In 2022, State of Life conducted a research study to evaluate the wellbeing impact of Tearfund’s Church and Community Transformation (CCT) process across Rwanda, Sierra Leone, Tanzania and Zimbabwe.
Here, State of Life’s Rose Fawcett explains how the evaluation was conducted, with a focus on the adaptation of the UK derived Wellbeing-adjusted Life Year (WELLBY) to estimate social and welfare benefits of the programme.
What is the WELLBY?
WELLBY, short for ‘Wellbeing-adjusted Life Year’ is a way to consistently measure and value improvements in wellbeing, first introduced in 2020.
The UK Treasury’s 2021 Wellbeing Supplementary Guidance defines a WELLBY as: a change in life satisfaction of one point on a scale of 0-10, per person per year. It recommends a value of £13,000 per WELLBY, based on two valuation methods (detailed below).
At State of Life we have been using this method for years, and are named advisors in the aforementioned guidance.
How is the WELLBY calculated?
The two valuation methods used to inform the recommended value of £13,000 rely on evidence from UK data, as outlined in Annex 2 of the Wellbeing Supplementary Guidance.
- Using the relationship between health and life satisfaction
Approach 1 takes the Green Book value of a Quality Adjusted Life Year (QALY) and converts this to a WELLBY. A QALY is the measure used by the UK’s National Health Service (NHS) and NICE to assess the cost-effectiveness of drugs and treatments. . A QALY represents the value of one more year of high quality life (Brazier et al. 2016). Based on various research using UK data (outlined in Annex 2) it is assumed one QALY is associated with a 7-point change in life satisfaction, obtaining the value £10,000.
- Using the relationship between income and life satisfaction
Approach 2 is based on analysis of the relationship between income and life satisfaction. Fujiwara 2021 estimates this relationship using data collected in the UK in October 2019. This estimated relationship is then applied to the average earnings in the UK in 2019, using median income, to obtain the value of £16,000.
Is the WELLBY specific to the UK?
The UK is leading the way in wellbeing valuation.
The New Zealand Treasury has adopted the approach, and suggests upper and lower monetary estimates of a WELLBY, which builds on the UK methodology.
As far as we are aware there is no equivalent research in the African countries we worked in, or other developing countries, that estimates a WELLBY value using national data.
So we asked ourselves ‘how appropriate is the WELLBY value in other contexts?’. The answer seemed to be: it could be, if we converted the value using an appropriate method.
Firstly, due to the lack of a WELLBY value in other countries, we must assume the relationships between health, income and life satisfaction found in the UK are also held in Rwanda, Sierra Leone, Tanzania and Zimbabwe. i.e. positive relationships such that, as health and income increases, so does life satisfaction. This assumption might not hold in some cases, for example extreme wealth, or over the long term. However, for now it is considered a fair assumption, and noted that testing these assumptions could be another interesting research topic.
Secondly, we can consider how absolute values of health, income and life satisfaction vary across the countries.
In the UK:
The aggregate averages across Rwanda, Tanzania, Sierra Leone and Zimbabwe are:
- Life expectancy = 61
- GDP = $1,044
- Average life satisfaction = 5.6 (as measured by Tearfund/State of Life)
Why not convert using exchange rates?
It’s clear absolute values of health, income, and life satisfaction are quite different. An exchange rate conversion of £13,000 into local currency would likely overstate wellbeing impacts and assign too high a value for a WELLBY. We therefore used median income to scale the WELLBY figure proportionally, to obtain a figure appropriate for the four African countries.
Why median income?
We want to reflect typical individuals. Although reliable sources of mean income are available, median income conveys far better the material wellbeing of the typical individual in a country because there will always be extreme wealth that skews the mean.
Unfortunately, there is no one clear source of median income data across countries. The Centre for Global Development’s calculation of median income points out its absence in the World Bank’s global poverty database. However, it was the ratio rather than the amount that was important for our calculation. So the fact that our various data sources use different methods, years and parameters – such as daily rather than annual – didn’t stop us.
Our method of converting the WELLBY value
We calculated the ratio between median personal income in the UK and other countries (see table 3 and appendix 4).
We then took the average between Rwanda, Tanzania, Sierra Leone and Zimbabwe, at 0.0577. From the three sources, Sierra Leone had the lowest average ratio at 0.0514 and Zimbabwe had the highest average ratio at 0.0682.
This ratio is then used to convert £13,000 into a WELLBY value (for moving one point on the life satisfaction scale, for one year) that is appropriate in the four countries: £750.
It is then converted to US$1,033 – a currency more often used for international comparisons – using yearly average exchange rates to year end 2021, and rounded.
What about our assumptions on the relationship between income, health and life satisfaction?
We’re relying on the assumptions made when calculating the WELLBY in the UK context; about the relationship between health and life satisfaction (approach 1) and the relationship between income and life satisfaction (approach 2).
Replicating this research would require large-scale studies of nationally representative data sets, which included all the relevant variables. Whilst costly, this would present huge potential for research and understanding of wellbeing in such countries. We’d love to speak to you if you’re interested in doing this.
What about the appropriateness of the UK’s life satisfaction measure in Africa?
The WELLBY relies on one of the UK’s ONS four personal wellbeing questions, which have been thoroughly tested. Was this an appropriate measure to use in (sometimes remote) African communities? Our survey was translated into local languages, so there would inevitably be some nuance in how it was interpreted. We concluded it was the best measure available.
Using regression analysis, our research showed that various demographic factors influenced our life satisfaction outcome in the way we would expect, given other wellbeing research studies. For instance, having better education and being in paid work has a positive association with life satisfaction, while having a disability or living in relative poverty has a negative association (appendix 6). We were reassured that life satisfaction was a sound measure in this context.
Measuring wellbeing in this context was challenging and thought-provoking. We had to make some calls that strayed from what had been tried and tested before – because no one else had done it.
We’re pleased to say this pioneering study resulted in some compelling results. We think the findings speak for themselves and look forward to seeing the evidence base grow.