Executive summary
We need better data, stronger checks, and tighter rules to rebuild public trust in our political system.
Over the last few months, fifty volunteers helped the TheyWorkForYou team go through the Register of Members' Financial Interests (RMFI), line by line, for all 650 MPs. We were looking for specific bits of information, but also to more generally understand the state of the Register and how rules on transparency are working in practice.
We have many ideas on how to improve that transparency but the goal is not good documentation of office holders' conflicts of interest; we want as few conflicts of interest to exist as possible. To better align politicians' behaviour with public expectations, there is no substitute for a much less permissive set of rules around MPs' financial interests.
As such, we are making four categories of recommendations, stepping from incremental change to improve data collection, to systemic reform of the funding landscape.
- Better data collection to achieve more accurate interests information
- Stronger checks to make sure the interests information is reliable
- Tighter rules so there are fewer unacceptable interests in the first place
- Systematic reform to decrease the role of money in the political system.
On the last point, our key recommendation is that a citizens assembly on funding politics (convened either by Parliament or civil society) would be a useful tool in breaking the deadlock on public funding, either by directly informing reform, or creating a stronger and more coherent approach to arguing for reform.
As part of this project we have added two new features to TheyWorkForYou. Enhanced election summaries for each current MP, adding organisation descriptions and summaries of different groups of donations; and a highlighted interests page, pulling out support from governments of not-free countries, oil and gas companies, and the gambling industry.
In the next few months, we will continue this work to release better data about devolved Parliament registers, and APPG groups. You can sign-up for updates about our work at: https://www.mysociety.org/democracy/who-funds-them/
All recommendations
These recommendations aim to improve the quality of data released. Adjusting how data is collected and published would better capture common interest types.
- A review of the register and its categories to further differentiate between different kinds of interest.
- Rewrite fields to make ongoing employment more distinct from employment which has ended, and ad-hoc employment
- Add new fields to capture relevant details about elected positions such as Mayor or Councillor
- Clarify guidance on standards of disclosure expected of new MPs regarding previous work and interests
- Make ‘date received' and ‘purpose' fields required for all donations.
- Improve fields in Category 4 to identify other individuals that join MPs on trips abroad, and the costs incurred.
- Supported staff need to be separated out in the RMFI with custom fields to capture information about the role they are performing.
- Those performing freelance roles with relevant public sector clients, or clients that lobby the public sector, should declare who those clients are
- Require public release of contracts (or similar) for second jobs.
- Require additional statements about if MPs' other employers' activities ‘intersect' with Parliament (commercial or lobbying relationships).
These recommendations are for Parliament as an institution to stop seeing poor quality disclosures as the MP's problem, but instead treat it as something that affects the standing of the institution as a whole.
- Random quarterly audits of a small batch of MPs' RMFI submissions each quarter, checking for missing donations and overall compliance.
- MPs to be prompted quarterly to update their registered interests, with options to either update the Register or respond explicitly to say ‘no updates'. Non-response should be published by the Commissioner.
- The Chair should regularly remind MPs of the rules around declaring interests fully when speaking in the chamber.
- Parliament should publish the nature of the interest declared when they publish written questions, not just whether or not an interest was declared.
These recommendations aim to achieve more disclosure through lowering thresholds, and to make more interests impermissible.
- Lower the gift registration threshold in line with wider public sector/civil service thresholds (e.g. £20).
- Consult and adopt new rules and guidance on when gifts shouldn't be accepted by MPs.
- Parliament should enforce rules to require income and expenditure statements from all APPGs, and proactively publish those statements on the Parliament website.
- In the absence of wider finance reform, better passthrough declaration rules are needed for organisations who donate more than £50,000 across all MPs.
- Lower the disclosure threshold for shareholdings to £1,000 and adopt the US rule on stocks.
- Lower reporting threshold for donations to £1,000 from £1,500.
- Quarterly declaration of aggregate donations below the donations threshold.
- Collect spousal and close family interests alongside MPs' own interests in the relevant categories, as happens in local authorities. These should cover at a minimum: income and employment, directorships, and shareholdings.
- The Register of Interests of MPs' Staff needs a fuller overhaul, better capturing information about the source of a staff member's funding.
These recommendations go beyond reform to Parliament's own rules, to how other regulators and civic society can act to improve control of money in politics.
- The Electoral Commission's Political Finance Database should be lower, to match the RMFI reporting threshold of £1,000.
- Any individual donor donating £1,000 or above should register with the Electoral Commission for a unique identifier.
- A citizens' assembly should be convened on money in politics, to unblock wider arguments about reform and inform civic action.
About mySociety and TheyWorkForYou
mySociety is a charity running innovative digital services that break down barriers to democratic participation, giving people the tools to collaborate and act, building a fairer society. Our main work is in the UK, but we have a wider role creating and championing pro-democracy tech around the world.
From parliamentary monitoring to Freedom of Information, our services make our democracy and governments more understandable and transparent — both to the public and to people working inside those institutions.
In the UK, we have run TheyWorkForYou for twenty years, and in doing so have been a major force in increasing the transparency of how the UK's Parliaments operate — as well as inspiring more and better transparency from the Parliaments themselves.
Introduction
As part of our WhoFundsThem project, we have been exploring how we can combine TheyWorkForYou's sizable public platform and mySociety's technical skills with volunteer time and effort, to create new information and pressure for reform on how politics is funded.
This report covers what we have learned through our crowdsourcing project looking through the Register of Members' Financial Interests (RMFI). The RMFI contains a list of disclosures MPs are required to make of financial interests or benefits which "others might reasonably consider to influence his or her actions or words as a Member of Parliament".
Both the initial creation of the register in the 1970s, and the current wave of improvements occurred after scandals that brought specific MPs and Parliament more widely into disrepute. Transparency was a big concession, but was also consciously included as a means to avoid more restrictive rules, and especially enforcement through new laws and outside regulators. As such, disclosure has had long running issues with incompleteness, lax rules, and data quality.
We have used a volunteer-supported approach to help connect the Register to other publicly available information, and work around poor quality data issues. We conducted an initial literature review and then brought together a group of volunteers to answer a set of questions about each MP based on their declared interests in the RMFI from September 2023 – September 2024.
Originally, we were planning to start the crowdsourcing project in early spring 2024, and update TheyWorkForYou to inform people before a late 2024 election. When an early election was called, we shifted the focus of our work and deferred this project until it was over.
After the election, there was a new government and more than half of the MPs elected were brand new. We've used our volunteers to help us build a picture of what's in the Register, helping us draw systematic conclusions about where our work is best directed, and where wider reform is needed to a) improve the standard and understandability of what is being disclosed and b) change the rules so that what happens is in line with public expectations.
Accompanying this report are two new features on TheyWorkForYou:
- Enhancing official data: An election summary for each MP, creating enriched summaries of donations/support received.
- Highlighting a subset of data: A highlighted interest pages showing support and gifts given to MPs from 'not free' countries and from a set of industries with low public support.
Additionally, we cross checked with Companies House data to find details of where MPs hold positions in external companies. We haven't published this data, but did suggest additional disclosures to affected MPs, and will report on updates.
This report has a number of recommendations for Parliament and policy makers in fixing problems of funding and transparency. But core to the TheyWorkForYou spirit is the idea that we should look for differences we can make from the outside. In the final section of this report, we'll discuss what we (and wider civil society) can do to build on this research and data to make a substantial change by the next general election.
New TheyWorkForYou features
A key way we think we can make a difference is to use volunteers and data analysis to create new datasets, and then use the platform and reach of TheyWorkForYou to help this data shape wider conversations about MPs' financial interests.
As such, we have added two new features to TheyWorkForYou through this process; enhanced election summaries and a highlighted interests page.
One of our goals for this project was to use volunteers to create summaries that make it easier to understand individual MPs' interests, rather than creating summaries of all interests.
When we started the volunteer crowdsourcing project, we were working with an older version of the output from Parliament, and a big part of the work was extracting relevant information like ‘donor name' from free text, before going on to research who that donor was and what they did.
The good news is that while we were in the early stages of the project, Parliament started releasing more complete structured data. As such we were adding less value through the data extraction, and needed to think about how we could recombine the crowdsource data and the better quality official data. Additionally, we knew that by launch the data would be getting towards six months out of date, and needed an approach that wouldn't immediately outdate.
Our approach has been to focus on Category 2 interests (donations) and create an ‘election summary' of donations in the run-up to the 2024 election. Here we have used the new official data to create automated summaries of the split between company/individual donations, and dig more into in-kind vs cash donations. We've also enhanced this with a description of each company, and grouped companies into categories (based on similar companies that have donated rather than the broad Standard Industrial Classification (SIC) codes given by Companies House). For individual donors, we've split summaries by whether one person has donated to multiple MPs, using the crowdsourced data to help reconcile different forms of the same name.
As described in the What's Next section, this is an approach we'd like to build on – and the availability of better data makes it easier to focus just on the bits where we can add more value from the outside.
A key idea in this project was trying to apply to the data a sense of what the public finds more or less acceptable, to provide focus and shift behaviour. Our analysis did this in two ways. We wanted to identify links to unpopular industries (either directly through companies or through the individual donors), and we wanted to identify when MPs were receiving gifts/work from the governments of 'not free' countries (based on the Freedom House listings). Part of what we were exploring in both cases was the extent to which it was possible to use the register to identify systematic influence of money - and the limitations we found inform our recommendations.
We informed our choices of which industries are unpopular from polling data, and further condensed this on reviewing the data (see ‘About the crowdsourcing process section for details). Volunteers first needed to identify companies and individuals who had made donations, and then flag if they were related to one of our target industries or countries.
Our process on this was to ask volunteers to flag registered interests that fell into a set of five industries we'd identified as having low public support. Internationally, we flagged trips to and paid from 'not-free' countries.
Based on the results, we've reduced this to a smaller set:
- Oil and gas
- Gambling
- Trips funded by governments of 'not free' countries
We then asked the flagged MPs whether they had any further comments as part of a right of reply.
In the end, we flagged interests from 40 MPs (6%).
| Category | Interest count(3 MPs have donations associatedwith both gambling, and oil and gas) |
|---|---|
Gambling | 19 |
Oil and gas | 10 |
Not Free Country | 19 |
We feel this relatively small group reflects that these are not donations/gifts that most MPs would seek or accept. By applying a lens to this group, we want to encourage what is now the norm in Parliament of not accepting donations or gifts from these groups.
Improving data quality
A key obstacle to producing better summaries of the Register of Interests is variable data quality (and suspected missing data) in many areas.
Prompted by issues our volunteers identified, and what we've learned reviewing the data, we are making a number of recommendations for improving the data quality of the disclosures in the Register.
We're highlighting specific problems, but in general we recommend there should be a review of the register and how it is divided into different categories. The goal of this should be to further differentiate between different kinds of interest, both in terms of better data collection on practical features but also being responsive to different democratic problems posed by different kinds of interest.
This would have the benefit of making the register easier to understand for the general public, and make available clearer information on different kinds of interest for systematic analysis and validation.
In this section we have three general recommendations:
- Improved data audits and validation are needed to fix poor data that makes it hard to answer basic questions about external employment.
- Information about contracts that must be made available to the Parliamentary Commissioner on request should be gathered and made publically available.
- Transparency (and restrictions) on employment should pay attention to features of the employing organisation as much as the defined role.
A minimum goal is that second jobs and other employment should be very clearly declared.
However, the data collection and overall compliance in this category are some of the worst across the Register. Our volunteers had to do huge amounts of research to understand what was being declared, because the fields have been used in very different ways by different MPs, and few are compulsory. This means there are regular data gaps (such as non-disclosure of income) that we're just not able to fill in. There is a common problem where income seems to have been declared when first prompted to complete the Register, but not updated since, leaving huge amounts of income information out of date.
As such, we did not feel we could produce useful comparable data, and so have focused our efforts on producing recommendations to improve the clarity of what is disclosed. We wanted to be able to build a comprehensive dataset to answer the simple questions:
- What else are MPs doing to earn money, in addition to their MP's salary?
- How much time are they spending doing it?
- How much are they earning?
- Is it regular and ongoing, regular but now ended, or ad-hoc employment?
Even with hundreds of hours of volunteer time, the extent to which the data in this category is fragmented and inaccurate made a complete dataset of this nature unachievable.
The ad-hoc question particularly is an area that could be usefully separated from ongoing/regular employment. Media appearances are a fairly common source of external income for MPs. These are a different kind of disclosure (relatively small, ad hoc, and plausibly an extension of the MP's role) from sustained employment in a second job, but are currently declared and displayed in a similar way. Separating these out would enable better data collection on both.
Improved data collection needs to happen at source (we cannot add information we do not know), but beyond that, there are options for different actors to assert pressure to improve the overall quality of the data. There need to be new processes/audits to ensure that the data that is produced is suitable to answer these questions, flagging inconsistencies in data provided, and promoting new disclosures from MPs who are likely to be out of date (e.g. MPs who have previously made monthly payment disclosures but stopped without declaring an end date).
At the end of the report, we discuss how we might implement this from the outside, but ideally Parliament should take an interest. It's not enough to say it's MPs' responsibility to disclose accurate information: the reputational costs of poor quality information is a collective problem that needs an institutional response.
After a huge influx of new MPs to Parliament at the last election, it is thought that 10% of MPs are also councillors, but whilst the number is changing regularly, with many MPs resigning councillor positions in the autumn or waiting until the next local election, the RMFI data is not keeping pace. The fields in this category are not well understood or used in a consistent way, making it hard to understand whether or not MPs are continuing in their role as councillors or whether they are still receiving payment.
In our research we had to supplement our analysis of the Register with additional research from council websites and local news to see whether or not an MP remained a councillor. We found examples of MPs with final salary payments listed and no declared hours who were still councillors; and MPs who had not recorded their final salary payment — and declared 30 hours work after the election — but who were not councillors any more. Many MPs who were councillors when they made their first RMFI submission in July seemed to have resigned their council position in the autumn, but not updated the Register since.
Whether "dual mandates" should be allowed is a democratic question, but it is also clear that these are a fundamentally different type of interest (where constituents may be overlapping). Data collection on dual mandates should collect relevant information related to this kind of role (ward/constituency, term of office, etc). (Notes)
Pre-election jobs and income
The Guide to Rules suggests that new members should be declaring all current registrable interests and all past registrable interests from the last 12 months aside from earnings. This has been understood in a variety of different ways. Some MPs have disclosed their previous jobs and only declared income when it has come in the form of a final salary payment after the election. Others have declared their previous jobs but not the salary. Most new MPs have declared nothing at all.
Sometimes a declaration didn't line up with our volunteers' research. A 27 August 2024 press release says a new CEO took over from Andrew George at Cornwall Community Land Trust on 1 August 2024, shortly after his election. However, Mr George registered this same position on 4 August 2024. The entry has no end date and no salary information, and has not been updated since it was registered in August 2024.
Other MPs are making full declarations about previous income. Blair McDougallis an example of best practice, registering in detail his ad hoc work from February 2024, before the election was announced.
There are broadly two democratic problems with MPs having outside employment:
- The external role interferes with being an MP, who then provides a worse service to constituents because they are doing work elsewhere.
- The external role intersects with being an MP — outside interests are trying to access(and MPs may be willing to sell) the special access or rights they have as legislators, that have been given to them by their constituents.
We discuss in our earlier literature review the general shift (in practice and public perception) towards an MP being a full time role that makes the first kind of second jobs less accepted.
The 2024 shift in the rules on second jobs were aimed at this second group, restricting MPs from giving advice on public policy or on how Parliament works as part of paid employment. However, the cynic would say that this was not actually what the exchange was for, just one of the more plausible reasons to have MPs with knowledge on the payroll, and that other semi-plausible ways to extend credit to MPs remain.
We have two recommendations in this area: make contracts transparent, and new disclosure requirements to evaluate the risk that an external organisation's work intersects with parliamentary work.
By the Code of Conduct, MPs are supposed to have a written contract (or similar) listing their duties which the Parliamentary Commissioner for Standards can ask for:
3. A Member who takes on any formal paid employment with an outside body must obtain a written contract or written statement of particulars detailing their duties. This contract must be made available to the Parliamentary Commissioner for Standards on request (but Members do not need to lodge a copy of the contract with the Registrar when registering employment).
4. Any such contract, or letter of undertaking from the employer, must specify that the Member's duties will not include lobbying Ministers, Members of Parliament or public officials on behalf of that employer, nor providing paid parliamentary advice, and that the employer may not ask them to do so.
There is no reason why these should not be required to be made publically available. This would both ensure that MPs are compliant with the rule above on the content of the contract, and provide greater transparency to constituents on exactly what the role entails.
The next problem is that role descriptions, and even contracts, may be technically compliant while in practice posing a corruption risk. An MP might not have a contract that says "I will ask written questions on behalf of my employer" but it might be mutually understood that this is part of the deal being made.
As such, we cannot rely on looking at the role description alone. The best way to manage risk is to look at the organisation and understand if there is a risk this is an intersecting second job - which requires additional scrutiny.
For instance, the rules could require MPs to answer with second jobs:
- Is the organisation a supplier of government services?
- Does the organisation in general lobby Parliament?
These questions are not exhaustive of all corruption risks, but regardless of the specific role the MP is being paid for, they would help narrow in on potential risks and shaping tighter rules.
In practice, this division should also be reflected in declarations of interests made in debates or when asking questions. While there may be reasons to be opposed to ‘interfering' second jobs, they should more rarely come as part of normal parliamentary activity. Later sections discuss how we can improve these disclosures. (Notes)
Recommendations:
- Review the categories of the register and its categories to further differentiate between different kinds of interest, both in terms of their nature, and of any potential democratic problems.
- Rewrite fields in the submission form so that ongoing employment is more distinct from employment which is finished or ad-hoc.
- Add new fields to capture relevant details about elected positions such as Mayor or Councillor.
- Clarify guidance on standards of disclosure expected of new MPs regarding previous work and interests.
- Require public release of contracts (or similar) for second jobs
- Require additional statements about if MPs' other employers' activities ‘intersect' with Parliament (commercial or lobbying relationships).
MPs are supposed to declare all donations (whether they are cash or in-kind) worth over £1,500. This applies whether the donations are intended for fighting an election, or for any other reason. The rules apply whether the donation came via their national party or local party branch (category 2a) or through any other route (category 2b). All donations over £1,500 should be reported within 28 days.
In total, more than £150,000 in donations were reported in September 2024 by MPs who reported £0 donations in the August 2024 Register (the first Register after the general election, when election donations were due to be declared). This figure excludes Conservative leadership candidates1, who bring the total to more than £275,000. This suggests a problem in timely donations reporting, but also raises questions as to whether all donations are being reported at all.
Notes ():
1: For the purpose of producing numbers around the general election rather than an internal party contest.
Parliament does not verify or validate the returns from MPs in the process of producing the RMFI. If an MP does not submit a return, Parliament does not publish anything. If they do submit a return, it is automatically published verbatim (spelling mistakes and all). Some high profile party leaders have no donations declared in the run-up to the general election, whilst other MPs fighting safe seats have declared dozens of donations. Parliament's staff team is stretched and cannot verify each submission to the Register, but the current situation where no checking takes place does nothing to support public trust.
The Electoral Commission is the independent body which oversees elections and regulates political finance in the UK. They publish a Political Finance Database of MP donations, but this is sourced directly from the RMFI. This data is not checked, just reproduced but at a higher threshold: it only lists donations above £2,230, whereas Parliament's threshold is £1,500.
This £1,500 threshold is itself a source of debate. We found a case where an MP justifies not reporting a £1,500 donation because the rules state the threshold is over £1,500. Meanwhile, 18 MPs have declared donations of exactly £1,500.
The Electoral Commission also oversees the process of candidate spending returns, which are administered by Returning Officers in local authorities. These forms collect information about donations given to candidates, and what they spent the donations on, in the five weeks before a general election.
After the election, these forms are made available to members of the public willing to visit their local authority in person: only headline total donations figures are made available online by the Electoral Commission, and months after the election. In the candidate spending returns we checked, 94% of the donations listed were from the candidate's local party organisation, with the original source of the donations remaining unclear.
To properly analyse donations made above the £1,500 reporting threshold, it would be useful to better understand where a number of smaller donations are, in aggregate, reaching this figure. For instance, in our election summaries, we calculate the proportions of donation income made by organisations and by individuals. These figures might be massively inaccurate if there are large numbers of donations in total below the threshold.
It is difficult to reconcile the headline figures reported by the Electoral Commission and the figures in the RMFI as they cover different periods. As such, we recommend a monthly/quarter declaration of the aggregate income below the threshold. Similar to items in the ‘misc' category, this may not represent a financial interest or conflict in itself - but it helps put other information disclosed into context.
When donations are reported to the RMFI, it can be difficult to assess when they were received and what they were used for. Category 2A RMFI declarations (which make up 80% of the donations reported) don't contain received and accepted dates, only reporting dates (usually the date the Register was published), so it's not possible to tell when the donation actually arrived. As such, we can't match this type of donation up with candidate spending returns, which only cover the five weeks before a general election.
In many cases where donations are declared in the months following the election, it's impossible to know whether the donation was in fact towards the election, or some other purpose, because the ‘description' category is not mandatory and the received dates are not usually available. We know there is public interest in what money is spent on as well as who it is donated by.
Individual donors are hard to identify by name alone, and aggregate analysis is made difficult by several instances of the same person's name being spelled in many different ways. When attempting to identify donors, our volunteers found the biggest factor in how successful they were was how unique the name was: not a recipe for good transparency. (Notes)
Recommendations:
- A lower donations reporting threshold of £1,000 for both the RMFI and Electoral Commission.
- A monthly/quarterly declaration of aggregate donations below the threshold.
- Make ‘date received' and ‘purpose' fields required for all donations.
- Randomly audit a small batch of MP RMFI submissions each quarter, checking for overall compliance.
- Any individual donor donating £1,000 or above should register with the Electoral Commission for a unique identifier.
The period we were looking at (September 2023 - September 2024) included a general election, which is likely to have limited the number of foreign trips MPs were taking.
The source data here is some of the best and easiest to understand across the Register, as the fields do a good job of capturing required information: where did MPs go, why, for how long, and who paid for it.
The ‘who paid for it?' question is hard to answer when the ‘donor' is an All Party Parliamentary Group, and the groups are not fulfilling their own reporting requirements. We found that 14 APPGs appear in the February RMFI Register, and in the January APPG Register, not a single one of them has published an income or expenditure statement at its last AGM. This is despite the Guide to Rules stating that "All groups must prepare and agree an income and expenditure statement, even if it is a "nil return"." (Notes)
Example: Who paid for this visit to China, and who went on it?
Two MPs (one Conservative, one Labour) declare in their Registers that they were a part of a trip to China funded by the China APPG on 19-23 March 2024, arriving back in the UK just two days before the group's AGM on 25 March 2024. The total value between the two MPs was over £5,000. The AGM took place two days after the funded visit to China. Nonetheless, the group did not declare any benefits at that AGM.
The descriptions and information supplied about the visits varied, and they were declared months apart from each other (one in April, one in July). One of the two RMFI declarations states that the donor for the APPG China trip was The Great Britain China Centre, which is a non-departmental but Government-funded public body. The other was named the "All Party Parliamentary China Group" but selected ‘true' for "Is Private Individual".
If the Great Britain China Centre funded the trip, it may not need to be registered as a benefit, as the money is coming from inside the Government. However, the rules state that it should still be detailed in the group's income and expenditure statement. These statements are not made available except on request. We have requested a copy of this statement, but not received a response.
If this trip was funded by a UK Government body, this raises further questions about the expenses incurred. The itemised costs of the two MP visits are identical in hotel and meal costs; however, one MP listed flight costs that were more than twice that of the other, and also declined to answer the ‘Is Sole Beneficiary?' question (which is not currently required). This suggests this MP might have taken another unknown person on this trip to China.
Recommendations
- Parliament should enforce rules to require income and expenditure statements from all APPGs, and proactively publish those statements on the Parliament website.
- Improve fields in Category 4 to identify other individuals that join MPs on trips abroad, and the costs incurred.
Directorships of companies should either be declared as income-generating work, shareholdings, or in "misc" if unpaid (e.g. where the MP is trustee of a charity).
This is an area where we can cross-check against another dataset: the Companies House register, which our volunteers did. Working with Any One Thing's tool to find likely entries that were a match for MPs, we and our volunteers went through MP by MP to weed out false positives and expired entries.
This tool grouped potential matches for MPs in Companies House register for volunteers to investigate and discard false positives. It is difficult to reach a correct balance with an entirely automated process, in one instance an MP with a relatively common name shared a birth month with someone else with the same name in the register. Whether or notwe have an official birth month for MPs at all is inconsistent – and having this dataset would help reconcile MPs against both Companies House data, and other datasets of beneficial ownership.
While we found companies that should by the rule be declared, generally we found that most MPs were declaring well (or where significant entries were missing in September, this improved over the rest of 2024 when we checked later versions of the Register).
In the 37 entries we felt merited passing back to the MPs concerned for more information, common themes were:
- Charity trustee/school governorship positions
- Directorships of active companies
- LLP memberships (which should be declared in the shareholdings section, but may fall below the threshold)
- Personal companies that were dormant or not being used since election
- Plausibly dormant companies or those where the MP had only peripheral involvement
While many of these need to be declared under the current rules, there were only a few where we felt the absence was a potential problem. We wrote to these MPs, seeking to find out their understanding of the rules as part of our right of reply. We received a range of responses:
- Some stated simply that as the interests were non-financial, they did not need to be registered.
- A few responded to make us aware that they would now either chase Companies House to clarify dormant status.
- The majority of responses we received were to say that they had checked with the rules, and they understood that they should have declared the interest, and that they would update their register.
Responses in the first group reflect a debate about the changing nature of the Register, where it is broadening beyond purely financial interests in what is covered. It will always be a fluid document, but where the ‘misc' section is picking up significant interests, these should be moved to new subsections with clearer field descriptions. If charity trusteeships are worth recording, they are worth recording correctly.
Additionally (although we ended up discounting these ourselves), there is inconsistency from MPs about whether they should declare memberships of residential management companies (organisations formed to collectively manage a block of flats). As these can sometimes reveal where an MP lives/has lived, we recommend that these do not need to be declared. (Notes)
Recommendations:
- Where companies are mentioned in the Register, company IDs should always be required.
- Parliament should publish the birth year and month of MPs to assist reconciliation against datasets of directorships of registers of beneficial ownership.
- Rules should clarify that residential management companies do not need to be declared.
- Guidance should be improved on how MPs can check that all of their registered directorships are present in the RMFI.
Something we asked our volunteers to look out for is where donors are supporting staff (either through funding or secondments) in MPs' offices, and especially for members of the Shadow Cabinet (many of whom are now ministers). As this is an emerging use of donors, we wanted to understand more about what is happening, and whether we wanted to make any recommendations about the practice.
In general, we've concluded this is a more specific version of the general problem of funding politics — and that our later recommendations effectively address this practice either by making it rarer or more benign. We do have several data quality recommendations on improving capture of this kind of donation.
Our volunteers tracked when MPs were being donated staff capacity in the September 2024 Register. Reviewing the results, we found around £2 million in donations (cash and in-kind value). This broke down into the following categories (multiple payments for the same role by the same donor were grouped):
| Category | Count | Overall value in £ |
|---|---|---|
Payment to the party to support staff | 16 | 905k |
Direct payment to MP for staff | 13 | 316k |
Services provided by external staff to an MP | 11 | 500k |
Secondment | 11 | 421k |
External staff have been seconded to the party, who spend some time supporting one MP | 2 | 37k |
Paid interns | 3 | 57k |
Leaving aside paid interns (the ones that we found were supported by the Aziz Foundation e.g.in Chi Onwurah's register), this almost entirely reflects additional staff support given to the Labour Shadow Cabinet. These categories are inexact, and the way MPs describe the situation may not fit into them precisely. There is a fuzzy boundary between donations direct to MPs, and those made indirectly through the party.
The general groups are:
- Access to external staff (services), where teams in think tanks are made available to an MP.
- Donations to hire staff: generally, this is the central Party pooling donations and directing them towards MPs.
- Secondments: staff at external offices becoming based in the MP's office (receiving parliamentary passes, etc).
A problem in properly tracking this set of donations is that they are not clearly separated in the data, and identifying all relevant interests currently requires a highly manual approach.
The new detail fields in the RMFI are an excellent improvement over the previous free text entries. However, there is not a category with a field to usefully capture donations of staff or secondments. Staff donations may appear as cash, or in-kind support. It may be unclear who is responsible for staff. As such, our volunteers needed to look through the data manually to pick up a few different variations.
To be able to better surface and track these donations, we want information to be captured that is relevant to donations of staff time. As such, we recommend a new subsection under section 2 explicitly for supported staff, that includes fields such as role title, start date, end date, FTE, and a corresponding ID for the Register of Members' Staff.
There is a much less well known register, the Register of the Financial Interests of Members' Staff (RFIMS).
In general this register is not in good condition. It's hard to access (split over multiple webpages, making it hard to search), with one free text field per member of staff (and a number of MPs' names misspelled).
Ideally, this register should be telling the reverse story of the RMFI, about staff members who are unusual because they are paid for by the party or are a secondee. For some MPs it's doing just that: looking at the March 2024 Register for Darren Jones or Rachel Reeves, you can identify the staff that are paid for by the Labour party, and secondees paid for by other organisations. However, we've found at least one example where a staff member mentioned in the RMFI is not then present in the relevant RFIMS, and the reverse where a secondee mentioned in the RMFSI is not in the RFMI.2
Notes ():
2: Jo White declared a researcher funded by Palace Yard (a thinktank), but this wasn't declared in the RMFI; John Healey declared an employment of a Grace White in the RMFI but is absent from the RFSMI.
This register has a problem of not cleanly separating financial interests inherent to the role (for example, "I actually work for a private company and am just here for a few months") and interests from activities ("I received some tennis tickets"). Ideally for each staff member, there would be a field specifying if they were funded from MPs' staffing costs, supported by the party, supported by an external donor, or secondee from another organisation.
As part of the Parliamentary Digital Service (PDS)'s ambition to transition all Parliamentary registers to a new model, we recommend a substantial overhaul of this register to better capture this category of conflicts, ideally with clear crosslinking between relevant entries in both registers.
Many secondments to the Shadow Cabinet were funded by the group Labour Together. This is an unusual new arrangement where, rather than donations being made to the Labour Party to support roles tied to the party, high net worth donations support Labour Together, which in turn put staff in the offices of eight or nine members of the Shadow Cabinet, with additional service support provided to others.
Without getting into factional Labour Party arguments about the role of Labour Together, there is a transparency difference between Labour Together and Labour Party funded roles. If you give £50k to the Labour Party and they use it to fund a position, the source of the money should be declared in the RMFI as a pass through donation. If you give £500k to Labour Together, and they hire 10 people and offer them to shadow ministers, the RMFI no longer has that direct link: all it shows is the current source of Labour Together, rather than the original donor. As they were registered campaigners, we do know who funds Labour Together in general (this is declared to the Electoral Commission). But a similar intermediary organisation that did not campaign would be even more opaque. (Notes)
Recommendations
- Supported staff need to be separated out in the RMFI with custom fields to capture information about the role they are performing.
- The Register of Interests of MPs' Staff needs a fuller overhaul to - better capture information about the source of a staff member's funding.
- In the absence of wider finance reform, better passthrough declaration rules are needed for organisations who donate more than £50,000 across all MPs.
Parliament needs to enforce its own rules
A key issue when carrying out this project was a sense that we cannot trust that all of the data that should be in the Register is there, and that the data that is in there is accurate.
We undertook this work as a crowdsourcing project rather than a straightforward data analysis project because we knew the data was not good enough in itself — but what we learned is that even with supplementary research and analysis by volunteers, a huge amount of the data is unfixable. We can't discover through Google how much someone got paid for a job they did not declare, or whether they received a donation that we cannot even find the date of.
We believe the rules need to be strengthened, but that will be of no use if they are not enforced. Across this and the APPG side of the project we have noticed a set of rules that technically exist, but are poorly enforced.
The new Parliamentary data releases, which separate information into different fields, are a massive improvement on the free text that came before. However, this has then revealed further issues in the data quality — it is now much easier to understand where contradictory or incomplete information has been declared.
For example, an MP registered "Consultancy work" in category 1, with a generic sounding company name, and no company ID. ‘Nature Of Business' is listed as "Strategy consultancy for business", with no further details, salary or payment information. The nature of this work is unclear, but so is its duration; there is an end date provided of 31 July 2024 — but the "Is until further notice" field is answered "true".
Fixing this needs an institutional focus on data quality. Parliament needs to stop treating poor quality disclosures as the MP's problem, but instead treat it as something that affects the standing of the institution as a whole. Poor compliance with the rules contributes towards a negative spiral of lower standards and lower public trust in politics. Clerks should feel empowered by the collective support of MPs (most of whom want this system to work) to support them to make good declarations, and have clearer reject-and-improve standards for poor quality declarations.
More enforcement of reporting on a regular basis would help make sure the register is up-to-date, but would also allow for faster checking of interests when they do arise. In the case of donations and other high-value financial interests, there is no substitute for regular audits to mitigate the risk of non-compliance.
In the final section, we suggest ways that we could exert pressure to improve data quality from outside Parliament. (Notes)
Recommendations:
- MPs to be prompted quarterly to update their registered interests, with options to either update the Register or respond explicitly to say ‘no updates'. Non-response should be published by the Commissioner.
- Random quarterly audits of a limited sample of MPs' Register returns to improve compliance.
When MPs speak in debates, they are supposed to declare any interests. Where this happens, it tends to be of the form "I refer Members to my entry in the Register of Members' Financial Interests" – which does not say what the interest was.
This should just never happen. The Rules for MPs are clear that "a reference will not suffice on its own, as the declaration must provide sufficient information to convey the nature of the interest without the listener or the reader having to have recourse to the Register or other publication."
This should be an opportunity for the chair to remind the chamber of the rules on full disclosure. Instead, it is just accepted that the Chamber will not enforce its own rules on being clear about conflicts. (Notes)
Recommendations:
- The Chair should regularly remind MPs of the rules around declaring interests fully when speaking in the chamber.
When submitting parliamentary questions, MPs should say if they have any relevant interests, and if so what they are. In the guide to submitting Written Questions online, MPs are told "If you have an interest to declare, click the box saying ‘yes' and explain what the interest is." Similarly, on the offline form, members are told to email the Table Office to say what the interest is.
However, only the ‘yes/no' bit of this is transferred onto the Written Questions section of the Parliament website. The actual interest being disclosed is never released. This is a lower level of public disclosure than contributions to debates, where the standard is now (in theory) that it should be clear what the interest is as well as the fact that it exists. For oral questions, that an interest needs to be declared on the form is explicitly given as a reason there is no need for further declarations in the chamber. This would be reasonable if the information was available to be added to Hansard later — but as it stands, it is not.
We made an FOI request for this information, which was refused, but the Speaker's Office said they had "commissioned an urgent review of the publication policy" in October 2024. We are still waiting for further information about this review. (Notes)
Recommendations:
- Parliament should publish the nature of the interest declared when they publish written questions, not just whether or not an interest was declared.
Changing the rules
One of the challenges we've faced a few times during this project is around the question of political trust.
2024 polling by Survation and the UK Anti-Corruption Coalition found that "two thirds of voters believe UK politics is becoming more corrupt, while just one third had faith in their elected officials to put the public interest ahead of their own. And 80% of respondents said they think "some or all" of the main political parties are corrupt and untrustworthy". Given this starting point, is it responsible to do work that might decrease people's trust in politics and increase the perception of corruption?3?
Notes ():
3: We discussed actual as opposed to perceived corruption in our initial literature review, highlighting both that the UK has low corruption by international standards, but also that our high trust society makes corruption look different. There are few transactional bribes, but there are clearly signaled potential rewards that shape incentives for politicians and public officials.
The quick response to that has been that it is the politicians' job to build trust —ours is to be truthful. Increasing transparency does not generate trust directly, but encourages more trusted behaviour. We know from studies of the introduction of FOI laws that the initial result is an increase in perceived corruption because more of it is being found. But then this decreases as the increased likelihood of being caught decreases actual corruption4. Creating more transparency is part of the process to get to appropriate trust, where people's belief in the system is not because they are being kept ignorant of its faults.
Notes ():
4: Cordis, A. S., & Warren, P. L. (2014). Sunshine as disinfectant: The effect of state Freedom of Information Act laws on public corruption. Journal of Public Economics, 115(March), 18–36. https://doi.org/10.1016/j.jpubeco.2014.03.010
But to engage with the spirit of the challenge: that answer also shows why we need to look beyond transparency. Transparency isn't the end goal we're after, it's a means, and transparency is not sufficient in itself to achieve that goal. If we increase transparency and nothing changes, then we've accomplished very little at all. In this report we make a number of practical recommendations about what we've found and how transparency can be improved. But more important than that is how we get to the next steps on solving the problems revealed by this process.
Greater transparency has been a hedge against changing the rules that would stop behaviour that makes politics less trusted. In this section we go beyond "publish better data" to propose how the rules need to change to prevent the behaviour that brings politics into disrepute.
The Guide to Rules states that MPs need to declare gifts and hospitality worth over £300 from a single source. Both gifts and hospitality can come in the form of multiple instances that together reach that total.
The Survation / UK Anti-Corruption Coalition polling asked respondents "how MPs' acceptance of £180,000 worth of ‘freebies' from corporations in the summer of 2023 affected their confidence that policymakers will make impartial decisions". This found that "an overwhelming majority (70%) says it gives them less confidence that policymakers will make impartial decisions". The problems with the overall scale have been well noted, but to work through the problems with both the practice and the data, we're going to go in depth on a relatively small donation on the list.
In Rachel Reeves' Register is a 2023 Christmas gift of a £30 bottle of wine from the Betting and Gaming Council. This is a relatively unusual declaration tracking a low value gift. There were only three entries that mentioned ‘wine' in the September Register. The other two were Helen Grant declaring a contribution from Kingsley Capital Partners LLP to a fundraising event, and Neil Shastri-Hurst declaring an £133 gift of wine for association volunteers.
It seems unlikely that this was the only Christmas gift of wine an MP received, and this shows a blind spot in the Register. The reason this £30 wine needed to be declared is because Reeves had received a separate gift from the Betting and Gaming council in March 2023 ("three tickets to a musical performance"), which means the aggregate value of gifts from the same donor crosses a threshold and the later, smaller gift needs to be declared. If Reeves had not previously accepted free tickets, we would not know about the free wine.
There are two points to make here. The first is that Reeves should not have accepted these gifts. She was far from alone in doing so, but to state the obvious, people were giving gifts to the Shadow Chancellor in the hope of access or favourable treatment when she became the Chancellor. Should a government policy emerge around softer-touch regulation of betting companies to promote growth, it's a problem that Reeves has personally accepted hospitality from groups promoting that, regardless of the origin of the policy idea. The goal here is not just not being corrupt, but avoiding the perception that this might be what's going on.
The second is that the declaration threshold for gifts is too high, creating a missing data problem. It is good practice across the public sector to be tracking and declaring small gifts, but for MPs especially, it would fix the problem of aggregate spend being publicly available. There could be a huge amount of money spent on gifts to MPs in aggregate, but without it crossing the threshold for any particular MP.
While the Code of Conduct for MPs only talks about disclosing gifts, guidance for civil servants is much more rooted in offences under the Bribery Act. The Civil Service Code more directly says that civil servants should not "accept gifts or hospitality or receive other benefits from anyone which might reasonably be seen to compromise your personal judgement or integrity".
2009 Cabinet Office Guidance breaks this down further with points around purpose, proportionality and conflicts of interest. Individual departments give more precise guidance, for example the Single Source Regulations Office (loosely influential because they publish their guidance online) talks about gifts being of a nominal/notional value, for an appropriate reason, not at an inappropriate time, and being one-off or irregular. In that case they also talk about a nominal amount of £50 for a single gift — and no more than £200 from the same source. There is no lower threshold for registering gifts. All gifts made, even if declined, need to be stored in the SSRO's registry. The ICO's policy allows that sub-£5 gifts do not need to be recorded. In some areas, councillors have far stricter rules on disclosures and non-acceptance than MPs, with Merton Council requiring gifts of £25 or greater to be registered.
The basic ideas that "smaller gifts need to be recorded" and "there are some gifts you should not accept" are not radical ideas: they are standard practice through the public and private sectors. Code of Conduct should be revised to lower the declaration limit and introduce, for the first time, principles that guide whether they should accept certain types of gift at all. (Notes)
Recommendations:
- Lower the gift registration threshold in line with wider public sector / civil service thresholds (e.g. £20).
- Consult and adopt new rules and guidance on when MPs shouldn't accept gifts.
We have tried to categorise shareholdings that fit into our highlighted interests categories, but due to the insufficiently strict rules in this category, we have a strong belief that relevant shareholdings are not being disclosed.
The UK rule sets a high threshold on shares,- which can have a value of £70,000 before they are declared. The Financial Times' Chris Cook complains, "MPs have to declare tickets to a football match, but can have up to £70,000 in stock in one company without needing to tell anyone." He contrasts this with the equivalent US House of Representative rule:
"The Financial Disclosure Statement must include a brief description, the date, and category of value of any purchase, sale, or exchange of real property, stocks, bonds, commodities, futures, or other forms of securities (including trust assets) that exceeds $1,000. The category of value to be reported is the total purchase or sale price (or the fair market value in the case of an exchange), regardless of any capital gain or loss on the transaction."
In the UK, when interests appear and disappear we have no idea if they were sold in full, or sold in part/lost value so they fell below the threshold. While in principle, interests would need to be declared if they became relevant to how an MP spoke in a debate, this is at the MP's discretion as there is no independent way to check whether this has happened. Further to this, Cook points out a gap around insider trading, as the rules "are designed to catch an MP whose vote or position may be intended to make money — but not one who trades on information they have gathered from their public office."
This rule currently represents a significant hole in our understanding of MPs' financial interests — and means we cannot with confidence say MPs who make no declarations have no financial interests. Either adopting the US rule in full, or significantly lowering the disclosure threshold to £1,000, would be a significant improvement. (Notes)
Recommendations:
- Lower disclosure threshold for shareholdings to £1,000, and adopt the US rule on stock reporting.
The data quality in Category 10 (Family members engaged in lobbying) is poor, and the current fields are not capturing enough information about the range of relevant interests that MPs' spouses and family members might have.
This is not just a form design problem: this data is flawed because asking for spouses and family members who are engaged in ‘lobbying' is not capturing all relevant interests. In the Guide to Rules, lobbying is defined as "undertaking activities in a professional capacity and on behalf of a third party or client in an attempt to influence, or advise those who wish to influence, the UK Government, Parliament, devolved legislatures or administrations, regional or local government or other public bodies on any matter within their competence".
This is a broad definition; however, we know that MPs are not declaring spouses who work for thinktanks, which even under a narrow view of the rules could be assumed to fit in this category. In other cases, MPs are declaring family members who are journalists, which less squarely fits this category. There are other professions which might not fit into this category, or may go in and out of it depending on the legislative agenda of the time. For example, farming was an area of interest after budget announcements, but may not be as relevant in future.
For spouses and key close family members, we want to know: what they do, how they earn money, and what shareholdings and directorships they hold.
In local authorities, councillors must list their own interests alongside their spouse's across all categories. In the Ministers' Register of Interests (which requires its own overhaul), declarations made to the Permanent Secretary regarding a Minister's spouse or family interests across any part of the Register then appear in their own category, "7: Relevant interests of spouse, partner or close family member".
Finally, under the current system, many declarations in this category state that an MP's family member is ‘freelance', or in some cases the company listed was hard to identify, as no registration number is required. (Notes)
Recommendations:
- Collect spousal and close family interests alongside MPs' own interests in the relevant categories, as happens in local authorities. These should cover, at a minimum: income and employment, directorships, and shareholdings.
- Freelance roles with relevant public sector clients, or clients that lobby the public sector, should declare them.
Funding politics
The Register of Members Financial Interests sits at the intersections of many arguments about the structure and funding of UK politics. Transparency has been a compromise reform, where rather than agree rules on unacceptable interests, the solution has been to intermediately (and sometimes half-heartedly) improve transparency of those interests.
We have many ideas on how to improve that transparency. The goal is not good documentation of office holders' conflicts of interest, but their minimisation and elimination.
To properly engage with this, we need to engage with the wider system of money in politics. As long as UK political parties are overwhelmingly reliant on donors, conflicts of interest are part of the fabric of politics.
This section explores the problem, why previous attempts at change have failed, and new approaches we can use to break through the debate.
Our key conclusion is that a citizens' assembly would be a useful next step for either official or civic progress. Public opinion is marshalled as evidence against reform, but there is reasonable reason to believe we can break through that with a different format.
The key parts of this argument are:
- The question of higher public funding is tied to imposing limits on donor donations (popular, and restricts the influence of money on politics).
- Reform of public funding is frustrated through the lack of cross-party agreement, and gridlock is furthered by evidence of public opposition.
- Public polling in this area shows both lack of knowledge (in ‘don't know' responses), and a lack of structure to consider trade-offs (i.e. the donor-funded status quo is also pretty unpopular).
- Citizens' assemblies using a representative group to learn and deliberate on the question are an effective way to push past knowledge problems and get a better sense of trade-off priorities, helping unlock movement on both civic and parliamentary action.
- Citizens' assemblies are in general a useful anti-corruption device to enhance elected democracy, lacking the conflicts of interest that occur when politicians set their own rules, and creating a process to better develop public views.
There is a strong case for greater public funding of the work of politics. The core principle here is that if you want a policy to benefit the public, you don't want it to be funded by someone else. It is a false saving to let wealthy people and companies pick up the bill if the access they get in return is to our collective disadvantage. When we want politicians to work in the public interest, we shouldn't be comfortable when they become dependent on people with deeper pockets.
By public funding of politics, we mean the electoral and parliamentary work of political parties. Political parties fit an odd role in our political system, since as individual parties they are vehicles for partial interests (justifying requiring support from members and people who support those goals), but collectively they are how we facilitate populations of millions being able to participate in elections (justifying public funding to support this collective democratic function). It is appropriate they receive both public and private funds — what is in question is the overall mix of these funds.
Most political funding systems are a blend of different funding models - but the proportions can vary considerably. The UK has low public subsidies from both an European and Anglophone perspective. Because of the different electoral systems, these two spheres tend to have different approaches in subsidising political parties. European countries tend to provide political parties with funding directly based on votes received in the last election, while Anglophone countries tend to have a wider range of subsidies (tax credits, rebates, state assistance) or spending limits. In both cases, the UK is towards the bottom of the chart - with low subsidy per head compared to Anglophone countries5, and low dependency on state funding compared to European countries6.7
Notes ():
5: Karl-Heinz Nassmacher - The Established Anglophone Democracies, IDEA Handbook: Funding of Political Parties and Electoral Campaigns, p. 271
6: D.R. Piccio - Northern, Western, and Southern Europe - The Established Anglophone Democracies, IDEA Handbook: Funding of Political Parties and Electoral Campaigns, p. 225
7: Methodology note: These two charts are not cross comparable (because of the analysis choices of different sources), but both make the point that the UK is low by different metrics in different contexts. The OECD public integrity indicators have a similar 'public share of political parties financing' metric that ranks the UK low, but also has a problem in not having correct data for at least Canada, and disagreeing on whether Australia has more than the UK. As actual funding varies over time and by party in different countries, this is always going to be an approximation — but overall, we have three different data series broadly ranking the UK as an outlier on public support.
The dominance of donors in the UK (roughly 60% of MPs' funding) is both unusual and leads to negative public perceptions of the political system. A 2011 study found a "substantial majority of respondents believe that large party donations are motivated by hopes for access and influence or special favours from the political party".
The current dominance of donors is self-reinforcing, with parties having made big investments in a known system, making even small reforms difficult. Labour's decision not to adopt a (widely supported) policy on banning foreign donors was opposed by their fundraising chief Lord Alli (himself a major donor). Even without most people involved personally benefiting, a system that is at best corruption-adjacent can continue on inertia.
Our view is that the problems transparency is trying to solve can be more directly addressed by shifting the balance of funding of parties. The core obstacle to those changes is this collective inertia, rather than an impassable divide between politicians and the public. We need approaches that help get the ball rolling, and start to unblock some of these conversations.
The ways in which political activity should be funded (and the public/private split) has been a recurring political argument, while change happens very slowly.
While aspects of public funding have appeared around the margins (such as the introduction of short money in the 1970s), the modern era of party finances arrives with the Political Parties, Elections and Referendums Act 2000 (PPERA). This framework created a regulator in the form of the Electoral Commission (EC), was in favour of disclosure and transparency of large donors, and capped spending rather than capping donations. It introduced a small amount of more public funding in the form of the Policy Development Grants Scheme.8 The cross-party talks in the late 2000s included a more direct element of public funding, pitched as a temporary bridge while parties switched to relying on greater membership dues or smaller donations.9
Notes ():
8: This distributes funding to parties for manifesto development according to a formula.
9: Phillips, H. (2012). The Funding of Political Parties. The Political Quarterly,83(2), 318–324.
A need for more state funding has been the practical conclusion of each major round of review. The House of Commons' Committee for Standard in Public Life in 2011 both recommended capping donations at £10,000 and that "an inevitable consequence of the gap will be an increase in support for the parties from public funds." In the government's response to this recommendation, it said that "The Government believes that the case cannot be made for greater state funding of political parties at a time when budgets are being squeezed and economic recovery remains the highest priority."
In 2021, Lord Evans of the current Committee restated the trade-off:
The basic premise of the argument made by our predecessors remains true. If small donations made by party members are unable to match the costs incurred in running a major party, there are only two possible alternative sources of funding: big donors or additional funding by the state. If we want to reduce reliance on the former, then parties – and the public – would have to accept the latter. [...] Without the political will or public appetite for major reform of party finance, due diligence, transparency and scrutiny must continue to act as the first line of defence.
The current system reflects this legacy of transparency and scrutiny not as the best way of solving a problem, but one that avoids making trade-offs between different funding models. Among political elites, there are problems in reaching cross-party agreements because different parties are dependent on a different balance of donors — and a lack of public enthusiasm for any particular mix makes it hard to break that dynamic.
In practice, the perceived problem of public support is described as the major obstacle, but this sometimes seems to be used as cover for this lack of elite movement. To shift this, we need to work to create a coherent public perspective that can unlock reform. Our view is that this is a problem a citizens' assembly is well equipped to handle.
The usual way of making democratic decisions — delegating responsibility to MPs — leads to a conflict of interest when those decisions are around MPs' own standards and behaviour. MPs are public servants and shouldn't be setting their own rules on weighing up conflicts between personal and the public interest.
The problem is when you follow that logic, you then need to answer who should set the rules. Opinion polling on individual questions around funding politics often gives high levels of "don't know" responses on alternatives, and reveals ignorance about the scale of money in politics. This is unsurprising because generally people have low levels of information about how politics works, and do not walk around with a fully coherent alternative system in their head.
This is where a citizens' assembly can be a powerful tool. By a citizens' assembly, we mean a group of around 100 people, chosen by sortition to be a balanced selection of the general public. This group is given an opportunity to hear evidence, have arguments, and reach conclusions over potential options. Here what we are trying to create is a set of decision makers who do not have the self-interest of politicians, but also are given the same opportunity and structure as political elites to fully understand options and consider trade-offs.
Ideally this assembly would be convened by Parliament itself to inform its own standards. Involve, the public participation charity, have argued that a citizens' assembly should play an important part in setting the roles and standards for MPs' behaviour, and potentially as part of the way that MPs are judged. Introducing deliberative democracy into this process helps fix the problem of processes where MPs both set and mark their own homework (where assemblies are part of the anti-corruption toolkit rather than a replacement for MPs).
But this doesn't need to wait for Parliament, and could also be convened by civil society as a way of developing a firm set of marching orders that aligns research, lobbying and transparency work. The UCL Constitution Unit commissioned an independent assembly, Democracy in the UK, which provided a number of useful directions about ethical standards set by MPs10, and can work as a model that informs civic action.
Notes ():
10: This briefly includes a statement supporting a donation cap, but otherwise had little content about funding directly.
In this project we've tried to stick close to principles we can identify in existing polling and research, but building a better picture of public expectations (and especially considered trade-offs) would help align our work to a civic agenda. Joining civic power to deliberative democracy provides power in one direction, and legitimacy in the other — a powerful force to engage with conflicts within Parliament to shift official rules and responses. (Notes)
Recommendations
- A citizens' assembly should be convened on money in politics to unblock wider arguments about reform, and inform civic action.
Low public understanding of the current way politics is funded is a constant theme across several decades of this debate.
In their 2006 focus groups IPSIS Mori found common misconceptions in participants:
- Confusion between political parties and the government, which led to confusion between public funding of parties and taxes collected by the government to fund public spending.
- The perception that politicians ‘siphon off' a significant amount of public money to fund their own salaries and benefits, that could otherwise be used to finance public services. Examples of the types of benefits cited are expensive cars, smart houses and travel in helicopters.
- The belief that parties spend their money almost solely on election campaigns, and a lack of awareness about parties' day-to-day running costs.
- The perception that the main donors to British political parties are large corporations, rather than individual donors.
Summarising polls as of 2013, Jennifer vanHeerde-Hudson and Justin Fisher said that "[t]he central finding is rather pessimistic: political finance is plagued by widespread and pervasive misunderstanding by the public", before going on to use new polling to find that respondents overestimated the amount of money spent, and were unaware of the average size and source of voluntary donations.
Some of these responses reflect that there is a wider problem in perceptions of scale in public spending. In 2024, 22% of the public ranked "MPs' expenses" as one of the top three items of public expenditure, while in practice actual spending is a tiny proportion of public expenditure. Looking at the proportions of party funding sources, Danielle May's YouGov polling found a perception that parties are already 29% funded by public money. There is a consistent belief that spending on politics is higher than it is.
While this is seen as a blocker to reform, it's also possible to see low public understanding as an opportunity. This means we could triple state funding (with corresponding clamp downs on large donations) and end up where the public thinks it already is anyway.
One of the appeals of citizens assemblies, from a public policy perspective, is that they are a shortcut to the problem that the public (on average) does not know a lot about any given area. Through a relatively intense process of bringing people up to speed, you get results that aren't necessarily "public opinion" but "public opinion if everyone had more information".
When talking about public attitudes to public funding, we can start with the idea that it is not very popular, but also that you get different answers depending on the framing, and absolute opinions rather than trade-offs.
For some example polls, a 2017 IPSOS survey found "64% of British citizens said it is wrong to fund political parties by means of public money and 13% of people said it is right" (23% don't know). 2014 polling from the Electoral Reform Commission found "41% of people agree that a publicly-funded political system would be fairer than the one we currently have, compared to just 18% who disagree" (with 41% Don't Know).
Researcher Danielle May of Aarhus University ran a survey experiment in 2021 with UK and Denmark panels to test if different framing improves respondents' support for state funding of parties. She tested if framing around "worthiness" (political parties play an important role in democracy) and "fallibility" (dependence on big donors distorts the public role of parties) made a difference to support for state funding. She found that both did, with the effect on fallibility being stronger in the UK than Denmark, increasing support from 40%(ish) to 52%(ish). This possibly reflects the greater awareness of the problem of big donations in a UK context.
This provides good support for the idea that there is a complicated signal coming from the public. A policy choice that involves several unpopular options, has high numbers of "don't knows", and where framing can shift in significant ways is not a good candidate for making policy based on polling. But it is a perfect candidate to develop more consistent views through a citizens' assembly.
In general, it's a mistake to advocate for a citizens' assembly on the belief that the public would secretly agree with you if only they were educated better. There are genuine question marks about what a representative group of people would support if given time to think about it. Those "don't know"s could go either way. They might be fine with more money going to parties based on election results, support only specific programmes, or they might just be opposed to increased funding. But each of these options would tell us more about where we are, and unlock new paths for reform.
This is already the approach used to support ‘short money' for opposition parties in Parliament, and has practical benefits in being very easy to calculate and administer (it is the main form used in Europe's party based systems). The downside is it is unresponsive to change over a Parliament, and can reinforce incumbent parties.
Another option would be to treat donations to parties and candidates as a signal for greater public funding through top up funding — which picks up changing political winds. A version of this proposed by the 2011 Committee on Standards in Public Life was tax relief equivalent to gift aid for donations to charitable organisations.
The scale of match funding could be adjusted to the desired outcomes. If seeking to give more weight to smaller donors, match funding could be combined with quadratic voting (where larger donations would receive less public support)11. Alternatively, thresholds could be set: in Canada there are bands for tax credits for political donations.12
Notes ():
11: For example, a match funding formula of ‘sqrt(value/100)*100' would give £31 match funding to a £10 donation (300%), £100 to a £100 donation (100%), but only £1,000 to a £10,000 donation (10%).
12: The first $400 receive 75% tax credit, which decreased to 50% up to $750, and 33% over $50 - with a cap of $600.
A variation on the donation signal is to give people money that they can give to candidates. Seattle gives residents democracy vouchers worth $25, which they can give to eligible candidates. Uptake of this is low in absolute terms (5% used in 2023).13
Notes ():
13: Eligibility for Seattle's funding scheme also requires candidates accepting additional restrictions on fundraising from non-voucher donors and overall campaign spending, reflecting how public funding can be used to shift wider factors.
This is the primary Canadian approach. Through ‘election expenses reimbursement', parties that reach a relatively low threshold of votes (2% nationally or 5% across constituencies where they ran candidates) can have 50% of their election spend reimbursed (there is an overall spending gap creating a ceiling on this). Similar rules with different thresholds apply to individual candidates.
Ireland similarly reimburses a significant fraction of the spending limit if a minimum vote total is reached.
The appeal of this approach is that it comes from similarly candidate-based election systems (unlike vote-share models from European examples), and is tied to fundraising and spending in the current election rather than the results of a past election. It has a multiplier outcome on the effect of tax credits or other donation top-ups. Given this, it is interesting that it is missing from the list of options created by the CSPL in 2011.
But there is also a complexity issue, a reimbursement structure could incentivise loans to parties in a different way. Tying the maximum reimbursement to a fraction of the spending cap creates a public cost to increase the spending cap (which is not necessarily bad, but would be a change). Generally, this approach solves a specific problem in systems like ours, but the mechanics are less straightforward.
The other side of the equation is restricting what political campaigns can spend money on, to reduce a competitive need for funding. We already do this in the UK, for instance through banning broadcast advertising except for a limited number of party political broadcasts. An increasing cost of UK elections is online advertising. This demand side pressure could be reduced by extending the ban on political advertising from broadcast to online.
One way of squaring increasing public funding of politics with public concern about how this is spent would be to provide it for specific purposes rather than open-ended funding (this was one of the conclusions of IPSOS MORI's 2006 focus groups).
To some degree, this is already what happens. The two main forms of state funding are Policy Development Grants and funding for opposition parties' parliamentary activities (Short Money). One option for greater public support would be to increase these programmes, removing the amount of donations needed to support basic activities.
Currently the Electoral Commission distributes £2 million in research grants to parties to develop manifesto ideas. This could be scaled further to avoid party dependencies on thinktanks that themselves have external donors.
In this report we talk about a growing trend of external funders supporting staff (especially for the Shadow Cabinet before the last election). This serves a legitimate purpose, but also provides potentially unfair access to future decision makers. The "need" for this could be removed by just increasing existing short money to support more support staff.
And there are lots of ways of spending money that make politics work better. The IPSOS Mori groups suggested attempts to increase turnout and campaigning at the local level as possible uses of money. Along related lines, Access to Elected Office and EnAble schemes have been used to make it easier for disabled candidates to run for office. This approach could be revived and expanded. Compelling arguments for greater public funding may in practice be through a thousand programmes rather than a big principle.
There are lots of different options for increasing the headline amount of public funding, which in turn reduces donor dependence/enables donor caps. None of them are mutually exclusive, and a diversity of approaches would cover their respective blind spots.
The goal is not necessarily an immediate transition to a European "majority state funded" model, but changing current decision making incentives, and presenting more options for unlocking the debate.
Currently, movement on this issue is blocked both by the lack of cross-party agreement, and the lack of a clear lead from the public on the ‘correct' path. With the current system, you get arguments like "it might not look pretty, but people are stepping up to fund the work the taxpayer is not willing to". As Labour donor Dale Vince put it: "If we have a system in our country, and we do, whereby politics is funded by donations, then we have to accept that it's a good thing to do. And if it isn't a good thing to do, we should fund it publicly."
This is why a citizens' assembly is a logical extension of this debate. There are frequent statements that "we need to have a real conversation about party funding" but they are not followed up with a process that makes that conversation possible. There has been a repeated idea that the public is not willing to make a trade-off, but no one has actually tested this hypothesis. Reforms stall, not because they are popular or not, but because of opposition from those who already have power. A citizens' assembly is a tool to shift this debate.
In a previous section, we discussed the trend towards more secondments and staff supported by donors, especially supporting members of the Shadow Cabinet.
This works as a microcosm for the wider funding debate, as it is less the activity that is the problem, than where the money comes from.
Opposition parties need staff and capacity both to fulfill their current scrutiny role of the government, and to lay the groundwork for governing in the event they win the election. That there is no transition period in the UK makes this need especially sharp. Shadow ministers go from being a spokesperson on an issue (with no additional salary over their MP salary) to being in charge of a complex bureaucracy within a day. This is a very sudden switch, and so in the interest of hitting the ground running, it is good if the opposition has collective resources and staff who can transition into special advisor roles if they win the election.
The issue is less what the money is spent on, but that in bulk it starts to represent a large amount, notably from large donors. While individual appointments may be defendable, systematically, it shows the role money plays in helping connections and networks form (and conversely, that potential hires less to the liking of outside funders are at a disadvantage). We need to have a healthy scepticism of public spirited defences of big money in politics, where there are clear mechanisms for this having private benefits to donors.
From a ‘follow the money' perspective, we don't need complicated rules to split into good and bad secondments. Some form of donor cap would make the large sums involved in secondments less problematic, and would in practice reduce the money spent on this approach to donations. Secondments would be much shorter, multiple donors needed to support full roles, and intermediate organisations would be less appealing as a way of challenging support to politicians.
Alongside this, there is a specific argument to increase Short Money as it is the line of public funding best aligned with the underlying issue. In FY 2023/24, the Labour party received £7.5 million in Short Money to support Parliamentary activities. This would be higher if not for a 2016 decision to switch annual increases to being indexed against CPI (which was itself a climbdown from a planned large cut to Short Money). In general, given that so many of the current Cabinet benefited from extra staff, it would be interesting to hear them make the case that the current funding is sufficient.
In most areas, fiddly issues of good and bad activities become less complicated if we can more directly shift the balance and composition of funding. The next section explores how we can build on this report to have a stronger hand in shaping a better political funding ecosystem.
About the crowdsourcing process
One of the things we wanted to achieve with this project is use the crowdsourcing expertise and tools we've developed in partnership with other organisations to improve our core democracy services. We see working with volunteers as being the key to how we can make a much bigger difference with our work.
In our Effective Crowdsourcing framework, mySociety is generally a partner to other organisations, providing crowdsourcing technology. In this project (which we are grateful to the JRSST Charitable Trust for funding), we wanted to bring lessons learned from our work from other organisations home to our core services.
We built on our in-house knowledge and expertise to shape the questions we asked, and then recruited 50 volunteers to help us answer questions about all aspects of the Register of Members' Financial Interests for all 650 MPs. We are enormously grateful to our volunteers for all their work, and see this project as an important milestone in learning how we can best work with other people with complementary skills to make things better together.
We ultimately dropped several categories of industries because it was making it harder to draw a clean line between industries.
We originally had a mining category that effectively became mining/nuclear (nuclear energy needs plutonium mining). Drawing a line between the two was difficult: a growing proportion of this category was ‘nuclear' rather than pure ‘mining' and we didn't have good information on public perceptions. For the sake of this first exercise, we chose to exclude it.
While we were clear from our research that we wanted to cover gambling, an immediate problem that then results is the strong overlap with any declarations to do with organised sports. Volunteers, not incorrectly, flagged donations from football clubs and the FA as being adjacent to gambling. In the end, we restricted the scope to donations from companies directly engaged in gambling, and horse-racing. This is a judgement call — and there's an argument for both tighter and looser focuses.
One of the benefits of a volunteer based approach is that humans can identify fuzzy connections between a donation and an industry where programmatic approaches might miss them. However, the more fuzzy a connection, the less likely we are to be consistent (e.g. one volunteer might work out a connection, but another misses it for a different MP) and there is a general question about how many ‘hops' we want to follow a specific group through.
For instance, Clive Jones MP received a donation from Lloyds Bank, which a volunteer flagged as "fossil fuel investments". Should we count this as an oil & gas donation? Andrew Griffith has a free Sky subscription from his previous employer - but Sky also operates Sky Bet and Sky Vegas. As the outcome we want is fewer of these donations being accepted, we wanted in a first attempt to focus on donations where the connection should have been obvious without much investigation.
But this approach is also a bit of a systematic gap when looking at donations from organisations and multilaterals with a range of interests. For instance, donations from philanthropists Daniel Luhde-Thompson and Cressida Pollock were flagged because of "links to Quadrature, which is a hedge fund firm, and may have positions in fossil fuels and energy". This felt like too many hops, but we're referencing it here because mySociety has had funding from the Quadrature Climate Foundation in the past and we didn't want to remove a point about one of our own funders.
In general, we could improve this by budgeting for additional reconciliation stages (eg working backwards from fuzzy connections that one volunteer found, to identify others). We have incorporated this into our ideas for follow-up work in the final section of this report.
The process of identifying individual donors got some good results but was a difficult process.
For instance, it would be difficult for a purely automated approach to match a gift from Michael Dugher (former MP and chair of Betting and Gambling Council) of free tickets to a Premier League game, to other gifts from the Betting and Gambling Council. This falls in the box of things that are somewhat known by politically engaged people, but would not be obvious to the general public, and an automated approach would produce a huge number of false positives. This is the type of area where using volunteers to answer questions really pays off.
However, as most donations are from not especially notable individuals to a single MP, it is a much harder task to map them reliably to industries. A lot of volunteer time was spent trying to identify individuals — and often the result could not be confirmed from the data alone. In future, we would use project time more efficiently with a greater focus on making sure notable individuals were tracked, while not doing this general exploration.
In this kind of project there are choices to be made between making fuzzy associations that others can dig more into, and drawing harder and faster lines at the cost of losing sight of important things in ambiguous spaces. We've opted towards the latter in this iteration.
What's next?
By the next election, we want to see a Register of Interests where tighter rules mean less is declared in the first place, and what is declared is more understandable and of a higher quality.
We have made a set of recommendations for Parliament to take up and improve their transparency. But also core to the spirit of TheyWorkForYou is that we, the public, shouldn't wait for a better democracy and policy system to be given to us. We need to look for things that we can improve from the outside without waiting for permission.
The following ideas build on what we've learned from this project, looking for opportunities where a combination of TheyWorkForYou's public platform, technical skills, and our volunteers' effort and time can have a real impact.
We currently only have funding to look more at devolved Registers of Interests, APPGs, and the ‘Interests declared' blog series. We intend to build on these both to make pitches for institutional funding, and make the case to our supporter base.
A key issue we have identified is data quality. Even before embarking on this project, we knew this was a problem, but were still surprised by its scale. Improvements in data quality is a prerequisite for efforts by us, or anyone else, to create better automated summaries of the Register.
The challenge is that there is a class of problems we can identify from the outside but not fix. We heard from an associate who tried to pass a list of errors back to Parliament that there is not a good institutional route for corrections rather than complaints.
We want to use TheyWorkForYou's position to create a route for external data corrections, and exert pressure on both Parliament and MPs to improve data by giving faster feedback on data errors.
There are several possible approaches to this problem:
- We could build automated approaches using validation rules to detect likely issues (eg. Category 1 disclosures without any payment information attention, monthly payments that have stopped, invalid company IDs, misunderstandings how to fill in the lobbying section), regularly passing these back to Parliament, and scoring them on whether the data is then updated.
- We could also highlight data validation errors on MPs' TheyWorkForYou pages if we run into obstacles passing back feedback.
- We could enable volunteer-led partial audits, where a small cohort would examine a random subset of disclosures for clarity, and pass this feedback back to MPs individually, as well as producing general reports to improve standards.
This would be a smaller undertaking than this project, but help us continue to drive home the overall conclusions and recommendations from this project — and build improvements in data quality that enable a wider range of accurate analysis to be performed more cheaply.
Some of the problems we're looking at, (such as persistent under-disclosure of interests in debates, benefit from prompt feedback to MPs and amplification of the problem.
To this end, we are planning to do a pilot run for a few months of fortnightly summaries of MPs' declarations in debates. This uses our vector detection approach to help us quickly identify speeches for follow-up, and we may enhance with further fuzzy generative AI suggestions of possible matches with the Register.
The goal of this is to give feedback to MPs who are not following Parliament's rules on disclosing interests in debates. In principle the Chair should do this, and we could also give feedback to Chairs who fail to enforce the rules. We also want to highlight MPs who do a good job at this, to reflect that this is a rule that's possible to follow.
Our view on this problem is that it is a sticky part of parliamentary culture rather than conscious resistance — and with a bit of a push, we can help bring practice into line with the rules Parliament has adopted relatively quickly.
While we're pushing for better data quality from the Commons, a big part of the value TheyWorkForYou already adds is making it easy to see differences in financial interests over time, and providing easy-to-use spreadsheets of the data. We can continue to add value through applying this approach to other elected officials.
We will soon be adding the Registers of Financial Interests for members of the Scottish Parliament, Senedd, and Northern Ireland Assembly to TheyWorkForYou, and creating spreadsheet downloads to make it easier for journalists and civil society to explore changes in these registers.
Where we could go further is by extending this approach to local government. With 10,000 councillors in the UK and over 300 councils, this is a much larger technical problem — but if we can make progress on funding for Democracy Club's councillor database (part of our combined WhoDoesWhat pitch), this would be a good foundation for a wider interests database.
Ahead of that, we could pick off devolved areas: for instance, a combined Manchester registry for the Greater Manchester Combined Authority and its 10 constituent councils; or a London Register for the Greater London Assembly and the London 33 local authorities. By picking off areas in turn, we could do targeted outreach with local journalism and civil society to ensure our work had an impact of increased scrutiny.
One of our key recommendations is a citizens' assembly to unblock the arguments on how politics can be funded. This doesn't need to wait for Parliament, and could also be convened by a coalition of civil society organisations. The goal is to develop a firm set of marching orders that aligns research, lobbying and transparency work.
A key output of an assembly is to reduce the uncertainty on where the public would stand if there were greater understanding of the issues overall. Democracy in the UK, an independent assembly commissioned by UCL Constitution Unit in 2011, provided a number of useful directions about ethical standards set by MPs14. This was especially useful in reaching some clearer lines on issues where public polling on similar questions had high numbers of "don't know" responses. Where lack of public understanding and consideration of trade-offs is used as a way of dismissing public discontent, building a stronger understanding of the public position builds a stronger set of arguments.
Notes ():
14: This briefly includes a statement supporting a donation cap, but otherwise had little content about funding directly.
But the impact of this wouldn't be limited to better lobbying: it can shape how a range of civic actors approach their work, including how we run TheyWorkForYou. In this project we've tried to stick close to principles we can identify in existing polling and research, but have run into a lack of information in some areas. Building a better picture of public expectation would help align our work to a civic agenda. For instance, by agreeing what would be a desirable Gifts and Hospitality policy, we could flag through TheyWorkForYou (and our millions of visitors) how well this was or wasn't reflected in reality, helping shape wider coverage and debate.
Joining civic power to deliberative democracy provides power in one direction, and legitimacy in the other — a powerful force to shift official rules and responses.
The election summaries on TheyWorkForYou are a proof of concept that we might decide to remove as we get further from the election. We want to learn from feedback how we can build on this approach to add value to the live Register.
The value we think we are adding in the election summaries is:
- Volunteer-sourced short descriptions, categories, and links for organisations.
- Automated calculations and summaries enabled by those categories.
To make this more sustainable, we need to have better sequencing of automated and human approaches for this specific task.
There are some easy optimisations. Our volunteers worked through a list of MPs, so we had different descriptions for some organisations that appeared multiple times in the Register. Identifying new organisations and categorising them once would be more efficient. We could also add a ‘this entry/organisation is unclear' button to help focus resources on where more investigation can add the most value. To keep up to date, we could run smaller quarterly crowdsourcing efforts, which might either be sourcing information directly, or reviewing user submitted suggestions.
Extending from adding context to the registry to make it clearer, we could do basic rearrangements of the content. One of our recommendations is to split Category 1 (outside earnings) reporting into multiple sections based on the type of work, in part to capture better kinds of information about these different interests.
While this would not improve the underlying data, we could do this kind of rearrangement ourselves to better group different kinds of interests. We're cautious of using generative AI approaches to create summaries (see our AI framework), but adding structure to content without fundamentally changing it could be a good match for the technology.
The highlighted industries crowdsourcing is the most labour intensive part of the work: most organisations/individuals investigated are not relevant to the result. But we also see it as one of the more valuable outputs, aligning public views on acceptability with a public platform.
There is an option to do a slightly worse job for less effort. Looking at relatively few industries and investigating individuals who donated less would simplify the analysis. But this is a narrow version of the potential of this project.
Building on improved data quality, better research on public acceptability, and rolling updates from the enhanced Register, this approach could instead become broader, building on work done elsewhere in stages.
Acknowledgments
We're extremely grateful to our wonderful volunteers, who put in countless hours to make this project happen. Thank you to:
Abigail Page-Rumsey
Aleksandar Zdravkovic
Becca Watts
Celeste Gómez Romero
Charlotte Street
Connie Sullivan
Gary Godfrey
Inigo Surguy
Jenny Pollex
John Knox
Kevin Nunan
Léon McGregor
Lloyd Shepherd
Matthew Haydon
Natalia Griaznevich
Nikki Dekker
Pete McGaughey
Rosamund Williams
Ryan Harris
Sarah Halford
Tom Irwin
William Buller
And to all our other volunteers who didn't want to be named publicly.
Thank you also to all members of the mySociety team (past and present) who have contributed to this project and TheyWorkForYou.
Thank you also to our funders for allowing this work to happen:
JRSST Charitable Trust,
Patrick J. McGovern Foundation
Indigo Trust
Porticus UK
Related research
Footnotes
1 For the purpose of producing numbers around the general election rather than an internal party contest.
2 Jo White declared a researcher funded by Palace Yard (a thinktank), but this wasn't declared in the RMFI; John Healey declared an employment of a Grace White in the RMFI but is absent from the RFSMI.
3 We discussed actual as opposed to perceived corruption in our initial literature review, highlighting both that the UK has low corruption by international standards, but also that our high trust society makes corruption look different. There are few transactional bribes, but there are clearly signaled potential rewards that shape incentives for politicians and public officials.
4 Cordis, A. S., & Warren, P. L. (2014). Sunshine as disinfectant: The effect of state Freedom of Information Act laws on public corruption. Journal of Public Economics, 115(March), 18–36. https://doi.org/10.1016/j.jpubeco.2014.03.010
5 Karl-Heinz Nassmacher - The Established Anglophone Democracies, IDEA Handbook: Funding of Political Parties and Electoral Campaigns, p. 271
6 D.R. Piccio - Northern, Western, and Southern Europe - The Established Anglophone Democracies, IDEA Handbook: Funding of Political Parties and Electoral Campaigns, p. 225
7 Methodology note: These two charts are not cross comparable (because of the analysis choices of different sources), but both make the point that the UK is low by different metrics in different contexts. The OECD public integrity indicators have a similar 'public share of political parties financing' metric that ranks the UK low, but also has a problem in not having correct data for at least Canada, and disagreeing on whether Australia has more than the UK. As actual funding varies over time and by party in different countries, this is always going to be an approximation — but overall, we have three different data series broadly ranking the UK as an outlier on public support.
8 This distributes funding to parties for manifesto development according to a formula.
9 Phillips, H. (2012). The Funding of Political Parties. The Political Quarterly,83(2), 318–324.
10 This briefly includes a statement supporting a donation cap, but otherwise had little content about funding directly.
11 For example, a match funding formula of ‘sqrt(value/100)*100' would give £31 match funding to a £10 donation (300%), £100 to a £100 donation (100%), but only £1,000 to a £10,000 donation (10%).
12 The first $400 receive 75% tax credit, which decreased to 50% up to $750, and 33% over $50 - with a cap of $600.
13 Eligibility for Seattle's funding scheme also requires candidates accepting additional restrictions on fundraising from non-voucher donors and overall campaign spending, reflecting how public funding can be used to shift wider factors.
14 This briefly includes a statement supporting a donation cap, but otherwise had little content about funding directly.