Foreword
We are at a critical juncture in the structural transformation of the UK economy - and in building a world-leading responsible and thriving innovation ecosystem. This transformation is already having profound societal impacts, not only on access to work, but also on the nature, conditions and quality of this work. The opportunities people have to lead fulfilling working lives are fundamentally intertwined with the ways in which innovation systems are designed and support the fulfilment of individual, firm and regional-level capabilities. These capabilities are conceptually distinct but related and, together, shape the wellbeing and prospects of people and communities across the country.
The Disruption Index (DI) presented here provides the first panoramic overview of the scale and trajectories of this technological transformation and offers a deeper understanding of the innovation ecosystem beneath headline national statistics. This approach not only enables exploration of the likely drivers and trajectories of transformation, but also each region’s readiness for it, helping policymakers to identify regional bottlenecks and strengths, and to access the most impactful points for policy intervention.
Building on the research paper ‘A Disruption Index: the geography of technological transformation across England’ (Rohenkohl, Clarke and Pissarides, 2024), this web-based report integrates other sources with this to elicit new insights and inform key policy areas, showcasing how the DI can be used in practice by policymakers at national and regional levels.
Empowered by these insights, this work invites policymakers to challenge traditional boundaries and assumptions about the environment in which work is created and shaped, distributed and disrupted. With more clarity on the mechanisms through which these changes are taking place, obstacles can be identified and policy steered so that the potential of people and places can be unlocked before deep-seated regional inequalities become entrenched.
Our work on the scale and variation of transformation, with its wide-ranging consequences, and spill-over effects, suggests that a focus on the identification of bottlenecks and broadening access to new opportunities at a regional level is now critical.
Accompanying this report, we have created a Disruption Index dashboard as a visualisation tool to allow policymakers and researchers to compare measures from the index across the country, over time, identify factors and individual indicators that contribute to the differences observed and zoom in on key points. To explore the dashboard, please follow this link.
Supporting good work and inclusive regional innovation ecosystems
This report is directed towards supporting thriving, inclusive regional innovation ecosystems in which good, local work can be created and sustained. Homing in on gaps and inconsistencies, the DI provides a panoramic but critical overview of the ecosystem at local and regional levels and invites bespoke, place-based and context-sensitive policy mixes to address the growing divisions experienced every day, in different ways, across the country. Our report is aimed at national and regional policymakers, but will also be useful for investors and businesses wishing to identify untapped competencies and capabilities.
Our analysis broadly supports the hypothesis that a sharper focus on creating and sustaining and improving pathways to good work within regional innovation ecosystems is key in shaping better outcomes. In particular, creating a pipeline of opportunities for people to learn new skills, develop their capabilities, exercise their agency to innovate and practice good work is required – whatever their role in the innovation ecosystem.
We also propose that indicators of ‘good work’ should begin to serve as measures of - and likely proxies for - responsible, well-functioning innovation ecosystems.
Supporting future work and public data infrastructure
Organised by theme and illustrated with examples, our purpose here is also to demonstrate how the DI can be used and developed in the future. This means flagging – where necessary - the limits of our approach and highlighting some significant evidence gaps. It also means making the case for systematic, ongoing monitoring to understand and shape an increasingly complex innovation system with cascading economic, social and wellbeing impacts that cannot always be anticipated. This case is made more pressing in the age of AI as the traditional processes of innovation are disrupted, additional components of the system are introduced, interactions change and impacts scale - demanding agile, intelligent, collaborative and cross-departmental responses at different tiers of government to new risks and opportunities, as they arise.
These factors point to the need to reframe government roles to build on the DI and create, maintain and govern a new public data infrastructure combined with monitoring and evaluation functions dedicated to tracking readiness, technological transformation and its social, economic and wellbeing impacts, especially on work and wellbeing. This should help surface new insights, help target marked gaps and bottlenecks, model policy options and scenarios, and rebalance and refine resource allocation across the country.
Innovation, like automation, is a multi-dimensional, dynamic and evolving phenomenon - and these features are becoming more pronounced as innovation processes are disrupted, boundaries blurred and traditional assumptions challenged (McCann, 2021). This means that a systems approach is needed to understand and shape the drivers, mechanisms and effects of our 'innovation ecosystem' – one that spans the entire technology value and supply chain and captures the tendency of the digital economy to concentrate, divide, scale and spill over (Haskel and Westlake, 2017; Hemphill, 2018; Gilbert, 2023).
To meet these challenges, the Disruption Index is divided into two dimensions. The first dimension - Technological Transformation - describes the extent of technological transformation that is being experienced across ILT2 regions (counties and groups of counties) in England. This sheds light on the capacity of regions to invest in, adopt, and adapt to new technologies. The second dimension - Readiness - measures the role of human capital and infrastructure as pillars and enablers of this transformation.
The Disruption Index is the first time these variables have been mapped, together, to understand the relationships between innovation and readiness within the UK.
The focus of the Technological Transformation Index (TTI) is understanding and measuring different types of investment in innovation and innovation activity, alongside measures of technology creation and adoption.
It goes beyond the conventional focus on innovation and R&D, emphasising the significance of other factors, such as the funding possibilities for firms, and the adoption and diffusion of technologies in practice. The TTI is divided into two dimensions: i) Investments, and ii) Technology Creation and Adoption.
Table 1: Technological Transformation Index structure
Technology design, development and adoption thrive in environments that support and nurture entrepreneurs, businesses, technologists, and workers, empowering them to learn, create, apply new knowledge and grow.
The Readiness Index (RI) highlights the crucial role of human capital and the firm’s ability to absorb, act on and develop new knowledge, in the context of the surrounding connectivity infrastructure, as enabling factors – or obstacles – to the technological transformation.
Our approach recognises that the success of innovation activities depends on the interplay of a range of technological and non-technological factors at the individual, firm, regional and national levels, showcasing a ‘socio-technical’ approach to innovation and the governance of technology.
The RI covers two dimensions: i) Human Capital and ii) Infrastructure. Each indicator has been selected for its own importance but also its ability to stand as a proxy for factors of a similar nature.
Table 2: Readiness Index structure
The DI demonstrates significant, regional concentration of technological transformation, reflecting a broader tendency of AI and other automation technologies towards concentration at firm and macroeconomic level is well documented (Philippon, 2019).
Regional differences in technological transformation are stark and accelerating and appear to be scaling along various dimensions of the DI, including venture capital investments, R&D expenditure and the creation of patented technology (see the full DI report for a more detailed discussion). These huge variations, which are moving in similar directions, tend to be missed in national statistics.
The established narrative of growing polarisation between adopters and laggards at a firm level is therefore borne out across multiple domains at a regional level. Given the existing deep regional inequalities in the UK (Resolution Foundation, 2022, Economy 2030, 2023), the Disruption Index – alongside other Pissarides Review outputs - highlights the growing role that technology and innovation have - and could have in the future - in either exacerbating or reversing current trajectories.
In fact, our analysis here suggests that there is risk of further concentration of regional inequalities and that we are at an inflection point in shaping and governing regional innovation ecosystems as well as our national ecosystem. The granularities of change at a local level, and the multiple, intersecting factors require new architectures which can respond to how individual, firm and system dynamics interact and enable a place-based but people-focused approach.
Figure 1.1: Geographical distribution of Technological Transformation scores in 2020
As discussed in more detail in our full Disruption Index Report, these regional disparities observed in the Technological Transformation Index (TTI) are particularly driven by the skewed distributions of venture capital investment, R&D expenditure and patents. These are concentrated in the ‘golden triangle’ - i.e. central London, East Anglia, Berkshire, Buckinghamshire and Oxfordshire.
Figure 1.2: Evolution of Technological Transformation scores 2016-2020
Figure 1.3: Technological Transformation Index score changes 2016-2020
TTI score changes in the period 2016-2020 are positive for almost all regions analysed, which suggests that adoption is progressing. However, the nature of this requires further unpacking because of the extent of variation in scores within the index.
The region with largest gains in terms of total scores is Inner London East, going from a score of 0.30 in 2016 to 0.48 in 2020. This region saw large increases in the number of jobs demanding technology-related skills and venture capital flows. We note that the indicator of ‘demand for technology skills’ captures the use of technology skills across sectors. Other regions that improved in terms of total TTI scores are East Anglia, Greater Manchester and Merseyside, all gaining more than 0.04 points.
When examining the rank of regions in terms of TTI scores, we observe high persistence, with the highest-scoring regions remaining unchanged between 2016-2020. At the other end of the distribution, the three lowest-scoring regions also remain in the same position. We discuss decreases and improvements in particular areas, overall scores and trajectories, further below.
Figure 1.3: Decomposition of Technological Transformation scores 2016-2020
Regional differences in Readiness also exist but these are less pronounced than those observed for Technological Transformation, suggesting that innovation is not following opportunity.
Inner London West and Inner London East are again leaders in Readiness total scores, but the margin of their lead is much smaller, pointing to untapped regional and human potential in some areas. The disparities observed here mostly relate to differences in the human capital indicators, such as the number of postgraduates and the level of on-the-job training in a given region. This contrasts with digital infrastructure, the most equal of all dimensions in the index.
Our more detailed analysis below demonstrates how information can be drawn about particular indicators, which may be more or less relevant to particular stages of the innovation cycle in a particular area. This can point to gaps, bottlenecks, strengths and weaknesses that invite a place-based, context-sensitive approach, paying careful attention to different types of readiness and the stage, degree and potential for technological transformation. Often obscured by headline national figures such as the level of R&D as a percentage of GDP investment or the level of digital skills, the DI shows that this varies hugely by location and that regions have particular challenges that may need to be addressed in order to unlock the potential of their innovation ecosystems. Importantly, the DI identifies likely ‘bottlenecks’ by area and indicator for the attention of policymakers.
Figure 1.4: Geographical distribution of Readiness scores in 2020
We see that the total scores for Readiness, which measure human capital and infrastructure, have improved across all regions from 2016 to 2020. This was driven largely by improvements in IT infrastructure, with marked improvements in areas such as internet speed, ultrafast availability and 4G coverage. However, disparities in educational attainment and investment in education are still pronounced and continue to influence differences in regional readiness levels.
Figure 1.5: Evolution of Readiness scores 2016-2020
This increase in total scores is observed across all regions and rankings in 2016 is largely preserved in 2020. Large changes in ranks from 2016 to 2020 are observed for West Midlands moving up 11 positions, from 23rd to 12th. In second place of top movers, we have Herefordshire, Worcestershire and Warwickshire, which went from 26th to 18th, moving up 8 positions. Among the regions that have seen the largest declines, Cheshire fell 11 positions from 18th to 29th, and East Anglia fell from 21st to 30th.
Figure 1.6: Decomposition of Readiness scores 2016-2020
Supporting their work creating the Disruption Index, Dr Bertha Rohenkohl, Dr Jonathan Clarke and Hummd Ghouri have also created a Disruption Index dashboard as a visualisation tool to help policymakers and researchers to compare measures from the index across the country, over time, identify factors and individual indicators that contribute to the differences observed and zoom in key points. To explore the dashboard, please follow this link. The data on which this dashboard has been created can be found here. It is shared under a Creative Commons license and must be cited if used.
In this chapter, we begin to examine what individual indicators - and groups of them - may suggest about innovation bottlenecks - or, more specifically, ‘good work’ bottlenecks - within regional innovation ecosystems as a route to unlock the potential for inclusive growth (Raquel Ortega-Argiles et al, 2016 and 2020).
An extensive body of literature discusses the relationship between innovation and productivity at firm, sectoral and national levels (Coyle and Mei, 2023). While there is broad consensus that knowledge-related research and development, ICT and related technologies and innovation activities are associated with economic growth (Ulku, 2004; Maradana et al., 2017), the link between innovation and productivity is not linear and cannot be assumed (Jones, 2023).
Where innovation is a systemic phenomenon, with innovators embedded within a particular regional context, it should be understood and measured to capture contextual influences and conditions within the innovation ecosystem. By introducing the concept of regional bottlenecks and demonstrating how the DI can help policymakers identify them, we contribute towards unlocking the potential of people and places across the country.
We focus here on two areas to illustrate this, both of which invite regional deep-dives to examine the bottlenecks identified further: R&D expenditure and start-ups. In the final phase of our Pissarides Review, deep-dives will pick this up and go further, introducing, for example, multidisciplinary perspectives and factors such as workers’ perception of risk and opportunity, their capabilities, aspirations, quality of life, and their sense of belonging (Soffia et al, 2024).
When looking at how innovation impacts economic growth, investments in R&D play a central role (McCann, 2022). The Disruption Index contains data on both business and non-business gross domestic R&D expenditure, which further highlight regional disparities in how these are concentrated, how private and public investment are related, and what policy interventions might be needed to ensure these investments can be converted into opportunities elsewhere in the innovation ecosystem.1
In 2019/2020, the top five regions with the highest R&D expenditure accounted for approximately 42% of total investments in R&D, a notable increase from 35% in 2016. We see that many regions - Greater Manchester, and Tees Valley and Durham, for example - saw large increases in business R&D expenditure, with Tees Valley and Durham experiencing a remarkable 61 % growth during the period analysed (from £110M to £177M). Meanwhile, other regions have seen a decline. In Merseyside, for example, business R&D expenditure dropped by 21% in the same period, from £499M to £394M.
Table 3: Business R&D expenditure, selected regions as examples
Non-business R&D expenditure (i.e. performed by government, higher education and non-profit institutions) is also highly concentrated – and largely in the same regions that have high business R&D. In 2019/20, the top five regions accounted for 24% of all non-business R&D expenditure (an increase from 16% in 2016), with Inner London West and Berkshire, Buckingham and Oxfordshire in the lead. In addition to this, the growth in non-business R&D varies significantly across regions, which is leading to a further concentration of R&D investment at a regional level. For example, from 2016-2020, Inner London, East Anglia and South Yorkshire recorded large percentage increases in investment - with East London recording a 41% increase - whereas Shropshire and Staffordshire and Merseyside faced a large decline of 20% (from £40M to £32M) and 6% (from £303M to £284M), respectively.
Table 4: Non-Business R&D expenditure, selected regions as examples
With reference to the current policy approach to R&D funding allocation across regions, variations in sectoral mix and other environmental factors do not appear to have been considered or adequately considered. Importantly, nor have local innovation bottlenecks such as access to skills and capital, which may be impacting the ability of businesses to undertake or apply R&D activity in a given region, commercialising this in the form of products and services. This is likely to affect its value and positive spillover effects for local people and communities.
For instance, a human capital indicator that has particular importance in the context of R&D investment is the number of people completing ICT apprenticeships. Here, data in the Disruption Index shows this declined across regions during the period analysed, in spite of job vacancies data pointing to a significant increase in demand for tech skills within firms. Given that over £3.3Bn of unused apprenticeship levy funds from large businesses were returned to the Treasury last year (Lewis Silkin, 2023), this suggests that there is a pressing need to address challenges with the technical education system around incentives and career pathways, which may be simultaneously impacting both regional R&D intensity and the career development opportunities for people looking to work in technology-intensive sectors.
The DI helps pinpoint areas where this challenge may be responsible for a ‘bottleneck’ - ie where other enabling factors may be in place, but there is a drop in take-up of ICT apprenticeships nonetheless. These areas include all of Yorkshire, including North Yorkshire (a drop of 44%), South Yorkshire (a drop of 39%) and West Yorkshire (a drop of 33%). This may point to the sectoral mix, size or absorptive capacity of firms in this area and therefore a need to adapt apprenticeships to make them viable, attractive and tailored to the needs of smaller businesses in this area.
The sectoral mix in an area and the regional presence of large R&D businesses - such as those in the automotive and aerospace sector - is an important factor shaping the level of business R&D activity in a region, which government has fewer direct levers to influence. One of these levers, for example, is R&D tax credits, which are national in scope and operate based on the discretion of firms. The government has more control over direct R&D funding, including investments in higher education, non-profit organisations and government research institutions.
Through support for regional universities and research centres - including supporting the creation of new research centres such as the Francis Crick Institute for Biomedical Science in central London - government can shape the development and trajectory of regional, as well as national, innovation ecosystems. The distribution of technical research centres in other countries - such as the 76 Fraunhofer Institutes located throughout Germany which undertake advanced applied research - supports this.
This suggests that new approaches to public R&D investment in the UK are likely to be needed, including the development of new institutional models, networks and funding mechanisms that can capitalise on regional strengths and local partnerships in more responsive and innovative ways, informed by a deeper understanding of place, and embedded in it.
Regional disparities in other indicators in the Disruption Index, such as business innovation activity and the level of employment in technology-intensive sectors, point to other underlying challenges, opportunities and bottlenecks. For instance, Merseyside has seen significant growth in terms of businesses undertaking innovation activities, which can be understood as the number of ‘broad innovators’ combined with high employment in technology-intensive sectors, with employment increasing from 2.7% to 3.5% in the period under review. This contrasts with changes in the level of employment in technology-intensive sectors seen in other regions, such as Tees Valley and Durham, which fell from 3% to 2.4% in the same period.
Viewed in the context of Merseyside’s strong - and improving - indicators of human capital and positive trends related to broader innovation activity in the region, the simultaneous reduction in both public and private investment points to the likelihood of bottlenecks associated with underfunded and poorly targeted R&D, combined with under-developed career pathways for those working in technology-intensive sectors.
By contrast, success in Tees Valley and Durham may reflect generous devolution deals combined with additional support obtained from central government departments, enabling initiatives which span the different stages of the innovation system in a context sensitive ways, from regeneration projects and to new skills initiatives for electric car technicians and healthcare workers. These potential bottleneck areas and the role of specific policy measures in driving these indicators invite further research and deep-dives.
The DI also surfaces a marked difference in the number of tech start-ups across regions and their trajectories, underscoring that regional disparities are seeded at the start of the innovation process - as well as through later stages when larger rounds of investment take place, technologies are adopted by other sectors, and startups look to scale. The Disruption Index data highlights a marked drop in the number of tech startups in all regions, with the smallest fall in Inner London West (6%). Much of the rest of the country saw falls in excess of 30%, although there was substantial global investment in the digital economy in the period under review.
Here, tech startups can be considered as an indicator of wider innovation activity and potential in an area linked to other factors in the ‘entrepreneurial’ innovation ecosystem, such as the number of STEM postgraduates, access to 'beachhead' customers that enable businesses to gain initial traction in the market and the presence of startup incubators and accelerators, to help nurture and grow startups applying R&D undertaken in a local area.
In 2020, the top five regions in the country accounted for well over 20% of the tech start-ups observed, a proportion that other indicators suggest are not linked to talent or readiness. In particular, it cannot reflect the distribution of the human capital indicators in the Disruption Index, including the number of postgraduates and the proportion of people with technical qualifications. For example, whereas West Yorkshire and the West Midlands have 7% and 5% of the UK's postgraduates, respectively, they account for a substantially smaller proportion of the number of startups (about 2%). Where research suggests that higher education substantially increases the probability of raising investment for a technology venture in the UK (Ratzinger et al., 2018), this suggests that there could be important innovation bottlenecks around support for university entrepreneurship and academic spinouts in the regions, given the UK’s world-class university and FE system outside London.
To what extent are the regional differences observed in Technological Transformation related to the industrial composition of these regions? We link our data to the Business Register and Employment Survey (BRES) - the most reliable source of data on employment by sector at a local level in the UK.
Comparing the proportion of a region’s workers employed in each industrial sector with transformation scores in 2020, we see that having a higher employment share within a sectoral mix in ICT, scientific and financial industries is strongly associated with higher Technological Transformation Index scores. Conversely, areas with higher shares of employment in manufacturing, retail and healthcare generally have lower transformation scores. These findings persist across all years from 2016 to 2020. These positively correlated sectors – ICT, science and finance, are likely to be the most intensive in terms of technological transformation, driving the aggregate relationships we see here.
This reinforces the known associations between certain types of innovation and technology-intensive sectors, and the nature and pace of their transformation, reflecting established different processes and existing priorities for innovation.
Figure 2.1 Correlation of TTI score with percentage of employees in each sector in 2020
Real-time data on how investments and innovation activity vary by sector within a region would provide an even richer understanding of Technological Transformation and the ability to identify and measure relationships with other indicators, particularly with regard to pinpointing particular blockers and unlocking opportunities. As an example, we note that the data from the UK Innovation Survey (UKIS) – used in the Disruption Index to examine the percentage of businesses who are undertaking innovation activities - does not allow for a simultaneous breakdown by region and by sector because it is based on a sample of British businesses that was not designed to be representative at the ITL2 or more granular geographical levels.
With other workstreams of the Pissarides Review, our research points to the role of good work as a forerunner and mediator of better outcomes – and understanding the relationships between work, technology and productivity. In particular, the need to be able to learn, develop and apply people’s skills and capabilities on an ongoing basis, from research and development to innovative application of R&D, is important. Without this, our analyses suggest that the process or ‘cycles’ of innovation, as well as career pathways, may be interrupted and the potential of people and places thwarted. The role of Good Work within the regional innovation ecosystems is examined further in the next chapter.
Everyday transformation of the economy is led by firms, as they make choices to design, develop and deploy new technology, and to change work as a result. This shapes the creation of new jobs, their nature, and the distribution and access to good work.
But these technology adoption choices are related to wider conditions. The distribution of skills shapes firm adoption practices, as well as being an outcome of them. The Readiness of a local area is linked to choices about technology adoption and practices of human resource management (Hayton et al., 2023). Firms in areas with weaker human capital investment and outcomes have been shown to make adoption choices that could exacerbate regional inequality – deploying new tools in ways which reduce demand for skills, and the quality of work – as well as being more likely to eliminate positions overall. In turn, a lack of investment in skills can reduce ultimate opportunities for learning and the development of human capabilities within a local economy.
In this context, we must begin to understand transition not in discrete units (region, firm, individual) but also through the lens of development pathways, interrogating the role of work and career cycles. At present, tracking and unpacking these dynamics in full is beyond the scope of this report - or currently available datasets - but this chapter starts to explore the particular role of good work in forerunning, mediating and creating ‘good’ automation and well-functioning innovation ecosystems.
Our analysis broadly supports the hypothesis that a sharper focus on creating, sustaining and improving ‘good work’ – in particular opportunities for learning, development and application of people’s capabilities – can convert vicious cycles of stagnation, poor life quality and poor wellbeing into virtuous ones. In this way, ‘good work’ serves to focus policy attention on some of the most important goals, roles, interventions and partnerships to address risks and maximise the best results of the new technological revolution.
Following on from this, we have also begun to see measures of ‘good work’ act as proxies for readiness and resilience, and to capture and evaluate some of the vast social, economic and cultural impacts of technological transformation in tangible, practical ways (Thomas, Nash, 2024). Without this focal point for multiple, interconnected factors - and a fresh approach to measurement - these impacts may be missed or embedded, when our analysis shows that they can be shaped.
‘Good work’ is more than employment. As set out in the Good Work Charter it is work that promotes dignity, autonomy and equality; work that has fair pay and conditions; work where people are properly supported to develop their talents and have a sense of community (Sekhar, 2011; Fang and Tilcsik, 2022; IFOW, 2022). IFOW publishes an annual Good Work Monitor to track access to six dimensions of good work across 203 local authorities in England, Scotland and Wales.
Table 3.1: Good Work Monitor Indicators
Here, by combining data from our Disruption Index with the Good Work Monitor, we go beyond changes to employment levels and job creation to highlight how access to good work relates to the geography of technological transformation.
The figure below shows the distribution of our Good Work Monitor (GWM) scores by region. The Good Work Monitor builds a detailed sub-regional map of access to good work across England, providing scores for 148 local authorities. On the y-axis are the ITL2 regions, ranked from highest to lowest by their Technological Transformation scores. The regions are classified into ‘low’, ‘high’ or ‘very high’ in each year analysed according to their TTI scores.2
Figure 3.1: Regional and subregional disparities in good work in 2016 and 2020
This suggests a generally positive association between Technological Transformation and Good Work Monitor scores. That is, regions with high TTI scores tend to also have higher GWM scores. We see that almost all local authorities in regions with ‘very high’ Technology Transformation scores are in the upper part of the Good Work score distribution. For example, in 2020, Inner London West - the region with the highest Technological Transformation scores - has 4 out of its 5 local authorities in the top quartile of the distribution of Good Work scores in 2020.
We also see great variation in access to good work within these large ITL2 regions. For instance, in Berkshire, Buckingham and Oxfordshire we find both West Berkshire - which had the highest Good Work Monitor score across all areas in 2020 - and Slough, which scores on the bottom half of the distribution. The positive relationship observed here between TTI and GWM scores is likely to reflect the fact that regions with high TTI scores have higher employment, a higher share of people working in professional jobs and higher median pay. They also have a lower share of people working in routine occupations.3
Importantly, over time, most regions are improving in terms of both scores although the pace and variation of the scores is large and driven by different components. This can be seen the figure below, which illustrates the change in TTI scores from 2016-2020 against the change in GWM scores in the same period. Only one local authority in Shropshire and Staffordshire (Telford and Wrekin) experienced a negative change in both TTI and GWM scores over this period.
Figure 3.2: Change in TTI and GWM scores, from 2016-2020
As we found earlier in the review, ‘for employers and employees, the adoption of, and engagement in, a Human Resource Management (HRM) philosophy with high-involvement practices will unlock a positive attitude to technology adoption and the perceived benefits of AI. This HRM philosophy reflects the Good Work dimensions of autonomy, support and participation in particular, which open the way to a better range of work outcomes including the creation of new jobs, augmentation of human skills and improvement in work quality’ (Hayton, 2023). Here too, good work indicators correlate with, and may drive a well-functioning innovation ecosystem. This is consistent with our analysis of the GWM and TTI.
The relationship between technological transformation at work and worker wellbeing has been the subject of a growing body of research. There is no academic consensus on how the introduction of new technologies at work, particularly novel automation technologies, may affect workers (Rohenkohl and Clarke, 2023). Instead, a complex, highly context-dependent picture emerges where some workers stand to benefit from new technologies, while others will likely see their experience of work deteriorate or may lose their jobs entirely.
These conclusions align with the results of a recent survey of over 5000 employees across the UK conducted as part of the Pissarides Review, which finds the use of digital information and communication technologies is associated with improved wellbeing, while newer and more advanced technologies were associated with reduced wellbeing (Soffia et al., 2024).
Building on this work, this chapter examines whether the TTI scores from the Disruption Index are correlated with the life satisfaction of the adult population of each region. At a high level, this allows us to begin to ‘triangulate’ analyses of technological transformation, good work and wellbeing, although we highlight further research needed in this space.
We are conscious that ITL2 regions are large areas that contain within them localities that may differ markedly in their sociodemographic and economic features. To make use of data on wellbeing, which is available at finer geographic resolution, for local authorities, here we plot each of these areas against their corresponding Technological Transformation score ranking. This is presented in the figure below. The regions are classified into TTI ‘low’, ‘high’ or ‘very high’ in each year analysed.4
The wellbeing data comes from the Annual Population Survey (ONS, 2023). It presents wellbeing at Local Authority District level through responses to the ‘ONS Four’ wellbeing questions, mapping to shorthand terms of ‘Life satisfaction’, ‘Worthwhile’, ‘Anxiety’ and ‘Happiness'.5
Focusing on life satisfaction, in 2016 and 2020, in the figure below we see a very weak negative correlation with TTI scores for the same year (-0.04 and –0.11 respectively). We also find that the variation in life satisfaction within regions is far greater than the variation observed between regions.
Figure 4.1: Life Satisfaction against TTI scores, 2016 and 2020
Life satisfaction scores generally decrease from 2016 to 2020, partly reflecting the Covid-19 pandemic (ONS, 2023). The greatest decreases in scores are found in areas with the highest life satisfaction scores in 2016, reflecting a reduction in spread in life satisfaction scores.
Between 2016 and 2020, it is possible to see that life satisfaction increased for a small number of areas. However, none of these areas were in regions with high TTI scores and were instead found in some of the areas with the lowest technological transformation scores. That said, the greatest decreases in life satisfaction over this period were also in areas with lower TTI scores, suggesting varying wellbeing trajectories for regions with the lowest TTI scores.
Overall, this analysis indicates that, while there is extensive variation in wellbeing across England, positive trends in life satisfaction and wellbeing are not associated with the TTI scores of regions. This is not unexpected, in itself, given the scale of the regions analysed, the aggregate nature of these analyses, the multiple social determinants of wellbeing and the need for multi-disciplinary analyses.
Our research shows that policymakers cannot assume technology will ‘automatically’ result in improvements to quality of life for most people. Instead, this must be consciously ‘designed’, developed, monitored and adjusted for as impacts emerge and accumulate, some of which may not have been anticipated or may be hidden. This is consistent with the findings arising from our individual level survey of workers (Soffia et al., 2024), which highlights the varied impacts of technology on job quality and wellbeing.
In this report we begin to unpack the interdependencies between regional conditions for innovation, firm choices about adoption, and individual outcomes and identify the virtuous cycles which can result from better, integrated thinking.
Our approach demonstrates why and how a socio-technical approach, with careful regard to variable social and economic contexts and impacts, anticipated impacts and policy choices, is required to innovate and govern AI and automation technologies. In particular, our analyses broadly support the hypothesis that a sharper focus on creating, sustaining and improving ‘good work’ - in particular opportunities for learning, development and application of people’s skills and capabilities across the innovation life cycle – have the potential to transform vicious cycles of stagnation, poor life quality and poor wellbeing into virtuous ones.
Through our analyses and other work, we have begun to see measures of ‘good work’ act as proxies for readiness and resilience, and to capture and evaluate some of the vast social, economic and cultural impacts of technological transformation in tangible, practical ways. In combination with the work, we have seen ‘Good Work’ act as a forerunner, mediator and impact which can help steer better outcomes from technological transformations.
Without this focal point for multiple, interconnected factors and a fresh approach to measurement, these impacts may be missed or become embedded; our analysis shows the extent of variation across the country – and thus that they can be shaped.
In this way, ‘good work’ serves to focus attention on some of the most important goals, roles, and partnerships needed to address risks and maximise the best results of the new technological revolution - and help policymakers move from crisis management to long-term, mission-oriented planning. Our analyses highlight the particular significance of several dimensions: participation and autonomy - and their application through skills and the opportunity to learn and develop skills and capabilities across the innovation ecosystem. These dimensions are likely to be key at each stage of the innovation process from initiating and engaging in innovation, to imitation, translating externally sourced knowledge into innovative outputs, adopting or adapting innovations and, maximising positive effects for local communities.
The DI also invites closer attention to our social and political choices, including the responsibilities and the consequences of innovation. Is the real purpose of R&D and other government initiatives to shape the UK’s innovation ecosystem for short-term growth, narrowly defined, in the high-tech sectors? Or is the public interest better served by initiatives which may promote patient, inclusive growth and new types of innovation beyond well-recognised recognised high-tech areas? This might include those which involve higher levels of risk, extending to the adoption of sector-appropriate technology and the transitions of other sectors across the everyday economy, such as healthcare. These questions are sharpened as all sectors, firms and occupations are exposed to transformative technologies, which can no longer be regarded as a discrete domain.
It is suggested that these questions may be balanced by considering prospects for the creation or improvement of good, local jobs by interventions which take careful account of the potential for places and people to flourish.
Note: these initial policy implications represent our current level of analysis. Rich work is ongoing in the Pissarides Review, and further analyses coming from that will iterate policy work as we build towards the final report of the Review.
Autor, D. et al. (2020) ‘The Fall of the Labor Share and the Rise of Superstar Firms', The Quarterly Journal of Economics, 135(2), pp. 645–709. Available at: https://doi.org/10.1093/qje/qjaa004.
Coyle, D., Ark van, B. and Pendrill, J. (2023) The Productivity Agenda – a blueprint for boosting the UK’s productivity. Available at: https://www.productivity.ac.uk/news/the-productivity-agenda-a-blueprint-for-boosting-the-uks-productivity/ (Accessed: 11 June 2024).
Coyle, D. and Mei, J. (2023) ‘Diagnosing the UK productivity slowdown: which sectors matter and why?’, Economica, 90(359), pp. 813–850.
Fang, R.T. and Tilcsik, A. (2022) ‘Prosocial Occupations, Work Autonomy, and the Origins of the Social Class Pay Gap’, Academy of Management Journal, 65(3), pp. 903–929. Available at: https://doi.org/10.5465/amj.2020.1564.
Fraunhofer (2024) Institutes and Research Units, Fraunhofer-Gesellschaft. Available at: https://www.fraunhofer.de/en/institutes.html (Accessed: 10 June 2024).
Gilbert, A. (2023) ‘Reframing Automation - a new model for anticipating risks and impacts’. Available at: https://www.ifow.org/publications/reframing-automation-a-new-model-for-anticipating-risks-and-impacts
Haskel, J. and Westlake, S. (2017) ‘Capitalism without Capital : The Rise of the Intangible Economy’, pp. 1–288.
Hayton, J. et al. (2023) 'What drives UK firms to adopt AI and robotics, and what are the consequences for jobs?' Institute for the Future of Work. Available at: https://www.ifow.org/publications/adoption-of-ai-in-uk-firms-full-working-paper (Accessed: 10 April 2024).
Hemphill, T.A. (2018) ‘Capitalism without Capital: The Rise of the Intangible Economy’, Cato Journal, 38, p. 750.
IFOW (2022) The Good Work Charter. Available at: https://www.ifow.org/publications/the-ifow-good-work-charter (Accessed: 9 April 2024).
Jones, R.A. (2023) 'Productivity, Innovation and R&D.' The Productivity Institute.
Lewis Silkin (2023) '£3.3 billion of unused apprenticeship levy funds have been returned to the Treasury – should you be thinking about apprenticeships?', Lewis Silkin. Available at: https://www.lewissilkin.com/en/insights/33-billion-of-unused-apprenticeship-to-the-treasury-should-you-be-thinking-about-apprenticeships (Accessed: 11 June 2024).
Maradana, R.P. et al. (2017) ‘Does innovation promote economic growth? Evidence from European countries’, Journal of Innovation and Entrepreneurship, 6(1), p. 1. Available at: https://doi.org/10.1186/s13731-016-0061-9.
McCann, P. (2022) ‘Levelling Up: The Need for an Institutionally Coordinated Approach to National and Regional Productivity’, The Productivity Institute, Insights Paper, 11.
ONS (2023) 'Annual personal well-being estimates - Office for National Statistics.' Available at: https://www.ons.gov.uk/peoplepopulationandcommunity/wellbeing/datasets/headlineestimatesofpersonalwellbeing (Accessed: 11 June 2024).
Ortega-Argiles, R. and McCann, P. (2021) ‘The Innovation-Productivity Paradox’, in. Background Paper for the OECD-EC High-Level Expert Workshop Series “Productivity Policy for Places.
Philippon, T. (2019) ‘The economics and politics of market concentration’, NBER Reporter, (4), pp. 10–12.
Ratzinger, D. et al. (2018) ‘The impact of digital start-up founders’ higher education on reaching equity investment milestones’, The Journal of Technology Transfer, 43(3), pp. 760–778. Available at: https://doi.org/10.1007/s10961-017-9627-3.
Resolution Foundation (2022) ‘Stagnation nation: The Economy 2030 Inquiry’.
Rohenkohl, B. and Clarke, J. (2023) ‘What do we know about automation at work and workers’ wellbeing? Literature Review’, Institute For The Future Of Work [Preprint]. Available at: https://www.ifow.org/publications/what-do-we-know-about-automation-at-work-and-workers-wellbeing.
Rohenkohl, B., Clarke, J. and Pissarides, C. (2024) 'A Disruption Index - the geography of technological transformations across England.' Available at: https://www.ifow.org/publications/a-disruption-index---the-geography-of-technological-transformations-across-england (Accessed: 31 May 2024).
Sekhar, S. (2011) ‘Correlates of work autonomy’, The IUP Journal of Organizational Behavior, 10(3), pp. 26–40.
Soffia, M. et al. (2024) ‘Does technology use impact UK workers’ quality of life? - a Report on Worker Wellbeing’. Available at: https://www.ifow.org/publications/does-technology-use-impact-uk-workers-quality-of-life.
Taylor, M. et al. (2017) 'Good work: the Taylor review of modern working practices.' Available at: https://www.voced.edu.au/content/ngv:77181 (Accessed: 11 June 2024).
Ulku, H. (2004) ‘R&D, Innovation, and Economic Growth: An Empirical Analysis’. Rochester, NY. Available at: https://papers.ssrn.com/abstract=879010 (Accessed: 10 June 2024).
