Subsidies: why we need good data on how Government spends £4bn a year
The Government is currently developing a new, post-Brexit subsidy regime for the UK. This will cover how public authorities give grants, loans, tax waiver and other benefits to businesses to support national priorities like R&D.
Subsidies currently make up about £4 billion of UK public spending a year, covering grants, loans and tax waivers to business (but not contracts, which are regulated separately). Flagship new policies like the Government’s Levelling Up Fund will be covered by the new rules.
Good public data on which businesses are getting subsidy, where, and for what, will help us prevent potential harms like rent-seeking, and to help us understand which subsidies really do successfully level up or create innovation.
Unfortunately, the Government’s proposals about the subsidy data that’ll be collected from public authorities reveal pre-digital assumptions about the costs and benefits of collecting data.
Here at the Centre for Public Data, we’re calling on the Government to collect and publish comprehensive data on how subsidies are being awarded.
This would help support an effective subsidy regime, and it would have many benefits for UK business.
Download our short briefing or full consultation response (pdf), or read on to learn more.
About the proposals
Subsidies help governments support new national priorities, but they can also be harmful, e.g. if they’re used to support failing industries. For these reasons, subsidies are governed by law. Until last year, subsidies were covered by EC law, but now the UK is developing its own system.
We’ve looked at what the Government proposes on what data will be published on which companies are awarded money, how much and for what.
Unfortunately, the data proposals are very weak, and not suitable for the digital era. While BEIS has set up a new public database of subsidy, it proposes that only total subsidies worth over £175k over three years to a business (in most cases) should have to be reported.
And worse than this, individual subsidies under £50k should be exempt from reporting.
This would mean that the data published on subsidies would be very incomplete, and would not include many very large subsidies paid by public bodies to businesses.
Why this reveals an out-of-date view of data
The consultation is positive about data publication in general, saying transparent data can be used to challenge harmful subsidies, and quoting evidence that transparency supports good governance.
But the impact assessment (pdf, paragraphs 79-85) is concerned that the downside of reporting smaller subsidies would be higher administrative costs for public authorities.
This reflects an out-of-date, 1960s view of the costs and benefits of data reporting.
Firstly, the impact assessment prices the cost of reporting each single data item at £31 - or one hour of administrative time. This might have been true in the era of paper. But now, public authorities will store their subsidy data in spreadsheets or databases, and upload them in bulk to the BEIS database. The marginal cost of uploading five subsidies vs 500 subsidies is zero.
Secondly, the consultation doesn’t address that excluding some, very large subsidies will make the whole dataset far less valuable for analysis purposes.
Oddly, the consultation itself acknowledges that it’s hampered by a lack of data. It says that it hasn’t been able to obtain good data on past UK subsidies, and asks respondents to suggest data sources that would give it a better picture of who’s been paid what.
But under its own proposals, the same confusion would remain in future.
The case for publishing comprehensive data on subsidies
We think BEIS should set the threshold at which public bodies should have to report their subsidies to the transparency database at zero.
In other words, all public subsidies should be reported - even subsidies of just 1p.
Creating a comprehensive public dataset on subsidies would have many benefits:
It would encourage evidence-driven policy development - to know which subsidies are working we will need good data on subsidies, joined with data on outcomes. This analysis is best done both inside and outside Government, so the data needs to be public. This is the only way we’ll know how effective the £4 billion of annual spending is.
It would minimise the harmful side-effects of subsidy, since increased transparency will help challenge inefficient subsidies and deter fraud.
It would give all market participants equal access to information, enabling efficient markets, supporting new entrants, encouraging overseas investment, and increasing public confidence in the transparency of the UK market..
And it would have no harmful side-effects:
It would not increase administrative costs - the cost assumptions in the consultation are out-of-date, as data reporting can and should now be automated.
It would actually decrease rather than increase admin burden for public bodies, as if they report comprehensive information to BEIS, they will not also need to report it separately to other departments or the new subsidy regulator.
There are no privacy concerns, since subsidies are by definition public funding. (In the rare cases where subsidies are awarded to individuals, we’d support smaller subsidies being anonymised, but still recorded.)
We think this is a clear opportunity for the Government to modernise its data processes, with no downsides, and major potential upsides for supporting its own policy goals.
How you can help
We’ll be talking to stakeholders about this over the next few weeks, and aiming to meet with decision-makers to make the case for better data.
We’d love to hear from organisations who support these goals, or who might work with this data. If you’d like to talk, please get in touch: contact@centreforpublicdata.org.