Dear Mike


As part of our regulatory work programme for the first half of 2019-20, we’ve been carrying out a compliance check of ONS’s statistics presented in Business Demography, UK. We’d like to thank your team for the commitment it has shown in supplying us with the necessary evidence and documentation for us to complete this work. We also welcome the energy and ambition the team has shown in wanting to further innovate and develop these statistics. This letter provides an update on our findings from this review.

Understanding the way that businesses are formed, develop, grow, merge, split and die is vital to supporting government policy aimed at supporting business and generating economic growth. Business demography and related statistics can really add value to our understanding of the dynamics driving business start-up, longevity and impact on economic growth.

Through our own analysis, and discussion with the business demography statistics team, it is clear that since we last assessed these statistics – we published our report in March 2012 – many changes have taken place in the UK affecting the number and nature of new businesses and of business deaths and how and when we measure these. Examples are:

  • expectations around the timeliness of the business demography statistics (ONS presents the data around 11 months after the end of the reference year) may have shifted since we last assessed the statistics[1]. The lag in publishing UK business demography statistics restricts their use more towards research purposes rather than analytical or modelling uses as potential lead indicators
  • in recent years there has been a significant increase in people moving from being wholly employed to being either fully self-employed or a mixture of being employed and self-employed. People consequently have set up their own businesses. When these businesses meet certain criteria[2] they appear on the Inter-Departmental Business Register (IDBR) and are counted as new business births in the business demography statistics. Those who chose to incorporate their own businesses tend to have relatively high incomes compared to the unincorporated self-employed[3] and, consequently, have a greater potential to add value to the economy and to employ other people. ONS’s business demography statistics count business births and deaths without regard to their legal status so users interested in the growth potential from new business births would find this an important limitation of the statistics
  • an alternative statistic now being produced by ONS is a measure of employer business demography based on businesses with at least one employee. We welcome the work that the ONS business demography statistics team has undertaken to provide an alternative measure as the employer demography statistics offer better understanding of entrepreneurial activity. ONS told us that there seems to be little user interest so far in the employer business demography statistics.
  • some users use business demography statistics as a measure of the rate of new business start-ups[4]. Such users are interested in new start-up rates because they are key to understanding what is driving economic and employment growth. However, we need more evidence around the extent to which new entries onto the IDBR are good proxies for business start-ups within the reference year.

There is evidence that measures of new business registrations at a local authority level may be distorted by multiple registrations at a single postcode. We spoke to the business demography statisticians about their analysis on such multiple registrations and the potential to distort people’s understanding of new business activity at a local level. Positive steps are being taken to develop greater insight in this important area and we welcome this initiative.

There is evident potential to enhance the value and insights offered by these statistics by taking better account of the many changes that have taken place in recent years. We can see that the business demography and IDBR statisticians in ONS also see significant potential in the further development of these statistics, in order to enhance understanding of the business cycle and the explore in more detail the profiles of key business types such as start-up and employer businesses.

It is my view therefore that in order to fully explore the potential innovations and enhancements that can be made to business demography statistics we should undertake a full assessment of these statistics. This will allow us to take a detailed and comprehensive view of the ongoing development of these statistics in line with all aspects of the Code of Practice for Statistics. An assessment will also give us the opportunity to engage with a wide range of users and stakeholders who have an interest in the insights and narrative these statistics could offer. In carrying out the assessment, we will consider the broader range of data that are published in this space, with a view to considering what opportunities there are for greater insight to be offered to users across the whole suite. Such data include ONS’s UK Business: activity, size and location statistics and the Department for Business, Energy & Industrial Strategy’s Business Population Estimates. The statistics in Business Demography remain designated National Statistics pending their impending re-assessment.

Again, we appreciate the statistics team’s engagement with us on this compliance check. We will work with the team to determine best timing for the assessment and look forward to continued engagement with yourself and the wider team as we begin the assessment of these statistics. I am copying this letter to Andrew Allen, Rhys Hopkins and the statistician responsible for Business Demography Karen Watkins.

Yours sincerely

Mark Pont
Assessment Programme Lead


[1] In our last assessment in 2012 we asked ONS to look at publishing the statistics more timeously. Consequently, publication of the statistics was advanced by a few weeks. Countries such as France and the USA have found means for transforming their business demography statistics into short-term economic lead indicators of entrepreneurial activity.

[2] They would only get counted in Business Demography if they came onto the IDBR (Inter-Departmental Business Register) either through registering their business for VAT or PAYE

[3] paragraph 5.68



Related Links:

Assessment Report 187: Statistics on UK Business Population and Demography