In a speech introducing his political agenda, the (in-all-likelihood) future Prime Minister Andy Burnham discussed changes to the whipping system in Westminster, with promises of “letting MPs be authentic representatives and not using the whip system to create fear or close down debate”.

This is a theme he covered at more length in his book (co-authored with Steve Rotheram) Head North. Their argument is that the whipping system fits into the wider problem of centralisation in British politics, where tight central control over how MPs vote moves power towards an even smaller number of people, who make worse decisions because the wider expertise of the party’s MPs are excluded from the process. Moving power away from MPs throws away the knowledge of place that those MPs represent. 

Here are three practical reforms to parliamentary whipping that address this complaint:

• Define the whip – give a clear scope and bounds to acceptable and unacceptable approaches
• Co-produce the whip – involve the parliamentary party in shaping the party line
• Publish the whip – produce clear public justification for voting instructions

These are reforms that can be implemented by any party without changes to parliamentary processes, but would be improved by changes to parliamentary processes to entrench and support them. 

What is the whip?

In the House of Commons, most votes are “whipped”, meaning MPs are expected to vote in accordance with the instructions the party has given.

There are different schools of thought on what the balance should be between parties and MPs. MPs are elected as individuals, but usually as part of a party with manifesto promises. As such, party discipline is not inherently illegitimate: there is a democratic argument that parties act as an intermediary that can keep their MPs aligned with the public mandate they got in the election. 

In this argument, MPs acting as completely free agents is bad for public accountability. Unlike parties, voters have no mechanisms between elections to keep MPs aligned with election promises, and very poor ones at elections, which work better as an informed choice between party agendas rather than 650 separate elections. 

That said, there is a big gap between this justification for the whip and how it works in practice. Burnham’s complaint is that in practice the whip is not forcing unruly MPs in line with their constituents: it’s doing the opposite. Party discipline goes far beyond just issues that were part of the manifesto and centralises decision making power in ways that make how MPs vote more distant from their constituencies:

“The minute [MPs] walk through the door to take their seats, the power given to them by the public is removed and effectively handed to a small number of their elected colleagues and many more unelected advisers who set the party lines and positions for which they are required to vote. 

“As a result, the whip system disempowers MPs, diminishes their status and, over time, disconnects them from their constituents. It makes you say things you’re not sure about and vote for things you don’t believe in. At its worst, the whip system unfairly makes good people seem like frauds.” 

There is room for these different perspectives on the whip to be reconciled through approaches that remain disciplined and cohesive, but that are also more transparent and emerge from more collaborative approaches to politics. 

Define the whip

A simple first step would be to write down what the expected “bounds” of party management are. 

This would be a statement that itemises political consequences the whip may impose (e.g. an ongoing pattern making it less likely that an MP will be chosen for government roles), versus approaches to party discipline on voting that would not be acceptable (such as denying absence slips, suspension from the party, losing funding for constituency projects, or intimidation). 

This could also set the scope for when the party has the clearest legitimate interest in whipping (votes related to manifesto issues), and confirm rules that there are certain votes (conscience votes, votes on house matters) where either the party might not take a position, or where there are no sanctions attached to not following that position. 

This should be an easy win, and is mostly restating what should already be good practice, or returning to an earlier set of norms. As Philip Cowley notes there’s been a substantial increase in MPs being suspended from the party for voting against the whip, with none in ten years of the Blair government compared to eleven in the first year of the Starmer government. A statement returning to that Blair-era norm would be a substantial reset in relations between the leadership and the parliamentary party. 

Writing the rules down is also part of making Parliament more normal. The “Whips” are a combination of what could boringly be called “party managers”, and a more secretive role, where the ambiguity allows them to contribute to the worst aspects of Parliamentary culture. The threats, bullying and blackmail are always said to be a thing of the past, but where “the past” becomes ever more recent as new stories surface. Greater transparency helps shift this balance, establishing party managers as a legitimate part of the Parliamentary system, whose actions are aligned with an effective and professional working culture. 

This step would, however, not be enough to address Burnham’s objections to the outcomes of the whipping system. To engage with the substance of the complaint means changing not just how whips behave, but how the decisions on the party line are made. 

Co-produce the whip

The substance of Burnham’s complaint is not just that he was told what to do, but he was told to do the wrong thing because the parliamentary party was not involved in setting the policy. 

As he put it, “the government I served in would have helped itself if it had trusted the majority feeling in the [Parliamentary Labour Party] rather than continue to force on it the preferred position of a small number of people at the heart of government.” What we need are approaches that mean the parliamentary party is more involved in setting the line it is then expected to follow. Here we can look to the German Parliament for useful approaches. 

The “free mandate” of German MPs to vote in accordance with their conscience is constitutionally protected. However, in practice, this ends up in a very similar place to the UK Parliament, where MPs almost always vote along party lines. The big difference between these systems is that in Germany the “party line” is not purely dictated by the leadership, but emerges from processes involving party MPs. 

In the German system, MPs in each party form working groups that shadow parliamentary committees. These are substantial policy-making bodies for the party, provided with parliamentary resources and researchers to support their work. 

If we shifted towards a similar system, MPs would develop deeper policy expertise and be more directly involved in shaping the party position on that area or piece of legislation, while MPs not part of that area defer to the process and wisdom of their more engaged colleagues. This reflects the party discipline that is important to clear lines of accountability to voters, while addressing the complaint that centralisation of decision-making throws away the intelligence and moral sense of MPs. 

Aspects of this happen already, and a step a new government could take is to take more formal notice of working groups, encouraging them to form through expectations that the whips office and ministers will engage with them. In time, more parliamentary resources could be made available to parties to support these groups. Another intermediate approach would be to allow the parliamentary party to select its membership on public bill committees, which in theory review legislation but in practice the whips select against members with specialisms in the area to maintain central control of the process. 

MPs are responsible for how they vote, but are judged collectively. The goal for MPs should be higher than the right to say “I didn’t agree with that decision”. This might be satisfying personally, but doesn’t help the people affected by bad decisions (and is unlikely to be electorally rewarded). The goal is to advocate and demand processes that lead to good collective decisions. 

Whether that’s the German model, an evolution of bill committees or something entirely new, the question is not just how we move away from centralising structures, but what new ones we need in their place that act as a counter pressure to government departments in the legislative process. 

Publish the whip

If parties are going to instruct MPs how to vote, they should tell us they’re doing that. Obviously we know in the abstract that most votes are whipped, but in the public record only the votes of individuals are visible.

It’s not that knowing the party instructions would excuse a bad result (MPs are ultimately responsible for how they vote), but it would play an important public education role in both how Parliament works and the quality of the decision making process. 

One of the long running effects of TheyWorkForYou has been to encourage MPs to make more public justifications of how they vote. Publishing the whip would be an extension of that, encouraging parties to make full descriptions of their reasons for votes. This would be a good incentive to move towards a more co-produced process, where the work of producing recommendations has public as well as internal outputs. 

We think greater transparency is the friend of better processes — helping draw power away from centralised and secretive processes towards those that need to explain and justify themselves in public. It’s also popular: 62% of the public agreed that parties should be public with how they instruct their MPs to vote. Committing to publish the whipping information would represent a public commitment to a new way of working, that helps encourage and entrench better practice. 

Building what comes next

Party discipline is a key feature of modern parliamentary democracy, but the specifics of “the whip” in the House of Commons are not. It has changed over time, and can be changed again.  Other places do things differently, and we can look there for ideas about how things can change.

The challenge for Burnham (and for Labour MPs who want to seize the opportunity) is to go beyond just minimising the drawbacks of the current system. What’s needed is to define the new institutions and processes that deliver the goals of a collaborative approach to politics, making the most out of the judgement and skills of MPs, to raise the standard of how decisions are made in Parliament. 

 

Header image: House of Commons Division Lobby | @House of Commons | House of Commons | Flickr 

Representatives in the UK’s Parliaments and Assemblies do many things, but one of the most important is that they make decisions. These decisions shape the laws that govern us, and can affect every aspect of how we live our lives. 

On TheyWorkForYou, we list and display the individual votes of representatives, and for the UK House of Commons, we also create voting summaries that group a set of decisions together into a policy, like ‘preventing climate change’ or ‘increasing stamp duty’ — so for any specific MP you can see how they have generally voted around that issue.

As of this update, the summaries now contain all relevant votes up to  the end of April 2026. This blog post logs changes we’ve made to what’s included, and our ongoing thinking about how new votes suggest broader changes to our approach. 

We’re always working to improve our coverage and how we display and share our information. If you think this work is valuable, and want to help us bring greater transparency, please make a donation to support our work. 

About our summaries

Our summaries group a set of votes into a ‘policy’ (like ‘removing hereditary peers from the House of Lords’) where votes either agree or are against that policy . We label votes as either “scoring” or “informative”: only the scoring votes contribute to the headline of whether an MP is for or against a policy. We also compare an MP to a cohort of similar MPs (MPs of the same party who voted on the same votes) to give an indication when MPs’ voting notably differs from their party. 

We also sometimes include ‘agreements’ (where a decision was made without a vote) as part of policies — but do so conservatively to balance the difficulty of interpreting these decisions with the blind spots their complete exclusion would lead to.  

In drafting new policies and including votes, our main criteria for inclusion are:

• Substance – votes using the powers of Parliament
• Cohesion – we include votes that are mostly, if not completely, about the specific policy
• Uniqueness  – votes that have limited overlap with scoring votes in different policies
• Noteworthiness – where this might reflect impact of the legislation, news coverage, or public interest

The goal of these principles is to create policies where the top level summary is a good guide to the votes it summarises. When votes fulfil some but not all of these criteria, we may include them as informative votes.

You can read more about our approach.

New live policies

This update introduces the following policies:

• Universal Credit (Removal of Two Child Limit) Bill — [Person name] voted for/against scrapping the two-child limit so that families on Universal Credit get extra support for every child, not just their first two.
• Railways Bill – [Person name] voted for/against bringing the railways into public ownership by creating Great British Railways, a single body running most train services and the track in England.
• Northern Ireland Troubles Legacy 2025  — [Person name] voted for/against scrapping the scheme that let people avoid prosecution for Troubles-related crimes in exchange for cooperating, and removing the ban on bringing Troubles-related claims in the civil courts. 
• Northern Ireland Troubles Legacy 2023  — [Person name] voted for/against setting up an independent body to investigate Troubles-related deaths and injuries, ending most other criminal investigations and inquests, and offering people immunity from prosecution in exchange for cooperating.
• NIC Changes – National Insurance Contributions (Employer Pensions Contributions) Bill — [Person name] voted for/against taxing salary-sacrifice pension contributions above £2,000 a year (from April 2029).
• English Devolution and Community Empowerment Bill – [Person name] voted for/against standardizing and increasing powers of devolved mayoralties, replacing two-tier local authorities with larger unitary authorities, and introducing a community right to buy.
• Sentencing Bill – [Person name] voted for/against making judges normally suspend prison sentences of a year or less, expanding suspended sentences and tagging, reducing independence of Sentencing Council from government. 

Votes added to existing policies

• A vote on government regulations around Greenhouse Gas Emissions Trading was added to the ‘measures to prevent climate change’ policy line. 
• Two votes on government regulations around reception conditions and provision of accommodation to Asylum Seekers were added to the ‘stricter asylum system’ policy line. 

Notes

Fiscal drag

TheyWorkForYou has a number of ‘policies’ on changes to tax rates. Over the last few revisions, we have been conforming these where a vote ‘for’ a policy is a vote for a higher rate of tax. 

Where this immediately runs into trouble is the distinction between ‘more tax’ and ‘higher tax rate’, where a period of inflation has made this sharper than would have previously been the case. 

This update shows the problem in two changes. The rate of alcohol duty was increased inline with RPI (uprating): an increase in cash terms, but not in ‘real’ terms. Separately, basic rate and personal allowance has been frozen, which is fiscal drag. As the House of Commons explainer puts it “[n]ot increasing the value of tax thresholds (‘freezing’ them) increases people’s taxable income without tax rates actually increasing”. 

So, with a “change in tax rate” lens (how we currently evaluate this), income tax has stayed the same, while alcohol duty has been increased. But in terms of inflation adjusted tax (roughly) the opposite is true. 

An alternate framing would be to shift to a “tax impact” approach, where a vote is evaluated on whether it increases/decreases tax relative to an uprated baseline (where that baseline would typically be which of the varying possible indexes was defined in law). 

This is slightly more complicated, but is usually clear from the commentary where it matters most. The main problem is consistency retrospectively (although low inflation for the majority of the period covered generally makes this a bit of a wash).

The main impact of this is that as well as reversing the direction of our income tax threshold policy, we are changing some of the language used. 

Moving from:

[Person name] voted for/against raising the threshold at which people start to pay income tax

To:

[Person name] voted for/against increasing income tax (in real terms)

For the alcohol duty vote, we have added this to the existing policy line as a ‘neutral’ vote, which means we have cataloged it as relevant, but it doesn’t contribute to for/against scoring. 

Diago Garcia and votes that go nowhere

We have tagged the votes on legislation around ceding the Chagos islands to Mauritius  for the moment. While our usual policy would be to include votes after the third reading (which has happened), this decision has now effectively been put on pause in terms of the final steps of the legislation.

With the voting summaries, we have a focus on votes that have impact — but substantive votes that end up going nowhere can happen.

Ten minute rule bills (which can pass our criteria on being a substantive vote, but then get no extra time so have no prospect of becoming law) also merit a different approach. Here, we’ll include them if we have an existing policy line, but do not create new policy lines for them.

This means we do not create inaccuracies in existing content (if we say there is a list of votes on a policy, we will include the new vote on that policy), but also do not create new policies for a specific bill when that bill is unlikely to become live legislation. 

This is in line with our approach on legislation that is moving through Parliament while also a matter of substantial public discussion, like the recent Terminally Ill Adults (End of Life) Bill. Here we added the votes to an existing policy around assisted dying as they were happening (because otherwise we would say an MP had not voted in an assisted dying vote), but in the end the legislation stalled in the Lords.

NI troubles legacy

The 2025/2026 bill votes reverses (or affirms the striking down by the courts of) elements of the 2023 Legacy Act.

As such, votes on the original legislation are read as being against the new policy that describes this bill. However, the original act had more elements, and so we’ve retrospectively created a policy line for the 2023 legislation. Generally these pieces of legislation have been coherent on party lines, so this should work cleanly for MPs present for both sets of votes. 

Lord Mandelson

Despite it being a hugely significant Parliamentary action, we have not created a new policy for the humble address (a parliamentary process to compel the release of government documents) on the release of Lord Mandelson documents and communications. 

This was passed by agreement (without a vote), and because of the difficulty of interpreting these, we include them conservatively where they fit into an existing policy.

We do have a currently unpublished policy that includes votes around openness and transparency which we have logged this in: this is waiting on sharpening up the description and/or splitting into smaller policy groups. 

Separating out older policies

We now display policies in voting summaries in five year groups – pushing policies that were last voted on longer ago further down each section. 

Here we are trying to balance a few different considerations:

• Long standing MPs have voted on a lot of things – some of these are now probably not what people are looking for when looking at the voting records – we want to have some recency bias. 
• Some of those long ago votes are still hugely significant – laws on the books continue to affect people’s lives even if the vote was a long time ago. 

Splitting policies into five year buckets based on when the last vote was will help keep votes that keep coming up in different forms present. At the same time, more historical policies are separated out, while still being in the list.

Uncontentious legislation

One of the signifiers of ‘noteworthiness’ we’re using is second/third stage votes on primary legislation (something that will become a law). 

This in general fits in with a shift towards describing these votes under policies that cover a specific bill (with a clear but simple description of what that bill covers) rather than covering the same vote under several different policies (which create a false impression that voting decisions are being made independently in different policies, when in practice it’s the same vote). 

Where we have been creating policy lines is where all major stage votes have been passed by agreement. In this update, for this reason we are not covering the ‘Industry and Exports (Financial Assistance) Bill’,  ‘Medical Training (Prioritisation) Bill’ and ‘Cyber Security and Resilience (Network and Information Systems) Bill’.

This reflects that these are uncontentious legislation that are typically smaller in scope  — although it also raises the complaint that we’re not reflecting progress on areas of cross-party agreement. 

This is coherent with our general approach of not having ‘all agreement’ policies, but a switch to describing bills in policies means the voting records can double (if currently a bit uncleanly) as a record of legislation passed during a Parliament. 

One approach in future might be to better distinguish in the voting record display between ‘thematic’ and ‘legislative’ policies, which could in principle then include separately ‘uncontentious policies’ which are entirely agreements. For the moment we have marked these three as draft policies in case we want to revisit this in future. 

Motions to annul (and similar)

This is adjacent to votes, but as part of our signatures update to TheyWorkForYou we started to list Early Day Motions (formally a draft motion system that in practice works as a petition system) and open letters MPs have signed in their profiles. As part of this, we separated out ‘motions to annul’ from the wider set of EDMs, as these are motions that, if promoted to a debate and vote, would object to and annul government regulations passed through secondary legislation. 

We can pick these out because Parliament’s API marks them as different; however, they are not the only kind of substantive motion that can be proposed through the EDM process. 

The current EDM on the draft EHRC code of practice is substantively similar to a motion to annul, but is following a form set in the legislation rather than the general motion to annul process. As such, we have manually added this motion to this group and it is displayed in the more substantive section on the signature pages of MPs who support it.  We have created a tag to capture this sort of motion, and amended copy on MPs profiles to reflect a slightly broader scope on what we’re including under motions to annul. 

—

If you appreciate our work to make TheyWorkForYou a clear, objective and up to date resource for checking MPs’ voting records, please consider making a donation.

Header image: Photo by Vimal S on Unsplash

Every two weeks there’s a new release of the House of Commons’ Register of Members Financial Interests. 

As part of the work that went into our Beyond Transparency report, we got together a team of volunteers to go right through the register and understand more about what was in it. From that we learned about patterns, and things we wanted to pay more attention to going forward. 

Every time the register comes out, we get it uploaded onto TheyWorkForYou; then we go through and make a BlueSky thread picking out interesting bits (where ‘interesting’ might just be noteworthy, or a new pattern, as well as “this is bad”). 

This is sharpening our thinking, helping us draw out trends over time, and think about how we can best use platforms and data to argue for changes to the rules, and higher standards in politics. 

I’m going to keep this blog post updated as an index of these threads, with a quick summary of the key things we’ve found. 

If you’d like to get them as they happen, follow @theyworkforyou.com on Bluesky. 

2026–06-30: thread

In this register, Nigel Farage’s gold, Rishi Sunak’s speeches, 23 freebies including Ascot, the Chelsea Flower Show, and an unnamed music concert.

2026-06-16: thread

This register featured payments from the social media platform X; the opening of the Labour leadership contest; and ministerial freebies — including a David Lammy football ticket declaration that shows how Parliament’s rules lag behind the civil service’s.

2026-06-03: thread

The gifts register is viewed through a ‘third parties handing MPs nice tickets’ lens, plus the recurring charity-offset convention.

2026-05-19: thread

The dominant theme is MPs accepting money from a source and donating it to charity, against a backdrop of widespread sporting freebies.

2026-04-29: thread

The last register of the parliamentary session, with continuing X payments, calls to ban them, and the developing charity-offset convention.

2026-04-14: thread

Gifts received in a private capacity versus those with a professional justification — and a trend toward MPs giving fuller justifications.

2026-03-24: thread

Several notable entries, focused on Farage’s earnings from the Cameo website; and whether freebies from third parties are worse than those from venues.

2026-03-10: thread

Who went to the BRIT Awards, plus an FOI request on civil service gift rules.

2026-02-24: thread

Foreign trips and freebies, including APPG travel and how the cumulative £300 threshold works.

2026-02-03: thread

A few notes on the new register, centring on a gambling company freebie and the problem of structuring the PM’s security as a gift.

2026-01-06: thread

New year, new register: payments from X; gifts from ‘not free’ countries; and big foreign-funded trips.

2025-12-16: thread

A short note on rare gifts from non-UK sources.

2025-12-02: thread

Small gifts tipping over the £300 disclosure threshold, and hospitality from ‘not free’ countries.

2025-11-18: thread

Sunak’s Hoover Institute payment, party donations, and freebie taxonomy.

2025-11-04: thread

Continuing Labour deputy-leader donations, an APPG foreign trip, and Ascot freebies.

2025-10-21: thread

Deputy-leader donations, a National Lottery operator’s Rugby World Cup freebie, and how APPGs open doors to gifts.

2025-09-17: thread

MPs’ outside legal work and the difficulty of naming clients, plus more freebies.

2025-08-05: thread

A register heavy on freebies, including Google/YouTube taking six MPs to Glastonbury; free tickets for the PM’s wife and a gambling tax lobbying push.

—

Image: Guzz Alkala

Through Parliamentary written questions, MPs and Peers can ask for information and documents from government departments. But as part of that, they’re supposed to declare if they have any relevant financial interests that relate to the question they’re asking. This provides (incomplete) transparency on where the access they’re given in the public interest is also mixed with potential private benefits. 

We’ve recently raised a successful complaint about a set of questions tabled by Nick Timothy MP, who has apologised for the missing declaration.  This blog post covers what we learned from making that complaint, and the potential for expanding the scope of our work on highlighting missing interests. 

Exploring written questions

As part of research investigating parliamentary written questions, we discovered that a set of similar questions had been asked without an interest being declared, but where we thought there were relevant interests that should have been declared.

We want to explore how we can use our data to improve compliance with Parliament’s existing rules. One of our concerns is that Parliament periodically adopts stronger transparency rules, but in practice there is a lack of proactive enforcement. As we are in favour of stronger rules, understanding how to encourage better enforcement of existing rules is important for our goal of raising standards. 

There is an existing avenue to address this. The Parliamentary Commissioner for Standards takes complaints from members of the public, and will then decide whether to open an investigation. 

This is not something we’d done before at mySociety, and we reviewed previous complaints to have some idea of the expected form and outcomes. We’ve learned more about the threshold for a complaint to be accepted, and this is already informing how we approach our future work. 

Our complaints

We raised two complaints about MPs from different parties. One was rejected as we had not “adequately explained” the breach of the rules. The other, about Nick Timothy, was accepted. 

In our report about we raised issues with three sets of his parliamentary questions:

• A declaration about an oral question;
• Two questions we thought should have been associated to a donation he received from the racehorse auctioneers Tattersalls;
• Four questions we thought should be associated with a gift of honorary membership he received from  the Jockey Club. 

The Commissioner only accepted the last of these three:

“I will only be investigating Mr Timothy’s alleged failure to declare an interest when tabling his four written Questions of December 2025. I will not be investigating your allegation about his oral Question of 6 February 2025. This is because, by virtue of paragraph 7(a) of Chapter 2 of the Guide to the Rules, oral Questions do not need to be accompanied by a declaration. I also will not be investigating your allegation that a declaration was needed in respect of Mr Timothy’s written Questions of 10 June 2025. This is because I am not satisfied on the evidence available that the test of relevance was met.”

The first issue we raised was around Parliament’s rule about clear declarations of interests: Timothy had referenced his register of interest in an oral question, but not specified which interest he was referring to (which is the requirement of the current rules for debates). In this case, knowing which interest he meant would have been useful in judging if he was aware of the relevance of the same interest at other times. 

The rule the Commissioner referenced in response was that an interest does not need to be declared “[i]f to do so would unduly impede the business of the House; for example, during oral Questions, when asking supplementary Questions, or when responding to a Ministerial statement.” This reflects the general problem that declarations on questions are held to a lower standard in the rules than speeches (and oral questions to a lower standard than written questions)  —  but fair enough, that’s the rule.

The second set is interesting and tells us how directly connected an interest and a question needed to be. In this we connected a pair of questions about horseracing and then pending gambling reform to a donation from Tattersalls referenced in the register.

Tattersalls is not in itself a gambling company. Our view is that there is a cluster of related industries that make up the horseracing industry, all of whom depend directly or indirectly on the connection between horseracing and gambling. Indeed, in our complaint, we quoted the chairman of Tattersalls from their annual report, explicitly highlighting how gambling regulation affects them as part of the wider industry. 

However, this was judged not to pass the test of relevance, which from the Code of Conduct is “whether those interests might reasonably be thought by others to influence his or her actions or words as a Member”.

From our perspective, looking at the questions and interests in aggregate, we see a set of donations from the horseracing industry, and in parallel a set of questions that relate to the big push by that industry around gambling reform at the time. This is not to say one happens because of the other, but that they’re a relevant fact to highlight. However, to engage the investigation, we need to draw a connection between a specific interest and a specific question. In this case, the commissioner judged there were too many hops.

For the third set, we had a more direct argument. This highlighted a set of questions that relate to racehorse training yards or racecourses,and highlighted that the Jockey Club (of which Timothy has an honorary membership) describes itself in its SIC code and accounts as operating sports facilities and training facilities for racehorses. 

This was judged to be relevant, and at the conclusion of the investigation, Timothy has apologised and said there should have been an interest declared:

On a point of order, Madam Deputy Speaker. I am an honorary member of the Jockey Club Rooms, which provides accommodation and function rooms in Newmarket. It is an offshoot of the Jockey Club, and I accepted the membership to support a valued local institution. I declared this in the Register of Members’ Financial Interests, but did not repeat the declaration when I tabled four written parliamentary questions regarding the taxation of training yards and racecourses. The Parliamentary Commissioner for Standards accepted that I tabled the questions because of their relevance to the local economy, and understands that my error was inadvertent. None the less, I accept his advice that I should have declared this interest a second time when I tabled the questions, and I apologise to the House.

We’re happy with this outcome, because we do not think the rules for declarations for written questions are unreasonable, or an especially high bar to meet. If MPs want to accept gifts to support local industries, then yes, they do then need to pay more attention when asking questions that are relevant to those local industries. We hope this apology raises awareness of the rules and encourages others to get the basics right. 

More details of the communications between the Commissioner and Nick Timothy can be read in the evidence pack the Commissioner published. 

What have we learned from this?

Our goal here was not specifically about Nick Timothy, but that this undeclared interest is part of a pattern we had found around undeclared interests in written questions, and we wanted to understand more about making complaints to engage the parliamentary standards process. 

Our key takeaway is that we get a different perspective on the threshold for an investigation from our failed approaches than we did from our initial work reviewing existing cases. 

From reading the Commissioner’s previous decisions it seemed like there was a low bar for the reliance test. However, these are the decisions that were adopted, and may have passed some other criteria to reach that point. Often, the MP themselves have self-referred after receiving an accusation — which may signify from the start that there is something to investigate. 

The key point we took from this is that a member of the public making a complaint needs to draw a clear and direct connection between a specific interest and a specific question. While more indirect arguments can be made (and we’d argue, can be valid), it is at the discretion of the commissioner whether these are adopted. 

Expanding the net

We’ve extended our approach to review 82k written questions asked since June 2025. Based on this we have candidates for future investigation and complaints.

But the initial data matching is the easy part. To go through the complaints system, we need to put research into developing the sharpest examples. Our experience with these first attempts gives us a sense of both the level of detail required, but also where complaints (which might reasonably still highlight violations of the rules) are unlikely to succeed. This helps us prioritise and shape our approach, as well as exploring alternate approaches where we can make better use of our information. 

AI and complaints processes

This also interacts with our thinking about AI and demand on public services, and especially complaints mechanisms. 

We did not use AI to write complaints (which has clear risks in hallucinating specific rules or connections) but did use AI technologies in the data matching — which has dramatically lowered the costs and time of a complicated fuzzy match between the questions and interests datasets. 

Our view is that this makes it easier for us to make legitimate complaints. An exercise of systematically detecting (and complaining about) questions without relevant declarations is much more viable than it would have been a few years ago. 

But resourcing at the other end has not changed, and engaging the standards system requires going through the Commissioner. Alongside that, the Commissioner (alongside other complaints mechanisms) might be seeing an increase in complaint volume from more direct use of AI complaints.

An increase in volume of reports from us (or AI-driven reports in general) might result in the kind of rationing of time where only the most direct examples proceed — which is not what we want: many of the examples that concern us most require a lens that operates with a slightly broader view than that.  

As such, we need to think about when complaints are the right mechanism, versus where we want to be building broader cases about the intersection of public roles and private interests. 

How you can help

Our goal is to raise standards in our political system. What we need most at the moment is time to understand the data we now hold and to develop new pipelines for having an impact with it. 

Your donations help give us that time to dig, research, and build new approaches – that make a real difference to how politics works.

Read more about the work we’re doing. 

—

Header image: Photo by Lorenzo Spoleti on Unsplash

Back in May, I gave evidence as part of a panel to the UK Parliament’s procedure committee investigation into parliamentary written questions (WQ), with a focus on the overlap with FOI. 

You can watch the video, or read our written evidence and the transcript of our oral evidence, but in this blog post I’m going to dive into how this issue interacts with issues in the wider FOI/complaints system. 

Why are WQs worse at getting good responses?

We’ve written about the differences between FOI and WQ before, and the basic thing the committee is digging around is that despite on paper they are mostly equivalent regimes, FOI is sometimes more effective at getting information. In theory, this shouldn’t happen. WQs are allowed to take up slightly more official time to answer, and should be answered sooner.

In practice though, FOI can get information released that was denied through WQs — which fits into a general complaint about insubstantial responses to WQs, and gives external validation to the idea that the system is falling short. 

Our theory on this is that it follows back from a key difference between the two regimes. FOI has an appeal system: you can ask for an internal review, and then appeal to the relevant information commissioner and the courts. Additionally FOI and WQs are dealt with by different teams, so while both make decisions based on FOI rules, one has a better sense of external feedback through appeals, and the other does not.

Our pitch as a solution to this is it should be possible to convert a WQ to an FOI after first refusal to access the appeal mechanism. This both gives an immediate point of appeal to parliamentarians, but also grounds the parliamentary affairs team in the practical experience of FOI in the wider ecosystem. 

The FOI appeal system is in trouble

However, tying the WQ process to the FOI appeal system raises the issue that all is not well in how appeals are working at the moment. 

Our research on FOI across Europe really brought home to me the importance of the appeal system as something that keeps the wider system working: the commissioner should not be fixing problems everywhere, but regular and timely decisions and interventions help bring practice at public authorities into sync. In the absence of this, there is little to constrain bad actors in delaying or obstructing a request. 

That’s the theory — but the practice can still present significant obstacles. As Democracy for Sale’s Jenna Corderoy told the committee: 

Unfortunately, though, I want to draw attention to the current FOI appeal system; it is a long and drawn-out process. If you are requesting an internal review, you are looking at 20 to 40 working days to get a response. If you are unhappy and go to the ICO, you are looking at a long time to get a case officer to conduct an assessment of the handling of your request. By the time it goes to the tribunal, you are looking at about a year into the process. In theory, it is good to have an appeal system, but it can unfortunately take a long time, and it takes the sting out of the newsworthiness of your original request.

And this is getting worse, not better. As we heard from both information commissioners at FOI Fest (an FOI conference we ran with the wider FOI Network), the appeals system is under significant strain at the moment, with a massive increase in appeals, likely due to AI. This is not unique to FOI: Martin Rosenbaum has found that there is an increase in volume across ombudsman and complaints processes.

Warren Seddon (ICO)’s evidence to the committee showed the problems this leads to for timeliness of responses:

In the last 12 months, we saw our intake increase by 16% in Q1, 38% in Q2, and then 60% in Q3 and Q4. That is in a context where, historically, over two decades of the FOI Act, you see a 20% to 25% increase every five years. We have seen a massive surge in the number of complaints in the system. Although the evidence is not there to pull that apart in detail, I think AI is probably driving that.

[..]In terms of the impact on performance and how the system is working at the funding levels that we are seeing coming in right now, there is a real concern from our perspective about where we will end up. We are projecting at the moment that, by the end of this financial year, it will take about 16 months to allocate a case, which is not good from our perspective.

A general increase in FOI is distributed over a lot of different authorities, but an increase in complaints means that high intensity casework is channelled in higher quantities to the commissioner’s office. The end result at the moment is an increase (and expected continued increase) in waiting time for cases to be assigned. This causes much wider problems through the system because authorities that want to drag their feet have a realistic understanding that the ICO is going to take a long time to overrule them.

Dealing with increased complaints

We’ve published our thoughts about dealing with AI-driven demand, and for FOI we think this should be a spur towards greater publication of response, converting what can currently be private benefit into a public benefit that helps reduce other requests, but also makes the act of answering a request far more socially valuable because the information is more widely available. 

But this is less the case for the complaints part of the process, which is more resource intensive than answering a request. In some cases, there may be efficiencies to be made, or triage to remove AI-generated complaints that are based on hallucinated ideas of the law. But in general, these complaints are not necessarily illegitimate, and can represent an increase in valid complaints made because the barrier to entry has been lowered. 

Seddon’s evidence brought up that the ICO’s FOI budget for the year (which is set by the government) was, in real terms, a decrease on last year,  which with an increased case load can only lead to more delays.

There isn’t a clever solution to this one — but there is a boring one. Our information commissioners, ombudsman and complaints systems need more money to process an increase in complaints. In the absence of this, the wider public systems they maintain will not have the anchoring of an effective complaint system. There are costs and benefits to AI, and sometimes we’re going to need to pay more as a society to maintain and improve these chokepoints in our redress systems. 

Taking an interest in wider FOI

So to bring this back to Parliament, one of the things we’d like is that Parliament pay attention to FOI. It’s not just a constitutionally good thing, but is part of the information environment of Parliament itself.

MPs make use of FOI themselves: Ben Worthy discussed how FOI and WQs represented different tools that can be used by legislators:

They spoke of using PQs as a very quick and convenient way to obtain some information—one interview described it as a ready-made press release—whereas Freedom of Information lent itself more towards long-term investigative work, digging deeper into an issue or building a national picture. 

But FOI is also part of the wider information environment, leading to both informed public and parliamentary discussion. MPs regularly highlight information released by FOI by journalists, academics and constituents as part of their contributions to debates. 

Along these lines, there were some interesting questions to the Cabinet Office witnesses around how performance data is gathered and judged. It’d be great to see more recognition from Parliament about FOI’s role in supporting public discourse, and how they can support that.

Cost limits

This evidence session also provided an opportunity to get some more information on the record about floated changes to FOI cost limits. Cat Smith (Committee Chair) brought up, unprompted, that the written questions cost limit is currently pegged to 140% of the FOI cost limit, so there is awareness that the two systems are linked. 

Jenna Corderoy pointed out that reduced cost limits would have a negative impact for users of FOI; t that AI shouldn’t be used “as an excuse to reduce cost limits” and that “any reductions on limits are most likely to have a serious impact on the ability to access information that is in the public interest.”. 

I made a point about the potential for improved AI search technologies to make it easier to discover information (not just for FOI, but for the government to better understand itself). 

This also saw the first public comment from ICO on this floated change, with the view that the current limit is broadly working fine:

Warren Seddon: The cost limit has not changed. There has been a bit of debate about it in the past, but it broadly works as it stands. It is well used; if you look at the central Government statistics, around a quarter of refusals are due to the cost limit, so it is in play a lot, and we see it a lot in the complaints that we get.

Interestingly the Cabinet Office witnesses, when given the chance, did not talk a lot about AI-induced demand as a reason to lock down: 

Chair: In your incoming FOI requests, do you see a lot of evidence that AI is playing a role in the input?

Clare Brunton: I do not think so. I asked that question of Eirian before this session. Obviously it is hard to tell because you cannot always tell what is being written by AI, but when you look at the trends of FOIs, there has not been a huge uptick as AI comes in. It has been quite a steady growth. I am sure some of that is AI, but actually it has been a fairly steady growth over the last 20 years or so anyway. We have not noticed any spikes.

Eirian Walsh Atkins: As the earlier panel said, where people are using AI is at the more complex end of FOI, in the complaint stages and in litigation to help them with those slightly trickier processes. I do not think that AI is being used significantly in initial requests to the Cabinet Office.

So this panel provided no strong evidence or opinions that cost limits would solve current problems of FOI, while focusing in on the issue of the complaints system, where a key issue is under-funding of the ICO (and equivalent ombudsman processes).

Better governance need better information

Fundamentally, Freedom of Information has been a massive success that has made the information environment (in public and in parliament) richer. It provides a benchmark to assess that the Parliamentary written questions system is not keeping pace with legal right to information that the Act gives us all. 

But at the same time, it’s not good enough. Both active pushback from government and wider shifts in technology and how public services are run risk making it a less effective and useful system. 

It’s great to see MPs asking questions about FOI, and we hope it’s followed up with more attention paid to this vital constitutional system.

 

Header image: Photo by Vadim Bogulov on Unsplash

LLMs can increase demand on public systems by removing the friction that previously limited access. One potential result of this is new forms of unconsidered rationing that recreate that friction. Instead, we should move away from zero sum systems and aim for technical and policy approaches that turn unscalable private benefits into efficient collective ones. 

Many kinds of citizen-driven interactions with the public sector are rationed through friction: fewer people engage in them than might do otherwise, because they feel the process is time-consuming or requires expertise. We can see examples of this in planning objections, correspondence with elected representatives, FOI requests and consultation responses. LLM technologies can lower the time or expertise required and also prompt people to engage in the processes in the first place: “Would you like me to draft a complaint about this?”

Systematic impacts

This reduced friction may be good for individuals, but the resulting increase in engagement can overwhelm the system itself. In response, it may slow down, collapse, or adopt new means of rationing or prioritising access. We can see indications of this across different kinds of interactions: journalist Martin Rosenbaum has identified an upward trend across public sector complaints organisations, and concerns are being raised across sectors about AI’s contribution to growth in volumes. 

So how should organisations that handle public submissions respond in an informed way? Here’s an approach to thinking about the problem. We can divide these interactions into three types:

• Private benefit – when an interaction has a benefit almost exclusively to the requester, either competitively (eg a grant or job application), or non-competitively (eg an application for a state benefit).
• Collective benefit – when an interaction has a benefit to the requester, and also to wider society (eg a public FOI request, reporting a pothole). 
• Zero sum interaction – when an interaction success for one person is a failure for another (eg planning). 

Private benefits

Some public services fall clearly in the first category: they are unavoidably a collection of private interactions. For these, there might be improved efficiencies to be found in delivery at scale but, particularly for non-competitive benefits, these are also likely to eventually run into decisions either about increasing provision (assuming a higher level of claims from those entitled going forward), or new forms of rationing.

As stands, AI inputs can both improve the efficiency of systems through sharper, more complete initial submissions, but can also make more verbose and complex submissions that cite non-existent law. To prioritise the former over the latter, systems can explore triage approaches that enforce or encourage the qualities that make input valuable: clarity, accuracy and concision. 

When running into real limits, it is important to be clear about the criteria you want to ration on, and that they are in line with the overall purpose of the system, rather than implicitly prioritising those with greater resources.  In their FOI complaints system, the ICO is using public benefit as a criteria for prioritisation. The British Academy uses partial randomisation above a scoring cutoff to ration randomly rather than requiring additional work (on both sides) to further differentiate.

Collective benefits

A bigger win is, where possible, to transform private benefits into collective benefits.  In these cases, reduced friction is self-regulating because spillover benefits from an individual’s case help reduce demand from others : the private benefit person B is looking for has already been provided by person A’s interaction. 

One of the key ways mySociety’s services help people is to harness the self-interest of individual users for collective benefits. Every public request made on WhatDoTheyKnow also adds to the pool of public knowledge accessible on the internet, reducing the need for duplicate requests (with a similar logic to reducing duplicate reports on FixMyStreet). This means we can effectively lower the bar to access while improving overall efficiency of the system. 

We come to this from a technology lens, but the same principles apply from an institutional-design approach. For instance, if MPs’ casework or complaints are  increasing, you want to shift towards more systematic rather than individual benefits from casework. This looks like support for better collective learning, and an improved ombudsman to support collective rather than individual fixes. This kind of approach works best where good statistics are collected at a system level to help identify what collective changes are needed: tracking the overall level of demand, level of demand to different parts of the system and nature of the demand, ie what are people asking for.

Zero sum systems

The biggest shift needed is in reforming zero-sum systems, where there is currently an incentive for both sides to escalate the volume. Reduced friction here just raises costs for all concerned rather than giving increased benefits to anyone. Individual use of AI to create submissions is individually enabling in these cases, but not collectively. So, in the words of the 1980s classic film War Games, “the only winning move is not to play”. The real innovation is in solutions that open up new, and more effective, ways of working out what everyone can live with, rather than recreating rationing through new means. For instance, rather than adversarial AI planning objection generators, we could aim for a collaborative planning system that through improved communication and coordination lowers costs and removes incentives to volumes of engagement. 

Red flags for zero sum interactions are when volume is implicitly being used as a proxy for strength of feeling, or popularity of a particular viewpoint, because its value as a signal is going to become increasingly degraded as AI use increases.   

Systems work better when the benefits are collective rather than atomised

Mass adoption of AI removes one set of bottlenecks, but this can create capacity challenges for public systems. Previous waves of civic technology have built on reduced costs of storing and sharing information to build systems that help share the benefits of people’s work and lower the barriers to entry.

The current wave of AI chatbots cut against this, encouraging atomised approaches, rather than collective ones. We need to explore technical and policy approaches that help systems better achieve their purpose, without giving up on the idea of lowering barriers to entry. We can do this both by exploring how the technological features of AI tools can be bent towards collective gains, and moving away from systems that incentivise these approaches. 

—

Image: Engin Akyurt

Generative AI is good at solving some kinds of problems, and bad at solving others. With the rush to apply AI approaches across the public and private sector, we want to encourage people to use the right tool for the right problem. This blog post proposes a test that makes it easy to understand whether or not the applications are genuinely beneficial for the job in hand.

Generative AI has no concept of truth. It is designed to create outputs that are internally consistent, and this might or might not coincide with true things when the training data and context are well aligned. By now, we’ve all heard examples of false-positive hallucinations, where AI has asserted that something exists or was said because doing so is internally consistent with the question — but which turns out not to be true. Depending on the application, if unchecked, this can have catastrophic effects, meaning that validation of outputs is essential.

How to assess your project for AI suitability

In our recent Shifting Landscapes report, we shared a simple matrix that helps to assess how useful it is to apply an AI approach to any given problem.

It asks how hard/expensive is it currently to produce a solution without AI, and how hard/expensive is to verify that the solution is correct, with four potential outcomes:

	Producing a solution is cheap/easy	Producing a solution is hard/expensive
Verifying the solution is cheap/easy	Weak AI benefits (which may increase at scale)	Significant AI benefits
Verifying the solution is hard/expensive	Get a human to do it	Break down the verification problem (and repeat)

Let’s look at each possible outcome in turn:

1. Weak AI benefits (which may increase at scale)
producing a solution is cheap / verifying the solution is cheap

This applies to tasks where AI tools might help people complete tasks more efficiently, but where the resulting impact or time savings are not significant. Over time/mass use, the benefits might increase.

Examples here include tasks like letter-writing and making summaries of documents or transcripts. If AI can do the initial grunt work, a human can take over and make tweaks to the output, nominally saving some time.

In our own field of civic tech, we can see this kind of tool being used to help people navigate bureaucracy: it might help format letters to representatives, or make effective appeals when FOI requests are refused.

Cheap processes at scale can also unlock new collective benefits. For instance, Muckrock uses LLMs to extract information and success/fail status from individual FOI responses. Doing this manually per request is easy for people, but requires lots of people to do the work to create a useful dataset across the entire corpus. An AI approach drops the costs further, which produces a small benefit on an individual scale, but collectively creates useful data.

As we note in our AI Framework, we have to recognise that a large number of small uses can build up into a negative effect. For instance, AI-created objections to planning applications might overwhelm a system that was built for a world in which there are higher hurdles to lodging an objection.

2. Significant AI benefits
producing a solution is expensive / verifying the solution is cheap

In this scenario, we’re thinking of situations where it is harder for a human to create a credible solution than it is to check if the outputs are valid. Conceiving a solution might be hard because it requires specialised knowledge, such as coding, or significant time and resources, like the analysis of a huge dataset; but it would be easy for a human to see whether or not the solution is working as intended.

One of the biggest practical uses of AI so far has been seen in coding, because coding problems fit so well into this category, and so provide potential benefits. The structure of computer code is often formally checkable (for at least syntax errors), and often there is a relatively short turnaround between “having code” and “checking the code is effective”. This isn’t to say that all coding fits in this box, but enough that a clearly productive set of tools exists.

There are strong potential benefits here because an expensive process can be made cheaper, while the quality of the output can be checked through relatively cheap verification methods.

This segment of applications can be impactful even where access to models is relatively expensive, as a relatively small number of LLM users can have a big impact through the products that emerge.

3. Get a human to do it
producing a solution is cheap / verifying the solution is expensive

Some LLM processes produce outputs that cannot be quickly verified by automatic or human means.

Here, using an LLM for the initial solution might be less effective than having a human do it from the start.  While tweaking an email that contains slightly poor wording is a cheap correction, adjusting a multi-page report written by an LLM (involving fact checking, correction, restructure, etc) might be more complicated than just having someone write the original work.

When humans approach a piece of work like this, the production and verification processes pretty much happen at the same time, because the skills required to produce the work are the same ones that suggest the work is valid.

“Use a human” is often most clearly the sensible approach for projects that need a high level of accuracy and confidence in the material produced. For example, we talked to OpenFun about their LawTrace site, which brings together legislative information in Taiwan. They made a point of choosing not to use AI at all in this project. Having accurate information was far more important to users than any convenience AI could introduce.

4. Break down the verification problem
producing a solution is expensive / verifying the solution is expensive

Sometimes solutions are expensive for a combination of reasons, and this can justify investment in trying to split the verification problem into smaller problems.

Through a sequence of different checks on LLM output, we can move problems towards being strong uses of AI, because it dramatically reduces the time needed to produce the solution, while the verification costs are manageable.

As an example, our APPG scraper sits in this category. We wanted to get accurate lists of parliamentary group memberships from dozens of different websites. Our original idea was that we would need to use a crowdsourcing approach, because we thought an LLM would be vulnerable to inventing lists of MPs.

But after some consideration, we invested time in a step where we could verify with code whether or the names extracted were actually listed on the relevant sites. We can see a similar example in the public consensus platform Pol.is – where category descriptions are linked back to concrete sources to facilitate easier double checking.

Similarly, you might find that aspects of your problem (if not the whole problem) are appropriate for mechanical checking. Could LLM code make a custom verification process easier? Can a series of automatic/human checks be made more efficient with a clear verification workflow? Each individual improvement moves your project closer to being a potentially strong use of AI.

Investment in the verification process might move the problem closer to having weak/strong AI benefits, where outputs can be derisked through cheap quality checks — but you’ll only know through systematically breaking it down in this way.

We hope that, by sharing this matrix, we will encourage more thoughtful deployments of AI technology in governments  and beyond. Please feel free to share it with those who will find it useful.

—

This blog post has been adapted from our report Shifting Landscapes – A practical guide to pro-democratic tech.

Image: Leo Lau & Digit (CC-BY 4.0)

Key points

• Lowering the cost limit time would reduce the scope of the Freedom of Information Act, giving government departments greater leeway to deny requests.
• This  will have a disproportionate effect on high-impact Freedom of Information requests made by journalists and researchers. 
• It represents a new restriction on public scrutiny of government, counter to promises around improved government transparency, such as the promised roll-out of FOI to contractors providing government services. 
• It is unlikely to significantly reduce the volume of work required to process requests – local governments also receive a comparable volume of requests at a lower cost limit, and there are administrative costs even if a request is rejected under a new, lower cost limit. 
• Transparency is not a nice extra to have that can be cut when the budgets are tight. Governments that think they cannot afford transparency will be surprised at the corruption and inefficiency they will need to afford in its absence.
• The actual solution to volume is improved government processes. Reducing the cost limit might increase admin burden on authorities (due to increased back and forth with requesters) whereas better proactive publication genuinely could reduce volume of requests by removing the need to request in the first place.

What’s being proposed?

A policy is being floated, around decreasing the FOI cost limit in order to address an increase in the volume of requests. 

Financial Times: UK considers FOI clampdown as requests soar:

British officials are considering a clampdown on the freedom of information system in a move that would spark backlash from transparency campaigners.

Government figures are discussing a reduction in the cost ceiling for processing a request as the number of annual submissions has spiralled, according to people familiar with the situation.

The soaring number of requests comes against a backdrop of heavily constrained Whitehall budgets, they added.

There are no further details beyond this briefing. Our assumption is that the proposal is for a reduction to the central government cost limit (see below), but with no details on the scale implied. 

As reflected in the FT story, because of central government statistics, we can see that this increase mostly relates to defence records being moved to the National Archives. It is also worth putting in the context of a separate attempt to justify restrictions based on national security. 

 

What is the cost limit?

The “appropriate limit” is the time allowed to deal with an FOI request. 

At the start of the FOI process, a cost is estimated for the likely time it will take to locate, retrieve and provide the requested information (but not time taken in doing public benefits tests or applying redactions). 

It has a value in cash, but this is pegged against a set cost per hour (£25 an hour in UK FOI, £15 in Scottish FOI).  So effectively this is a time allowed in hours:

• £600 (40 hours) – Scottish FOI
• £850 (34 hours) – Parliamentary questions
• £600 (24 hours) – Central government FOI
• £450 (18 hours) – Other public bodies FOI

A related part of the rules is that authorities can aggregate similar requests (for similar information by connected people and made within 60 working days) and apply the cost limit to them collectively. Authorities may interpret this quite broadly if the requests share an overarching theme or are handled by the same team. 

Another relevant system is parliamentary questions, where the search time is pegged to 140% the cost limit for central government. The resulting ceiling is £850 (34 hours). 

How are the cost limits changed?

The cost limits for UK FOI are set by The Freedom of Information and Data Protection (Appropriate Limit and Fees) Regulations 2004. 

A new set of regulations could be made without a vote in Parliament.  The cost limits are changed via a statutory instrument passed by the negative procedure. This means the government lays the change before Parliament, and it automatically becomes law without a vote. 

MPs can sign a petition to call for a vote to annul it, but there is no automatic threshold where a certain number of signatures requires a vote. Generally it requires support of the official opposition to get a debate. 

What would be the effect of reducing the cost limit?

The likely effect of reducing the cost limit would be to prevent a class of currently useful and productive FOI requests, without significantly reducing volume or administrative costs. 

Who would this affect the most?

As the existing cost limit already rules out very broad requests, the change in any reduction would fall mostly on the most complex requests allowed by the current rules – and as such is likely to disproportionately affect journalistic and researcher use of FOI. Exploratory requests would need to be framed more narrowly, and a lower limit combined with the aggregation rule would make it easier for authorities to chain related requests together and deny them.

Any reduction in the central government cost limit would also have a knock-on effect on parliamentary questions, as the search time is linked. 

Would it reduce administrative costs?

This change would have a mixed effect on administrative costs: marking a bigger set of FOI requests as invalid has costs of its own. 

Reducing the cost limit would give more leeway to authorities to refuse requests when the documents requested are difficult to provide, but would be targeting a narrow band between what was previously acceptable and the new limit.  A lower threshold invites more dispute about the threshold, and requires justification for it falling in a narrow range, potentially causing more back and forth with requesters. What should happen in these cases is that authorities give advice and assistance on reducing the scope of the request to help fit inside the cost limit. Failing to do this has been noted in ICO decision notices about whether the exemption was applied correctly. As such, administrative savings are likely to be disappointing, as a lower cost limit creates work of its own. 

The natural experiment of the two different cost limits also does not suggest reducing would have a large effect on volume. The lack of comprehensive FOI stats means we do not have an up-to-date figure, but in 2017, local and central governments had comparable volumes of average FOI requests – despite the difference in the cost limit. 

What is a better approach to FOI volume?

Increased FOI volume raises the importance of efficient discovery and publication of information. Rather than reducing public transparency, public authorities should invest in their own processes and data to better meet internal and external needs. 

Public authorities need to be good at managing information — not just to answer FOI requests, but in order to work effectively. The effect of improved technology should be to make it easier for authorities to understand the information they hold, both for their own purposes and for public transparency.

More value can be realised by each FOI request released through improved disclosure logs. WhatDoTheyKnow.com removes the need for future FOI requests by making previous requests easier to find, with far more users of the site viewing information that has been published in previous FOI responses rather than making new requests. Public authorities can help reduce duplicate requests by publishing disclosure logs that make information released available to search engines (including AI agents), delivering more impact to releases and reducing repeated costs. This also helps address the social cost of atomised AI approaches: information is released for public benefit. 

Building on this, authorities can also learn from the subjects about  which  FOI requests are frequently made, and use that to inform their proactive publication of information. Increased volume of requests represents people making use of their information rights: this should be encouraged, while trying to make the process of finding and publishing information as efficient as possible. 

Transparency isn’t a cost: it’s a necessary investment for the rewards of reduced risk of corruption, and improved quality of work through the deterrent effect of future transparency. Efforts to cut costs could instead focus on the cost of secrecy— the high legal fees government departments have paid to try and keep secret information in the public interest. Government and parliamentarians should be invested in making this system work well, for the public benefit, rather than restricting access. 

Read other responses

• George Greenwood on Relight My FOIA blog on the security arguments being made
• Jenna Corderoy from Democracy for Sale on the money the government spends fighting transparency cases
• Claire Miller discusses cost limits in her weekly FOI news roundup
• Sunday Times – Our secrecy-addicted state is at it again. It must be stopped
• The Times – Curbing our right to know would be a blow to democracy

Header image: Photo by Jr Korpa on Unsplash

Today we’re launching a new report: Shifting Landscapes: A practical guide to pro democratic tech.

This report builds on the conferences, seminars and conversations we’ve been having in our TICTeC programme over the last few years, to present a comprehensive picture of where pro democratic tech is now. We explore how technology can strengthen and defend democratic life, and how civic tech practitioners, pro-democracy organisations, and funders can make effective choices in a rapidly shifting landscape for both democracy and technology.

The result is a report of eight chapters in four thematic areas:

Pro-democracy tech: this extends our definition of pro-democracy tech, to explore how  technology can be joined to wider democratic movements working to both defend and extend democracy using technology.

Communities of practice: what we’ve learned about how we can best work together with our communities of practice around Access to Information and democratic transparency, balancing efficiencies of scale with unique circumstances and needs.

Shaping the landscape: Civic tech sometimes needs to adapt to changing times, but should be trying to shape the times. These chapters look at changing distribution methods (video and AI chatbots), but also how we can create infrastructure that makes democratic projects easier and more effective.

Using technology effectively: These chapters are aimed at practitioners thinking about how to use technology, with examples and frameworks for practical approaches to AI technologies, but also other examples of tools that can be effective in ways that AI approaches can’t. 

The report can be read online, or as a PDF. 

Header image: photo by Kalen Emsley on Unsplash

As part of our WhoFundsThem work we want to make better information available about money in politics. 

Last year we released a report Beyond Transparency – looking at the UK Parliament’s register of financial interests, and wider arguments about how we fund politics. 

Today we’re releasing a follow-up report: Leaky Pipes (read online or download as a PDF). This covers what we’ve learned (and what we think could be better) about the systems for reporting election donations. You can also re-watch the launch event on YouTube. 

This report started because we were a bit confused about the different ways data could be declared and reported.  And to be honest, we’re still a bit confused – but we have more diagrams to explain why. 

What we explore in this report are the multiple routes for declarations, different thresholds for disclosure, and uneven public access. This makes cross-checking difficult and leaves gaps where information can vanish depending on how a donation flows (direct to candidate vs via party), how large it is, and whether the candidate wins.

The result is that candidates and agents face complex reporting requirements, electoral administrators hold paper-heavy returns that are hard to inspect, and the public (and sometimes regulators) struggle to build a consistent picture of who is funding whom.

From this, we’ve made recommendations on making reporting easier to do correctly, faster to publish, and simpler to scrutinise:

• Move to a “report once” process that informs multiple systems
• Harmonise public disclosure at £1,000
• Create a comprehensive public database above that threshold
• Create a safe private database below the threshold for research and evaluation purposes

Building on this, we suggest three practical avenues for follow-up work that would strengthen the case for reform and help design better systems:

• User research and prototyping to map how a “report once” service would work for candidates, agents, administrators, Parliament, and the Electoral Commission. 
• Sampling local authority returns to demonstrate the scale and type of inconsistencies between routes.
• Exploring a data-sharing agreement for controlled research access to the Electoral Commission’s small-donor/return data.

The report can be read online or downloaded as a PDF.

Header image: Photo by Meg on Unsplash