Government and IT - "a recipe for rip-offs": time for a new approach: Further Report, With the Government Response to the Committee's Twelfth Report of Session 2010-12 - Public Administration Committee Contents

 
 

 

Appendix 2: Comments from Professor Helen Margetts, Oxford Internet Institute, University of Oxford, and Professor Patrick Dunleavy and Jane Tinkler, LSE Public Policy Group, London School of Economics

Note: These comments focus only on the Government responses to PASC recommendations which we judged to be of the most importance. Hence the coverage of PASC recommendations below is not comprehensive and covers only some of the Committee's suggestions and government responses.
PASC Report   Government's response  PPG comments 
PASC Report

Recommendation

On Suppliers: paragraph 30

 

30. Extremely serious allegations have been made about the behaviour of some large suppliers. There are clearly very strong feelings on both sides of this debate. We are not in a position to come to a firm verdict on this matter. Having described the situation as an "oligopoly" it is clear the Government is not happy with the current arrangements.

Whether or not this constitutes a cartel in legal terms, it has led to the perverse situation in which the governments have wasted an obscene amount of public money. The Government should urgently commission an independent, external investigation to determine whether there is substance to these serious allegations of anti-competitive behaviour and collusion. The Government should also provide a trusted and independent escalation route to enable SMEs confidentially to raise allegations of malpractice.  

 8.  The current in-depth reviews of large contracts are looking at both pure commercial issues as well as wider contract management areas including supply chain management. The opportunities identified are then taken forward by the customer or if necessary escalated to the Crown Representative for resolution. A key part of the role of the Crown Representatives is to change the relationship between government as a single customer and significant suppliers, including greater communication and the requirement for consistent and appropriate behaviour. Failure to resolve significant problems could ultimately lead to decisions being taken on whether a contract should be terminated to ensure understanding that behaviour that is anti-competitive or commercially unsound is seen as unacceptable.

9.  Key to the Government's fresh approach to procurement is developing a more proactive approach with the market; and a transparent and fairer system to allow small business and organisations to compete for government contracts. The Government Procurement Service (GPS) has successfully launched the "Dynamic Marketplace", a flexible, easy-to-use on-line tool making it easier for Small and Medium Enterprises (SMEs) to compete directly for ICT work under the £100,000 EU threshold. The appointment of Stephen Allott, as the Crown Representative (CR) for SMEs has enabled us to start to build a more strategic dialogue between government and smaller suppliers.  

The current review of large contracts is, we understand, being run from within the Cabinet Office (which facilitates taking an overall view across government), but which has historically proved weak in influencing government-contractor relations (since it is not itself a big IT spender). The government response here does not take up the Committee's point on commissioning an independent review of the issue of oligopolistic suppliers.

In-depth reviews are presumably taking place under the Major Project Authority, rather than as Gateway reviews. The government point here does not go far enough or give the impression that the government takes this issue as seriously as the Committee does. Although external reviews can be useful for individual projects or departments, they will not help to reduce the predominance of particular 'system integrators' (hereafter SIs) unless they are able to look across the government as a whole. For this central oversight is needed along with authority to influence departmental decisions on new suppliers as well as managing relationships with current ones.

An alternative approach is that used in the Netherlands, where every government department consciously maintains a competitive market for its contracts, typically aiming to keep 9 or 10 contractors involved. Yet there is no hint of a government commitment to major IT departments actively maintaining such competitive markets at the individual department level.

Re point 9 in the government response, proactivity as regards helping SMEs enter the market is welcome. However this will not make a significant difference to the overwhelming percentage of the government ICT market held by a small number of big suppliers. Previous administrations have attempted to reduce this percentage and to some effect. However, after years of effort mergers and acquisitions within the IT industry have again driven this figure up. In order to tackle this issue, legacy systems and contracts will need to be examined and possibly broken up.

The previous gains to SMEs from past, much-hyped, online procurement systems have in fact been very small, because the systems have actually been so complex to operate that highly specialist staff are needed to even understand (let alone meet) government requirements. SMEs cannot afford such high overheads. So it will be important that the new "Dynamic Marketplace" is independently evaluated for its ease of use, and that its operations are continuously reviewed to ensure that its intended effects are being realized.  

PASC Report

Recommendation at paragraph 34  

34. We recommend that the Government develop a strategy to either replace legacy systems with newer, less costly systems, or open up the intellectual property rights to competitors. Alternative means of dealing with legacy systems should be explored with the widest possible range of suppliers, including SMEs.  
 12.  The development of a common ICT infrastructure, with agreed open standards and open Application Programme Interfaces (APIs) will allow government to put a 'wrapper' around legacy systems so that they can continue to deliver business benefit by operating untouched in the background, with new systems developed and built, according to the principles of the ICT strategy, on the other side of the wrapper. This will allow for greater flexibility of policies and services delivered at lower cost and within a shorter timeframe by the widest possible range of suppliers.  Development of a common ICT infrastructure will have little effect on reducing the influence of legacy systems in government. Most legacy systems are already now patchworks of overlapping 'wrappers', reflecting previous efforts such as those described in the government's response. The historical problem has been that the underlying systems are too vital and too big to be shut down. Maintaining and integrating these systems is extremely costly for the departments involved and does not allow for much flexibility in service provision. This response does not seem to grasp the seriousness of the legacy issue in moving forward on government ICT provision.

Simplifying, modernizing and eventually replacing legacy systems has long-term gains in allowing expensive older systems to be switched off, but it has always tended to be postponed by previous decision-makers because of the upfront costs and risks involved. Has the government undertaken any review of major UK government legacy IT systems against benchmark arrangements in comparator OECD countries? Is there a long-run risk register of areas where investment to replace legacy systems will eventually be needed with action dates?  

PASC Report

Recommendation at paragraph 64  

64. The Government presumption in favour of smaller, disaggregated contracts should lead to more direct contracting with SMEs. This will require Departments to invest more effort in managing relationships directly with SMEs meaning that more systems integration work is performed in-house, but this will yield longer term benefits through increased innovation and lower costs. Ministers need to ensure their officials have the skills, capacity and above all the willingness to deliver on ministerial commitments to SMEs.  
 35.  The Government agrees with the Committee's recommendation. A programme of work is underway to develop the capability of procurement staff within government departments to ensure that they are equipped with the knowledge and skills to operate the new lean sourcing process which, as described in our response to the PASC paragraph 57 recommendation above, includes measures to ensure that SMEs are not disadvantaged in the procurement process. The training is being piloted in October and will be rolled out from January 2012.  In order to procure and manage ICT contracts, a wider range of managers within departments need to have a much firmer grip on the issues surrounding working with technology as an integral part of the policy process. The future of public management will be even more extensively digitally driven than at present - this is a long-run agenda which the UK civil service has heretofore marginalized for 'mainstream' managers, representing it as an issue only for IT staff 'on tap but not on top'.

This point should be linked with the recommendations on building up the IT profession within government.


 

PASC Report

Recommendation on Challenges to using Agile: at paragraph 87  

87. Agile development is a powerful tool to enhance the effectiveness and improve the outcomes of Government change programmes. We welcome the Government's enthusiasm and willingness to experiment with this method. The Government should be careful not to dismiss the very real barriers in the existing system that could prevent the wider use of agile development. We therefore invite the Government to outline in its response how it will adapt its existing programme model to enable agile development to work as envisaged and how new flagship programmes will utilise improved approaches to help ensure their successful delivery.  
 44.  The Government will shortly be setting up a cross-government network of Agile champions / practitioners to establish the foundations from which the necessary cultural and behavioural changes can be made to adopt Agile methodologies and make them succeed.  This initiative, with the support of the Agile SME community, will enable the adoption and application of Agile for appropriate projects and programmes.  When UK government uses the word 'champions' it is normally an acknowledgement that no one of sufficient seniority has in fact got the necessary expertise to push through changes. In the e-government era, e-gov 'champions' were often mainline managers with at best amateur-expertise in IT issues, taking on the issue as a secondary or tertiary activity besides their main day job.

So it will be important to know what weight, seniority and expertise these new set of 'Agile champions' will have. What level will they be, what training will they have - and how will they be able to incentivize the rest of their agencies to comply? Will competence in Agile be measured and incorporated into promotion and payment (as e.g. completion of different levels of Six Sigma training is in many private companies)?

UK government is also littered with past examples of 'complete handbook' solutions like Agile becoming a cult that is endorsed uncritically for a few years before becoming rather rigid and fetishized, and then being superceded by a new package of ideas. So how will progress on Agile be independently reviewed and critically appraised, so as to develop elements that work quickly, and drop those that will inevitably not prove useful in practice?  

PASC Report

Recommendation on Security and Privacy: paragraph 99  

99. Governments have learnt that they must secure both personal data and data relating to national security, whilst also guarding against gold-plating its security requirements - which can greatly inflate costs without delivering any tangible benefits. Overclassifying routine administrative and operational information causes unnecessary technology and operational costs, and prevents the public sector taking advantage of the economies and efficiencies of commodity software and new opportunities. It also acts as a further barrier to more effective use of SMEs in the supply of IT goods and services. Government must do more to demonstrate how a risk-based approach is helping achieve a better balance in information assurance.  
 45.  The Government agrees with the Committee's recommendation. Effective security is vital to the efficient and safe conduct of public business and the protection of personal data, but it needs to be applied appropriately and proportionately. The Cabinet Office is currently undertaking a thorough review of security policy as part of the broader drive to modernise and transform the way that government does business. This includes revisiting the way that we value, classify and protect our assets. Where there is scope to simplify policy in this area and reduce the burden on departments and delivery partners, changes will be introduced. A more straightforward approach to classifying assets should enable greater use of commercial-off-the-shelf (COTS) ICT products across government, and potentially unlock significant efficiencies.  This response appears to miss the gold-plating point being made by the Committee. Restricting access to online services behind unnecessary or overdone password-protected walls, such as in the Government Gateway, puts off users from accessing those services, making them less productive and less value for money. Departments need to consider where security and access are not necessary and to review whether highly secured electronic identifiers are needed or not. For instance, in DWP online applications could have been introduced for Jobseekers Allowance from 2001 onwards without electronic identifiers or security being a major issue, because all unemployed customers and all their paper documentations were seen face to face in Job Centres before any payments could be authorized. Yet the Department made zero progress on allowing online submissions of JSA claims until 2009. Similarly across the public sector there are local authorities that require people to register before paying their tax bills or consulting public library catalogues - for no apparent reason.

It is important that the government and public managers should recognize that often only parts of all transactions need to be kept secure. This needs to be examined carefully to ensure that government is not over-securing transactions and data, thereby restricting digital accesses by making services more difficult to use for customers, and making their service processes more expensive than needed.

There are also substantial opportunities for government to work proactively with key private sector areas (for example the main banks), so as to strengthen private sector security provision in ways that help government to minimize its own additional security or electronic identifier costs - as Sweden has done, using e-banking identifiers extensively across government services.  

PASC Report

Recommendation on "An intelligent customer?": at paragraph 108  

108. Managing suppliers is as important as deciding who to contract with in the first place. To be able to perform both of these functions government needs the capacity to act as an intelligent customer. This involves having a small group within government with the skills to both procure and manage a contract in partnership with its suppliers. Currently the Government seems unable to strike the right balance between allowing contractors enough freedom to operate and ensuring there are appropriate controls and monitoring in-house. The Government needs to develop the skills necessary to fill this gap. This should involve recruiting more IT professionals with experience of the SME sector to help deliver the objective of greater SME involvement.  
 45.  The Government agrees with the Committee's recommendation. One of the work streams of the ICT capability strategy, to be published in October 2011, will be to develop a common terminology and skills model for the ICT functions retained in house.

46.  One of the core roles within the Crown Representative function is the co-ordination of the views and requirements of HMG customer departments. This will ensure that government requirements are adequately represented, speaking with one voice, when undertaking important decisions regarding the management of suppliers, letting of contracts and placing of procurements.

47.  The government notes that lack of requisite commercial skills are equally problematic; technical expertise alone is not adequate.  

HMG outsourced almost all of its IT expertise in the 1990s and it is taking a significant time and resource push to reverse, especially in key policy-making departments like the Cabinet Office and Treasury. There is not yet the necessary cadre of high quality IT professionals within government to be able to compete in any effective way against large SI companies in providing in-house expertise on ICT issues. It remains unclear whether enough expert government staff can be retained to undertake the key IT functions by just doing procurement and contract regulation. Countries with better in-house government IT staffs, like Canada and Netherlands, often undertake IT operations and development roles that in the UK have been outsourced. Developments like cloud computing may also change the situation further.

Although the CIO Council provides leadership across government, few CIOs are on Departmental Boards, and therefore there is not a strong voice arguing for the importance of issues around being an intelligent customer of IT. And the other senior officials on Boards may not have the necessary expertise or acceptance that digital changes are fundamental to public management to deliberate effectively on IT-related issues.

The Crown Representative function will be useful but only if lessons learnt are disseminated and used across departments so that SI corporations who have been awarded contracts that have not delivered are reputationally and contractually disadvantaged in later contract negotiations with other departments. This has not been the case in the past. Instead each new contract negotiation has tended to reset to zero in 'groundhog day' fashion.

Overall, the governmental response to this particular recommendation is vague and does not engage with the Committee's points.  

PASC Report

recommendation on "Spread of skills": at paragraph 115  

115. Knowledge about how modern information systems and technology can be used to improve public services should not be restricted to the IT profession - this knowledge is essential to the work of all senior civil servants responsible for designing and delivering policy. The Government should explore how departmental boards and senior officials can best benefit from professional training and support in technology policy. A systematic programme to improve these skills across the senior civil service would also help support the Government's aim of ensuring public services become "digital by default" by improving the integration of technology and policy throughout the policymaking process.  
 49 The Government agrees with the Committee's recommendation. The Government is exploring with senior officials how we can best develop training and support for the senior civil service in technology policy. A progress update will be provided by the end of the year.  This is very vague and unspecific and it is not clear how 'exploring … how we can best develop training and support..' will progress the 'digital by default' agenda. Officials at a wider range of levels in mainstream service operations need to be incentivized to place digital technologies and innovations at the heart of their operations - for example through rewards for 'pushing out' dissemination of information via social media and digital channels. Similarly, driving up the usage of electronic services should be rewarded, especially where it makes possible the running down of other, more expensive channels.

? An indicator of how far government has to go in modernizing civil service attitudes here is this. Across most government websites, almost none will 'replay' to users any feedback on what webpages or e-facilities other people have found useful - a practice that has been standard in the private sector for more than a decade now.  

PASC Report

Recommendation on "User engagement in service design": at paragraph 162  

162. It is self-evident that the people using systems, be they frontline officials or members of the public are best placed to provide suggestions on how to improve them. User feedback should be directly integrated into the design of new systems and the development of existing systems and processes to ensure continuous improvement. We recommend that Departments exploit the internet and other channels to enable users to provide direct online feedback both in the design of services and in their ongoing operation and improvement.  
 73.  Many departments are already actively using user-centred design tools and principles and ensuring that they have detailed customer input at the start of the service design. Examples of departments who are engaging in user centred design include DWP for the Universal Credit. DWP has created a customer immersion centre where customers - be they claimants or operations users - will critique each part of the user experience as it is developed. Other recent online services such as the Skills Funding Agency for Next Step (National Career Service) and the Driving Standards Agency for driving tests where improvements based on detailed user testing on the current live services are currently being built.

74.  Users of the current Directgov site are provided with the ability to rate and comment on every article - eliciting 40,000 ratings and 10,000 comments a week across the breadth of government services. This feedback is passed onto all relevant departments and is used by many of them to improve the articles and services, supported by the customer input.  

The examples of user engagement in service design given here are useful. However these are very far from being usual practice within departments. In the past there has been little evidence that where user views are sought, significant changes are made as a result. It would be helpful for departments to have to report on how they have responded the comments about their services received on Directgov to on an annual or a bi-annual basis. Getting 10,000 comments per week - over 500,000 per year - is a huge amount of free information being provided to the government on the strengths and weaknesses of its service provision. It would be a shame to lose the value of this information through departments being able to 'cherry pick' for action only those comments that fit with their current thinking.

Furthermore, the emphasis in the governmental response are in explicit user input posted directly on the Directgov site and possible user consultations. What about analysing the patterns of transactional service usage under different 'experimental' modes of presenting information, issues and service opportunities - e.g. companies routinely run such online experiments to find out from objective metrics what works and what does not. Yet in government the emphasis has been on having only one form of presenting information or services, with experimentation being outlawed. Equally there are many, many opportunities for government to routinely harvest less censored or conditioned comments from social media produced outside the Directgov site - on Mumsnet or Patient Opinion, for instance. When will government departments begin to do the full range of intelligent digital research to ascertain how services are being perceived?  

PASC Report

Recommendation on "Open delivery of online government services": at paragraphs 167 and 168  

167. Government should open up online service delivery to non-public sector organisations and explore ways in which public services can be offered through other websites, applications, devices and providers. This should be developed by providing an open Government platform around which others can innovate and improve, built on the principles of open data, open standards and open source.

168. In doing so Government will need to address issues of liability for the external delivery of Government services. Moving to a model where third parties provide online Government services will require clarity about where citizens should turn for help when they encounter difficulties, as well as clarifying who is accountable for service delivery.  

 76.  The Government agrees with these recommendations. We are committed to delivering better public services digitally by default and the GDS has been established to transform government digital services, ensuring the Government offers world-class digital products that meet people's needs. The current areas of work for GDS include:
  • Single Domain for Government - GDS is developing a prototype "beta" of a single domain for government, which will include a public beta of citizen facing content and a private beta of a shared corporate publishing platform, as well as a first draft "Global Experience Language" for the gov.uk domain. The domain will be built on the basis of open APIs and focussing on delivering only what needs to be provided by Government, not what can be provided elsewhere.
  • Directgov - Directgov is the current single website for access and information about government services in the UK. Currently the content of the Directgov site is available to anyone who wishes to use it in five open standard formats. There are currently over 450 subscribed users of the Directgov syndication web services, which enable subscribed users to make Directgov news, articles and government contacts available through their own websites, intranets, widgets or applications. The new single domain will build on this approach to ensure open standards are maintained;
  • Digital by Default and Assisted Digital GDS is working with Departments to ensure that they deliver services digitally by default, in the places and forms that users want to access them, however, in doing so assisted digital approaches will be in place to ensure that there are appropriate forms of support for people who are unable, for whatever reason, to access or use digital services.
  • Digital Engagement - improving the way citizens can interact with government online through collaboration, conversation and consultation. This includes freeing up information and facilitating the use of new platforms and tools, such as third party communities and sites, to inform citizens and listen to their views as well as introducing digital tools into the day to day working of government.
 
The government response does not take up the Committee's Recommendation at paragraph 168. Opening up access to government data is a valuable initiative. This will however open up the possibility of suppliers who are external to government (whether private or voluntary sector) being able to provide services directly to the public using government data. This could be both cost-efficient to government and of use to customers. However, it would need to be carefully managed by government and other public sector agencies to ensure that they were fulfilling their duties to provide equivalent services or access to services for all customer groups. It will also be vital that government does not lose access to the information on its services and lose touch with its customers (hence our points above about much more digital research being needed). Government will also need to be a much more intelligent holder of data from digital services.

It is not yet clear in what ways the Single Domain for Government will differ from Directgov. How will these major digital resources be used together? How will other government sites be managed and closed?  




 

 
previous page contents next page

 

 
© Parliamentary copyright 2012 
Prepared 26 January 2012