The Business Population Estimates need a more practical way of accounting for self-employment

Self-employment and the fuzzy border between what is work and what is enterprise is back in the news, and the trend looks set for the quantity of these type of stories to grow. Over the past week, the big news story has been the industrial action taken by Deliveroo couriers in central London over new contractual terms.

It is important to note that Deliveroo couriers are self-employed, so from the perspective of enterprise research, we would expect to see these couriers in the Business Population Estimates in the field for ‘enterprise with no employees’. The couriers would be incorporated into this estimate because of the way the BPE methodology uses HMRC’s self-assessment count. However, the newly monikered Department for Business, Energy and Industrial Strategy (BEIS) — formerly known as BIS— released a statement in response to the Deliveroo strike saying ‘An individual’s employment status is determined by the reality of the working relationship and not the type of contract they have signed’.

This introduces an aspect of uncertainty to the Business Population Estimates (BPE), compiled by BEIS and the source for a wide number of claims about ‘entrepreneurial Britain’. Unsurprisingly, the BPE does not have a methodological approach to understanding ‘the reality of the working relationship’ for self-employed people. As was identified in an early blog post on How many business are there in the UK? the estimate simply uses a combination of the Labour Force Survey and HMRC’s self-assessment data as a proxy of entreprenualiasm. The theme of self-employment and its relationship is also considered in posts on Self-employment: The missing piece of the puzzle and Are the new self-employed ‘reluctant entrepreneurs’?

Increasingly the ‘gig economy’ and the changing business models being driven by the ‘platform economy’ are seeing an increasing use of self-employment contracts in what might have previously been considered employment-type relationships. It should be important then for researchers — looking at either the stock of entrepreneurs or the labour market — to be cognisant of the self-employed component in the BPE and to make necessary adjustments to the estimate accordingly. It is becoming increasingly clear that the proxy is over estimating the levels of entrepreneurialism in Britain.

While it’s easy to accommodate this in footnotes to caveat a piece of research, as we have seen in media coverage of the entire BPE, such nuance does not get reported. It is likely then that research that uses the full estimates is getting cited/quoted in other work and the nuance being stripped from reporting. We need a more practical solution that can be applied to the BPE to help with disaggregating microbusiness from the Non-Entrepreneurial Self-Employed component (NESE).

The initial inspiration for this blog was to explore the case for a coefficient that could be applied to the BPE to adjust for the NESE, however in thrashing out the preamble I realised this may be a little too ambitious a project for a blog post and, perhaps more importantly, that the data may not be available to construct it.

There are some obvious challenges such as accounting for regional and sectoral differences which would require some differentiation — possibly estimating the coefficient through a time-based shift-share analysis — or perhaps doing some new research using HMRC’s datalab, especially the self-assessment dataset, to identify other ways to estimate the NESE component, or looking at other survey data that are available to the research community.

It’s probable there isn’t a usable mathematical solution to this problem working with exiting data. However, exploring it in more depth should definitely add some value to the policy question. This might lead us to request more data, although I know there is general reluctance to add administrative burden in this area. It’s possible that a one-off sample based survey may yeild more information than currently exists. There’s much to be considered.

It may also be a case of reconsidering the tax rules in this regard — and this research investigation could contribute to this process. I was reminded of the issue that HMRC had in the early 2000s with IR35 Contractor status. Many contractors in professional services such as IT were leaving employment, establishing a Ltd company and re-engaging with their former employer as an enterprise. This produced a spike in the enterprise statistics as people took advantage of shifting income away from PAYE and towards dividends. In the end, HMRC had to issue an amendment to the rules and this eventually stabilised the statistics.

What is clear is that there does remain a need for a practical measure to keep BPE as a statistical product that is fit-for-purpose. It seems to be a research area that has some policy priority.

Garry Haywood, Microbusiness Research Portal advisory board

Published on 24th August 2016

Recent News

We need to rethink workers’ rights in today’s gig economy

The Guardian, 30th November, 2016. Theresa May has asked Matthew Taylor to lead a review of modern employment. The fast-changing... Read more...

Recent Blog

The Business Population Estimates need a more practical way of accounting for self-employment

Self-employment and the fuzzy border between what is work and what is enterprise is back in the news, and the... Read more...

Latest Tweets