By Josh Martin
Official economic statistics do not currently offer much insight into an important sector of the economy – one which educates, cares for, supports and homes people. The civil society sector, also sometimes known as the third sector or social sector, is largely hidden in the National Accounts and other important economic statistics. This is because it is spread across many industries and sectors of the economy. But we have reason to believe that it is large, important and warrants closer attention than it currently receives.
A new report by Pro Bono Economics and ESCoE, commissioned by the Department for Culture, Media and Sport (DCMS), offers a roadmap for constructing a satellite account of the civil society sector. The report explores the practical challenges of how a civil society satellite account could be constructed – answering questions including:
- How should the sector be defined?
- How should relevant organisations be identified?
- What variables should the account contain?
- How should the estimates be calculated?
What is civil society?
Part of the problem is that civil society lacks a precise definition in economic statistics. It broadly encompasses non-government organisations that serve a social purpose, and make no profit or have limitations on the distribution of profit. It includes many charities, but also other legal structures and organisation types. The activities of the sector are diverse, including education, health and social care, research, support for vulnerable people, environmental work, and more.
Those familiar with the National Accounts might be tempted to equate civil society with the “non-profit institutions serving households” (NPISH) sector. While there is an overlap, this is only part of the story. First, as the name suggests, non-profits serving businesses or government cannot be in the NPISH sector. Second, organisations making more than 50% of their income from sales will not be classified in the NPISH sector, but may still be not-for-profit and part of civil society. Third, NPISH includes most universities in the UK, which are not generally considered a core part of the civil society sector.
Research in other countries suggests that the civil society sector is usually equivalent to around 5-8% of GDP in other developed economies – not much smaller than the size of the manufacturing industry in the UK. And while we have tremendous detail on the size, composition and outputs of the manufacturing industry, we know little about the civil society sector.
So how do we identify civil society organisations?
This is a large part of the challenge. Given the diversity of relevant organisations, and their wide distribution across industries, sectors and legal forms, there is no silver bullet. The best solution is likely to be detailed work to link relevant registers (lists) of organisations, collected by government departments and sector bodies, to the Inter-Departmental Business Register (IDBR) – a comprehensive list of UK organisations used as a sampling frame for official surveys by the ONS. This would still likely be an undercount, as many small organisations are not on the IDBR, and are likely to be an important part of the civil society sector. There is already a “legal status” marker on the IDBR, including an option for “non-profit making bodies” – the origin and meaning of this marker is unclear, but it offers a promising starting point.
Once we’ve identified the relevant organisations, what variables should the account include?
As it stands, we cannot even answer basic questions about the civil society sector – what contribution does it make to the economy? How many people does it employ? Has it grown over time? So initially the variables to be included in the satellite account are similarly basic – gross value added, employment and hours worked (for both paid and unpaid workers), pay and compensation of employees, and number of organisations.
Given that the definition of civil society varies between stakeholders and users, the report recommends a modular approach to the account. All the variables should be available broken down by relevant industries, and various subtotals made available corresponding to different definitions of the sector. Geographic breakdowns are also recommended where possible.
Finally, the account should go beyond the concepts and variables normally included in the National Accounts. Notably, the volume and value of volunteering should be included, since this makes up an important part of the civil society. Some data on volunteering already exists and should be integrated with the economic data to give a fuller view of the sector.
How should the estimates be calculated?
Once relevant organisations are identified on the IDBR, new breakdowns of official business surveys could be compiled by re-examination of the microdata. More complex methods, such as post-stratification (a statistical technique used in sample surveys to improve the efficiency of estimators), could be used in some cases. The new estimates could be adjusted to ensure consistency with other economic statistics. Other data sources, such as those on volunteering, would also need to be integrated.
An ESCoE discussion paper by Josh Martin and Jon Franklin provides a proof of concept for some of these methods and approaches for the NPISH sector. Annex A of the new PBE-ESCoE report also contains some exploratory data analysis of the Annual Business Survey, IDBR/Business Structure Database, Annual Survey of Hours and Earnings, and Labour Force Survey.
What are the next steps?
The new PBE-ESCoE report provides a roadmap for developing a civil society satellite account, but putting that into practice will require investment, resources and some time. Ongoing ESCoE work on the measurement of the NPISH sector in the UK national accounts will provide further inputs into future development work. Until then, we’ll have to keep guessing about the contribution of this important sector to the economy.