Summary
Building on proposed changes in the Election Strategy 2025, this report examines the processes for reporting election donations in the UK, recommending how these processes can be improved for better coverage and transparency.
There are three different reporting systems that can catch election related donations in the UK (Local Authorities, Parliament, and the Electoral Commission). All of these systems have different thresholds for disclosure, and different levels of public accessibility. There is not a consistent threshold of when a donation is high enough to be publicly named between processes, or between donations made to parties or candidates.
We recommend:
- A "report once" process, with more aligned thresholds, a focus on ease-of-use and quicker transparency - making the resulting information more portable for both election and parliamentary transparency.
- A harmonised threshold of £1,000 for public disclosure of donations.
- A public database of donations above the threshold - a consistent database in one place covering donations to candidates and parties.
- A private database where information below this threshold can be made securely available for research and scrutiny.
A statutory instrument would be needed to change the threshold (as was previously used to raise it), beyond that we believe that progress can be made through changes to reporting templates and flow within existing legislation, if the Electoral Commission takes a pro-active role in coordinating the process.
We have identified three avenues for future research that could support these goals:
- A sampling approach of local authority returns to demonstrate problems between the two systems.
- Exploration of what a data-sharing agreement on Electoral Commission small donor data would look like.
- User research to map out more details of the "report once" system - including the possibility of bridging into current disclosure templates.
Introduction
Over the last year we have worked with volunteers to explore the donations, gifts, and the wider political finance system.
Our main report (Beyond Transparency) focused on improvements to the specific processes of the House of Commons' Register of Members' Financial Interests (RMFI).
Turning to the system of election finance, the same general approaches apply. We can improve how we capture data, we can improve enforcement of existing rules building on that better data, and we can push for more comprehensive rules that build on more functional infrastructure.
This report covers what we learned investigating how election donations are recorded in different systems. We have mapped the different reporting pipelines, and both the processes and looking at the quality of the data coming out was concerning. The current system is confusing for both scrutiny and those submitting information. There are multiple declaration routes, and places where information vanishes, or is not validated or checked.
As it stands, the system of recording election information serves no one well. There are complicated reporting requirements, but the result is often poor public transparency. While in principle a huge amount of information is collected — very little of it is easily accessible to the public or the Electoral Commission.
There is a clear common interest between candidates, campaigners, electoral administration and the electoral commission in developing better systems for capturing and disclosing donation and spending data. Building on this better data, there is a clear need both for more transparency, and for better tools and approaches to understanding the mixed public/private aspect of small donor information.
This report walks through the data available, the problems we have run into, what should change, and some practical next steps.
Reporting routes
There are three main reporting routes:
- Local Authority Electoral Services - Candidates for election make a paper-based submission of donations over £50 made in the election period.
- The Register of Members' Financial Interests (RMFI) - People who are elected as MPs must declare donations above a £1,500 threshold.
- The Electoral Commission (EC)- who both take disclosures directly from parties, and act as a repository for information disclosed through the other routes.
There is overlap between these different routes, but in aggregate have significant gaps.
To show the problem here, we will walk through, depending on where the donation was made and the size of the donation, if the donor is named publicly or not.
We have three donors, one gives to the local party, one to the national party, and one directly to the candidate. In all cases, the parties give the money for a candidate's local campaign.
There are different thresholds for disclosure to parties: £2,230 for candidates/local parties and £11,180 for national parties1. If a candidate is elected, parliament's threshold for declaring donations is £1,500. Spending returns do not have a passthrough rule to track donors to parties for the purposes of a candidate's campaign, Parliament's register does - but this only applies if you are elected. Diagram 1 shows this logic in a flow chart. The results look like this:
Notes ():
1: If multiple donations from the same source to the same destination collectively cross the threshold, all donations are then in scope (this is why there are donations in the EC database significantly below the disclosure limit).
| Amount | Electionresult | Named in: Local Authority Spending Returns | Named in: RMFI | Named in: EC public database | |
|---|---|---|---|---|---|
Donor B (direct) | 500 | Elected | ✅ | - | - |
Donor B (direct) | 2,000 | Elected | ✅ | ✅ | - |
Donor B (direct) | 2,500 | Elected | ✅ | ✅ | ✅ |
Donor B (direct) | 2,500 | Not elected | ✅ | - | - |
Donor A (through nat. party) | 500 | Elected | - | - | - |
Donor A (through nat. party) | 2,000 | Elected | - | ✅ | - |
Donor A (through nat. party) | 2,500 | Elected | - | ✅ | ✅ |
Donor A (through nat. party) | 2,500 | Not elected | - | - | - |
Donor A (through nat. party) | 2,500 | Not elected | - | - | - |
Donor C (through local party) | 500 | Elected | - | - | - |
Donor C (through local party) | 2,000 | Elected | - | ✅ | - |
Donor C (through local party) | 2,000 | Not elected | - | - | - |
Donor C (through local party) | 2,500 | Not elected | - | - | ✅ |
The problem is there are several overlapping systems trying to do transparency in slightly different ways. The aggregate effect is to make it quite hard to build a consistent complete picture of donors.
Diagram 1: Donation reporting flow
Elections in the UK are practically run by Local Authority Electoral Services departments. They are responsible for the administration of the election, including being the repository of legally required transparency of spending and donations. These candidate spending returns are primarily a "follow the spending/keep your receipts" form of transparency (targeted at embezzlement) — but within the packet, one form also covers donations.
There is a window between the day a general election is called and polling day ("the short campaign") where candidates must report every donation above £50 to their local authority via these forms. They are later sent to the Electoral Commission (more on this later).
Under section 89 of the Representation of the People Act 1983 anyone can go in-person to their local authority and look at a copy of the physical spending return of any candidate for up to two years after an election (for donations, this copy excludes donor addresses).
To understand more about this process, we visited Leeds Electoral Services Department to see more about the disclosure process. Here we were given a pile of A4 paper – more than 100 pages. You're allowed to look at it, but if you want to take it away, you have to pay 20p per side. We were (politely) told it's not possible to get a digital version, or a copy of one page: it's the whole thing or none of it.
In the first candidate we looked at, the only ‘donor' listed is the candidate's local party branch and this was common across other successful candidates covered by this local authority.
Across the seven successful candidates we looked at, there were 18 donors listed. In total, the value of the donations from these 18 donors came to more than £84,000. Of that total, £79,000 (93.8%) were donations made by branches of the political parties, and £2,000 (2.4%) was donated by the candidates themselves, to their own election campaign. This leaves just 3.9% of the total value coming from sources other than candidates and their parties.
| Donation Amount | % | Number of donors | % | |
|---|---|---|---|---|
All Donors | £84,336.86 | 100.0% | 18 | 100.0% |
Local party branch | £79,103.94 | 93.8% | 13 | 72.2% |
The candidate | £2,055.42 | 2.4% | 2 | 11.1% |
Trade Union | £3,000.00 | 3.6% | 1 | 5.6% |
Individual donor | £127.50 | 0.2% | 1 | 5.6% |
Venue | £50.00 | 0.1% | 1 | 5.6% |
Across 7 MPs, we found one named individual donation directly to them (rather than the party). As it turns out, this might have been because of our one local authority sample. In their study, Chris Butler and Mark Pack looked at a random sample of MPs, and found non-party donations to candidates represented about 23% of the total donations (more on their research in a bit).
Just because this data has a much lower threshold does not necessarily mean it is more complete. Given that this is buried somewhat in another disclosure process (that is far less public), means things might easily be missed and not corrected. We did a crude check to understand more about data quality. Looking at the Electoral Commission's publication of the headline figures, 34 MPs declared they received no donations, which is of course possible, but worth digging into.
Comparing this to their post-election register of interests we found there was a set who declared donations related to the election campaign to Parliament, but nothing in the spending return. 10 of these had specifically specified that they had received donations via their local party for the election, but this was not listed as a donation from the party in their spending return (in one case, there was also an individual donation that in our interpretation should have been declared separately).
It is a little unfair to highlight these ones because these are ones that we think filled out one source badly because they filled in another source well. There is another group of MPs where we suspect the donations in the RMFI should also be in the spending returns but is less clear because they don't specify the purpose of the donation (but either timing or source suggests it was an election donation). There is then the final group with no donations in either — which might be more suspicious but we cannot dig further. This is just looking at those with no donations, with the aggregate figures and different time ranges, we cannot strongly validate if those who declared above zero were declaring everything.
There are a few things we can take away from this. There is information available, if not necessarily easy or cheap to access. If we did get access, we'd be able to cross reference between different datasets to find problems in both — but we can be pretty sure there are some data quality issues from the information we already have.
In the next section we look closer at the register of financial interests - which is a much more practically accessible dataset, but with its own problems, and much higher thresholds for disclosure.
Once an MP has been elected, they retrospectively have to declare their donations through a parliamentary process if there were donations above £1.5k. This has a key advantage in being centralised and easier to access. Importantly, the rules apply whether or not the donation came via their national party or local party branch (category 2a) or through any other route (category 2b). However, we also see problems with data quality and suspect missing records. All donations over £1,500 should be reported within 28 days, and this data is then published in the RMFI every two weeks.
The 2nd September 2024 edition of the RMFI showed nearly £16 million in donations received by MPs in the last 12 months. £10.98 million in donations were received via local or national parties (2a), and £4.98 million in donations directly or via other means (2b). If we were to take an average MP who reported a donation, they reported around £31,000.
| 2A (via party) donations | 2B (direct/other route) donations | All donations | |
|---|---|---|---|
Number of MPs | 443 | 178 | 513 |
TOTAL value | £10,976,515.95 | £4,980,854.18 | £15,957,370.13 |
MEAN per MP | £24,777.69 | £27,982.33 | £31,105.98 |
This shows us that a huge number of donations are being reported to the RMFI, but also that many MPs are not reporting any donations over the declaration limit at all. This may be legitimate, and would be where all donations were below the threshold — but shows where you might be able to get more information in-person than through the centralised system.
Something else we noted here was late declarations. Comparing the September and August editions of the register, we found around a dozen MPs who seem to have missed the original deadline to report donations that came in during the election. In total, more than £150,000 in donations were reported in September by MPs who reported £0 donations in August (this figure excludes Conservative leadership candidates, who bring the total to more than £275,000).
In general, the RMFI is a "what they declare is what you get" dataset. As an institution Parliament does little to chase disclosure or validation of the content (we think they should). It is in principle possible to validate it against the local authority disclosures (looking for individual donors above £1,500) - but this currently requires considerable work and expense to bring it all together, given that the candidate spending returns are only available to view in-person (and as covered, some donations are missing in the opposite direction). A donation made of £1,499 before the short campaign directly to the candidate might not appear in either.
This brings us to the third disclosure route, and the only organisation that has all the data we need under one roof: the Electoral Commission.
The Electoral Commission (EC) is the ‘independent body which oversees elections and regulates political finance in the UK'. They collect donations data in their own right, but also collect and create aggregate analysis of other sources of data.
For our purposes, the three important features are:
- Parties directly report their donations to the EC.
- The EC collects a copy of the spending disclosures from all returning officers.
- The EC publishes a database of donations.
Political parties have to directly report the donations (above £11,800 if national, £2,230 for local) they receive to the Electoral Commission (both in the long and short campaign). This is part of what populates their donors database. If these donations are then passed onto individual candidates, they should also appear in the RMFI.
The Electoral Commission gathers the spending statements to collect the total spending figures and publish aggregate amounts. While it then holds a lot of information, it doesn't do much extra with this in terms of publication.
For instance, while the EC in principle has a set of (recorded) donations of more than £50 - this is disconnected from their donors database, and they are reluctant to make it available publicly.
As part of Butler and Pack's study, the EC shared a sample of the returns, but redacted personal information like donor names "which the Commission said was necessary in order to make them GDPR compliant (given that the law requires personal information of donors to be declared on these expense returns)". Butler and Pack point out that "the law also provides permission to inspect the returns in full". In the conclusion, we will discuss potential obstacles to obtaining this data, and propose better process to make this data available for analysis.
The EC's Political Finance Database combines the party level returns with information from the RMFI. However, they apply an extra filter to this - only publishing donations to MPs above £2,230, whereas Parliament's threshold is £1,500.2
Notes ():
2: HoC Library briefing: "Under the rules of PPERA 2000 that the Electoral Commission regulates, donations for political activity over £500 must be from a permissible source. MPs must report individual donations over £1,500 to the Parliamentary Commissioner for Standards, including multiple donations over £500 from the same source in a calendar year. Donations over £2,230 must also be reported to the Electoral Commission but this is done automatically from data provided to the Parliamentary Commissioner". The MOU can be seen in this evidence.
Why does this happen? In 2023, a statutory instrument was passed raising both the spending limit in the election and the effective disclosure limit from £1,500 to £2,230. This disconnected the two thresholds — meaning that there is information that is known and in the public domain about donations between £1,500 and £2,230 — but this is not included in the Electoral Commission database.
While the Electoral Commission in principle knows about every donation to every candidate above £50 (if in a big set of PDFs) — in practice they do not publish donations to non-successful candidates, and publish less information about successful donations than is available on the Parliament website.
Improving donor information
Our starting assumption is that most problems and under-disclosure are legitimate errors resulting from managing several different (confusing) systems. If there is purposeful under-disclosure, to find it we need to reduce all the other errors to make this more visible.
Our recommendations are:
- An improved "report once" digital reporting process.
- Harmonised and (lower) threshold for public disclosure of donations.
- An improved public database.
- Processes and procedures for allowing research access to mixed public-private data.
An underlying problem is that there are several different reporting systems, with different processes and thresholds. These processes are not productively redundant because differences in time period, threshold and accessibility make cross-checking difficult. The current system is both hard to cross-check, and accumulates different errors in different places.
What we want instead is a "report once" system that can be filtered for different requirements. The goal here isn't to catch candidates out, but to have a clear reporting flow that gets the best information in more places.
Improved data infrastructure around a candidate's financial data is implied, but not explicit, in the government's election strategy giving the Electoral Commission more powers of enforcement. In their review of the 2024 Election, Parliament's Public Administration and Constitutional Affairs Committee recommended a "single or integrated digital platform and system for electoral administration", as part of a range of measures to simplify the work of electoral administrators.
The Association of Electoral Administrators (AEA) have also called for a system along these lines to update a paper based approach:
"Candidates and their agents expect transactions to be instant and online as they are in most other areas of their everyday lives. Currently, election expenses returns are paper-based and must be submitted with original documents, including invoices and receipts.
An online facility for the submission of candidates' election expenses returns should be developed. It should include provision for both a candidate and, where appointed, an agent to signify secure approval of the return. Such a system should also provide a means for the public to inspect returns, declarations and associated papers including provision made available to purchase copies of the expenses."
The way this works currently is there is a spreadsheet template used for the wider spending and donation reporting. This should be updated to a digital reporting flow — with a strong focus on ease-of-use and data validation rules.
Rules on party passthrough donations should be reflected in this system. Having this as a requirement on the donation returns (rather than just ‘donation from party') would let MPs work from the same underlying data when reporting to both.
The benefits of this is then that Parliament could directly ingest data prepared in this format into the register (applying a "public" threshold over the raw donation information). This would mean quicker and more accurate public disclosures - rather than varying errors and delays through new staff working with disclosures in different systems.
A threshold is a blunt way of dividing between small ordinary donations and larger donations that justify more public scrutiny. Currently these thresholds are too high, and have got higher. This trend can be reversed through secondary legislation.
In our Beyond Transparency report, we recommended dropping the disclosure limit in the RMFI to £1,000. We would mirror this for donations to both local and national parties.3 Combined with a single reporting flow, this makes sure that all relevant databases have the most full picture of donations.
Notes ():
3: We have not covered this in this report, but would also recommend the threshold for coverage for unincorporated associations should also be dropped to £1,000.
In the meantime, we recommend that the Electoral Commission improve their own database by including all information released via the RMFI. This is already public information - and it helps analysis and transparency to have it all in the same place.
There is an inherent tradeoff between transparency of the political system and donor privacy. The largest donations have the clearest justification for transparency, as this can represent the over-representation of wealth in the political system. At the other end are small donations. In general, we want donating to political causes to be an essentially normal thing to do, that does not necessarily lead to great publicity or administration for very small figures.
A threshold can be a crude dividing line between these two — but it is ultimately a spectrum, and information below a publication threshold has important information for understanding the effectiveness and processing of the donation system.
As such, in addition to a public dataset at a set threshold, the Electoral Commission should work to build a private database of donations below this threshold (with a practical lower threshold of £50), and work to make this available for research and evaluation.
This would supersede the current system of local authority transparency - providing richer public transparency than photocopies, and provide more options for privacy-sensitive handling that is not constrained by the difficulty of accessing or processing the data.
One of our wider recommendations is that the Electoral Commission should run a service that can give a unique identifier to donors donating more than £1,000 to aid reconciliation between disclosures to different campaigns. This could more easily be integrated with a centralised reporting system. As part of a project investigating beneficial ownership we explored ways of creating useful public IDs for reconciliation without releasing the underlying data, and a version of that might be useful here (e.g combining names with part of addresses to help identify unique people, without releasing addresses). Information that is not suitable for public release can be converted into a variety of useful formats for analysis.
What can we do from the outside?
Key progress on this issue requires adoption and effort by the electoral institutions - but there are actions that can be taken from the outside to make reform and progress easier.
Getting more donations information in the open a) immediately improves transparency and b) reveals all the holes that make the case for reform.
Something this exercise has shown is practically possible (but logistically different/expensive) is getting over-£50 donations directly from the Electoral Services departments (with some concerns about data quality). The problem is both the number of Electoral Services Departments (about 360) and the cost of photocopying everything for the single page we need - the worst case is about £5k in photocopying if Leeds is a model case.
The Butler and Pack study also shows that in principle the information is present at the Electoral Commission in a big pile of papers. The obstacle here is both search time and the privacy objection which would present an obstacle to a Freedom of Information (FOI) request. While the electoral service departments have a duty to allow you access, the Electoral Commission does not.
Generally decisions in this area have held that if there was a reasonable expectation of privacy when a donation was made, privacy takes precedence over transparency. However, in this case this information is publicly available elsewhere (with a principle in law that this is correct to be publicly accessible), so there might be a successful argument to make here.
The more practical FOI obstacle is probably the reasonable cost limit because of work required to both extract the relevant information, and redact addresses. In principle the relevant redactions should already have been done by returning officers, but this version may not have been deposited alongside the unredacted version.
Given these problems, either instead or in tandem, we could focus on a smaller sample to make the case for reform. Building on the random sampling approach Butler and Pack use, but using volunteers at a selection of local authorities to get access to the underlying donations data rather than the redacted copy through the Electoral Commission. This would control time and cost.
Against this approach is that it would be a lot of work to find out something we already mostly know - there are systematic problems in how this process works. What we could learn, given our recommendation to drop declaration limits to £1k, is what we expect the number of donations to be covered is. Although a limit here is this would only cover donations to candidates, as opposed to parties.
Another alternative would be to explore a data-sharing agreement with the Electoral Commission for access to their copies of the records - either in terms of research products in this Parliament, or the principle that there should be research access under future systems.
This would involve exploratory conversations with the EC, clarity on the legal position, research on comparable approaches in the UK/similar jurisdictions, and understanding if enabling legislative change is required.
A proposed reform from several different groups is an improved and unified reporting flow.
Ideally such a system would be officially investigated and commissioned by the Electoral Commission (with an understanding from the government that any enabling legislation would be forthcoming). Under current plans, the Electoral Commission will for the first time have a comprehensive candidate database — a great enabling step for this work. However, work could be done from the outside to demonstrate what is possible.
At a basic level, exploratory user research could be done on the requirements to help shape any official future service. This would build a picture of use by candidates and agents — to understand how they can be supported — and how this could be used to produce fast transparency products of use to democratic scrutiny. This research could directly inform simple approaches that work within existing legislation and processes. For instance, a tool that produced the Excel templates that need to be deposited with electoral services as well as producing better data products, streamlining the interface in a way that could be built on in later steps.
This is not a solution that can be taken all the way without significant work from official institutions. Consideration of work in this area should be sensitive to emerging plans, looking for opportunities to keep the idea of an improved set of pipes in the conversation, and where there are opportunities for small amounts of research or prototyping that could significantly move the conversation from the outside.
Conclusion
If transparency is how we are going to manage the outsized influence of big donations, we need a transparency system that works. Election reporting in the UK is currently duplicative for limited reward, creating administration for campaigners and election administrators, while leaving significant holes in basic public transparency.
This is not an unsolvable problem. A range of different groups are affected, and while there is work involved in a solution, it is credible to have a significantly improved process (both in terms of administration and public transparency) in place for a 2029 election. Future work can further expand specific issues such as user research on better data entry systems, better understanding on disclosure gaps between different systems, or a strong pitch on the need for data sharing agreements that balance the democratic benefits of transparency with small donor privacy.
Related research
Footnotes
1 If multiple donations from the same source to the same destination collectively cross the threshold, all donations are then in scope (this is why there are donations in the EC database significantly below the disclosure limit).
2 HoC Library briefing: "Under the rules of PPERA 2000 that the Electoral Commission regulates, donations for political activity over £500 must be from a permissible source. MPs must report individual donations over £1,500 to the Parliamentary Commissioner for Standards, including multiple donations over £500 from the same source in a calendar year. Donations over £2,230 must also be reported to the Electoral Commission but this is done automatically from data provided to the Parliamentary Commissioner". The MOU can be seen in this evidence.
3 We have not covered this in this report, but would also recommend the threshold for coverage for unincorporated associations should also be dropped to £1,000.