Further detail on the data, formulas and methodology behind the tool
Design the future tool
Explore the tool
This weighting is the simplest. It assigns government support for R&D in proportion to population. If you choose only this weighting then every region of the UK will receive the same spending per capita.
This weighting uses scores from the 2014 REF to assign government support for R&D in proportion to research excellence in each region. Full calculations are documented on the Regional Research Scores GitHub repository. They largely match the formula used to assign QR money in the UK, except that only the output score is considered and no London weighting is applied. Removing London weighting as compared to the QR formula makes this a pure excellence measure rather than one with a regional component.
This weighting uses data on R&D spending by business. It assigns government support for R&D in proportion to business spend on R&D in 2016 taken from Eurostat table rd_e_gerdreg. A conversion rate of £1.00 = €1.22 is used to convert 2016 data from €/head to £/head. In 2016 the GDP of the UK was £1960 billion, which is the basis for our calculations.
This weighting uses regional GVA data from the ONS. It assigns government support for R&D in inverse proportion to the GVA/capita of the region. If you choose only this weighting then regions with weaker economies will receive more government spending on R&D per capita.
This weighting uses only the manufacturing component of the regional GVA data from the ONS. If you choose only this weighting then regions with more GVA from manufacturing will receive more government spending on R&D per capita.
This slider lets you model increases in government spending on R&D. As a rough rule, business spends £2 on R&D for every £1 that government spends on R&D. Following this rule the UK government's stated goal of reaching 2.4% of UK GDP being spent on R&D would require it to raise spending from the current 0.6% to 0.8% of GDP. This tool does not predict how business R&D would change over time in these scenarios.
Once you've selected your weights the following formula is used to calculate how much R&D should be spent in each region.
This tool works at NUTS1 scale (Yorkshire, London, East of England, etc...), but it would work equally well at NUTS2 scale (Greater Manchester, West Yorkshire, East Anglia) except for London which should generally not be split into parts for this kind of analysis. Since our code is open and documented such a tool could be built easily and would add to public discussion once the complexities of this tool were understood.
University spending on R&D is overwhelmingly funded by government and thus included. Charity spending is less clear cut. Our "private vs. public sector" split is probably more accurately described as "market-led vs. non market-led" spending since charities have in common with government that there is less market pressure to assign R&D spending for greatest economic return than exists for business. We created versions of this tool making that distinction but found them to be widely misunderstood and have opted for the more familiar language we use now.
We don't. All data and calculations are based on UK GDP in 2016 and expressed in constant 2016 £s. As UK GDP continues to grow faster than inflation the necessary corrections will grow. It is safest to consider the tool as modelling how spending would have been different in 2016 if different priorities had been used to assign R&D spending.