use a provider other than the Public Service Operations Agency has to make a case to a board made up of the deputy ministers of all main departments - the PSOA itself cannot compel its clients' loyalty.

Gordon clearly felt very strongly that these quasi-market disciplines were critical in ensuring that the objectives of PSOA remained effectively aligned with those of its customers.  It was less clear h0w those disciplines could be made to work where the reason for the shared service approach was not (or not just) efficiency, but about the wider customer experience, and in particular the delivery of integrated services to citizens.  Defecting from, say, an integrated online portal is clearly a rather different proposition from defecting from a shared payroll.

My overall conclusions from the presentations and the lively discussions which surrounded them were:

How to make shared services work

• Clear governance, with a very authoritative collective management - in our terms, probably needing to be at permanent secretary level
• Clear initiatives/programmes where a shared service approach is going to be pushed forward - otherwise it is too easy to sign up to the principle while keeping clear of the substance
• Strong leadership commitment - it will always be possible to find short term reasons for not taking this forward, and unless those are transcended centrifugal forces will outweigh centripetal ones.
• Piloting
• Consider both internal and external services - but be clear about the differences between them

How to sustain shared services

• Defined standards (operational and technical) against which performance of the service can be measured
• Plausible threats to withdraw - where ideally, plausibility is more important than withdrawal
• Shared service organisations and organisations which share services:  the former will often be simpler; where there are reasons for the latter there will need to be strong mechanisms to ensure that the provider organisation does not favour itself over other customer organisations
• Embedding cultural change:  the overall effect is to shift the boundary of the make or buy decision:  without confidence in the sustainability of the 'market' for services, the tendency will be to creep back towards the apparent control offered by making rather than buying
• Measuring success:  if shared services don't deliver benefits, they are not worth the overhead.  If they do, the benefits should be identified not just for their own sake but as a further way of embedding change.

Can everything be shared?

• Services for efficiency: yes
• Services for customers: less straightforward.  In principle there may be stronger reasons for sharing, because there are external as well as internal benefits to be captured (which in principle suggests that just looking at the internal benefits will lead to sub-optimal levels of sharing).  But in practice, there may be a trade off between more specialised expertise and more integrated, but more generic, service provision. 

Stockholm: Mapping medicine

A completely serendipitous encounter on a bus led to a fascinating conversation about the Map of Medicine with one of its inventors.  The original idea came from the Royal Free, and is part of the Connecting for Health programme, but it is now being developed commercially (though apparently costs to HMG are zero, because of the origin).  Essentially it's a structured best practice guide for doctors:  it's not a tool for diagnosis, but picks up after that point, and guides the clinician on what to do.  Currently it covers 250 'patient journeys' (itself a phrase which resonates with our channels thinking), and the idea is that it can be updated to reflect the latest approaches and research, rather than having to rely on the distribution of paper with the hope that it is read, filed and remembered.

The parallels with the guidance given to our staff is obvious:  it would be interesting to establish whether the underlying engine is sufficiently generic to support decision-support tools for our business.

Stockholm: Participant led design

William Heath was a last minute addition to the Stockholm programme, giving a rumbustious presentation based on some of the themes of Ideal Government, and a variant of a session he led at the e-government ministerial conference last month.

His main themes were participation and trust - which he sees as inextricably linked.

Trust is not the same as control and is inherently mutual:

if you approach your citizens as if they were criminals, don't expect their trust

if you distrust people, don't expect their trust

His most challenging line, though, was 'users innovate better than producers'.  While we congratulate ourselves for having got as far as 'Let's design public services for users', we should instead be moving on to 'Let's get users to help design'.  The basis for this is that in other areas of design, there is evidence that user involvement leads to better results.  William referred to a Design Council study (which I think must have been this one).

He also talked about the competition run recently by Ideal Government for using public sector data and mapping tools to create new information services - the assessment of the entries is here.  The point William was making was expressed very clearly by Stefan Magdalinski (who, for the avoidance of doubt  is the Stefan referred to in that assessment):

I feel strongly about this. OS charges developers £500 even to have a play with any of the data...then you have to share your business idea with them. This policy fails to create any decent sites or apis or interfaces. It's generally monopolistic and has effectively destroyed a vast amount of value for the nation.

Britain was absolutely the world leader in cartography 10 years ago.

In 5 years, we'll be nowhere, and I think it's OS' fault for trying to extract monopoly rents out of taxpayers' data. This chokes innovation until the eyes pop out of its head. It's as bad as GCHQ sitting on public-key cryptography, letting Whit Diffie get all the credit, and losing Britain the opportunity to lead in what became a mult-billion dollar industry.

Both were phenomenally bad economic decisions.

The public sector sites the OS might point to are disappointing. Now they could be improved overnight and save a bob or 100 just by re-implementing with Google maps. Oh the irony.

There's been more innovation using Google maps in the last *week* than OS has 'driven' since 1994. I have to say that it's not for lack of quite a few of us trying.

Talking to William after his session, we got to the interesting question of whether it was better for public bodies which weren't doing user-centred design (a) to start designing for users, then to move on to designing with users, or (b) to leap straight to designing with users.  I was arguing for (a), but I am not sure I was right.  Views?

 

Stockholm: Conservative radicalism and communicating knowledge

John Hennessy is the president of Stanford University.  PDP11 fans everywhere will also know him as the inventor of RISC.  He was talking about the 21st century university, extolling the virtues of collaboration and mixing cultures, supported by modern communications. 

Hennessy saw four drivers for change in universities:

• Globalisation: the need to appreciate the culture and views of others.  Students should spend time abroad and should learn how to do cross-border collaboration
• The growing importance of science and technology: the absence of real understanding risks a bias to conservatism, so the key question is, what should all students understand about scientific principles and their possibilities?
• How has the world become so ignorant about science and technology?
• by ignoring the importance of science and technology in, for example, the teaching of history in schools [by which I assume he was looking for more of a Guns, Germs and Steel approach]
• by downplaying its importance, which leads to lack of interest, which leads to people not following scientific careers
• The need for a broader range of skills and knowledge, and for new skills and knowledge over time:  quoting Alvin Toffler's line that 'the illiterate of the 21st century will not be those who cannot read and write, but those who cannot learn, unlearn, and relearn'.  Continuing education and life-long learning then become central, and the notion of being a full time student doesn't make much sense.
• Universities should do transformational innovation:  universities excel at technology transfer, because they systematically transfer people who understand innovative technology into a wide range of places it can be useful, through a process known as 'graduation'.

There was a lot of good stuff in this, but in some ways it was curiously conservative:  he put a lot of weight on students mixing in dormitories and spending time on international exchanges as sources of cultural exchange.  That’s an approach which may work well for Stanford, but is hardly scalable and there was quite a lot of muttering from the audience that this was very elitist.

He also talked about using new technologies both to support communication and to provide alternative channels for teaching.  He said proudly that the materials and the assessments – assignments and exams – were all available online and were identical  with their offline equivalents.  At one level, that's a very powerful statement.  But at another, it's rather odd:  given a different medium with different audiences, different economics and necessarily different approaches to learning, why assume that the two should be the same?

The relevance of all this to us is less in what Hennessy said, more in some of the implications and parallels it prompted with our own dissemination and sharing of knowledge, on which I will write a separate note shortly.

Stockholm: Identity management

One of the more interesting breakout workstreams was on 'the trust cluster' - which unfortunately was not the one I was in.  I did though have a very interesting conversation with one of the presenters in that group. Malcolm Crompton, who was until recently Australia's privacy commissioner and is now the MD of an outfit called Information Integrity Solutions.

Malcolm's paper for the summit is well worth looking at, both for its content and for its comprehensive links to other sources. His argument is that 'With care, vigorous debate, and a lot of hard work, we will still have private lives in 20 years from now and inconceivably good government services.'

One of the things Malcolm is currently working on is setting up a private sector provided identity management arrangement for Australia through the banks which draws on the BankID system developed in Sweden.  they in turn describe what they are doing as:

The Swedish Government has embarked on a programme to support the development of so-called “24 -hour authorities�, and is working with its own departments, municipalities and other authorities to facilitate the introduction of this service. This service generally presupposes a secure electronic identification and signature system, such as that offered by BankID.

For example, Rikskatteverket (The National Tax Board) and Riksförsäkringsverket (The National Social Insurance Board) websites are now fully operational, and both offer their customers access to BankID.

There is increasing interest in the facilities offered by the BankID concept and provided via the member banks.

Members of the public can identify themselves and sign documents from authorities, companies and other organisations on the Internet by using BankID’s electronic identification and signature system, the customer’s identification having been previously guaranteed by the member bank.

Authorities, companies and other organisations can check the customer’s identity and signature by using computer software developed by a specialist company working in conjunction with the member bank.

Stockholm: Reinventing Government

David Osborne wrote Reinventing Government over ten years ago, and has suffered the fate of many of his ideas becoming received wisdom, without any particular recognition that he had them (or at least gathered and published them) first.   His current big theme is The Price of Government, the core idea of which is that government is in permanent fiscal crisis - and that even more than before, traditional bureaucracies will not be able to survive in this changed world.

Industrial era bureaucracies v Information age realities
Stable bureaucracies	Rapid change
Traditional work processes	Information technologies
Mediocre services	Public expects quality
One size fits all	Public that wants choice
Slow-moving monopolies	Global competition
Ever-expanding budgets	Severe financial constraints

 

IT, he says, can make things a little faster - but that just makes the customers frustrated faster.  More holistically, he has five Cs which need to be addressed:

• Core - purpose
• Consequences - rewards and threats
• Customer - empowering and accountable to
• Control - moving power away from the top and the centre
• Culture - changing habits, hearts and minds.

To get to the core, his solution is a radical approach to public sector budget setting, moving away from questions such as what to cut and what to tax, which tend to ignore the vast bulk of spend, and get answers focused on managing pain before the next election.  Instead, he suggests a four step process:

• Decide how much revenue to raise - easy to identify what citizens are willing to pay from historical data.
• Identify which outcomes matter most to citizens
• Divide the money between outcomes - on the basis of how much they are worth,  not how much they cost.
• Deliver the outcome
  
  • Experts go to buy the outcome, based on key indicators and a cause and  effect map
  • Challenge is to improve the outcomes at a set price

In effect, this is a slightly specialised zero-based budgeting technique.  David was pretty gung-ho about existing institutions having no right of continued existence, and no presumption that they would get all or any of the work.  The technique has been applied in a dozen or so US jurisdictions - state, county and municipality.  One example he gave was that lifetime health outcomes are only 20% attributable to healthcare (lifestyle is by far the most important determinant), with implication that budgetary provision should be massively switched to preventative measures.  It wasn't clear how this addressed the needs of people whose life had already been unhealthy and who therefore needed medical intervention, not how acute care worked in this model, but the state of Washington was said to be very happy about it all.

I asked him how he thought this would work in practice where budgets were committed as a result of long-term contracts in areas such as IT and buildings. Perhaps not surprisingly, he didn't really have an answer, beyond suggesting shorter contracts.

Laurence Millar, the New Zealand government CIO said that NZ had been doing this for years, and felt that they had got most of the efficiency benefits from improving outputs - so the question was now how to improve outcomes.  Osborne effectively said the approach didn't change, to which Millar responded that outcomes were often multi-causal and involved a number of different agencies, and that identifying and apportioning the contribution of each was far from straightforward. NZ hadn't yet cracked that problem and was looking for help - but I don't think Millar thought he was getting any.