The big implications of revising R&D spending methodologies

Previously on Wonkhe we have covered whether the UK has already hit its target of spending 2.4 per cent of GDP on research. There is ongoing work to revise the accounting mechanisms for R&D spend but effectively this particular revision came about because research and development spending by smaller businesses was underrepresented within the ONS’ surveys that then fed into the overall accounting for R&D intensity.

The Science and Technology Committee that took place this Wednesday provided further insight into why it is so difficult to establish a single measure of R&D activity. In effect, ONS’ previous methodology assumed the majority of R&D activity took place within larger businesses and was easy to define within a few sectors of the economy. This was potentially the case through the late 80s and early 90s when this methodology was developed.

Today, the R&D landscape is much more complex. On the one hand there is an incentive for businesses to categorise a wide range of activities as R&D because of the tax benefits. On the other hand there are fewer single large industrial employers, particularly within the public sector, where R&D is easy to account for. More widely this raises the questions as to whether measurements and the incentives that flow from them are appropriately geared toward university and business research partnerships.

What gets measured gets managed

For example, the most recent Autumn Statement announced a revision to R&D tax credits which in the crudest explanation increased incentives for large businesses to undertake R&D and reduced incentives for small businesses to undertake such activity. The aim is to develop a more simple system which is less open to maladministration but it has a significant real world impact.

SMEs, statistically, account for 99.9 per cent of the UK’s business population. Wider government investment into Catapults, place based innovation, and gearing of existing funds toward regional development, rely on this broad business base. The reduction of tax credits not only harms the viability of SMEs undertaking R&D activity, and therefore the research partnerships that go with them, but may also inhibit the growth of small but innovative companies where margins are already tight. This decision is made politically and economically easier where there is an unclear picture of R&D activity within SMEs.

On the plus side, it does not seem likely that public R&D spending will decrease any time soon but policy decisions on things like tax credits raise the question as to how well R&D activity in SMEs is understood. Therefore, an impact may be that while public investment may rise it does not flow in a manner that reflects the real economy. For example, a revision in statistics may reopen a debate as to whether an innovation economy is best served in betting on the already big and successful or through investing in the small and high potential. The way this is measured and the way incentives like R&D tax credits flow from this are key matters of public policy and crucial to university partnerships.

Thinking beyond SMEs, the underlying measurements also impact how universities may interact with business and how they may anticipate receiving funds. To think of a few examples. It is harder to make a case for specific tax reliefs for innovative collaborations if the potential scale for collaboration is obscured by statistical methods. The second is that there are possible inconsistencies within regional data if national data is an aggregate of business surveys. In turn, this impacts considerations on where investment may be best served for levelling up through R&D, and where regional programmes and incentives may contribute to any lingering levelling up agenda. The final area is whether the composition of R&D intensive industries is reflected within current ONS data. This may mean there are strengths in areas of research that are unaccounted for, and assumed strengths that are over represented by dint of being concentrated within large businesses.

As Chris McDonald, Policy Chair for Innovation and Enterprise at the Federation of Small Businesses, noted at the committee, previous methodologies

Relied on an implicit assumption understanding that there was a correlation between business size and innovation intensity or innovation capabilities.

If this proves not to be true it could have significant implications for businesses, public policy decisions on R&D investment, and by extension universities and their partners.