The value of voluntary activity in the UK for 2012 is estimated to be £23.9 billion, approximately 1.5% of GDP. This paper discusses the methodology and the assumptions used in this estimation. It is part of the Household Satellite Account, which attribute monetary values to household production activities that are not included in the National Accounts. These estimates are an update of the voluntary activity estimates published by the Office for National Statistics in 2002 for the year 2000. This paper also provides possible options for valuing the intervening years.
In 2002, the Office for National Statistics (ONS) published the first Household Satellite Account (HHSA) for the UK. This measured and valued the goods and services that are produced by households, but are not included in the National Accounts (ONS, 2002; Holloway, Short & Tamplin, 2002).
Following the publication of the Report by the Commission for the Measurement of Economic Performance and Social Progress (2009), there has been fresh interest in estimating the value of household production. As part of the Measuring National Well-being programme, ONS is currently updating the HHSA which measures the aspects of household production in the UK that are not included in the National Accounts.
Household production, in the context of the HHSA, is all the unpaid production of goods and services by households in the UK. Conventional National Accounts measurements, such as GDP, do not fully take into account these goods and services. However, if these goods and services are paid for, they contribute to GDP (for example, paying for clothes to be ironed rather than the task being done by a member of the household).
The HHSA provides a means by which the influence of changing patterns of unpaid work on the economy can be measured. The valuation of unpaid work could prove useful in assessing the impact of policies on households and the voluntary sector. This work falls outside the scope of the UK National Accounts. This is because the inclusion of all activities which are productive (in the economic sense) but which do not have a monetary value would swamp the monetary flows1, obscure what is happening in the markets, and reduce the usefulness of National Accounts data for analysis. This significance is demonstrated by the HHSA results for 2000, which show the total monetary value of these flows to be close to GDP for that year. The HHSA remains separate from the UK National Accounts; however, they are conceptually consistent.
The HHSA extends the National Accounts boundary to include all activity that could be delegated to another person. This activity is divided into several principal functions:
ONS aim is to update all the above eight modules as part of the HHSA. This work is being completed as part of the Measuring National Well-being Programme. ONS has published the estimates for Informal Childcare in the UK and Informal Adult Care in the UK in 2013. This paper updates the third module – voluntary activity.
The approach being taken by the ONS to value HHSA is to focus on the outputs of the principal functions except voluntary activity, where the focus has been on input approach due to data restrictions making the output approach challenging. The methodology remains under development and any estimates reported should be considered experimental and interpreted with caution. ONS welcomes comments and feedback on all aspects of the methodology used and the assumptions made, and seek suggestions for further/alternative data sources.
System of National Accounts (SNA) 2008 – Section 1.41: Household Production.
As there is some crossover between volunteering and care it is important that the definitions are clear. Volunteering is defined by the International Labour Organisation (2001)1 as ‘unpaid non-compulsory work; that is time individuals give without pay to activities performed either through an organisation or directly for others outside of the household’. This definition shows that voluntary activity can be classed as formal or informal depending on the nature of the activity. This paper only estimates the value of the activity that occurs through a formal institution – a group or organisation. Informal volunteering, such as helping friends or family members, is included either in the childcare or adult care projects (separate modules within the Household Satellite Account). By not including estimates of informal volunteering we may be underestimating the value of volunteering in the UK.
The output, or product, of voluntary activity is difficult to measure. The goods and services produced by volunteers are generally not sold and, therefore, there is no readily available market price. The voluntary activity was originally published as part of the HHSA by ONS in 2002 and was valued at £13.2 billion for 2000. As with all the other modules of the HHSA the goal was to measure the output independent of the inputs, for example, the number of hours children spend in uniformed organisations rather than the time spent supervising them. However, due to data limitations, the 2002 account valued voluntary activity using the proportion and hours of the people volunteering – an input based methodology.
Outlined in Appendix 2
There are a number of ways to value the hours of formal voluntary activity:
1) The opportunity cost approach values the hours volunteered at an individual’s wage rate. This method assesses the value of what the person is willing to give up in order to volunteer: the wage that they could be getting from a job. Therefore, the volunteering must be worth at least that much to the individual for him/her to give up that wage. However, using this method an identical service may have a different value depending on who provides it. For example, a leaflet being handed out by a lawyer could be worth more than the same leaflet being handed out by a pensioner.
2) The well-being approach gauges the positive change in a person’s personal well-being that is associated with frequent voluntary activity. This change is then given an approximate value equivalent to the rise in income that would cause the same change in personal well-being. This method again approximates the value to the individual of volunteering - how much someone has to be paid to give up volunteering and still be as happy. A well-being approach to valuing volunteering was published in 2013 and is examined in Appendix 2.
3) The replacement cost approach values each individual hour at a market wage rate to show what would be the cost of paying somebody to do the same job. This wage rate could either be the minimum wage, the mean wage, the median wage or a market wage for the voluntary work. This method assumes that the quality of voluntary work is the same as the quality of paid work and that the volunteers would be replaced with paid workers. However, as we are trying to measure the product of volunteering, the value of the work being done, this method is the most appropriate for valuing voluntary activity. This methodology uses a specific wage from the ONS Annual Survey of Hours and Earnings to value voluntary activity.
Both the opportunity cost approach and the well-being approach value the gain to the contributor, rather than to the recipient of the output. They do not show the value of the output being produced, but the replacement cost approach shows how much the recipient would have to pay to replace the volunteering and therefore the cost that the volunteers are forgoing. This makes the replacement cost method the best available method to value voluntary activity. The replacement cost approach has been recommended by the International Labour Office and is the most comparable to the National Accounts.
This section estimates the number of hours volunteered in the UK in 2012/13 and values them using the replacement cost approach.
Those who volunteer at least once a month are characterised as frequent volunteers. This paper only estimates the value of frequent volunteering. The exclusion of infrequent volunteering is due to issues with data quality. The infrequent hours data tend to be less consistent and by just using the frequently volunteered hours, the estimates could be more accurate.
From the Community Life Survey1 it is possible to work out the proportion of the sample who volunteered formally and informally and the hours which they volunteered in 2012/132. The survey estimates that about 29% of those who are over 16 years old frequently take part in formal voluntary activity. This equates to nearly fifteen million people in 2012/13.
The Community Life Survey provides data on twelve specific types of activities which are then split into three general groups. These groups show the type of activity taking place for valuation purposes. All of the volunteering valued within the HHSA is through a formal group or organisation. These are outlined in Table 1.
|Citizenship Categories||Type of Activity|
|Raising Money||Clerical and Secretarial|
|Leading a Group||Professional|
|Organising an Event||Clerical and Secretarial|
|Visiting People||Personal and Protective|
|Befriending or Mentoring People||Personal and Protective|
|Giving Advice, information or Counselling||Personal and Protective|
|Secretarial, Admin or Clerical||Clerical and Secretarial|
|Providing Transport||Personal and Protective|
|Other Practical Help||Personal and Protective|
|Any Other Help||Personal and Protective|
Median wage rates are used to reduce the effects of outliers which could cause distortion. The data, which have been split into groups dependant on the type of voluntary activity, are valued using gross median hourly earnings3. An individual can take part in one form of activity or a combination of two or all three. The median wage is then applied to the number of hours depending on which activity or combination of activities the individual takes part in. The median wages are outlined in Appendix 4.
Table 2 shows the volume and value estimates for Voluntary Activity in 2012/13. UK residents formally volunteered 2.12 billion hours in 2012/13. As mentioned earlier, this paper focuses only on the frequent voluntary activity – that is those who volunteer at least once a month. The value of frequent volunteering was £23.9 billion in 2012/13. This value is an estimate of what the same number of hours would cost if they were being provided by a paid person, in the market.
|Year||Total Hours (billions of hours)||Frequent Hours (billions of hours)||Total Value (£billion)||Frequent Value (£billion)|
It is more meaningful to analyse the value of volunteering activity as a proportion of GDP. The value of frequent volunteering as a percentage of GDP4 was just over 1.5% in 2012.
For a full explanation of the data sources please see Appendix 1.
The later years of the Citizenship survey (2007/08, 2008/09, 2009/10 and 2010/11) and the Community Life Survey (2012/13) span 2 years.
Annual Survey of Hours and Earnings (ASHE).
This is Current Price Annual GDP from the ONS UK Economic Accounts.
Voluntary activity was valued for the year 2000 (in the 2002 Household Satellite Account) using the 2001 ONS Omnibus Survey with the replacement cost approach. However, no estimation exists for 2001 – 2010/11. This section discusses options to estimate the number of hours volunteered in the UK between 2001 and 2010/11 which are then also valued using the replacement cost approach.
Between 2001 and 2010/11, the data available on voluntary activity is not as detailed as in the Community Life or Omnibus Surveys and therefore there are challenges involved in estimating the number of hours volunteered between these years. One of the best data sources that could be used is the Citizenship Survey, which preceded the Community Life Survey. This survey measured the proportion of the population that formally and informally volunteered as well as the frequency of volunteering, but not the formal voluntary hours. To value the proportion of the adult population that is volunteering, a proxy needs to be used to estimate the number of hours volunteered.
Three options to estimate the number of hours volunteered are discussed below. All of them use the Citizenship Survey but with different approximated hours. The sources for these three different sets of approximated hours are the 2001 Omnibus Survey, the 2011 Understanding Society Survey and the 2012/13 Community Life Survey. In all three cases, the replacement cost method is used to derive values from the volume estimates.
It should be noted that the Citizenship Survey dataset does not cover all the relevant years. Figure 1 shows the years that are covered by the Citizenship Survey.
Options to estimate the number of hours volunteered
Creating proxy hours for each frequency of voluntary activity - weekly, monthly and so on – for each different age group allows the use of hours data in conjunction with the proportions data to create volume estimates for the number of hours volunteered in the UK. The hours from the Community Life survey are given in Table 3.
|16 to 29||171||125||5|
|30 to 49||213||83||4|
|50 to 69||283||90||5|
|older than 70||226||93||6|
Table 4 shows the average number of hours that people from different age groups worked as volunteers. These are split into the frequency of volunteering and average number of hours is calculated for the whole population from the Understanding Society Survey1. These data are used to approximate the number of hours worked by the individual, depending on their age and regularity of voluntary activity, in each of the years of the Citizenship Survey.
|16 to 29||241||89||10||9||87|
|30 to 49||223||73||7||7||77|
|50 to 69||314||100||8||12||109|
|older than 70||332||106||10||4||113|
In 2002, HHSA used the 2001 Omnibus Survey to estimate the number of volunteers, the hours volunteered in the last four weeks and the type of voluntary activity undertaken. The average number of hours per year in the Omnibus Survey for each activity is given in Table 5. The hours provide a separate estimation to proxy hours provided by the 2011 Understanding Society Survey.
|Average Annual Hours|
|Personal & Protective||165.1|
|Clerical & Secretarial||161.2|
|Personal and Professional||136.5|
|Personal and Clerical||163.2|
|Clerical and Professional||134.6|
Table 6 shows the total frequent hours estimates from each of the three options discussed in the previous section. The range in the estimates is noticeable. All three possible options create estimates of around two million voluntary hours over the decade. The pattern between the years reflects that of the proportions seen in Figure 1.
|Year||Understanding Society||Community Life||Omnibus|
These hours are valued using the same replacement cost method used to value the 2012 estimates discussed earlier - splitting the hours into types of activity and using a market replacement wage. This wage, as before, is taken from the Annual Survey of Hours and Earnings and is explained in Appendix 4.
Figure 2 shows a comparison of the valuation of the three different options and the value for 2012. The value of each year is shown at the wage rate of that year. This means that an increase in value is due to wage increases rather than growth in volume.
The highest and the lowest value of voluntary activity derived above using the three options discussed earlier could be used as a range to value voluntary activity. With each of the options, the value of voluntary activity follows a similar pattern. The Community Life Survey is the most comparable to the 2012 estimates due to the proportions used in the methodology. Each of the options gives an experimental estimate of the value of voluntary activity over the decade. Any future improvement in data sources might lead to more accurate estimates.
Data Sources explained in Appendix 1.
The Household Satellite Account (HHSA) is an experimental dataset and therefore the methodology used in this paper to value volunteering activity is under development. This section discusses another data source – the British Household Panel Survey (BHPS) – as opposed to the three data sources used above for valuing formal voluntary activity for 2001 – 2010/11 to show how big the change would be if this data source is chosen.
Conducted until 2008, the British Household Panel Survey (BHPS) was a UK wide biennial survey that provided data on the proportion of the sample that formally volunteered. It was replaced after 2008 by the Understanding Society Survey. Both surveys form an alternative data source to the Citizenship and Community Life Survey. The proportion of those volunteering was much lower than that indicated by the Citizenship Survey (see Figure 3). The BHPS goes into much less detail in the questions about voluntary activity than the Citizenship Survey. By asking more questions about volunteering, the citizenship survey collects more information and a more reliable number of those volunteering.
According to the BHPS, the proportion of the sample that volunteer frequently and formally averages just under 11% in the BHPS (2000, 2002, 2004, 2006, 2008) presented in Figure 3. The methodology used to create the estimates for 2012/13 discussed in the earlier section is replicated to create the volume estimates in this case.
As the BHPS does not differentiate between the different types of activities, the value of volunteering must be estimated using an average wage of the three activity types1 and the volume data. The results are shown in Table 7.
|Year||Total Hours (billions of hours)||Frequent Hours (billions of hours)||Total Value (£billions)||Frequent Value (£billions)|
A comparison of the BHPS estimates with the other three options is shown in Figure 4.
The BHPS estimates are notably lower than that shown by the three options using the Citizenship Survey. This analysis demonstrates how two different data sources could produce different results and shows the importance of using reliable data sources. As mentioned in the first section, ONS welcomes comments on the different data sources used and seek suggestions for further / alternative data sources.
Clerical and Secretarial, Personal and Protective and Professional.
The estimated number of volunteers varies considerably, depending on the survey used, with different surveys asking different questions at different times of the year. The primary source of information for the estimates in the HHSA is an unpaid work module within the Citizenship Survey.
The Citizenship survey only has data for the following years: 2001, 2003, 2005, 2007/8, 2008/9, 2009/10, and 2010/11, so it does not present full estimates. The survey covers only England and Wales and was run by the Home Office until 2006. The final four waves of the survey were run by the Department for Communities and Local Government. The survey states that about 40% of the survey’s sample formally volunteers at least once a year, and 25% at least once a month. This is higher than proportions from the original data source (Omnibus Survey 2001) and other comparable surveys such as the British Household Panel Survey (BHPS). The possible data sources have been examined and with the input of experts, it has been decided that due to its survey design, the citizenship survey has created the most accurate estimates for the proportion of individuals that volunteer. This is because the questions on the Citizenship and Community Life Surveys go into more detail and have more information about voluntary activity than one question would, as on the BHPS.
The Citizenship Survey has a nationally representative sample of 10,000 adults (16 and over) and an ethnic minority boost sample of 5,000 adults.
This survey covers only England and is run by the Cabinet Office. It is the successor to the Citizenship Survey and covers many of the same questions and, therefore, variables. The survey was carried out for the year 2012/13 and is ongoing. The proportion of the sample that volunteer has slightly increased in 2012/13 compared with the results found within the Citizenship Survey. This may be due to a change in methodology or as a result of the ‘Olympic’1 effect. The Community Life Survey has a representative sample (of England) of 6,600 interviews that take place over three quarters, of 2012/13.
The survey has published data from two different waves, 2009 and 2011. The Understanding Society Survey is currently being updated for 2013. Only the second wave (2011) contains a full module on volunteering. Understanding Society is a longitudinal survey that captures information annually about the social and economic circumstances and attitudes of people living in 40,000 UK households. It is run by the University of Essex and is UK wide.
Understanding Society has a total of 38,271 households– 26,035 in the General Population sample, 3,959 in the Ethnic Minority Boost sample and 8,277 previously in the BHPS sample. The samples are representative of the UK population.
The most comparable data source to the Citizenship Survey in terms of coverage is the BHPS. This survey was the predecessor to Understanding Society Survey and, biennially, ran a question on the frequency of volunteering. The proportion of those frequently volunteering is much lower than in the Citizenship Survey and there is no information about the type of activity taking place. This paper compares the results for years 2000, 2002, 2004, 2006 and 2008 with the results gained from the Citizenship Survey. It was also run by the University of Essex and was UK wide. The BHPS originally had a sample of 8167 households. All members of the households were sampled and ‘boost’ samples were collected from Wales and Scotland. The samples were representative of the UK population.
The Olympic effect can be characterised as the uptake of volunteering after the use of volunteers during the Olympic and Paralympics Games.
In 2011, the International Labour Office in Geneva published a manual on the measurement of volunteer work. The purpose of the manual was to create ‘a guide for countries in generating systematic and comparable data on volunteer work by means of regular supplements to labour force or other household surveys’. The target of this guide was to produce comparable data that could be used to help produce cross-national outputs that would be easy to compare. This would help to fulfil the recommendations made by the United Nations Secretary General’s International Year of Volunteers report that aims to ‘establish the economic value of volunteering’.
The manual discusses a number of definitions and the use of different variables and measurements in measuring volunteering.
Although data collection is not covered extensively in the manual, the Labour Force Survey comes closest to meeting the recommended rationale and strategy for measuring volunteer work. This is due to the frequency of the Labour Force Surveys, the sample sizes and the fact that they facilitate an accurate valuation of the work itself. The UK has a number of current and recently discontinued household surveys that collected informative data on volunteering.
The manual outlines an interesting ‘conservative’ estimate that ‘the value of contribution of time, i.e., volunteer work, is approximately double the value of contribution of money’. This highlights the need for comparable cross-national information on volunteering becoming readily available, as volunteering is ‘an enormous renewable resource for social problem-solving’.
Targeted variables are discussed within the manual. Having similar, if not identical variables, is an important part of making cross-national economic values of volunteering comparable. The Manual pinpoints three key target variables: number of volunteers, number of hours volunteered and the occupation (type of work performed).
The main objective of the manual is to provide a basis for estimating the economic value of volunteering. To ensure international comparability, there should be a common approach by all countries so that all economic values are measured in comparable output units.
The manual recommends using the ‘replacement cost approach’ rather than the ‘opportunity cost approach’. The replacement cost approach is seen as a better method of measuring the economic value of volunteering to the overall economic view point of society, as it measures the value for the recipient of the voluntary effort. The replacement cost approach estimates a value that the recipient would have to pay to hire a person to perform the same job as the volunteer. This is compared to the opportunity cost approach, which measures the economic value of volunteering to the volunteer. More information about the two methods can be found in the manual.
In 2013, a paper was published by the Department for Work and Pensions and the Cabinet Office by Fujiwara, Oroyemi and McKinnon on ‘Estimating the Value of Volunteering using Subjective Well-being Data’.
The paper uses the British Household Panel Survey and analyses four waves of the survey; 2000, 2002, 2004 and 2008, for people aged 16 years and above.
The paper is unique in the UK because it develops a Well-being Valuation (WV) approach. The WV approach is being developed as an alternative to the opportunity cost approach and the replacement cost approach, which are currently used in much of the literature surrounding the valuation of non-market costs and benefits.
This approach estimates the increase in self reported well-being associated with a particular good or a service and then calculates the equivalent amount of money necessary to give the same boost to well-being.
To summarise, the paper’s key methodological points are that volunteering and well-being are positively correlated and that the WV approach has a number of advantages over traditional preference-based valuation methods for the topic of volunteering.
As the Citizenship Survey covers only England and Wales and the Community Life Survey only covers England, it was important to gauge if the proportions were appropriate for use in Northern Ireland and Scotland before applying the UK population figures. The British Household Panel Survey (BHPS) gives the proportion of those that formally volunteer biennially until 2008. This covers the whole of the UK and can be split into each separate country. This shows us if the difference in the results between England and Wales and the whole of the UK are statistically significant. If they are not, the use of the England and Wales proportions (from the Citizenship survey) as a proxy for the whole of the UK is an acceptable methodology. Table A1 provides the proportions.
|England and Wales||10.90%||10.60%||9.90%||9.60%||12.20%|
To analyse whether the difference between the proportions in the UK and in England and Wales is not statistically significant, a linear regression was performed separately for both the proportion of the sample that frequently volunteers and all of those that volunteer.
The proportion of people who volunteered in England and Wales was found to be a good approximation of the proportion of people who volunteered in the whole UK and, therefore, data from the Citizenship Survey can be used.
|Year||Professional||Clerical||Personal||Personal and Professional||Personal and Clerical||Clerical and Professional||Personal, Professional and Clerical|
Stiglitz, J. Sen, A. Fitoussi, J.P. (2009). Commission on the Measurement of Economic Performance and Social Progress.
Holloway, Short and Tamplin (2002). Household Satellite Account (Experimental) Methodology. Office for National Statistics.
Office for National Statistics (2002). Complete Household Satellite Accounts.
Fujiwara et al (2013). Estimating the value of volunteering using subjective wellbeing data.
International Labour Office (2011). Manual on the measurement of volunteer work.
Salamon, L M et al (2011). Measuring the Economic Value of Volunteer Work Globally: Concepts, Estimates, and a Roadmap to the Future. Annals of Public and Cooperative Economics, Vol. 82, Issue 3, pp. 217-252, 2011
Gaskin, K. Dobson, B. (1997). The Economic Equation of Volunteering. Joseph Rowntree Foundation.
Details of the policy governing the release of new data are available by visiting www.statisticsauthority.gov.uk/assessment/code-of-practice/index.html or from the Media Relations Office email: firstname.lastname@example.org