Response to Impact UK concerns
19 February 2026
We are pleased our publication of the size of the impact economy has generated interest and even challenges about what was included and what was excluded. The debate is welcome because the purpose of the report was in part to stimulate discussion about what we mean by the impact economy. We noted this is difficult and that ‘different analysts might reasonably draw boundaries in different places’. This blog responds to some concerns we have heard. If such methodological arguments excite you, dive right in!
The first concern is we excluded co-operatives from the impact economy. That is incorrect, though we can see why the list of omissions on page 15 might seem to suggest that. We neither excluded nor included co-operatives by default. They are counted (directly or as part of extrapolation from a sample) if they met the central definition, namely whether they prioritise public benefit over private gain. For example, co-operatives that are Community Interest Companies are included because legally they must serve the public interest. Our approach was to be neutral about organisational structure and governance. This allowed us to include private businesses that purposely seek to generate public benefits. In other words, we wanted to test the assumption that underlies much economic theory and thinking, namely that businesses (and the people in them) are driven by profits. Our estimate is that (at least) 15% of the economy is driven by impact.
Advocates of co-operatives argue that ownership is central to being impact driven. In practice they might be right. But conceptually we chose the position of neutrality of ownership to avoid including cooperatives—as well as other civil society organisations and limited liability companies such as football supporters clubs, amateur dramatic organizations, and professional bodies–that exist for the benefit of their members. We did not count creating benefits for members or owners as equivalent to serving a community interest. Our inclusion criteria favoured businesses that contributed profits to the mission.
As an aside, based on a comment about not including cooperative societies regulated by the Financial Conduct Authority (FCA), we dug deeper into the regulations governing cooperative societies. We found something we overlooked. While not required by legislation, FCA guidance on how it decides what counts as a cooperative society notes they take into account the International Cooperative Alliance Statement on cooperative identity, values and principles, which includes the principle that ‘Cooperatives work for the sustainable development of their communities through policies approved by their members’. If we take this as a regulatory requirement, we should have included approximately 800 cooperative societies in our calculations, which we estimate would add approximately £80M to our estimate of GVA.
A second and related concern is that we do not apply our ‘prioritise public benefit over private gain’ test consistently to private companies for whom shareholder interests are legally paramount. If benefits to shareholders cannot, legally, be overridden by anything else, then no business could pass a strict reading of the test. We did not consult lawyers and did not take a strict definition of whether private companies (and whether cooperatives are different in this respect) are legally allowed to prioritise public benefit over private gain. If not, we should relax the test. This could and should be the subject of further discussion.
A third concern is why we excluded the value of volunteering and public giving. This follows from choosing Gross Value Added (GVA) as our measure of the impact economy which is explained in the report. We reviewed the estimate of the value of volunteering on GVA made by Pro Bono Economics but excluded it as we wanted to relate our estimate of the impact economy to Office of National Statistics estimates of GVA for the whole economy. As we note, excluding this (and other factors) means our estimate is a floor for ‘true’ GVA of the impact economy. This is another topic that could be revisited.
These are three concerns we have heard. There may be others. The exercise of sizing the impact economy is thwart with difficulties not least because drawing a strict boundary as to what is in and out ignores the large amount of grey area. Our estimate is not designed to favour any part of the impact economy over another. We hope it is the beginning of a conversation, not the end. We look forward to continuing it!