Data sharing workshop II | 19 January 2016 | Meeting note

A version of this note for comments is available here: https://docs.google.com/document/d/1pRYfeLvr1KIqnPP9ANZyni-r5EID1RuUB1Eozhi6M2c/edit?usp=sharing

Introduction

Tim Hughes welcomed attendees and explained Involve’s role as facilitator of the discussions for the Data Sharing proposals.

Sue Bateman explained that the aim of the session was to work through the proposals and to understand support and on-going questions or concerns. The views of the Open Policy Making (OPM) group would be reflected back to the Minister for the Cabinet Office as part of further advice on the consultation package and process. The session allowed for participants to voice their views on the proposals that most interested them.

Principles

It was agreed that the group continue to abide by principles, but remain open to changing the day’s schedule to run better at a majority of attendees’ behest.

  • Transparency– non-attributed comments but everything will be written up and made available.
  • Accessibility – everyone is open to engage if they have interest.
  • Collaboration- all participants are constructively participating in the process to improve the proposals.
  • Independence – there is no expectation of reaching consensus as an outcome but to know where the areas of broad agreement lie is the aim.

The first half of the workshop focused on the core package of proposals to closely understand the positive points for each, and where the OPM group felt that more work is required.

During the second half of the session, the group considered the new proposals along with the consultation process and next steps.

Presentations

Each of the leads gave a short presentation to recap the main points and safeguards for their respective proposals. The full proposals paper is available here.

Identified Data

Key proposals:

  • To give the National Statistician the right to securely access identified data for the purpose of producing aggregate national statistics – from public sector bodies, charities and businesses.
  • To securely share data with Devolved Administrations to meet their statistical needs arising out of further devolution.
  • Reaffirm strict penalties for misuse of data
  • Create an obligation for data suppliers to inform/consult with the National Statistician before changing underlying stats collection and transmission systems so as to protect the integrity of official statistics outputs based on these data

Safeguards:

  • UKSA’s independence from ministers, and direct reporting lines and accountability to parliament.
  • Standards, a code of practice and ethical principles will be consulted upon publicly.
  • Statistical disclosure controls exist and will apply to these proposals.
  • The National Statistician’s ethics committee will have an important role in evaluating relevant ethical questions.

De-identified Data

Permissive power proposed to:

  • Apply ​to all public authorities, with the exclusion of health and social care providers, to link de-identified data for the purpose of research in the public interest​.
  • In order to use this power, a process for ​the de-identification of data must be followed (removing ​of ​personal identifiers).
  • The legislation itself ​has been drafted with a certain level of detail and ​seeks to strike a balance between transparency of process and robust legal definitions on the one hand and flexibility (using measures such a delegated Ministerial power to amend through secondary legislation) on the other​.

Safeguards include:

  • Use of accredited access facilities for data use.
  • UKSA to govern the accreditation process
  • A code of conduct or similar​ will be published,
  • ​with sanctions for non-compliance by participants including withdrawal of accreditation​
  • Criminal sanctions will apply to individuals and organisations for unlawful disclosure  of data, with a maximum penalty of up to 2 years imprisonment, a heavy fine, or both.

Fraud

Permissive power proposed to:

  • Act across devolved administrations.
  • The piloting of the power will be detailed in legislation and be subject to ministerial oversight and review.

Safeguards:

  • A review will take place after three years to determine how well it is working, at which point it could be sunsetted.
  • Public bodies will have  conduct guidelines.
  • Unlawful disclosure rules will apply.
  • A clearly defined process for public accountability will be built into the power.

Tailored Public Services

Permissive power proposed to:

  • Ensure citizens receive the services they need, when they need it
  • The power will enable public authorities to share data to allow better delivery of public services.
  • Power is limited to a number of social policy outcomes, specific objectives will be listed on a schedule.
  • The power will only be used in specific circumstances.

Safeguards:

  • There are definitions of when it could be used, ensuring that the objectives are intended only to benefit citizens.
  • Objectives could be amended in a schedule but this would go through an affirmative process, public bodies can be removed or added.
  • A code of practice will exist to detail how this power is intended to be used, including transparency measures about describe intent about how data is used.
  • Use of power must comply with Data Protection Act and Human Rights Act.

Discussion groups

The room split into four groups to discuss each of the proposals:

De-Identified Data discussion

Positive Points Raised in Table Discussion:

  • Proposal builds on existing good practice.
  • Standardises possibility of data sharing (eliminates the “if it’s external then probably not” argument).
  • The proposal specifies ‘public interest’ as a requirement.
  • There are criminal sanctions for misuse of data.
  • There is clear provision for accreditation.

Issues/Questions Raised in Table Discussion:

  • ​The need for clear and consistent messaging in consultation (language)
  • ​Do we need an appeals process, if not a ​mandatory gateway?
  • Does the current absence of ​health data cause a significant problem​?
  • ​How is ​”​unlawful disclosure” to be interpreted? Should there be a specific offence for “​re-identification​” of de-identified data​?
  • ​A definition of public interest, or public benefit, or how these will be defined, would be helpful.
  • A related point on defining the public interest is whether this would include benefits to private sector, i.e. research which led to increased profits, which could be a potential red line for some participants.

Identified Data discussion

Positive Points Raised in Table Discussion:

  • The proposals will lead to more robust data and National Statistics.
  • This will remove the need for ONS to make ad hoc data requests.
  • Proposals will aid research.

Issues/Questions Raised in Table Discussion:

  • Clarification is required about ‘obligation to consult National Statistician’, what does this mean?
  • How does this fit with EU data sharing regulation developments?
  • Clarity is required around terminology of data sharing/linking, where are the differences?
  • There needs to be read-across in policy documents between these proposals and the de-Identified strand.
  • How will the relationship with private companies work?

Fraud

Positive Points Raised in Table Discussion:

  • This proposal will help to simplify a very complex legal landscape.
  • The proposal will Improve transparency.
  • Pilots and sunset clauses are sensible (but details need to be fleshed out further).

Issues/Questions Raised in Table Discussion:

  • How, when it comes to the evaluation of the gateway and the pilots, can we be sure this is being looked at objectively and counter the risk?
  • Is the three year window sensible? Is this long enough to make a sensible evaluation?
  • We need to be forward looking to new legislation like the GDPR on the horizon.
  • Lots of detail about which organisations, but there could be more information on what they could share, and where the boundaries lie.

Consultation Themes Raised in Table Discussion:

  • Detail required around code of practice
  • Evaluation criteria needs work
  • Exploring more about private company covered in the legislation – what are the safeguards around that?

Tailored Public Services

Positive Points Raised in Table Discussion:

  • Theme about transparency and streamlining the processes – increase the transparency for everyone.
  • Safeguards are well embedded.

Issues/Questions Raised in Table Discussion:

  • How to define benefits to citizen.
  • Cannot publish the code of practice but should explain what is going into it
  • Future-proofing and accidentally breaking future DPA?
  • Is there work around funding and training for new data skills?

Consultation Themes Raised in Table Discussion:

  • Define what ‘benefit’ means in this context.
  • Detail how best to engage with public both during and after consultation to ensure they are aware of the powers and their intent.

At the end of this first session, Tim asked the group whether they agreed that the identified strand should be brought back into the core set of proposals subject to addressing questions and points raised in discussion, this was agreed.

Subject to addressing issues of detail that had been raised, participants were in agreement with the approach of the respective core proposals.

New Proposal Presentations

After a short break, the room reconvened to discuss the new proposals being reviewed for inclusion in the package.

Debt Presentation

Proposed to:

  • Enable public sector bodies and private bodies who carry out a function on behalf of a public body, to share data for the purpose of better debt management across government,
  • Identify people with multiple debts to government, to enable better repayment structures, provide appropriate advice and support to vulnerable debtors who owe multiple debts, and to help create a more ‘single debtor view’.
  • These debts are legally collectable and enforceable, and all appeals processes and disputes have been concluded.
  • This does not introduce anything new. Public authorities already have the ability to share debt data. However, these gateways are restrictive, often misinterpreted, and are complex and time consuming to use.

Safeguards:

  • A schedule will list bodies who can share with one another.
  • Pilots will enable minimal amount of data to be shared and will help to create a single debtor view, must be able to identify vulnerable debtors, and will follow best practice around reclaiming debts.
  • There is a review clause of three years, although concerns have been acknowledged around the process of this.

DECC Proposals

Proposed to:

  • Provide a public service to citizens living in fuel poverty by enabling a wider group of households to receive an automatic energy bill rebate, provided by their electricity supplier.
  • This policy is currently only available to those in receipt of pension credit and takes place every winter.
  • This power would allow an expansion of the policy, allowing energy companies to provide automatic rebates to more people who need it, notably low-income working age individuals and families.
  • It would also allow this assistance to be prioritised for those in greatest need – i.e. those who live in the coldest homes.
  • Recipients could be made automatically eligible for other forms fuel poverty support, such as energy efficiency.
  • The ability to data match would remove the need for the customer to provide evidence, including copies of sensitive documents such as benefit entitlement letters i.e. it would be safer, securer and less intrusive than current methods.

Safeguards:

  • A minimum amount of data is involved – no data would be shared outside government except for an eligibility flag to inform energy companies which of their customers are eligible. Steps are also taken to minimise the overall number of customers involved in the data match.
  • Criminal sanctions would be brought against companies misusing data.
  • The scheme administrator, currently Ofgem, would set up strict data protection procedures.
  • The Data Protection Act applies.

GRO Proposals

Proposed to:

  • Introduce a discretionary power that would allow the sharing of records, such as a birth record, with public bodies to ensure they can fulfil their respective functions.
  • The overall proposal is discretionary, allowing for civil registration information to be shared on a case-by-case basis, or in some areas on a bulk basis.
  • This proposal would not change governance of civil registrations or override any restrictions on sharing data of this nature.

Safeguards:

  • The introduction of a statutory code of reference developed with the ICO,
  • A Code of Practice to be laid before parliament.
  • MOUs and data sharing agreements would be required.
  • Ministerial notification of data sharing in relation to GRO data is required.

Group Discussion

Attendees were given 10 mins to note points against the various new proposals Tim then summarized for the room in plenary. The room then broke into three tables for further discussion of the proposals. Attendees could choose to discuss the two proposals they felt most strongly about.

Debt Discussion:

Positive Points Raised in Plenary:

  • The idea that a single communication from a person who understands the entire case to a debtor, in principle, is good debt management.

Issues/Questions Raised in Plenary:

  • Most of the proposals are about data in relation to a policy, the problem is that this is the data sharing aspects of a policy that is yet to be made. We do not have a final proposal or policy outline, just a gateway without the debt management proposals to go with it.
  • Criticism that this is not a data-sharing proposal, it is a debt management proposal.
  • Without a centralised debt management system, the data sharing aspects apply to an undefined outcome.

Responses to Issues Raised in Table Discussion:

  • The policy proposal is to enable better debt management across Government, that is fair to our debtors.
  • The power to allow departments to share data currently exist, this power would simply enable a quicker process, and a more joined up approach.
  • It is recognised there are significant problems with the single debtor view, without piloting in the first instance these cannot be worked through to begin delivering benefits to government.
  • Pilots will enable departments to explore different ways that the power could be used, in line with the constraints set out in the Code of Practice.
  • A centralised debt management system could form one of the pilots. This would help identify overlapping debtors and enable recovery in a unified way. This is an objective that the Government, and NAO and many departments have.
  • The macro view of debt and department interactions is required through piloting otherwise we can never get to the point where we have a single debtor view.

DECC Discussion:

Positive Points Raised in Plenary:

  • It will achieve maximum possible uptake of the benefit by those eligible.
  • Removes the burden on vulnerable groups to actively apply.
  • Better use of limited resources to support young people.
  • Improved access to data for specific purposes.
  • Without this measure, suppliers would still promote energy efficiency to their customers (and the customers of others), but the marketing would be untargeted. This would result in a great many households, including the vulnerable, getting information about offers for which they turn out not to be eligible.

Issues/Questions Raised in Plenary:

  • Needs to be more safeguards around onward data usage (to prevent marketing, at least for non-fuel poverty purposes etc.).
  • Need clarification on whether the proposers have considered how this data sharing will potentially interact with the Internet of Things in the future, for example, with the greater use of Smart Meters and what their conclusions were
  • Reasons required that set this apart from Tailored Public Services.
  • What is the role of OFGEM in these proposals?
  • Is there an opt-out mechanism for people who do not want to receive the benefit?

Response to Issues Raised in Table Discussion:

  • These proposals are separate to TPS because they extend to private companies (which is out of scope for the core TPS proposals).
  • It is out of scope for sensitive data to be shared with the private sector using this power.
  • Energy companies would only be permitted to use the eligibility information for the strict purposes for which it was shared – i.e. for alleviating fuel poverty.

GRO Civil Registrations Discussion:

Positive Points Raised in Plenary:

  • Proposals will offer better services for citizens, and streamlining for some services.
  • Proposal clearly states policy application and how it will benefit citizens.
  • Wider benefits outside of government if used by organisations such as banks.
  • Clear benefits for digitising government services.

Issues/Questions Raised in Plenary:

  • Digitization of data is patchy and there will be gaps which need resolving.
  • Will there be an opt-out for individuals?
  • Individual data sharing cases are one thing, but the details of bulk sharing need further detail.
  • There are broader implications of this work in creating a de-facto ID database which need to be mitigated against.
  • Will this work have an inherent increased workload for Local Authorities or will it be covered by central government?
  • Are there links to Verify with this work?

Response to Issues Raised in Table Discussion:

  • Data  will be held by central government.
  • Sharing will take place via a data matching process.
  • There are no intentions to link up the records data in order to create an identity database.
  • The purpose of the data share will be specific to the public function of a department or body  and restrictions  on sharing information will continue to be in place.
  • Consent will be required in certain circumstances.
  • There are potential benefits from sharing information wider than government, however this proposal is focused on sharing within government.

High level narrative

Sue shared a two minute summary of the draft narrative in the spirit of the OPM sessions, in order to gather feedback from the group and help ensure future iterations reflect the :

Data is a vastly untapped resource –  with the potential to change and improve our lives and underpinning next wave of digital service reform.

By using information in Government more effectively we can revolutionise the way we interact with people,

 

  • seamlessly use information better to provide better services every day,
  • develop fit for purpose policy, whilst at the same time reducing costs for organisations and saving money for the taxpayer.

But,

The current data sharing landscape is very complex

Restrictions are often placed on how information government can access from other parts of government, even where it is the greater public interest to share it.

And understanding the application of legislation is difficult – leading to practices that are over-cautious and inflexible, with data untapped or locked away in silos.

There are challenges and barriers in accessing data across organisational or departmental boundaries, often resulting in duplication of data collection and storing, which can be costly to the taxpayer.

While some gateways exist to permit the  disclosure of data between specified public bodies, there are many examples of where they have been created over-cautiously, often significantly limiting what information can be disclosed and to whom.

We recognise that the complexity of the current landscape means it would be nigh on impossible to try and tackle everything – so we have been running the OPM process and set up a consultation look at how we can remove barriers to data sharing within Government in a way that provides a better experience or service for the public.

Better data sharing offers us vast opportunities, in particular around policy development by providing up to date and fuller evidence on issues such as social mobility, economic growth and crime prevention. This consultation forms part of an open policy-making process we have been running together with government and civil society organisations since March 2014.

Our conversations have led to a focus on access to data for the purpose of

 

  • Better research and statistics -whereby  more complete and up to date information would be a rich source for researchers  to use  and would create a clear and accurate evidence base to inform critical decisions
  • The prevention of fraud and recovery of debt – meaning we can put taxpayers money to the purpose it was intended
  • Providing Tailored Public Services  – providing the right services at the right time – and we would see GRO and DECC proposals as part of this broader strand

Our proposals for new legislation will not give the Government fundamental new powers.

They will provide greater clarity in what can be shared as well as flexibility for government to respond to policy and public needs as they emerge.

Data will continue to be protected, using both data protection principles and safeguards and embedding them into any proposed legislation.

Proposed powers would provide  specific and immediate benefits across government to enable better and more efficient services unlock data as a driver of further public service reform. Furthermore, they will allow government to better understand and develop the right policies to support the modern digital economy and the public

Feedback

Everyone was asked to stand up and show their levels of support for the narrative.

Feedback from the room:

  • “The proposals fit into the context of the narrative”.
  • “Proposals save money, more effective service delivery, fits with the narrative”.
  • “more enthusiasm and specificity, e.g. ‘this could improve people’s lives in concrete ways’”.
  • “The knives seem to be out for privacy campaigners in this process, no real detail or why/how the benefits work, other than just restating ‘it’s good to share data’, not much has been shared today, across the board, it’s a learning experience, but crucially a sharing experience”.
  • “we’re missing the context that these discussions are taking place,  positive stories like Google flu trends are being used. Other contexts are ignored, this data revolution that is taking place is out of control and everyone is tracked online etc. we are in the age where we are becoming digital profiles, a spectrum is emerging between tech driven authoritarianism on one extreme to citizens having digital autonomy on the other, the state should reflect on this”.
  • “It doesn’t reflect on who we need to consult with, it’s not addressed to the public or drawing out the benefits by conflating too many things and making them unclear, the consultation respondents should have questions like outsourcing, private companies, funding etc. should already be covered in the narrative, of Local Authorities are to give a meaningful response, this should be framed with this view”.
  • “The official stats, de-identified data and some of the GRO things are about research and anonymisation, with lots of detailed safeguards. None of the questions around debt/fraud etc. maybe it would be better to divide up the package between research elements and efficiency elements, more detail is required in the bill about who is going to appeal, who to, and what are the legal rights?”.
  • “Central issue about the benefits of sharing data, but there are lots of additional proposals bolted on and there is a huge amount of things being consulted on. The debt question jars with the public benefit narrative”.
  • “Debt and fraud sound more like research with the piloting and testing of the system, just looking at a legal gateway in itself is not enough for people to decide upon. In some of the areas,  it needs to be made clear that some parts are for policy development and others for different things”.
  • “When this goes to consultation, separate audiences will have specific interests in specific areas”.
  • “There are fundamental areas that need to be addressed, stating what is trying to be achieved”.
  • “Greater clarity is required, as people new to the process are asking questions that have arisen before, this would indicate a lack of coherence and communication with the proposals.”

Close

Sue explained that taking the various proposals through as a single package is partly to ensure that things were brought to the OPM group that were relevant, rather than as single, disparate work strands.

Tim drew the session to a close – asking guests to think about who needs to be engaged in the process of consultation to try and broaden the process out to as inclusive a group as possible.

Suggestions for engagement in consultation:

The following groups and organisations were highlighted by the OPM group as valuable to contribute towards the consultation process:

  • Health and Social Care Information Centre (HSCIC)
  • Range of local service providers (e.g. for troubled families)
  • National Police Chiefs Council (NPCC)
  • UCAS
  • Schools (time issues – could go through specific vehicle e.g. UCAS)
  • Catch22, Prince’s trust
  • Association of Directors of Adult Social Services (ADASS)
  • Association of Directors of Children’s Services (ADCS)
This entry was posted in Meeting notes on by .
Tim Hughes

About Tim Hughes

Tim is Involve's incoming director, taking over from 21st January 2017. Tim has led campaigns and advocacy on open government; advised national, devolved and local governments, civil society organisations and multilateral institutions; and researched and written on topics including public participation, open government, democratic reform, civil society advocacy and public administration.

Leave a Reply

Your email address will not be published. Required fields are marked *