How many businesses are there in the UK?

The Business Population Estimates for the UK & Regions (BPE), published by the UK’s Department of Business, Innovation & Skills (BIS), are currently receiving increasing attention from the enterprise research community. The BPE is the principal official source of statistics on business population and includes an estimate of businesses which are not registered for VAT of PAYE, along with a count of those that are. The former, which includes non-registered businesses, is increasingly important because of the rapid increase in unregistered business numbers in the UK.

In the past, data based solely on ONS’s Inter-Departmental Business Register (IDBR) were the key source used to understand changes in the business stock. However, these data were limited because they only included businesses that were registered for VAT and/or were operating a Pay As You Earn (PAYE) scheme. Businesses that had no employees, or where their turnover was less than the VAT threshold – which is the level of turnover at which a business is compelled to register for VAT and currently stands at £82k per year – were not included. As most academic researchers, and others operating in the realm of enterprise research, are aware, it can take businesses many years to reach the VAT threshold or take on employees, and many, or even most, never do.

Consequently, using VAT and PAYE based statistics misses a critical – and growing – number of businesses. Ultimately this leads to a serious underestimate in the number of businesses, a probable shortfall of several million undetected businesses. Essentially, using VAT and PAYE records, the number of businesses that go undetected is larger than those that actually are detected and counted. Obviously, this is unsatisfactory. However, the question is, how large exactly is this shortfall? In other words, what is the likely total number of businesses in existence?

Producing an accurate estimate of this total is complicated, but also rendered increasingly important as other evidence demonstrates increasing numbers of micro-businesses engaging in economy activity. Consequently, BPE has become an important tool to help the research community respond effectively to the changing composition of the business stock.

One of the main motivations for establishing the Microbusiness Research Portal is to identify and understand key issues relating to these changes. In this post, the first of a series exploring the BPE, we want to examine the composition of the estimate and draw on some important nuances that researchers and practitioners should consider in using this dataset.

Over the past three years it has been widely quoted that there are 5 million businesses operating in the UK. The claim is directly sourced from recent editions of the BPE. Since October 2015, the current estimate is 5.4 million businesses. Yet it is worth noting more than 3 million of these businesses are not registered for VAT or PAYE.

Since the estimate was first published in 2011 there have been more than 500,000 new unregistered businesses added to the stock accounting for 70% of the total increase over that time. So we can see that having these data available is important as it captures a significant component of change, but to what extent is this new figure of 5.4m more accurate? Are all of these 3 million non-registered businesses really businesses? And is this solely representative of new entrepreneurship in the UK, or are there other economic factors that might be affecting this estimate?

Let us consider how the BPE arrives at this estimate of 3 million businesses. BPE uses the Labour Force Survey (LFS) estimate of 4.28 million self-employment jobs, less 0.99m that are either registered for VAT or PAYE and, therefore, already accounted for in the IDBR. Then, using HMRC self-assessment tax return data, the remaining 3.3 million is allocated into 0.2 million partnerships (of 2.33 self-employed persons per partnership) operating under the VAT threshold and 2.8 million self-employed sole traders giving us 3 million businesses.

Increasingly self-employment is occurring in relation to roles which were previously undertaken by employees of companies. Among these are, for example, delivery drivers who may have previously worked for a parcel company, but who now work as a self-employed person for that same company, which supplies them with piecemeal delivery work, thereby reducing risks and costs for the company. Essentially, if there is little demand then the self-employed ‘employee’ takes a hit to his/her income and the ‘employing’ business does not lose money, thereby shifting the risk away from the company and onto the individual. In this respect the coordination of the driver’s activity looks more like that of an employee, yet their exposure to the market is more like that of an entrepreneur.

Business models in many sectors in today’s economy operate in this way. Currently, such situations count as self-employment and come under the category of ‘entrepreneurship’ even though many – perhaps even the individual delivery driver themselves – would not consider it as such. This type of trend means that the total number of ‘businesses’, including self-employed ‘entrepreneurs’ of this type, is rapidly increasing, rendering accurate estimation increasingly difficult but increasingly important.

In addition, 0.5 million of the self-employed in the LFS estimate have self-employment as a first job and self-employment as second job. So, both of these count towards the estimate. A further number of ‘businesses’ consist of those who have a second self-employed ‘job’, but whose main job is as an employee. While these are indeed all individual businesses, the economic activity of those working in these self-employed businesses is not being applied full-time, since they are businesses run in addition to traditional employment; yet, that is what might be inferred from an uncritical assessment of the 3 million firms. So there are many complexities associated with knowing how many businesses there actually are in total, and these estimates should not be taken uncritically.

These issues are of obvious importance for researchers. Some of the key research questions today, regarding such important topics as innovation, competition and productivity depend on knowing whether entrepreneurship and a broader business population is beneficial or detrimental. Answering these questions requires an accurate estimate of both the total number and the composition of businesses. While BPE has provided welcome initial insight into the numbers of unregistered micro-business in the economy, it still requires some careful interpretation.

Over the coming years, those of us who have contributed to establishing this Microbusiness Research Portal intend to explore these issues further, and in detail. We hope we can count on your interest and support in helping us to do so, as we feel that important insights into the nature of our modern economy, of use to both government and researchers, will be forthcoming.

Garry Haywood – Market Evaluator & Advisory Board Member
James Derbyshire – CEEDR, Middlesex University & Microbusiness Research Portal

Published on 17th December 2015

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