Table of contents
- The production boundary – defining the labour share
- Data sources for numbers of self-employed in UK
- The sources of income for the household sector and how they contribute to the labour share boundary
- Labour share estimations
- International comparisons of the labour share
- Further work
- Authors and acknowledgements
The article is part of an analytical series undertaken by Office for National Statistics (ONS) to develop our understanding of trends in the economy, which are driving changes in labour productivity. In this article, we explore the labour share of national income, and how changing patterns of employment in the modern economy, specifically the number of self-employed persons, may lead us to reconsider how this is calculated.
The growth in self-employed persons has been a prominent feature of the UK labour market in recent years. The number of self-employed increased from 3.3 million people (12.0% of the labour force) in 2001 to 4.8 million (14.8% of the labour force) in 2017. While self-employment income, or “mixed income” as defined in the UK National Accounts has become a larger share of total household income, this increase in relation to the growth in the number of self-employed is small. Mixed income as a share of total household income only increased by 0.5 percentage points to 8.4% between 2001 and 2017. In comparison, compensation of employees, which makes up the largest share of total household income, increased from 61.9% to 66.2% over the same period, showing that compensation of employees grew faster than mixed income, and continues to increase its share of total household income.
This divergence between self-employment numbers and mixed income could be partly attributed to the classification of the self-employed in the UK National Accounts. The self-employed are a broad group with varied characteristics, which makes classifying the self-employed and their respective income, within the Accounts complex. This classification is dependent on various factors, placing different types of self-employed worker either into the household or non-financial corporation sector. There are practical difficulties in completing this allocation, which can will have an impact on the estimation of mixed income.
The labour share of income estimates the income received by labour in the generation of value added, which includes both compensation of employees and mixed income. The heterogeneous practices across countries in allocating the producer or enterprise units to different sectors of the national accounts can lead to difficulties in making international comparisons of the labour share.
This article begins by explaining the role of the labour share in the production boundary. It then presents the sources of income for the household sector, where the self-employed are classified, and how there incomes are positioned under the labour share boundary. It then discusses the treatment of mixed income in the labour share and analyses alternative approaches for calculating this share based on the labour contributions of mixed income to the production boundary and gross value added (GVA). The methods compare the UK, France, Italy, Germany and Spain and discuss the heterogenous approaches that countries experience when recording mixed income from self-employment. The article concludes by proposing further research into understanding the labour share in the UK economy, which would reflect the adapting evolution of employment compensation and its contribution within the production boundary, within the economic framework of accounts.Back to table of contents
3. Data sources for numbers of self-employed in UK
Estimates of the number of the self-employed are sourced from the Labour Force Survey (LFS). This is a household survey that collects data on employment status, where individuals are asked to classify themselves either as employees, as self-employed, in government-support training or as an unpaid family worker. However, the survey also covers employers, own-account workers, members of producers’ co-operatives as well as those engaged in the production of economic goods and services for their own and their household consumption. This broad definition does not identify the legal framework of the business activity and whether it has a legal accountability for the losses and debts, neither does it distinguish the independence and autonomy in decision-making between the business activity and the household, nor does it clarify whether the business activity can produce a separate set of accounts, which would all be important criteria for classifying an economic producer in the correct sector.
Individuals working in the medical profession are a good example of the broad differences in the self-employed boundaries between the national accounts and the LFS. The LFS classifies General Practitioners (GPs) and dentists as self-employed. In practice however, they can either earn an income from self-employment or they can earn a salary as compensation of employees. Whilst the number of people working as GPs from the LFS will record these individuals as self-employed, the national accounts will source data on their earnings from different sources, depending on the nature of their employment: income of those GPs who are self-employed will be sourced from HMRC Self-Assessment data and form part of mixed income, whilst income of those GPs who are employees will be sourced from the HM Revenue and Customs (HMRC) Pay As You Earn (PAYE) system and will therefore be included in compensation of employee, and correctly excluded from mixed income.
To make things even more complex the LFS also asks self-employed individuals information about other methods of payment received aside from receiving a salary or wage direct from an employer. This gives the self-employed the option to select the following categories:
- Paid a salary or a wage by an employment agency.
- A sole director of your own limited business.
- Running a business or a professional practice.
- A partner in a business or a professional practice.
If a self-employed individual is receiving a salary or a wage, that income is likely to be recorded on the PAYE1. In 2010 to 2011 the National Insurance Lower Estimate Limit was £97, meaning individuals earning over this amount were required to be registered on the PAYE system and included in compensation of employees. The remaining three other categories do not clarify whether the producing unit they work in, or they own, as self-employed individuals is an autonomous business with independent decision-making capabilities from the household it operates in, which may be classified as a private non-financial corporation.
Mapping these responses of people recording themselves as self-employed from the LFS to their economic activities as producer units, and therefore the classification of the “self-employed” is very important to the labour share calculation. Understanding whether individuals receive a regular wage from the business or whether they receive a dividend or a combination of both can provide a picture of how income may be distributed as compensation for the employment provided or whether it is re-distributed because of ownership of financial or other assets, that is capital income. It is important to maintain consistent classification of these transactions across the non-financial and financial accounts to prevent distortion of our labour share estimates at an aggregate level. To examine whether and how the self-employed can impact the overall labour share in the economy, it is also important to understand the household income components and where they are sourced from. 2.
Notes for: Data sources for numbers of self-employed in UK
In 2010 to 2011 the National Insurance Lower Estimate Limit was £97, meaning individuals earning over this amount were required to be registered on the PAYE system.
SNA Reference chapter 24 and chapter 25.
The labour share is an important measure of how value added has been generated from the production process. Although most of the income components of gross domestic product (GDP) are straightforward to allocate to labour and capital shares, the same cannot be said for mixed income, which has an impact on the income of the self-employed.
When accounting for mixed income within the labour share, it is important to apportion it into what is a return for labour input and capital investment. While most methods use a fixed factor share to adjust the labour share for mixed income, these methods do not account for the changing composition of the economy over time. In comparison, the ONS approach to apportion mixed income between labour and capital does account for this, and tends to have a factor share slightly higher than the two-thirds approach advocated by Johnson (1954). While most of these methods adjust the labour share for mixed income experience a level change, the overall trend remains the same. However, method 5, which assumes that the self-employed earn the same as an employee, does show an upward, and divergent trend from other adjusted measures from 2003 onwards. This can be partly attributed to the interaction of self-employment growing strongly from 2003 onwards, and the higher income imputed for the self-employed, leading to the imputed income being higher than actual mixed income estimates. As this approach falls outside the production boundary, this cannot be a reliable estimate for the income of self-employed.
While separating the income of the self-employed into a labour and capital component is a complex process, the diverse approaches used by countries to allocate incorporated enterprises, and therefore their income, into the UK National Accounts framework also leads to international comparisons of the labour share to be heterogeneous. The allocation of unincorporated enterprises into the National Accounts framework differs among countries, which are allocated based on a criterion to decide if they are in the households or non-financial corporation sector. Coutries which allocate all unincorporated enterprises into the households' sector experience a larger increase in the labour share than those which classify them based on employee numbers, leading to most unincorporated enterprises being recorded into the non-financial corporations sector. They show a lower increase in the labour share.Back to table of contents
8. Further work
This article is part of a continuous development programme in Labour Productivity in the ONS. Whilst the information and data sources which are currently available for estimating the labour share in the UK utilise the latest UK National Accounts aggregates and labour market variables, they provide the best available estimates for measuring the labour share. Further development work could be taken forward with the availability of microdata from administrative sources and particularly HMRC such as self-assessment forms and PAYE records would be particularly advantageous. They would enable a detailed analysis for tracking changes in recording individual’s salaries and entrepreneurial income over time.
In the medium term, reviewing the conceptual differences in defining the self-employed in the labour market statistics, compared to the UK National Accounts sector delineation would be a good starting point. This would help align the labour concepts to those of output, in the longer term if further administrative data were available it may also be possible to differentiate the labour share by industry in the economy and by taking into account this lower level granularity we may be able to present a more detailed analysis and capture the evolving labour contributions to the production process by industry.
Appleton, J (2011). Revised Methodology for Unit Wage Costs and Unit Labour Costs – Explanation and Impact. Newport: ONS, pages 1 to 10.
Atkinson (2009), “Factor Shares: The Principal Problem of Political Economy?”, Oxford Review of Economic Policy, Volume 25, pages 3 to 16.
Cho T, Hwang S, Schreyer P; (2017), OECD Statistics Working Papers, ‘Has the Labor Share Declined?: It Depends’,
Eurostat, Task Force on the Treatment of quasi corporations, National Accounts Working Group (November 2014) Item 8 – CN841.
Guerriero, M. (2012). The labour share of income around the world. Evidence from a panel dataset. (Working paper No. 32). Manchester, NH: University of Manchester, Institute for Development Policy and Management.
Johnson, D. (1954) The Functional Distribution of Income in the United States, 1850 to 1952, Review of Economics and Statistics, 36, pages 175 to 182
Karabarbounis, L. and Neiman, B. (2013). The Global Decline of the Labor Share. The Quarterly Journal of Economics, Oxford University Press, 129(1), pages 61 to 103.
Kravis, I. B. (1959). Relative Income Shares in Fact and Theory. The American Economic Review, 49, pages 917 to 949.
1OCED (2001), Measuring Productivity: Measurement Of Aggregate And Industry-Level Productivity Growth, OCED Manual, page 44
Office for National Statistics (2018a). Trends in self-employment in the UK. Newport: ONS.
Office for National Statistics (2018b). Labour productivity, UK: January to March 2018. Newport: ONS.
Rognlie, M. (2016). Deciphering the Fall and Rise in the Net Capital Share: Accumulation or Scarcity?. Brookings Papers on Economic Activity, 2015(1), pages 1 to 69.
Turvey, A. (2009). Developing a unit labour costs indicator for the UK. Economic & Labour Market Review, 3(6), pages 51 to 56.Back to table of contents
Contact details for this Methodology
Telephone: +44 (0)1633 455086