Strategic Integrated Impact Assessment for the 2026-27 Scottish Budget: process evaluation of pilot activity

Findings of the Process Evaluation

Based on the evidence collected, this section summarises the main findings.

Is it feasible to integrate five legal duties into one assessment?

The main findings are as follows:

• There was a broad consensus amongst officials that integrating the legal duties in a single overarching analysis is both feasible and valuable.
• However, it remains important to acknowledge there are challenges in ensuring balance when assessing impacts of different duties, especially where evidence for some duties is limited, or not readily available.

Feasibility of considering impacts together

The SIIA brought together assessments for five sets of legal duties in a single ‘integrated’ assessment. Most officials agreed that bringing multiple statutory duties into one integrated process was preferable to running separate processes and that there is significant overlap in the information required across duties. A single process rather than multiple separate returns of information, supported a more joined-up view and reduced duplication. However it also brings some complexity, and it is important to establish roles, responsibilities, and routes to securing specialist input in advance. Overall there was a view that the request for all the impacts taken as a whole was very complex and detailed. Navigation around the requests was therefore important, and senior review and clearance could be simplified.

One of the risks identified in taking an integrated approach was the potential for ‘diluting’ analysis on specific impacts or characteristics. It was also noted that evidence appeared stronger for some types of impact than others. However, there was also a view that the integrated approach was useful because it is cross-government. This means that by taking together a number of different impacts, it draws together the big picture and can help highlight trade-offs, for example between the different duties. Additionally, it was noted that the budget has had an integrated approach for a number of years with the difference that this year, five duties were being integrated rather than two. Officials also thought the integrated approach helped stakeholders understand the breadth of work.

Views on integrated approach of the SIIA publication

Stakeholders supported the principle of integrating multiple statutory duties into a single assessment, noting that it reduces the effort needed to otherwise navigate multiple separate reports, whilst helping bring equality and rights considerations together. However, they also flagged a risk that analysis of some duties could become more descriptive than analytical. This could lead to more of the analysis focusing on the challenges facing the government, rather than on how decisions respond to these challenges.

Together Scotland said that they supported an integrated approach to Impact Assessments where appropriate, and indeed this reflects their case study approach in their recent State of Children’s Rights Report 2026^[23]. The Equality and Human Rights Commission (EHRC) was not opposed to an integrated approach but commented that, in its view, the assessment of impacts against the first two of the three ‘needs’ of the Public Sector Equality Duty (to have due regard to the needs relating to non-discrimination, and to equality of opportunity) was only covered at a high level. Furthermore, EHRC also commented that, in its view, the assessment of impacts against the third ‘need’ of the Public Sector Equality Duty (to have due regard to the need to ‘foster good relations between persons who share a relevant protected characteristic and persons who do not share it’) needed to be improved.

Is budget tagging feasible for the Scottish Government?

The main findings are as follows:

• Budget tagging is feasible but it has been resource intensive to build up as it requires detailed line-by-line assessment of the impact of budget spend on different criteria and an assessment of the evidence underpinning the tagging score. This needs to be replicated across a large number of spending lines. A moderation process across all lines needs to be undertaken to ensure consistency.
• Further work is needed to ensure the consistency and robustness of scores, as well as clarity on what the scores mean. This could be driven by more specific guidance and communication, and it was suggested there was an opportunity to involve external stakeholders.
• Generally tagging has been seen as an improvement which could help decision-making by taking a more strategic view.

Budget impacts database tagging

At a basic level, the pilot showed that budget impacts tagging is a feasible exercise. For example, a spreadsheet of 84 budget lines, was produced and published. This included tags (or scores) against each line in relation to eight equality characteristics, socio-economic impacts, islands impacts and consumer impacts – meaning that in this pilot, over 900 budget tags/scores were estimated by Scottish Government officials.

Officials broadly understood the scoring approach but highlighted that judgement-based fields were open to interpretation, creating the risk of variation across teams and portfolios. Definitions such as “exceptional” were often cited as requiring clearer explanation and more concrete examples to support consistent application.

Some officials also described a tendency for written contributions to sometimes provide general policy descriptions rather than directly addressing the contributions that budgets were expected to make, particularly where budget lines reflected continuing programmes or, for example, where reductions were being considered. Significant amounts of editorial work were required by those co-ordinating responses within portfolios. This included shaping content into the required format and aiming to strengthen the link between evidence, rationale and score.

Some officials also had mixed views on the tagging. They pointed out that the quality of evidence was limited in places, or did not provide all the evidence asked for. However, in the final SIIA report it was thought that evidence from other sources compensated for this somewhat. Overall it was suggested that there is a variable degree of depth and robustness of supporting evidence on impacts, with it being stronger where it could draw from an established evidence base.

Stakeholders generally thought that the tagging pilots were promising and worthwhile to develop. There were some areas for improvement, however that would enhance credibility.

To improve the tagging process, respondents:

• Generally felt better guidance and support could improve consistency and robustness. However, there was a belief that tagging could become more straightforward in future years now that a sample had been done, and there was a degree of familiarity.
• Suggested there is a need to be more explicit about what is being scored (for example whether assessing overall programme impact or the impact of marginal budget changes), and to strengthen consistency in how high-impact or ‘exceptional’ tags are applied.
• Suggested building a clearer feedback loop so evidence gaps identified through tagging lead to targeted evidence improvement with analysts and portfolios over time. This could include capturing and summarising reasons for ‘unknown impact’ assessments, for example identifying if there is a lack of evidence of long-term policy effects, which in turn could support proportionate follow-up and advice.
• Suggested maintaining or strengthening an ‘independent assurance’ function, where colleagues outside the immediate policy area can review and strengthen confidence in scores.

Child poverty tagging

A different model of budget tagging was used in relation to the government priority of eradicating child poverty. Rather than asking policy areas to submit tags and evidence as with the tagging of the budget impacts database, the child poverty tagging was more of a top-down approach driven by the Exchequer Analysis team, and built on extensive work already developed by those working on child poverty policies and programmes. Information was collected on budget programmes, including for example whether the programme was expected to have a direct or indirect impact on child poverty, whether it was a ‘primary driver’ of tackling child poverty (for example by generating income from employment or reducing the cost of living), what groups of the population it supported, and the quality of evidence underpinning the analysis.

On the latter point, officials reflected that evidence underpinning tagging was inevitably mixed across portfolios, and that it was unrealistic to expect consistently robust evidence in all cases. At the same time, this variability was seen as a useful diagnostic, helping to surface gaps where evidence is weak or missing.

Officials also emphasised the need for a clearer articulation of who policies are intended to reach (for example households or families, universal or targeted, with children). It was also suggested it should be acceptable, and sometimes necessary, to record ‘we do not know’ where evidence is insufficient.

In relation to the scoring of budget lines officials reported feeling reassured by the structured parameters and the checking by an expert group, which helped anchor judgements back to agreed criteria. They acknowledged some unavoidable subjectivity but felt definitions and steps in the process reduced arbitrariness.

They also noted that different policy areas can differ in how they describe direct impacts, with risks that some policy areas may overstate impacts or, alternatively, not fully recognise indirect contributions. Targeted follow-up from the Exchequer/child poverty team where there is strong disagreement was suggested as a pragmatic strengthening approach, rather than attempting to centrally reassess everything.

Officials valued having the clear parameters of child poverty impacts being low, medium or high but did not favour reducing the range of evidence collected. It was suggested, for example, that whilst analysis could prioritise looking at high-impact lines, retaining information about those budget lines that are seen as no-impact and medium-impact information is important for completeness. For example, it is useful to understand why some lines are assessed as having no impact, and to spot unintended or negative impacts.

Is it feasible to scale up budget tagging to cover more budget lines, or other outcomes?

The main findings are as follows:

• It is feasible to scale up budget tagging by increasing the number of budget lines being tagged for the budget impacts database but it requires significant resource and therefore would have to be undertaken incrementally over time.
• More work could be done on improving the evidence underpinning the tagging of the existing set of budget lines. A balance will need to be struck between expanding the number of budget lines tagged, and work to improve the quality and consistency of evidence underpinning it.
• In principle it would be possible to tag other outcomes or priorities, but the evidence from the child poverty tagging pilot indicates that the level of resource required would be very significant.

As indicated above, to build the budget impacts database^[24], 84 budget lines were each tagged against 11 characteristics or impacts. This meant that over 900 tags, or scores, were estimated. There are generally over 400 Level 4 budget lines. Arithmetically, therefore, tagging the whole budget for these 11 characteristics, at Level 4, would imply 4,000 to 5,000 separate pieces of analysis.

As indicated above, the tagging process for the impacts database was resource intensive, with analytical effort spread across a wide range of policy areas.

However, Scottish Government (SG) officials also indicated that once an initial tagging exercise has been completed, annual updates and revisions could become more manageable.

The child poverty budget tagging model used a more centralised approach, but involved a much more detailed breakdown of budget lines tagged. Prior to the work on the 2026-27 SIIA, there had already been two years’ worth of work by child poverty analysts. Several weeks’ worth of further work was then undertaken by Exchequer analysts to review, refresh and update the analysis for the SIIA. It was therefore also a resource intensive exercise. Developing budget tagging to cover other priorities, targets or outcomes would also be expected to require a significant degree of resource.

There is a risk of spreading analytical resource too thinly. Choices would be required about the balance between strengthening the consistency and depth of evidence of the existing budget tagging pilots, and increasing the number of budget lines to be tagged in the impacts database. There is also a choice about the range of priorities, targets or outcomes which could be tagged against, and the resource and timing to do so.

Can budget tagging add value to the budget process?

The main findings are as follows:

• In principle budget tagging can provide sharp, digestible analysis during the fast moving budget process, for example to identify ‘red flags’ on potential budget allocation decisions.
• Full strategic mapping, as was done in the case of child poverty, also provides opportunities in the future to guide strategic decision making.
• However, as this was the first year in which budget tagging had been deployed, and as it was done on an experimental basis, it is too early to carry out an impact evaluation.
• The value of tagging will depend on the quality of the underpinning analysis, and the greater the use of the analysis in the budget process, the greater the incentive for robust, well evidenced information and analysis.

How was budget tagging and other evidence from the SIIA used in the budget process?

With the aim of being able to provide briefing on potential impacts of budget allocations, information was collected from portfolios in a phased approach starting in August, with information requested again in two other broad phases.

This early collection meant that briefings could be provided to Ministers to support discussion and decision making during the autumn when the Budget and the Spending Review were being considered.

A key focus for the consideration of impact assessments was a cross-Ministerial workshop on Protected Characteristics and Socio-Economic Disadvantage (Equality and Fairer Scotland) and Children’s Rights and Wellbeing, held in November 2025, immediately prior to the UK Budget. This had a particular focus on the government’s priorities, especially eradicating child poverty. Briefing was provided, and this included analysis based on the information collected from portfolios. Evidence from the child poverty budget tagging exercise was also provided in support of the workshop. Stakeholders were very supportive of developing an approach where the evidence being collected was used to support the live budget process.

Did budget tagging add value to the budget process?

Officials highlighted the wide and varied range of considerations which feed into budget decisions, meaning that there are inherent challenges in linking any tagging activity directly to specific budget decisions. However, some expressed a view that budget tagging and other impact evidence was seen as a useful element to complement other evidence and discussions.

The Ministerial workshop, which drew on tagging and other evidence, was perceived positively by officials. For example, it was seen as sending a strong signal that systematic assessment was being undertaken, and that the analysis was supporting more strategic conversations about choices.

If budget tagging were to be continued, then it is expected that there may be more scope for an impact evaluation in relation to the 2027-28 Budget, once the tagging process has established.

Potential to add value to the budget process

In principle, one of the benefits of budget tagging is that it might support better evidenced decision making during the fast moving and complex budget process. Some stakeholders agreed this could be a benefit. However, this pilot was intended as a proof of concept and it is too early to assess impacts of the data produced. However, the evaluation did explore whether those involved saw potential in the approach.

Participants discussed future outputs that might potentially add value. These included making comparisons across lines assessed as having high, medium, and no impact, as well as tracking spend and budget change over time by impact category.

Some stakeholders felt that an up-to-date budget impacts database might be a useful resource to support pre-budget scrutiny work, and there might be greater opportunities to engage with and explore the data.

Officials considered that the child poverty budget tagging pilot would continue to be useful, both for individual learning and for strengthening governance and strategic conversations in a constrained fiscal context. For those newer to the topic, the process was described as straightforward to follow and helped build a clearer understanding of how different policy areas contribute to child poverty. From a governance perspective, colleagues highlighted the value of having a more explicit evidence-based view of what contributes to child poverty to support discussions with ministers about trade-offs and priorities.

Participants described the overall budget tagging analysis as good enough to act as a starting point that might identify ‘red or green flags’ but cautioned against treating the analysis as definitive without sense-checking and context. Quality of tagging was seen as strongest in cases where the criteria and quality assurance process were applied consistently. Budget tagging also worked best where the outputs generated from the data linked to decision-making moments, including where there was ministerial engagement.

A recurring point was that the assessment/tagging is only as strong as the information provided. This emphasised the need to loop back with portfolios to improve data quality and strengthen rationales and evidence over time, and create a ‘virtuous cycle’.

Can budget tagging improve the transparency of the budget process?

The main findings are as follows:

• In principle budget tagging might help provide a clear demonstration of the government’s view of the contribution different budget lines make. However, in this pilot, the tagging data was badged as experimental and this has limited the degree to which the impact on transparency can be demonstrated.
• There has been some uncertainty amongst stakeholders about the meaning of tagging, specifically whether the tags refer to the contribution of the budget lines (which was the intention), or the outcomes experienced in these policy areas. Greater clarity will be needed to improve transparency.
• The child poverty tagging work was described by officials as a practical way to illustrate cross-portfolio connections, to identify gaps and risk points requiring joined-up action, and to support preventative thinking.
• External stakeholders suggested that external challenge, scrutiny and participation on tagging could improve the quality and credibility of tags. Some external stakeholders also suggested that the evidence drawn upon and cited could helpfully include a wider range of external sources, beyond the Scottish Government itself.

In principle, some of the benefits of budget tagging are that:

• it might help provide a clear demonstration of the government’s view of the contribution different budget lines make;
• it could help to more directly link potential impacts to specific budget lines; and
• it could allow a cumulative, cross government ‘big picture’ to be built up, for example of impacts of budget decisions on one characteristic.

However, in this pilot, the tagging data was badged as experimental and this has limited the degree to which the impact on transparency can be demonstrated in relation to this year’s budget. There have been a number of caveats and limitations highlighted, which will require to be addressed.

One of the issues raised related to clarity and potential risks to credibility in the tagging. A central concern was that the tagging and labels (for example neutral, positive, exceptional) were not sufficiently clear to an external reader, and could be misinterpreted as statements about achieved impact (or outcome) rather than relative contribution or materiality of the budget line.

It was noted that the wording of the tagging labels shaped how the content was understood, and that the spreadsheet could give a different overall impression from the narrative in the published document. Stakeholders also raised concerns about apparent inconsistencies, including examples where ‘neutral’ ratings appeared in policy areas they viewed as inherently gendered, such as social care and unpaid care.

One example of this ambiguity was highlighted in the Budget scrutiny session of the Equalities, Human Rights and Civil Justice Committee^[25] (3 February 2026). In this case, one member of the committee highlighted an exceptional scoring in the budget spreadsheet, relating to Violence against women and girls, whilst another highlighted the exceptional score provided for Employability^[26]. In both cases there was uncertainty whether the score related to the importance of the budget measure, or the quality of service and outcomes achieved. The Cabinet Secretary for Finance and Local Government stated that the Scottish Government would reflect on its language with a view to being clearer “that we are talking about budget lines rather than outcomes for everyone”.

Stakeholders were generally positive about the child poverty tagging. As indicated above, for example, it was described as a practical way to illustrate cross-portfolio connections. It was also viewed as moving closer towards a rights-based framing, including a stronger focus on minimum core or foundational expectations and where rights may not be met. Stakeholders encouraged continuing the approach to improve consistency and usefulness over time, whilst being clear about caveats and limitations.

The Scottish Parliament Information Centre, whilst also noting a number of areas for improvement, said that “the pilot approaches on budget tagging and intersectional analysis are both useful and promising”^[27].

External stakeholders suggested work could be done on quality assurance and external challenge: It was suggested a stronger quality assurance or challenge step could test coherence between narrative evidence and spreadsheet ratings. They indicated that external scrutiny could play a constructive role and were open to support with quality assurance and improvement of the processes.

How feasible and valuable was piloting an intersectional gender impact assessment?

The main findings are as follows:

• The intersectional gender impact assessment pilot was broadly welcomed, and considered to be pitched at a high level. The format used required some resource from portfolios and in Exchequer analysis.

Following discussions with the National Advisory Council on Women and Girls (NACWG), a pilot exercise considered the impacts of the budget on minority ethnic women. This drew on research undertaken by the Scottish Government in 2024^[28].

There were limited comments on the intersectional pilot. Officials noted that whilst it was not the most time-consuming of the requests for information to portfolios, it did add to the workload. Stakeholders who commented were pleased to see the inclusion of the analysis but some also noted that it was at a high level and therefore limited. However, on the positive side, it was thought it did ask questions, identified areas where evaluation evidence could be improved, and provided a helpful place to build on.

How did readers find the overall readability and accessibility of the document?

The main findings are as follows:

• Readers generally considered the SIIA document had improved accessibility through a clearer structure.
• There is potential to explore other means of communications to increase engagement and this might include more visuals and, potentially, interactive dashboards.
• Some stakeholders indicated that it would be valuable to see a clearer link between the evidence on impacts and the decisions made in the budget process, and similarly would like to see a more explicit explanation of the trade-offs made in the budget process.

Readers (internal and external) generally described the SIIA as clearer and easier to navigate than previous formats, with a structure that supported readability and reduced reliance on broad, generic statements. The narrative style was seen as engaging and, in places, more relatable for external audiences, including in how it presented intersectional issues.

Stakeholders valued the time and effort invested in producing the document and recognised it as a meaningful marker of mainstreaming equalities and rights considerations within budget processes. For example, whilst highlighting a number of areas for potential improvement, Together Scotland said that overall the SIIA “feels clear, well-structured and broadly accessible”^[29].

The Scottish Women’s Budget Group (SWBG)^[30] discussed the SIIA with the Women’s Economic Empowerment Group. SWBG reported that although the SIIA tried to highlight how major budget lines aimed to support the achievement of the Scottish Government’s priorities, “it was not clear from the published information what impact these different lines had on the priorities, or how the information had been used to inform budget decisions”. The Fraser of Allander Institute also indicated it would be helpful to show “more evidence of the decision-making process to demonstrate more convincingly that the evidence cited in the assessment has actually informed the decisions”^[31]. The EHRC also suggested that the SIIA focused “predominantly on positive or neutral impacts and does not critically consider potential negative impacts of budget decisions”, and noted that in its view the SIIA did not “provide a coherent cumulative impact assessment”.

In order to strengthen transparency, some stakeholders also called for a greater focus on the trade-offs between different budget choices, including highlighting potential negative impacts. For example, whilst the Fraser of Allander Institute acknowledged, and welcomed, that the SIIA contained more evidence than previous assessments around the trade-offs (eg section 3.2), they also suggested this should be further developed. The Women’s Economic Empowerment Group (convened by the SWBG) were similarly “not able to understand what these trade-offs were”.

Other stakeholders found the document still to be a lot to get through, but also recognised that it was possible to go into more detail in a number of areas. Some said they would like to see more graphics and visuals to bring out key issues. Other options mentioned include use of more interactive dashboards.

Views on the easy read version differed slightly. One participant considered it broadly consistent with what they would expect from accessibility standards, whilst another found the layout off-putting, particularly the use of pictures for every sentence, despite finding the text logical and clear when followed carefully.

What was the impact of staggering the timing of the publication?

The main findings are as follows:

• Whilst there was no strong objection to publication of the SIIA a few days after the budget, there was some mixed feedback.
• Benefits identified by officials were the potential for higher quality analysis and closer integration of the assessments with budget decisions, especially any decisions made late in the process.
• Drawbacks included some potential knock-on effects on parliamentary scrutiny and not being able to include the SIIA in immediate briefing. However, some external stakeholders described the short interval between publications as useful.
• Alternative options were suggested, for example, more than one publication per year.

In previous years, the Equality and Fairer Scotland Budget Statement (EFSBS) was published on the same day as the draft budget publication. In this case the SIIA was published six days after the budget.

Stakeholders reported that publishing the SIIA six days after the budget created both benefits and drawbacks. It allowed more time for detailed reading after ‘budget week’, but reduced the extent to which the SIIA could feed into time-critical parliamentary scrutiny outputs and committee briefings.

Participants suggested that greater transparency about the stages and constraints in producing the assessment would help external users understand why earlier publication is difficult and could produce a less robust product, and to explore what could realistically be produced earlier in the budget cycle, noting examples elsewhere (including Wales) where a staged approach is used^[32].

Stakeholders discussed the option of publishing a pre-budget document (for example in the summer). It was thought this might support parliamentary scrutiny, help guide committees, allow some scrutiny of choices, and allow some testing of the data.

Officials highlighted substantial benefits of publication a few days after the budget, and suggested it would support greater accuracy, and sharper more focused analysis of budget changes, and so helped delivery of a higher quality document that is more closely aligned to actual budget decisions.

How did the overall process work, what resources were involved, and what are the opportunities?

The main findings are as follows:

• Overall, the process was found to be ‘manageable but time-consuming’ although it was acknowledged that efficiency gains could be achieved in the future due to greater familiarity with the process.
• Many of the officials involved in the SIIA process deal with competing demands during the budget process and highlighted the need for a proportionate approach.
• The integration of duties, the central support provided and the use of concrete examples as guidance were highlighted as positive elements of the process.
• Areas where improvements were suggested included: clearer assignment of roles and responsibilities at the start, early clarity on the purpose of the analysis and how it would be used in the process and clearer guidance.

As might be expected with a novel process, the time and resource commitment to producing the information used in the SIIA involved a substantial time commitment, often during busy periods with a number of respondents describing the process as ‘manageable but time-consuming’. Nevertheless, despite its complexities, most officials felt the overall process, whilst initially daunting, was clearly laid out and understood. There was a general view though that the three phase approach (asking for information in summer, autumn and winter) could be improved, and refocused or rationalised. Some of the most commonly reported barriers were: ensuring the most appropriate people were involved; coordination issues within portfolios (given the wide range of information being requested); and evidence requirements being unclear or hard to meet.

Each portfolio usually identified an official to act as the central co-ordinator, translating the central request into portfolio actions, identifying the right contributors, and consolidating inputs. Delivery often relied on informal networks and relationships to secure timely contributions, particularly where information needed to be drawn from multiple policy areas.

Approaches to coordination varied but commonly included small multi-disciplinary working arrangements (often involving finance and analysts) and a staged approach to gathering content. Where portfolios had established internal processes and clear ownership, completion was described as more manageable. Where ownership was diffuse, leads reported spending substantial time chasing inputs and resolving gaps.

A recurring theme was early uncertainty about the purpose and ‘end use’ of the information being collected, particularly how portfolio inputs would feed into decision-making, briefings and any published outputs. Several leads felt that a clearer initial narrative would have helped them brief contributors and sustain engagement across phases.

Some feedback on the quality and consistency of portfolio returns was provided centrally as part of the process. This was seen as an area where further work would be helpful to improve the value of that feedback so it was provided at the right time, in a suitable format, and in a proportionate manner.

Officials also noted that the integrated nature of the assessment became clearer during delivery rather than at the outset. This suggests scope to strengthen early framing, including what ‘good’ looks like for the return and how different elements connect across duties.

Interviewees emphasised that SIIA delivery sat alongside significant parallel budget activity. Tight turnaround periods were frequently described as challenging once internal coordination, drafting, quality assurance and clearance were accounted for. Several leads highlighted that the real constraint was not only completing fields, but securing timely engagement from colleagues under pressure, especially finance teams.

Prior experience was a clear enabler. Leads who had supported previous assessment processes, or who had stable teams, described greater confidence in interpreting the ask and knowing who to contact for specialist input. Newer leads or those experiencing staff turnover described a steeper learning curve and greater reliance on central support.