Which Occupations Should Get Skilled Worker Visas? Informing the UK’s Visa Reform

Abstract

The United Kingdom’s 2025 immigration reforms will restrict access to Skilled Worker visas for “mid-skill” occupations unless they are retained on a revised shortage list. This raises an important policy question: which occupations should continue to receive access, given the limits of domestic labour supply and the effects of labour shortages on the government’s industrial strategy objectives?

This paper brings together Home Office visa data, Department for Education apprenticeship data, and Migration Advisory Committee labour demand indicators to assess where Skilled Worker visas function as a critical source of new trained labour and where domestic training pipelines are least able to respond quickly to additional demand. We show that dependence on visas varies substantially across occupations and sectors. For many roles, the loss of visa access would create a significant supply shock that could not readily be offset through apprenticeships, particularly given long training lead times and weak completion rates. We further show that recent reductions in visa recruitment have not been matched by increases in apprenticeship starts, suggesting that domestic training may not automatically scale to replace visa flows, at least in the short run. In this context the government must consider its tolerance for demand destruction in the interim.

We finally construct an index ranking occupations for continued visa access across multiple factors, for use as an input in visa prioritisation decisions. The strongest case for continued visa access is where occupations combine high dependence on skilled migration, constrained domestic pipelines, strong labour demand, and strategic importance. These dynamics are especially salient in parts of the clean energy workforce, where labour shortages may impede delivery of wider policy goals.

Introduction

The United Kingdom (UK) is currently in the process of making major changes to its Skilled Worker visa. All “mid-skill” occupations, unless on a shortlist of exemptions, will no longer be eligible to access the visa. For some occupations, skilled migrants supply most of the new worker pipeline; a sudden ban on recruitment would pose a serious risk to policy goals. This is especially the case for visas related to the UK’s “Clean Energy Mission”, which faces severe workforce constraints (ESNZ Committee, 2025) and relies heavily on visas for many key occupations (Huckstep and Dempster, 2025).

This paper is intended to inform discussions regarding which occupations should be on the list of exemptions. To do so we harmonise Home Office visa data, Department for Education apprenticeship data, and Migration Advisory Committee (MAC) labour demand indicators to show where visas provide a pressure release valve for labour supply, and where domestic pipelines are least able to compensate quickly.

The government’s stated aim is to incentivise employers to increase investment in domestic training by reducing access to visas. This might happen, but it is not a given. Employers could respond in several ways: they may increase investment in domestic training, but they could also reduce activities, invest in automation, offshore tasks, or raise wages to compete for workers who are already trained. For many of the occupations under consideration, these responses—including increased domestic training—are either not possible at scale in the short term, would not alleviate sector-wide shortages, or would increase the cost of labour outputs to a level that might cause demand destruction.

Our analysis of new apprenticeship starts and visa use suggests that employers do not use apprenticeships and visas interchangeably: visa use for the occupations under consideration has fallen over recent years due to previous efforts to limit international recruitment, but apprenticeship starts have not risen commensurately to compensate. In some cases (where visa use is still high relative to apprenticeship rates that remain low) replacing lost foreign-trained workers with domestic workers would require an increase in apprenticeship starts of more than 100 percent.

Moreover, even if tighter migration policy does induce more training, additional trained domestic supply will arrive only after a lag, often of years. In this period labour supply shortfalls can be expected to negatively impact the UK’s ability to achieve the (often ambitious) goals set out in the Industrial Strategy.

We then draw out implications for how the MAC and government departments could prioritise between occupations for continued access, and construct an index showing occupations’ “deservingness” of visa access. We pay particular attention to visas needed for the green transition, noting that in the context of workforce shortages an individual worker can contribute significant carbon abatement and energy generation benefits.

Changes to the Skilled Worker visa policy

In May 2025 the UK government published a white paper on immigration (Home Office, 2025a). Several of the key changes are intended to reduce net migration to the UK. Notably, it significantly increased fees charged to employers; eliminated lower wage thresholds for shortage occupations; and reduced the list of jobs eligible for the Skilled Worker visa. Occupations assessed as “mid-skill”—falling between levels 3 and 5 of the Regulated Qualifications Framework (RQF)—would no longer be eligible. Employers could only sponsor foreign workers at these levels if their occupation was placed on a revised Temporary Shortage List (TSL) by the MAC. The TSL was reduced in length in July 2025 (Home Office, 2025c), maintained until the end of 2026; the product of the MAC’s current investigation will replace it beyond that (MAC, 2025b).

These changes were developed further in the June 2025 Industrial Strategy (DBT, 2025). This tied the revised TSL to the new set of eight sectors prioritised for government interventions (IS-8). It also reintroduced salary discounts for shortage roles via the Immigration Salary List (ISL) as a form of industrial policy (the MAC recommends no salary discounts on the TSL (MAC, 2025a)).[i] Occupations would be placed on the TSL if “an appropriate plan,” possibly including a formal sector workforce strategy, was adopted by employers to develop domestic skills and increase local recruitment (DBT, 2025: 71).

Using visa policy changes to incentivise training investments

The stated goal of the government is to incentivise employers to invest in domestic training by reducing access to international recruitment (Home Office, 2025a). It is unclear whether this will happen. The MAC (2024) has warned that there is no clear causal link between employers’ use of international recruitment and investment in apprenticeships, and that reductions in apprenticeships may be due more to changing funding structures and employer need than to access to visas.

While the government has required that sectors demonstrate “an appropriate plan” before retaining access to visas, its ability to verify the policy’s effectiveness is limited. As the Migration Observatory notes, a sectoral approach could allow free-riding, while it is hard to monitor commitments by individual employers (Brindle and Sumption, 2026). Moreover, the apprenticeship system is widely accepted to face severe challenges requiring lengthy reforms (The St Martin’s Group, 2024; Wolf, 2025; Field, 2025; Industry and Regulators Committee, 2024; British Chambers of Commerce, 2026).

An apprenticeship is a paid job combined with structured training for a specific occupation. The supply of apprenticeships is thus led by employers’ perceived long-term demand for workers. An employer must choose an approved apprenticeship standard and training provider, use the apprenticeship service, and agree a training plan. This structure means that the government has limited control over employers’ training choices: whether an employer takes on an apprentice is a function of their anticipated forward demand. Apprenticeships take between eight months and four (or more) years to complete, and thus the employer must have confidence in the long-term operating environment to take on the risk. It is also a function of the provision of training capacity; many sectors face shortages of trainers and other staff (DESNZ, 2025).

In this context, losing access to Skilled Worker visas may see employers limit activities rather than take the risk of investing to train an apprentice. Research from Korea, for example, finds that the suspension of a low-skilled guest worker programme led to firm exit rather than an expansion of domestic hiring (Lee et al., 2026); research from the United States similarly suggests that firms unable to access immigrant labour respond by curbing production, and that domestic hiring in fact decreases (Clemens and Lewis, 2024).

Employers could also, in theory, attempt to respond to lost labour supply by raising the productivity of their existing workers through investment in automation, or by offshoring work, rather than by curbing activities or increasing domestic training. Many jobs on the TSL, however, can only be automated to a limited degree, and are in addition highly location-constrained: they cannot be offshored. Finally, employers may also seek to increase wages to compete for workers already present in the country (Filippucci et al., 2025). We discuss this later: many of these occupations are shared across multiple sectors, and competition is likely to occur. In the short-term, this would increase the cost of obtaining labour, raising the cost of the marginal unit of labour output. This may lead to demand destruction, which might be an undesirable outcome where labour output is of strategic importance: the government must balance its priorities, and assess how much demand destruction it is willing to tolerate in exchange for a long-term rise in domestic training.

Finally, rising wage costs may indeed incentivise employers to increase supply via apprenticeships. There is little empirical evidence on the elasticity of training offers to wages. In Switzerland, a 10 percent increase in hiring costs is associated with a roughly 2.9 – 4.8 percent increase in the probability of training at least one apprentice, and an apprenticeship-count elasticity of 1.4. In Germany, by contrast, no link is found—possibly due to the higher cost of apprenticeship training (Aepli et al., 2024). In the UK, the probability of an employer offering an apprenticeship has previously been found to increase by 2.1 percentage points for each extra hard-to-fill vacancy for skilled trade workers (Mcintosh et al., 2011). Where this happens, starts may also respond with a lag—evidence from Germany suggests a possible lag of 3 to 12 months in which employers assess the situation (Lüthi and Wolter, 2020). Apprenticeship reforms or economic uncertainty may also delay employer responses (Majumdar, 2007).

If employers do respond to reduced supply of trained workers by investing more in training new entrants, there will therefore be a lag in which they: (i) recognise the changed supply environment; (ii) judge whether demand is likely to persist; (iii) expand their training capacity; and (iv) train workers to full competence. This outcome also depends on sufficient capacity among local off-the-job training providers with which the employer would need to partner; there may be a further lag in which they respond to new employer demand for training. During this period supply of labour inputs would be reduced, increasing costs and leading to some degree of demand destruction.

The purpose of the TSL is, in essence, to provide a recruitment “bridge” to allow employers access to key occupations—and thus avoid demand destruction—while incentivising them to scale up their training capacity in preparation for a future cut-off. The effectiveness of the “bridge” mechanism relies on employers believing that demand will persist long enough to justify investment in training; having the capacity to take on and supervise trainees; sourcing local off-the-job training provision; and being able to scale supply within a useful timeframe.

Where these elements are not available, restrictions on visas are likely not to induce rapid substitution towards domestic training, but instead to reduce labour supply, raise project costs, delay delivery, and suppress output. In designing the TSL, it is crucial that it: (i) includes occupations for which adequate domestic substitution will not be possible within relevant timeframes; (ii) includes occupations for which the loss of labour outputs would be strategically undesirable; and (iii) offers a long enough “bridge”, and is accompanied by enough supply-side support, that domestic training in those occupations can scale to meet needs before international recruitment becomes inaccessible.

The occupations under consideration for the TSL

The MAC is currently reviewing visa access via the new TSL for the 82 occupations previously on the ISL. These occupations are grouped according to the Industrial Strategy (IS-8) sectors to which they are relevant. An additional sector, “Foundational Industries / Critical Infrastructure”, is also used, and several occupations are not clearly relevant to a lead sector. Most occupations overlap between multiple sectors. They are all considered to be at RQF levels 3 – 5, so will automatically be excluded from Skilled Worker visa access (MAC, 2025b). These occupations include technicians; artists; welders; and sewage operators. The full list of occupations is available in the appendix. The distribution of occupations across sectors is given in Table 1; note, as discussed later, that many occupations are shared across multiple sectors. Clean Energy, with 43 occupations, has the most of any sector, although only 9.3 percent of these are not shared with another sector.

Table 1.

For most sectors and occupations, the Skilled Worker visa is the primary visa used for international recruitment. This is not the case across the board: the Creative Industries sector, unsurprisingly, makes heavy use of the Creative Worker visa, and thus is less exposed to changes to Skilled Worker visa policy. For priority sectors such as Advanced Manufacturing, Clean Energy, or Foundational Industries, however, Skilled Worker visas are critical (Figure 1).

Figure 1.

Assessing the implications for labour supply

As will be described below, many of these occupations will soon see a spike in labour demand at the same time as losing access to Skilled Worker visas. This will therefore require an increase in the capacity of the apprenticeship system. But how reliant are these occupations on Skilled Worker visas now? And can the domestic training pipeline scale, and scale quickly enough, to fill needs?

To answer these questions, we harmonise apprenticeship data with visa data (as set out in the methodology) to allow a comparison of labour supply channels over time. We take apprenticeships at National Qualifications Framework (NQF) Levels 2 and 3 as proxies for new entry, given that higher-level apprenticeships are often used for upskilling by existing trained workers.

For most of the occupations in question, it is safe to view these two supply channels as an adequate proxy for new trained labour-market intake. This is not the case for all, however: some occupations on the MAC’s list draw new entrants from degrees as well as apprenticeships; don’t draw from apprenticeships at all; or are poorly mapped to apprenticeship pathways. We therefore bucket occupations according to the reliability of apprenticeships as a domestic workforce entry proxy (per Skills England’s mapping of expected education by 4-digit standard occupational classification (SOC) code) (Skills England, 2025).[ii] We include these buckets across our analysis to aid interpretation:

• Predominantly apprenticeships: the highest expected education is NQF Level 3.
• Mixed pipeline: expected education includes low levels, but also NQF Levels 4/5 and over.
• Weak apprenticeship proxy: expected education is NQF Level 6 and over.

The loss of access to visas will come as a supply shock. Assuming that the flow of entrants reflects demand for qualified workers, the apprenticeship system would need to replace lost visas with new throughput of trainees if demand is to be met. For many areas this will be difficult: apprenticeship completion rates are stubbornly low. In Table 2, we show key data across all IS-8 occupations under consideration. For many occupations, the loss of visas would require huge increases in domestic training: in some cases, more than double.

Table 2.

Dependence on Skilled Worker visas varies across sectors and occupations, and by time. In Figure 2, we show visas as a percentage of total new entrants (across apprenticeships and visas) into individual IS-8 occupations, by sector from academic year 2021/22 – 2024/25.[iii] Across the Advanced Manufacturing sector, for example, visas peaked at 18 percent of observed new entrants in the academic year 2023/24, falling to 11 percent in 2024/25. Due to the long lead times of most apprenticeships, the domestic supply of fully trained workers lags visa policy changes by twelve months to several years.

Figure 2.

Assessing labour supply versus demand

The stated purpose behind the Skilled Worker visa reform is to incentivise UK employers to train more domestic workers. Looking at the data on apprenticeship starts for recent years, it does not appear that employers use apprenticeships and visas interchangeably. The total number of Skilled Worker visas obtained across the list of priority occupations fell from 14,317 in 2023/24 to 8,019 in 2024/25 (a 44 percent decrease) following policy changes intended to reduce international recruitment (Brindle and Sumption, 2025), but apprenticeship starts across those occupations remained very stable, rising by roughly 500—not enough to offset the loss. This is consistent with the idea that visas act as a release valve at the margin for employers: if this is the case they may not be substituted for by apprenticeship starts, which require long-term investment certainty.

As noted, apprenticeship supply may increase after a lag, or may only increase when a demand threshold is surpassed. For this reason, we consider how the supply of new apprenticeship starts interacts with labour demand indicators and access to visas. The MAC (2025b), in its call for evidence, supplies data on earnings and on labour demand as proxied by online job adverts (obtained from Lightcast). In Figure 3, we stack a set of labour supply indicators against these two demand indicators, which we index from 2019/20. From 2019/20 onwards, total employee job volumes across the occupations considered have remained very stable. (This is not the case for all occupations, however: some have declined over the period.) Given that total apprenticeship starts have also remained stable, this suggests that on aggregate a combination of access to visas and gradual improvement in apprenticeship completion rates has buffered against retirements. Completion rates are not, however, on an inevitable upward trend; the reported rate for air-conditioning technicians, for example, fell from a low 27.4 percent in 2022/23 to a very low 15.4 percent in 2023/24. This is a major challenge, and exacerbates workforce under-supply in the context of very low visa use and rocketing job advert intensity.

Figure 3.

As discussed earlier in the paper, migration could be used as a bridge while training increases. This will be especially needed where labour demand is anticipated to rise sharply to meet policy goals for occupations with long training times. Skills England (2025) have projected total employment increases for IS-8-relevant occupations to 2030; in Figure 4, we map these increases against the percentage increase in apprenticeship starts needed to offset lost visas and the time taken to complete an apprenticeship, showing projected employment increases by the size of bubbles. Several occupations occupy a worrying place on the chart. The “Construction and building trades (not elsewhere classified (n.e.c.))” occupation currently receives 56 percent of new entrants from visas; replacing them will require an increase in starts equal to more than 100 percent of current training levels, with workers taking a year and a half to complete training. At the same time, demand is expected to rise by 17,000 workers by 2030, thanks in part to a government housebuilding effort—throttling labour supply would worsen delivery challenges. A similar story is true of “Welding trades,” “Pipe fitters,” “Electrical trades (n.e.c.),” and “Engineering technicians,” among others.

Figure 4.

Skills England’s projections don’t cover all of the MAC’s considered occupations. They also, importantly, don’t factor in replacement demand generated by retirements; instead they assume that current supply fully meets replacement needs. This is a major gap, given that replacement demand is estimated by the Working Futures project to typically outstrip expansion demand by a factor of around seven to one (Department for Education, 2020). As can be seen in Figure 3, even with visas, total employment in some shortage occupations is currently falling. In our index for visa inclusion we use modelled replacement demand estimates created by the Skills Imperative 2035 project as a proxy (Department for Education, 2024), but these are only indicative at the occupation level and have not been updated since August 2024.[iv]

Competition between IS-8 sectors will also make it harder for any individual sector to meet demand: the same sets of skills are often flagged as required by multiple sectors, and may be highly valued for strategic purposes (including in the defence sector). At the same time, historically, many of the occupations on the MAC’s list are not, comparatively speaking, abnormal in their use of visas or their job advert intensity.

In Figure 5 we show use of Skilled Worker visas as a percentage of all new entrants (apprenticeship completions and visas) versus job adverts per 100 employee jobs across 2021/22 – 2024/25, colouring by pipeline and inter-sectoral competition. We define a quadrant (in pink) with a threshold at the average for each axis across all UK occupations in the “Predominantly apprenticeships” or “Mixed pipeline” buckets. For much of the time period, many listed occupations are not above the UK average in either metric. This should, however, be taken with a pinch of salt: Lightcast data is gathered from online job listings, which may not be a good proxy for all occupations; multiple occupations are listed as having job advert intensity near or at zero.[v]

Figure 5.

Occupations outside this quadrant might have less need of retaining discretionary visa access. It is also possible, however, that these may be strategically important skillsets whose loss, even for a short period, would have major ramifications for policy goals with spillover effects. In these cases, the effects of the lag between loss of visas and hypothesised domestic catch-up could outweigh the government’s interest in incentivising domestic training.

This is likely to be true, for example, for “Electrical and electronic trades n.e.c.” (SOC 5249), the occupation within which lineworkers sit (Skills England, 2026). Lineworkers are crucial for grid buildout, now ramping up (National Grid, 2026). They are reported by industry sources to be in critically short supply (Energy Networks Association, 2025; Millard and Kerr, 2026): new demand, combined with retirements, exceeds current supply (Harkness et al., 2025). Their work has major implications for decarbonisation, energy prices, and wider fiscal outcomes (Arup, 2026). This is especially the case in the context of spiking energy prices, when curtailment of generation due to grid buildout delay becomes relatively more expensive. The rapid increase in demand is reflected in an increase in Skilled Worker visas from 13.5 percent of all observed new entrants in 2021/22 to 46.6 percent in 2024/25. At the same time, they are also reported to have only 0.5 job adverts per 100 employee jobs, well below the market-average quadrant cut. Even if the data could be considered reliable, the importance of the occupation’s contribution to broader outcomes is such that ensuring that supply can respond rapidly to demand is likely to be a worthwhile policy choice.

Occupations within the Clean Energy sector

In its impact assessment of the 2025 visa changes, the Home Office (2025c: 14) argued that loss of access to visas would have “no meaningful impacts” on decarbonisation goals. If anything, the Home Office proposed, loss of visa access would have carbon benefits due to a reduction in travel.

This is far from given. The Energy Security and Net Zero (ESNZ) Committee of the House of Commons, in an inquiry into the state of the UK’s clean energy workforce situation, concluded that domestic supply alone would not meet needs in the timeframes required (ESNZ Committee, 2025). Targeted use of visas to supplement the domestic workforce, therefore, is crucial to the UK’s ability to achieve its decarbonisation goals. This conclusion echoes previous findings by the National Energy Systems Operator, the Climate Change Committee, and numerous relevant stakeholders (Huckstep and Dempster, 2025). In its response to the inquiry, the government reiterated its intention to incentivise domestic training, but also acknowledged the need to include relevant occupations on the new shortage list (HMG, 2026).

Modelling by CGD suggests that on a per-worker basis, access to visas can make a major decarbonisation contribution (Huckstep and Harnoss, 2026). Under the assumption that each additional worker alleviates a binding labour constraint on installations, an electrician who had arrived in the UK in 2024 and worked through to 2032 would allow the abatement of 700 tonnes of CO₂.[vi] This is, for comparison, the equivalent of planting over 3,000 trees and leaving them to grow for 50 years. A heating technician installing residential heat pumps would have a still bigger carbon impact (Figure 6). They would also allow household savings and reduce household use of imported fuel, an attraction given continued supply volatility (Rathi, 2026; Rocha and Mazneva, 2026). Contrary to the Home Office’s suggestion, moreover, the carbon costs of travel from likely countries of origin are far outstripped by the worker’s abatement contribution.

Figure 6.

It is therefore important that access to visas is maintained for sectors key to decarbonisation. At the same time, however, this is an area in which international recruitment could cause a “brain drain” from countries of origin, harming their own ability to decarbonise and expand energy access. In some cases, the UK’s recruitment of workers could have, on net, a harmful effect on global decarbonisation: higher grid carbon intensity in countries of origin mean that if workforce constraints are equally binding there as in the UK, international recruitment may harm country of origin decarbonisation more than it helps the UK’s (Huckstep and Harnoss, 2026). It is not inevitable that these workers’ departures leave a decarbonisation gap—but we recommend that the UK assess whether this is a risk, and partner with these countries to support mitigatory training if so (see Dempster and Huckstep, 2024). Given that decarbonisation is a global public good, bottlenecks elsewhere caused by UK international recruitment could be detrimental to UK interests.

Which occupations should remain on the TSL?

The outstanding question is that of prioritisation. The response must be informed by multiple factors. To answer it, we create an index synthesising a collection of key indicators to provide an indicative ranking of occupations for retaining visa access. We caution, however, that it is only indicative: key data is missing, which may outweigh the data included.

We build the index from two subindices: a ‘core visa need’ index comprising predominantly labour supply indicators, and a ‘forward pressure’ index made up of current and modelled labour demand indicators. This latter index contains less reliable data (the patchy Lightcast data and projections or modelling), so we set its weight in the default index to a modest 0.15.[vii] We summarise these indices in Table 3; the full methodology is available in the associated note.

Table 3. Components of the visa prioritisation index

However, as both the Immigration White Paper (Home Office, 2025a) and the Industrial Strategy (DBT, 2025) note, visa access is also a matter of securing labour outputs with strategic importance. This is fundamentally the approach taken by the new TSL system. We therefore apply a “strategic overlay” multiplier before the final ranking. This reflects that the more sectors to which an occupation is relevant, the more its labour outputs can be presumed to be valued; and that some sectors may be more strategically important than others. The initial weights applied to the multiplier are the following:

• A 0.03 multiplier for each sector after the first
• A 0.07 multiplier for relevance to either the Clean Energy or Defence sectors

These necessarily include a degree of subjectivity, hinging on how one prioritises between policy goals, how one assesses data reliability and inclusion, and how one weights different components. We assume that Clean Energy and Defence will currently be prioritised above other sectors. Users can edit these parameters, and all others, to model for other prioritisation and to weight components differently. There is, in addition, a concern regarding missing data: we do not, for example, have data on the fiscal effects of workers’ labour outputs.

We show the visa prioritisation index in Table 4. We flag two possible issues: some occupations have inadequate data coverage, and are excluded in the default setting; and some occupations have a small workforce (fewer than 10,000 employed), meaning that some indicators will have higher volatility and caution may be needed in interpreting results.

Table 4.

Following the prioritisation index in its default settings, “Metal plate workers” is the occupation with the strongest case for retaining access to visas (of the 64 with adequate data coverage): it has high visa intensity, long training duration, and high wage pressure, and is also a priority occupation for the Clean Energy sector. At the other end of the list is the “Science, engineering and production technicians” occupation, which despite its relevance to three sectors scores low across multiple indicators. “Marketing associate professionals”, the shortlisted occupation with the most visa grants since 2021, ranks only 43^rd (despite the fact that its mixed pipeline status may mean that domestic labour supply is underestimated). Four occupations have inadequate data coverage to be ranked at all.[viii]

The index is broadly stable under substantial alternative specifications, including changes to weighting choices and the removal of key components. It is useful as an indicator of broad bands of occupations likely to have a strong case for retaining access to visas. Predictably, however, there is volatility at the level of the individual ranking, and differences between adjacent ranks should not be over-interpreted. Sensitivity checks are provided in the methodology.

Conclusion

Without workers, the government’s Industrial Strategy priorities will not be met. Tightening visa access to incentivise domestic investment in training, though understandable, creates a material delivery risk. If visa access is restricted but domestic training does not scale to meet the triple pressure of policy-stimulated demand, replacement of retirees, and loss of visas, crucial targets will not be met. Moreover, it is by no means a given that the loss of visas will incentivise new apprenticeships: it may instead just lead to a slow-down in operations. (In its report, the ESNZ Committee (2025) recommended that the government set out options for conditionality, publishing studies of impacts; the government partially agreed but did not commit to do so.)

Our findings suggest that the strongest case for continued sponsorship (or recruitment via the TSL for a “bridging” period) is where (i) Skilled Worker visas provide a high proportion of inflows of new trained labour; (ii) domestic training has long lead times (and possibly low completion rates); (iii) demand is rising due to time-limited policy interventions; and (iv) the work that would go undone in the absence of workers is strategically important. In these cases—as, for example, for “Metal plate workers,” “Ship and hovercraft officers,” or “Welding trades”—visa use may be unavoidable if targets are to be achieved. Any configuration of migration and skills policy should take into account that domestic supply responses will lag the loss of international recruitment by several years.

This paper has set out new data to help prioritise which occupations retain access to sponsorship. The aim for government should be to balance two key goods: building domestic training pipelines while sustaining delivery of strategic outputs. In so doing, it should not forget that some—those related to the Clean Energy mission—have impacts beyond its borders.

[i] These occupations are “Dancers and choreographers” (3414); “Musicians” (3415); “Information technology trainers” (3573); and “Steel erectors” (5311).

[ii] Note that when the forward pressure module is toggled on and the Lightcast data is included, demand may be understated for some occupations that are poorly captured by online job-posting data, and the index may thus under-rank them.

[iii] This modelling assumes that this “marginal” worker alleviates a binding labour constraint to allow additional installations that would otherwise not occur; for more detail see the methodology in Huckstep and Harnoss (2026).

[iv] Note that these modelled estimates are highly uncertain; for most shortage list occupations the modelling suggests a replacement demand (from retirements, mortality, inter-occupational mobility, etc.) of more than 25 percent, and for some of more than 100 percent, of total 2024/25 occupational employment.

[v] This is because many manual occupations may not advertise posts on LinkedIn or similar platforms, limiting their legibility to web-scraping tools.

[vi]Apprenticeship data is held by academic year, and cannot be reliably converted to calendar year; we therefore harmonise visa data with apprenticeship data by academic year.

[vii] Note that Skills England’s mapping of expected education levels deviates from that used by the MAC.

[viii] The TSL provides time-limited access to visas for specific shortage roles. This list does not provide salary discounts. Salary discounts are only available to occupations on the ISL, for which the minimum salary is 80 percent of the usual minimum rate; not all TSL occupations are also on the ISL. The ISL is expected to be phased out in future changes and replaced by the TSL; as of March 2026 they are coexistent and do not fully overlap (Home Office, 2025b).