Fair Start Scotland: economic evaluation

Findings from an independent economic evaluation of the delivery and outcomes of the Fair Start Scotland employment service. The evaluation relates to the first three years of the service, from April 2018 to March 2021.

Inclusive growth and wellbeing

Inclusive growth

As mentioned, the Scottish Government defines inclusive growth as growth that combines increased prosperity with greater equality, creates opportunities for all, and distributes the benefits of increased prosperity fairly. The aims and design of Fair Start Scotland inherently led to some aspects of inclusive growth by increasing the total economic activity of low-income groups.

However, it might be the case that Fair Start Scotland may have benefited certain groups over others. Therefore, to understand the extent to which Fair Start Scotland led to broader inclusive growth, the analysis will be disaggregated across different characteristics. The results will show the performance of the policy under different dimensions of inclusive growth.

A limitation of inclusive growth analysis is that cost data is not disaggregated across different categories, limiting the scope for accurate disaggregation of results. The exception is Lots (geographic area) and participant groups (Core, Advanced, or Intense), for which disaggregated cost data is available (cost methodology for Lots and participant groups is detailed in Annex B). It is possible to look at the cost-benefit analysis for different groups assuming that the cost is constant across participants. This option is better suited for some categorisations than others. For example, assuming a constant cost across gender is reasonable, whereas the same cannot be said about costs across disability status.

Therefore, for the inclusive growth measures, the analysis will be disaggregated across Lot, participant group, and gender, the last of which assumes a constant cost. A detailed breakdown of the cost-benefit analysis across these three characteristics is provided in Annex D.

Starting with gender, Table 14 shows the results of the cost-benefit analysis separately for men and women. Since the analysis assumes a constant cost across gender, the results show only differences in outcomes. While the BCR for both is very similar, men achieved both higher costs and higher benefits. This is because 64% of those who achieved a job start are men. Indeed, looking at all participant data (not just those who achieved a job start), this mirrors the percentage of men who joined the service. In the Scottish population, men constitute 54% of those who are unemployed.[35] This means that some of the gaps can be attributed to the characteristics of the unemployed population, but there is still evidence that Fair Start Scotland worked disproportionately with men. However, it is important to note that this gender gap is declining, with women constituting 40% of participants in the 2020 cohort.

Table 14. Results of the cost-benefit analysis disaggregated across gender
Gender Women Men
Total benefits £54.7m £112.5m
Total costs £29.6m £53.2m
Total benefits (QALYs, redistributive effect) £102.0m £197.3m
Financial BCR 1.9 2.1
Total BCR 3.4 3.7

Source: Analysis of management information, Wave 3 survey data, cost data, and post-2021 forecasts.

Table 15 shows the BCR for different geographic locations. There is quite a high variation in the BCR across Lot, going as high as 2.2 and as low as 1.3 (for financial BCRs). The best performances are observed in Tayside and East, and the worst performances are observed in Forth Valley and Highlands and Islands.

Table 15. Results of the cost-benefit analysis disaggregated by Lot
Lot Financial BCR Total BCR Public Finance BCR
East 2.2 3.9 1.7
Forth Valley 1.3 2.4 0.9
Glasgow 2.1 3.7 1.7
Highlands and Islands 1.5 2.6 1.1
Lanarkshire 2.0 3.6 1.5
North East 1.7 3.0 1.3
South West 2.1 3.9 1.7
Tayside 2.2 3.9 1.7
West 1.7 3.1 1.3

Source: Analysis of management information, Wave 3 survey data, cost data, and post-2021 forecasts

The Lots with the lowest BCRs also had some of the lowest percentages of recorded job starts and the highest percentage of Intense participants. This is shown in Table 16. Having a lower percentage of job starts means that the non-supplier costs are divided across a smaller pool of participants, meaning a lower BCR is expected. Similarly, having a higher percentage of Intense participants implies higher costs since providers are paid higher for Intense participants than for other groups.

Table 16. Share of participant group and job starts across Lot
Share of participant group across Lot Core Advanced Intense Job starts
East 59% 32% 9% 32%
Forth Valley 38% 21% 42% 20%
Glasgow 65% 21% 13% 33%
Highlands and Islands 51% 17% 32% 27%
Lanarkshire 59% 23% 18% 34%
North East 54% 18% 27% 26%
South West 69% 23% 8% 33%
Tayside 61% 20% 19% 42%
West 49% 35% 16% 31%

Source: Analysis of management information.

To capture even more clearly the discrepancy in outcomes between Core groups on the one hand and Advanced and Intense groups on the other, Table 17 provides a breakdown of BCR by participant group. As expected, Core participants had a much higher BCR than Advanced or Intense participants. Again, the same conclusion can be reached; since the providers were paid less for Core participants, the BCRs for the Core group are significantly higher than those of the other two.

Table 17. Results of the cost-benefit analysis disaggregated by participant group
Group Financial BCR Total BCR Public Finance BCR
Core 2.5 4.4 2.0
Advanced 1.7 3.1 1.3
Intense 1.3 2.5 1.0

Source: Analysis of management information, Wave 3 survey data, cost data, and post-2021 forecasts

The main implication of this analysis is that discrepancies in the performance of different geographic locations is driven by participant groups; areas with a higher concentration of Core participants achieved higher value for money than areas with a higher concentration of Advanced or Intense participants. However, from an inclusive growth standpoint, the areas with lower BCRs could better capture the groups with more barriers to work that Fair Start Scotland set out to help.

There are several reasons that may explain the discrepancy between the Lots. Each of the Lots had a different delivery model in response to varying estimates of local need. For example, Forth Valley was entirely local authority led and their delivery model was focused on engaging participants in need of more intense support. Overall, their performance has been significantly poorer than the other Lots in terms of BCR, but they managed to capture by far the largest percentage of Intense group participants.

Additionally, while the majority of referrals are made by Jobcentre Plus, providers are able to generate their own referrals through their social media or other marketing campaigns. These are referred to as Third Party Organisation (TPO) referrals. The majority of TPO referrals are likely for Core participants. The rural Lots performed poorly with respect to TPO referrals and, therefore, had fewer Core participants. Relatedly, rural Lots are likely to incur disproportionately overall costs for delivery of services to all participants due to distance and logistical difficulties.

Finally, providers' interpretation of which categories participants should be placed in may explain some of the differences across Lots. As is evident from the categorisation in Table 1, the groupings are not strictly defined, and participants may fit into more than one. In that case, providers may make varied decisions regarding which groups the participants belong to, creating the differences across Lots. It is unlikely that demographic characteristics drive much of the difference between the areas.

Wider wellbeing impact

One of the key differences between Fair Start Scotland and other employment programmes is the focus on wider wellbeing measures and impact as it is built on the Scottish Government's key values for public services of dignity and respect, fairness and equality, and continuous improvement.

Cost-benefit analysis allows for the inclusion of some aspects of wellbeing, namely the positive impact of having a job on participants' QALYs. This was estimated to be around £61m. It also implicitly includes measures such as the wages of those who achieved job outcomes, the number of hours they worked, and the length of time they spent in employment.

However, there are other impacts to participating in Fair Start Scotland that are not captured by the cost-benefit analysis. For participants who achieved job outcomes, for example, the type of contract is not included, neither is the longevity of each job they keep.

The wider impact on all participants – not just those who achieved a job start – is also not captured by the analysis. This includes labour market outcomes such as increased labour market attachment or increased job searching and applying skills, as well as non-labour market outcomes such as increased confidence or improved mental health.

Quality of jobs achieved

Table 18 shows the average real wages in 2021 pounds and the average hours worked for different Fair Start Scotland participant groups. On average, Fair Start Scotland participants who get a job earn £8.70 per hour, slightly higher than the average real minimum wage across the three years at £8.60. As is evident in Table 18, there are minor differences in wages achieved between participants across cohorts, participant groups, and Lots.

Similarly, when it comes to hours worked, Fair Start Scotland participants work on average 30 hours per week. There is slight variation across groups, but again the difference is minor. Just under half of the participants (45%) work full-time (35 hours per week or more). This may explain lower levels of Intense participants; feedback from disability organisations shows that focusing on the provider payments for jobs that are 16+ hours per week may disincentivise disabled people and those with severe health conditions, especially mental health conditions from joining the service.

The similarities in outcomes between the different participant groups (Core, Advanced, and Intense) indicate that the discrepancy in BCR between areas with different group composition is driven by two factors. The first one is the costs paid to providers for the different groups, as discussed earlier. The second is differences in job start rates, with Core group participants being disproportionately more likely to start a job than Advanced or Intense group participants. However, the job quality that the Advanced and Intense groups achieve is as good as the one that the Core group achieves. This means that once participants with high barriers to entry achieve a job start, they perform just as well as those with low barriers to entry.

Table 18. Average real wages and hours worked across various characteristics
Cohort Real hourly wage Hours worked per week
2018 Cohort £8.40 29
2019 Cohort £8.70 30
2020 Cohort £9.10 30
Group Real hourly wage Hours worked per week
Core £8.70 30
Advanced £8.70 29
Intense £8.50 28
Lot Real hourly wage Hours worked per week
Glasgow £8.60 29
Lanarkshire £8.60 28
Tayside £8.60 31
Forth Valley £8.40 28
East £8.80 31
South West £8.80 30
North East £8.50 27
Highlands and Islands £8.60 28
West £8.80 29
Total £8.70 30

Source: Analysis of management information.

The types of job contracts and employment sectors are not consistently measured across different providers in management information. However, they are captured in surveys. The occupations with the highest concentration of Fair Start Scotland participants are shown in Table 19. A large percentage of participants work in elementary administration and service occupations and administrative occupations. There are slight variations across cohorts; for example, caring personal service occupations became more prominent in 2019 onwards, likely due to the onset of the Covid-19 pandemic.

Table 19. Share of participants in top employment occupations
Job description 2018 Cohort 2019
2020 Cohort
Elementary administration and service occupations 18% 30% 24%
Transport and mobile machine drivers and operative 10% 5% 6%
Customer service occupations 14% 10% 13%
Caring personal service occupations 2% 14% 11%
Administrative occupations 20% 10% 7%
Other 36% 31% 39%

Source: Analysis of Wave 3 survey data.

Table 20 shows the type of job contracts that Fair Start Scotland participants are on. The decline in permanent contracts across years is striking, going from 70% for the 2018 cohort to 43% in the 2020 cohort. Since Table 20 is based on Wave 3 survey data, the increase in temporary work across cohorts is likely due to more recent cohorts having spent less time at work and, therefore, being less likely to find permanent work. It could also be partially due to the onset of the Covid-19 pandemic and the accompanying economic decline.

Table 20. Share of participants in different contract types
Contract type 2018 Cohort 2019 Cohort 2020 Cohort
Permanent contract 70% 52% 43%
Temporary contract 14% 34% 33%
Zero hours contract 6% 9% 10%
Self-employed 9% 4% 8%

Source: Analysis of Wave 3 survey data.

When looking at wage rates, sectors, and contract types, it is evident that, on average, the jobs that Fair Start Scotland participants achieve are at the lower end of the quality and fair work spectrum. The wage rates, for example, are barely above the legal minimum and below the real living wage. Even for the 2018 cohort, Fair Start Scotland participants are still overrepresented in temporary work compared to the general population.[36] This is not surprising given that many of those who joined the service were far from the labour market and found work during an economic recession. While it is important to work towards improving these outcomes, this context needs to be kept in mind.

Another important dimension of job quality is job stability, which can be proxied by the type of contract as well as time spent in employment. As mentioned earlier, only 14% of the 2018 cohort achieved the 52 weeks job outcome. Additionally, many participants recorded several jobs start dates without achieving the 13 weeks outcome. This provides evidence for employment instability and job-hopping.[37]

Whether this is detrimental to the wellbeing of Fair Start Scotland participants, however, is unclear. Findings from the literature on work and wellbeing indicate that having a job is better for a person's health than having no job.[38] Having unstable employment or a low-quality job is associated with worse wellbeing outcomes when compared to stable and good quality jobs. However, it is unclear if this relationship still holds if the alternative is no job at all, which is the case for most Fair Start Scotland participants.[39]

Additionally, it is also unclear to what extent job-hopping indicates bad quality jobs. In some cases, quick changes between jobs could indicate instability or poor working conditions. However, it could also be the case that participants who re-joined the labour market are now able to use their new-found experiences and network to move into better paid or better-quality jobs quickly.

All in all, the quality of jobs that Fair Start Scotland participants have achieved is reasonably good given the barriers to work that many of them face and the deteriorating economic conditions due to the onset of the Covid-19 pandemic midway through the programme. On average, participants earn just above the minimum wage, work 30 hours a week, and are concentrated in services and care occupations. While there is some indication of job instability, especially from 2019 onwards, it is unclear to what extent this results in deteriorating wellbeing for participants, especially given the alternative of unemployment. In the Wave 1 survey, those in work reported higher wellbeing scores than those who were not at work and to the population average.[40]

Other labour market outcomes

A key aspect of employment programmes like Fair Start Scotland is improving the motivation to return to work and, more generally, attachment to the labour market. This is partially captured in the cost-benefit analysis by including the timeframe in which the benefits (having a job) persist for participants who recorded a job start date. However, improvement in labour market attachment and motivation to return to work for participants who did not record a start date is not captured in the analysis.

As discussed earlier, 25% of participants from the 2018 cohort indicated having a job three years after joining the programme. Additionally, 31% of the 2019 cohort indicated having a job two years after they joined the programme. The percentage of people who lost their job and did not find another one is 9% in the 2018 cohort and 6% in the 2019 cohort. Overall, this indicates a good level of labour market attachment.

The surveys capture Fair Start Scotland's effect on motivation to work for those who do not currently have a job. The results are shown in Table 21. Throughout the three survey waves, the majority of respondents said that they wanted to return to work to a great extent. The majority also stated that Fair Start Scotland increased their motivation to find full-time employment. This indicates that the programme positively impacts labour market attachment and motivation to work, even for participants who were unable to find a job.

Table 21. Results on the impact of Fair Start Scotland on motivation to work
Extent to which participants
would like to return to work
Wave 1 survey Wave 2
Wave 3 survey
To a great extent 69% 75% 80%
To some extent 17% 16% 11%
A little or not at all 11% 9% 5%
Extent to which Fair Start impacted
motivation to find full-time work
Wave 1
Wave 2
Wave 3 survey
Increased 65% 63% 61%
No effect or decreased 32% 37% 35%

Source: Wave 1, 2, and 3 survey data.

However, there is some indication that the motivation to return to work declines over time. For example, the share of participants in the 2018 cohort who said they were motivated to return to work to a great extent declined from 69% in the Wave 1 survey (in 2019) to 61% in the Wave 2 survey (in 2020). The perceived impact of Fair Start Scotland support on the 2018 cohort participants' motivation had also fallen, with 65% saying that the support increased their motivation in Wave 1, and 52% in Wave 2.

The same is also true for the 2019 cohort, with the share of participants indicating that they are motivated to return to work to a great extent declining from 75% in the Wave 2 survey to 62% in the Wave 3 survey. Similarly, the perceived impact of Fair Start Scotland support on the 2019 cohort participants' motivation fell from 63% in Wave 2 to 52% in Wave 3 (in 2021).

Fair Start Scotland is also designed to help participants develop a sense of self-efficacy.[42] Self-efficacy can give participants the confidence and ability to search for and find jobs and maintain employment. It also touches on wider wellbeing measures that go beyond labour market outcomes.

In the survey, respondents completed a nine-item measure of the strength of an individual's belief that they have the skills to undertake a range of job search tasks, known as the Job Search Self Efficacy (JSSE) Index. Table 22 shows the results across the three waves for people who do not currently have a job.

Table 22. Results of the Job Search Self Efficacy Index for Fair Start Scotland participants
Confidence in job search activities Wave 1 survey Wave 2 survey Wave 3 survey
Searching for jobs online 71% 72% 71%
Applying for jobs online 66% 66% 68%
Making a good list of all the skills that you have and can be used to find a job 60% 59% 67%
Talking to friends / other contacts to find potential employers who need your skills 60% 53% 63%
Talking to friends / other contacts to discover promising job openings suitable for you 61% 56% 60%
Getting help in order to become familiar with a new job 59% 56% 57%
Completing a good job application and CV 58% 61% 68%
Making the best impression and getting your points across in a job interview 52% 54% 58%
Contacting and persuading potential employers
to consider you for a job
45% 43% 58%

Source: Wave 1, 2, and 3 survey data.

Generally, the majority of participants who did not have a job still indicated having confidence in undertaking job search activities. While this cannot be entirely attributed to participating in Fair Start Scotland, it is likely to be a major contributing factor given the programme's type of support in improving job application and search skills.

Non-labour market outcomes

Fair Start Scotland is rooted in the principles of dignity and respect, and the programme is designed to treat individuals in a way that reflects these values. This manifests in a number of ways that go beyond labour market outcomes to encompass other areas of wellbeing.

As mentioned, one of the most important ways that Fair Start Scotland differs from other employment programmes is its voluntary nature, with participation being unconditional on receiving benefits. This is to ensure that people will not be driven to take part by fear of benefit sanctions. Table 23 illustrates how the service was perceived by participants.

Table 23. Results of participants perception of Fair Start Scotland across the three survey Waves
Non-labour market outcomes Wave 1 survey Wave 2 survey Wave survey
Aware that the service is voluntary 94% 95% 90%
Felt they were treated with dignity and respect 92% 91% 95%
The support took account of their individual needs and circumstances 80% 80% 82%
They felt they had choices about the support they received 80% 81% 83%
Felt the service offered support to improve their general quality of life and wellbeing 78% 81% 84%
Felt they were in control of their progress 79% 80% 83%

Source: Wave 1, 2, and 3 survey data.

In all three surveys, the vast majority of participants were aware that Fair Start Scotland is a voluntary programme. The vast majority also felt that the support that they received aligned with Fair Start Scotland's principles and values of providing a personalised service that treated people with dignity and respect and put wellbeing at the forefront of its aims. Indeed, the perception of Fair Start Scotland across most measures improved throughout the three years.

Fair Start Scotland had an overall positive impact on the wellbeing of its participants, both in terms of improved labour market outcomes as well as their experience of the service. There were some discrepancies in the experience based on certain characteristics. For example, those from minority ethnic backgrounds were less likely to know that the service was voluntary. More generally, across most wellbeing and labour market outcome measures, those who were unemployed for a long time, those with limiting health conditions, and those with no formal qualification were likely to have worse outcomes and perceptions.

Tying these findings to those from the inclusive growth section, it is important that, moving forward, Scottish employment services ensure that their programme benefits those who face barriers to employment in terms of their health and disability status. This does not mean that Fair Start Scotland did not have a positive impact on people with disability or limiting health conditions, but that more could be done to ensure that these services are better tailored to engage and help them.


Email: Stephanie.Phin@gov.scot

Back to top