Skip to content

Economic Prosperity

A Fair Deal for All: Delivering Flexibility and Protections for a Modern Workforce

Paper21st September 2022

Our Future of Britain initiative seeks to reinvigorate progressive politics to meet the challenges the country faces in the decades ahead. Our experts and thought leaders will set out a bold, optimistic policy agenda across six pillars: Prosperity, Transformative Technology, Net Zero, Community, Public Services and Britain in the World.

Chapter 1

Executive Summary

Technology and the new employment practices it gives rise to are radically changing the world of work in the UK. We estimate that as many as 20 per cent of workers are now in non-traditional roles, ranging from freelancers and sub-contractors to agency and gig workers.

These new positions offer valuable flexibility to both businesses and individuals – but the system of worker protection we have in place has failed to keep pace with these profound changes. As a result, individuals and businesses face more uncertainty over their relationship and mutual obligations. In turn, this leads to greater precarity for individuals and an unlevel playing field for business as good employers face being undercut by less scrupulous ones. Left unaddressed, both outcomes will sap our already dismal rate of productivity growth and damage prosperity.

The legal and institutional frameworks that underpin workers’ rights need to be modernised. They need to strike a more coherent balance between workers’ rights and the advantages of flexibility accruing to both firms and employees in a labour market being rapidly restructured by technology. Failure to do this will mean any likely progress is piecemeal and reliant on slow-burn action in the courts.

There is growing speculation that the government favours a radical deregulatory drive, even though the UK already has one of the least regulated – and least well-regulated – labour markets. Some on the left, in contrast, favour heavily interventionist approaches that inevitably restrict the flexibility of both businesses and individuals.

Instead, this report shows that the best way to support both businesses and workers is through a new set of minimum employment rights that deliver flexibility with protections. Providing a clear floor of rights for workers engaged in any form of work, independent of their employment status, will enhance individuals’ control over their own work. Crucially, it would also give more certainty to businesses, without which they will be reluctant to invest or train their workers, while preserving the flexibility they need to respond to rapidly changing markets.

A system of minimum employment rights will help those in most need while also having the virtue of simplicity. It avoids the complex legal battles and economic damage that would stem from a wholesale reclassification of employment status. But it also prevents a deregulatory race to the bottom on wages and standards that most British businesses do not want.

To deliver flexibility, fairness and dynamism in the economy and labour market, we advise the following steps:

Recommended Actions

  • Maintain the existing three categories of employment status (employees, self-employed individuals and workers): shifting to a two-tier system across the economy would hurt businesses, workers and consumers.

  • Introduce a pre-approval process to determine employment status and a fast-track process to resolve disputes, which would improve predictability for businesses.

  • Introduce a set of core minimum employment rights that apply to anyone engaged in work.

  • Create a single enforcement body with powers to advise businesses and individuals on employment status, recommend updates to the minimum employment rights, inspect workplaces, and bring prosecutions relating to the minimum wage, exploitation and discrimination.

  • Understand and address the complex interactions between the system of employment rights and the skills, welfare and tax systems.

Chapter 2


New technologies have created possibilities for new business models to be developed and scaled globally. They have also given individuals the opportunity to work far more flexibly than they would have been able to in traditional careers and business models.

The ability to choose where and when they work has helped some people access the labour market. For others, it has meant the ability to organise work around care responsibilities, earn money while studying or even switch careers.

These ruptures in the labour market have significant implications for the economy and society. But, if the government fails to take a strategic approach to these changes, workers risk getting a raw deal and businesses could be undercut by poor working practices, both of which would undermine productivity growth.

For many, non-traditional work has proved less flexible than they had hoped and has led to more precarious work. Economists have long understood that technological shifts, if left solely to market forces, typically strengthen winner-takes-all dynamics for the few firms and individuals that can leverage technology to operate at scale, and distort labour markets and competition for those that cannot.

Those firms that cannot compete on scale or value are forced to compete on lower wages and standards. For most individuals, this means depressed wages and less control over their working conditions and prospects.

Even as non-traditional work has grown, our system of determining and enforcing employment rights has adapted far too slowly, leaving individuals uncertain about their rights and businesses facing an unlevel playing field. The few pockets of progress have either taken too long, typically after lengthy court battles, or been too limited and concentrated in specific parts of a handful of industries (for example, voluntary, collective agreements with self-employed individuals in the food-delivery, arts, media and entertainment industries).

This piecemeal, patchwork of protections for non-traditional workers is creating problems for the economy and society, including:

  • Eroding worker power and so contributing to inequality and insecurity both in terms of pay and working conditions.

  • Increasing uncertainty for businesses and individuals, which is limiting their flexibility and dynamism.

  • Hampering productivity as firms underinvest in training and so constrain an individual’s ability to develop human capital over time.

The traditional solutions of the far left and the far right are economically damaging, as they both pit workers against businesses in a zero-sum game. Heavily interventionist approaches adopted in other European countries prefer to treat all people in non-traditional work as “employees”, restricting the flexibility of both individuals and business. In contrast, radical deregulatory approaches to employment law tend to create a competitive race-to-the-bottom on wages and standards, largely because the UK already has one of the lightest and least well-regulated labour markets. Neither of these approaches supports prosperity, provides fairness in work or allows businesses and individuals to reap the benefits of new technologies.

The following section categorises the various non-traditional occupations – from gig and freelance work to on-demand and seasonal jobs – allowing us to quantify the workers in each category as well as analyse their experiences of work and how these have changed over time. We then assess the factors affecting these individuals’ bargaining power. Next, we look at how our system of determining and enforcing employment rights is creating uncertainty for both businesses and individuals, and how this is affecting investment in human capital.

Lastly, we set out policy recommendations for strengthening our system of employment rights by creating a core set of minimum worker rights, providing greater predictability for businesses and individuals, and improving the enforcement of employment law in the UK.

Chapter 3

The Rise of Non-Traditional Work

The International Labour Organisation (ILO) and the Organisation for Economic Co-operation and Development (OECD) define traditional employment as work that is full-time, indefinite and part of a subordinate bilateral relationship.[_] Typically, permanent, part-time employment is excluded from such formal definitions of traditional employment, as researchers are interested in understanding the dynamics that lead to permanent part-time work.

Several studies have used a similar analytic framework to quantify specific aspects of non-traditional work in the UK, for example looking specifically at agency workers, gig workers, insecure workers or individuals on zero-hours contracts.

These analyses, however, provide only a patchwork and piecemeal view of the total number of people in non-traditional work. It has therefore not been possible to understand how non-traditional work is changing over time or how the total breaks down into discrete and distinct (that is, into mutually exclusive and completely exhaustive) categories.

We take a more nuanced view of traditional employment and non-traditional employment. In the UK, an indefinite, subordinate, bilateral employment relationship has clear implications in the legal and tax system that vary only marginally between full-time and part-time employees.[_] Due to the fact these entitlements are common, we consider permanent, part-time employment to be a traditional form of work as well.

In turn, we classify anything that does not meet this definition of traditional employment as non-traditional work.

Of all people in work over the age of 16, we estimate that 81.6 per cent were traditional workers in 2019–2020. As part of this group, we estimate that 80.4 per cent are traditional employees and 1.2 per cent are business owners: they are self-employed with employees.

Using this definition of traditional employment, we find that in 2019–2020, 18.4 per cent of adults – 5.5 million people – who did any hours of work in the UK and were aged over 16 were in non-traditional work.

The Different Types of Non-Traditional Work

To better understand the different types of non-traditional employment, we categorise it using the structures of working relationships, in particular the extent to which the relationships are non-subordinate, non-bilateral or impermanent. Using these characteristics allows us to zoom out from individual factors that have always affected a worker’s status and bargaining power, such as skills and experience, and investigate the wider control factors at play instead.

Using the latest wave of the Understanding Society survey,[_] we categorise non-traditional work into three groups based on how its structural dynamics – the degree to which it is non-bilateral, impermanent or non-subordinate – affect an individual’s agency over accessing work (see the Annex for full methodological details):

  • Hustlers (non-subordinate, bilateral working relationships): Hustlers identify themselves as self-employed but have no employees. When asked to describe the nature of their self-employment, they refer to themselves as sub-contractors, freelancers or solo-business owners with no employees.The essential characteristic of this group is that they are meant to have the greatest degree of agency over what work they do. In particular, they carry out work set by contracts struck between two parties (for example, two separate businesses) where they are one of them. Yet while there is no subordinate employer-employee relationship, there may often be a subordinate customer-client relationship that echoes traditional employer-employee relationships (for example, disguised employment) and therefore limits the individual’s agency.

  • Giggers (multi-party structure determines access to work): The defining characteristic of this group is that their access to work is typically determined through a multi-party structure, such as a digital labour platform, an agency or another tripartite-type relationship (for example, an umbrella company), and the third party often exercises a degree of control over the term of the work. We have allocated individuals to this group if (1) they are paid by someone other than the business or organisation they work for; (2) if they describe their work as agency work; or (3) if they work more than 50 per cent of their total working hours on digital labour platforms.

  • Impermanents (impermanent, bilateral and subordinate roles, creating uncertainty over the next job): Impermanents are typically in bilateral, subordinate but impermanent working relationships. As such, their work tends to be “on-demand” – for example, zero-hours contracts, for a fixed period (for example, academic lecturers) or seasonal (for example, farm workers).

Figure 1

Categorising non-traditional work

Source: TBI calculations using Understanding Society

A Note on Definitions

This paper is fundamentally about people and their relationship to their work. In our writing we try to refer to them as “individuals” and, if necessary, as “workers”. To avoid any confusion with the legal terms, where we want to refer specifically to an individual’s employment status, we attempt to use the terms “employee”, “worker” or “self-employed”. We do our best to avoid confusion over terminology, but confusion and uncertainty over terminology – and the rights that flow from employment status – are very much part of the wider problem.

Figure 2

Non-traditional employment has grown over the past decade

Source: TBI calculations using Understanding Society and ONS population statistics

What Kind of Work Do Non-Traditional Workers Do?

Hustlers tend to fall into four broad groups. The first are typically self-employed tradespeople working in construction, such as plumbers, electricians or roofers, or in cleaning occupations. The second broad group are those in creative, literary and media occupations working as artists, authors, translators, actors and entertainers. The third broad group are personal-service providers, such as drivers, therapists, hairdressers or caregivers. The fourth broad group are individuals dispersed across a range of professional freelancer and contracting services, including information and communications technology (ICT) professionals and consultants.

Giggers are in similar occupations, but this time predominantly drawn from road-transport drivers, individuals in teaching and education, and health and care professionals. These include, among other professions, care workers and home carers, ambulance stuff, nursing assistants, childminders and playworkers as well as nursery workers and assistants.

Finally, impermanents, the largest group of non-traditional workers, are those working on-demand, such as hospitality professionals including bar staff, waiters and kitchen and catering assistants; individuals working on seasonal contracts, including in agriculture; and junior academic staff such as social scientists and humanities scholars working to the academic season.

Zooming in on Digital-Platform Labour

A subset of our giggers group comprises individuals who access work through digital platforms. Over the past decade, accessing work through digital labour platforms has become a prominent part of non-traditional work. Typically, there are two forms of digital-platform labour: location-based platforms that allocate work to individuals in a specific geographic area (for example, Uber and Deliveroo), and web-based platforms where work is outsourced to individuals who can be geographically anywhere (for example, UpWork).

There has been a range of estimates on the size and growth of the platform-based economy, with the Royal Society of Arts (RSA) suggesting it was about 4 per cent of workers in 2018 and the Trades Union Congress suggesting giggers were about 14.7 per cent of workers in 2019.

According to the Understanding Society survey used in this analysis, the number of workers on digital labour platforms in some form is far smaller, with approximately 520,694 or 1.76 per cent of workers aged over 16 engaging in work through a location- or web-based platform. Figure 3 shows the different types of tasks performed by the UK’s digital-labour-platform workers. This estimate is broadly in line with a bottom-up analysis of the numbers of workers on digital labour platforms.

Figure 3

Different types of tasks performed by the UK’s digital-labour-platform workers

Source: TBI calculations using Understanding Society

Who Are Non-Traditional Workers?

Overall, we find that 18.4 per cent of all people in work over the age of 16 in the UK – 5.5 million individuals in total – are in non-traditional work. Of those 5.5 million individuals, 860,000 (16 per cent of all non-traditional workers) are in the South East, 720,000 (13 per cent) are in London, and 610,000 (11 per cent) are in the East of England.

As a percentage of regional or national employment, non-traditional work features not just in these three geographies, but also in the North East (at 20.3 per cent of regional employment), in the South West (20.2 per cent) and in Wales (18.7 per cent).

Figure 4

Different categories of work as a share of regional/national employment

Source: TBI calculations using Understanding Society

Looking more closely at the geographic distribution of non-traditional work, we see that the three categories of non-traditional work – the hustlers, the giggers and the impermanents – are concentrated in London, the South East, the East of England and the North West. Other regions, however, tend to have relative concentrations of specific types of non-traditional work. The South West, for example, is home to 11.3 per cent of all hustlers (typically solo-business owners and freelancers), while Yorkshire has 11.7 per cent of giggers (primarily agency workers).

Figure 5

Distribution of non-traditional workers across each region/nation

Source: TBI calculations using Understanding Society

Looking at where non-traditional workers carry out their work, we see that it is primarily hustlers who have embraced the remote-working trend, as their age and occupational profiles suggest they are more likely to have the experience and professional networks to access and conduct work from home. Despite having non-traditional roles, 63 per cent of impermanents and 50 per cent of giggers work at their employer’s location, reflecting the bilateral and subordinate (if impermanent) nature of their work. In addition, both giggers and impermanents are relatively more likely to work at more than one location.

Figure 6

Location of work

Source: TBI calculations using Understanding Society

Who Are Those in Non-Traditional Work?

When looking at the skills composition, 49 per cent of traditional workers, hustlers and impermanents have a degree versus 47 per cent of giggers. For this latter category, there is unlikely to be significantly different levels of bargaining power due solely to differences in education levels because there does not appear to be sufficient difference in the share of graduates for that to be the case. This is important because non-traditional work is often considered to be work chosen precisely because insufficient skills make it harder to find more permanent forms of employment.

The impermanents are relatively more likely than traditional workers to have attained their A-levels. Giggers and hustlers, meanwhile, are relatively more likely than traditional workers or impermanents to have other qualifications or have no degree at all. One of the reasons individuals in non-traditional jobs might have relatively lower educational levels is that such a job allows them to balance work with other activities such as education or training.[_] As discussed in the next section, impermanents and giggers are also more likely than both traditional workers and hustlers to be engaged in additional training, including at college or university.

Figure 7

Workers’ qualifications

Source: TBI calculations using Understanding Society

Figure 8

Age of workers

Source: TBI calculations using Understanding Society

Hustlers are on average 50 years old, seven years older than the average worker in traditional employment. Giggers are on average 43, the same average age as traditional workers, but the age distribution is much less even. There are more under-34s and over-65s in the giggers category than in traditional work. Impermanents, mostly engaged in on-demand and seasonal work, are disproportionately younger. Over half of those in seasonal and on-demand contracts are aged 16 to 24.

Figure 9

Ethnic composition in traditional and non-traditional types of work

Source: TBI calculations using Understanding Society

Ethnic disparities are also of interest too, as divides along ethnic lines when it comes to access to work may also have implications for racial disparities in income, wealth and career opportunities. While the ethnic composition of individuals in traditional work reflects that of the general population, there are differences across the three categories of non-traditional work. In particular, giggers and impermanents have relatively more mixed, Asian, black and other ethnic-minority workers. Given the earnings profiles of these roles, this could be a reason for concern.

Figure 10

Gender shares in traditional and non-traditional types of work

Source: TBI calculations using Understanding Society

Having classified and quantified non-traditional workers and analysed their experiences of work, we now will assess the factors affecting these individuals’ bargaining power, before looking at how our system of determining and enforcing employment rights is creating uncertainty for both businesses and individuals, and how this is affecting investment in human capital.

Chapter 4

Sources of Non-Traditional Workers Unequal Bargaining Power

Surveys of non-traditional workers suggest individuals choose their jobs primarily because they offer greater autonomy[_] and flexibility[_] than traditional employment, allowing them to organise work around other commitments.

However, despite non-traditional employment carrying the promise of additional flexibility, it is only hustlers who seem to be genuinely enjoying agency over their hours. Giggers and impermanents, on the other hand, seem to have no more control over their working hours than traditional workers.

Figure 11

Only hustlers have greater control over their hours

Source: TBI calculations using Understanding Society

The relative lack of control for giggers and impermanents is a particular issue because they are relatively more likely than traditional workers to be paid by the hour or by the task. Despite being almost twice as likely to be paid by the hour, 70 per cent of impermanents and 48 per cent of giggers say they do not receive additional pay for extra hours.

Figure 12

Salaried or paid by the hour?

Source: TBI calculations using Understanding Society. The question is not asked to individuals who are self-employed. Self-employed individuals exist in all three categories of non-traditional work and comprise the entirety of the hustler category. As a result, these numbers refer only to individuals in non-traditional employment who have employee status and there are no results for hustlers.

Here, the experiences of individuals in non-traditional roles suggests they have less control over their pay and working conditions than afforded by the employment status they might have been given. Our discussions with businesses, tech platforms and unions suggest individuals’ bargaining power is more likely to be determined by a variety of other factors, including their ability to navigate technological change, weak enforcement of labour laws and lengthy legal processes to determine rights. These bring sets of problems for workers and employers, and we explore each of these in turn.

The Individualisation of Employment Relations

The UK’s employment and industrial relations system has not kept pace with changes to the economy and labour market over the past 40 years, a gap increasingly exposed by the application of new technologies to working practices. A key problem is that the decline in trade-union power has not been compensated for by the emergence of a sufficiently robust alternative system for policing workers’ rights.

Prior to the 1980s, workplace relations had been dominated by powerful trade unions. These relied on their considerable industrial muscle to induce employers to bargain with them and eschewed efforts to build an alternative system based on the statutory positive rights common in other European countries.

Union membership reached a high of 55 per cent of the labour force in 1979 – the year Margaret Thatcher was elected prime minister – but the influence of industry-level bargaining and the wages councils at the time meant around 85 per cent of the workforce was covered by collective pay-setting mechanisms.

Thatcher saw overmighty unions as an obstacle to company restructuring, and therefore a brake on productivity growth, and oversaw a concerted attack with three main prongs. First was an assault on union power spanning several Acts of Parliament which restricted their ability to organise, recruit and take industrial action. Second was encouragement of individual – rather than collective – rights at work. Third was assertion of the primacy of relatively minimalist labour law over other mechanisms for employee protection.

The impact was profound. Union membership collapsed to around 24 per cent of the labour force in 2020. In the private sector, where the wage-premium from union membership disappeared for the first time in 2020, the proportion is less than half of this. In essence, the union-dominated system has been replaced by ever smaller pockets of collective bargaining in a rising sea of individualised negotiations and unilateral employer discretion. However, the minimalist legal framework that replaced collectivism is becoming increasingly detached from the realities of work in the new economy.

New Labour enacted two Employment Relations Acts, in 1999 and 2004, which strengthened employment protection and introduced a statutory route to union recognition. There was help for low-paid workers through tax credits, a raft of family-friendly policies, and a national minimum wage which introduced a pay floor for the first time.

Yet this has not reversed the long-term decline in union membership. Much progress was then undone under the coalition and Conservative governments, which weakened protection against unfair dismissal and clamped down on unions’ abilities to organise and take strike action.

Many of New Labour’s most significant reforms, for example on working time and part-time work, came indirectly through EU regulations imported after the UK signed the European Social Chapter, a route no longer available after Brexit. Indeed, worker protections enshrined in EU law, such as the 48-hour week, now face being ripped up as part of the government’s post-Brexit overhaul of regulations.

Thus, from the point of view of workers’ rights, the UK arguably has the worst of both worlds: neither strong, cohesive trade unions to protect workers through collective action nor a robust enough system of legal protections for workers whose bargaining power in the labour market may be weak.

Ability to Understand and Navigate Technological Change

A particular set of problems arises with the growing prevalence of algorithmic management, whereby human managers are replaced by artificial intelligence (AI) systems. Algorithmic management reduces costs and can prevent unconscious bias. But it entails increased surveillance of workers and drives them to complete their tasks in less time, resulting in the intensification of work, with uncertain impacts on quality and unknown impacts on unmeasured outcomes such as mental health and wellbeing. In response, unions such as Prospect have created campaigns to advise members and reps on both the positive and harmful uses of technology in the workplace.[_] However, most workers are unaware of the data being collected about them or how it is used to monitor their performance, and they lack the means of getting feedback or seeking redress.[_]

By eroding workers’ ability to control how their work is structured, algorithmic management can be demoralising for workers and impact their mental health.[_] For example, when platforms combine algorithmic management with demand-forecasting algorithms, the result is a “gamification” of work allocation that nudges people to work unsocial and disruptive hours through the use of targeted micro-incentives.[_]

Weak Enforcement of Existing Labour Laws

As workplaces have fragmented and lines of accountability become increasingly blurred, workers’ awareness of their rights at work remains low.[_] This information problem creates incentives for some employers to exploit individuals to gain a commercial advantage, putting competitive pressure on other companies to follow suit.[_]

For example, analysis by the Resolution Foundation[_] shows that workers in atypical work are particularly vulnerable to violation of their employment rights: 409,000 of all jobs were paid below the national minimum wage in 2019.[_] Even where underpayment is a genuine mistake, the onus is still on individuals to spot and challenge this and, in the interim, individual businesses are able to exploit weak enforcement.

The solution is better enforcement by public authorities. Yet enforcement is currently split between four separate bodies: His Majesty's Revenue and Customs (for the minimum wage); the Gangmasters and Labour Abuse Authority; the Employment Agency Standards Inspectorate; and the Health and Safety Executive. As well as the above, workers can also act through employment tribunals. However, the latter can be expensive to initiate, and funding for the aforementioned public bodies has been cut, with regulators told to take a “light-touch” approach to enforcement.[_]

Lengthy, Detailed Court Battles

Workers’ rights under English labour law are primarily linked to their employment status. But it is employers who decide what status is afforded to any given individual. If employers knowingly or inadvertently attribute the wrong status to an individual, the onus is on the latter to seek redress. In the event of a dispute, it is the courts that decide how a given individual’s circumstances should be treated in law.

Individuals have periodically used this channel to successfully overturn incorrect decisions about their employment status. For instance, earlier this year the Supreme Court decided to reclassify certain Uber drivers’ employment status.[_] Among other things, the judges found that Uber had exerted significant control over drivers’ pay, workflows and performance, and hence the Uber drivers in question should be classified as “workers” rather than as self-employed. This entitled them to key rights such as the national minimum wage, paid annual leave, rest breaks and protection against unauthorised deductions from pay. The decision, however, did not apply to all Uber drivers, just those named on the court case.

Several barriers prevent more widespread use of the courts to achieve similar ends. The first is knowledge. Navigating the nuances of employment law is difficult, and it is hard for individuals to know whether they have strong grounds for a claim.

Second, litigation is resource-intensive. Although the Supreme Court struck down as unlawful a move by the Conservative-led government in 2013 to introduce tribunal fees (which saw cases plummet),[_] claimants still face many other constraints, including instructing lawyers and spending time supplying appropriate evidence to support their claims. And while some claimants benefit from trade-union support, this is far from ubiquitous.

Third, litigation takes time. Even where individuals know their rights and can go the distance in court, it can take years to get a resolution, particularly when cases are appealed. For example, more than four years elapsed between the first tribunal decision and the final Supreme Court judgement in the above Uber case.

All of this means that many potential cases go unheard. This is bad for the individuals concerned because they lack due recourse and therefore appropriate bargaining power. But it also means the law evolves more slowly: our system of law affords courts a lot of discretion to interpret legislation and fine-tune legal precedents in the face of new facts and nuances – but if cases are not being heard, there is less scope for them to do this, which ultimately undermines individuals’ bargaining power too.

Sporadic Progress Via Voluntary Arrangements

Our system of employment rights has tilted more heavily towards individual rather than collective rights over recent decades, but collective action still presents workers with another potential source of employment protection (see boxout above).

Self-employed individuals are excluded from the legislation that bestows key bargaining rights upon employees and workers. There are, however, a handful of voluntary collective agreements between self-employed individuals and businesses. Following a recent major agreement covering more than 90,000 self-employed Deliveroo riders, for example, the GMB will now collectively bargain on riders’ behalf regarding pay, while Deliveroo will consult the GMB on matters including health and safety, diversity and insurance.[_]

These sorts of agreements have been tucked away in other corners of the labour market, too – not least in the arts, media and entertainment industries. For example, in 2010, the Broadcasting, Entertainment, Communications and Theatre Union (BECTU) and the BBC agreed to negotiate self-employed freelance workers’ pay, hours and holidays. In 2018, BECTU signed an agreement with the Producers Alliance for Cinema and Television (PACT) to represent freelance technicians in the film industry about matters including night work and overtime. The actors’ union Equity has signed deals with entertainment-industry employers including the BBC, the Society of London Theatre and the UK Theatre Association, covering rights such as pay, rest breaks, working hours and pension contributions. And the Musicians’ Union has made deals on behalf of freelance players across more than 60 orchestras, including on issues relating to performance rates, overtime and travel costs.[_]

Despite these examples, however, collective agreements on behalf of self-employed individuals and businesses remain sporadic. The risk is that, as a result, our labour market ends up with a patchwork of rights for one group of individuals fortunate enough to be in a particular business or sector and nothing for a similar group of individuals in another organisation or sector.

Having assessed the factors affecting these individuals’ bargaining power, we will now look at how our system of determining and enforcing employment rights is creating uncertainty for both businesses and individuals, before setting out how this is affecting investment in human capital.

Unions No Longer the Primary Recourse When Workers Need Help

Currently, unionisation is prevalent in specific industries and the public sector. Most private-sector workers think their employer treats them and their colleagues fairly, and that if they did have an issue, they would talk to someone more senior or to colleagues to resolve the issue. If things really went wrong, private-sector workers believe they have options and so would look for another job.[_] Ultimately, they do not see unions as being the primary way of dealing with their individual workplace issues.

If they do think of unions, the public tend to be positive,[_] but associate them with older white men from a different time, a different sector or a different occupation.[_] Today, private-sector workers want unions to help them personally, not just to think of the collective, and to work with their employer, not against them. Modern union movements such as Community[_] and WorkerTech platforms like Organise.Network[_] have shown how, in an increasingly decentralised labour market, unions can modernise by using data and technology to empower individuals through personalised advice and a network of support as well as through collective action.

Chapter 5

The UKs System of Employment Rights Is Fostering Uncertainty

Determining individuals’ rights under English labour law is not always straightforward. At the headline level, there are three types of employment status: employees, self-employed individuals and workers. Conceptually, “workers” sit between employees and the self-employed: they provide work personally but neither share enough of the tenets of employment to be considered employees, nor operate independently enough to be defined as self-employed. As Figure 17 in the Annex illustrates, each of the three categories confers its own very different cushion of rights, so deciding any given individual’s status has material implications for them and their employer.

However, determining the boundaries between the three employment categories is complicated. Because legislation does not codify each form of worker status, the courts have developed specific tests to help decide where the barriers between them lie. For example, these tests include:

  • Control – the extent to which employers can direct individuals’ actions.

  • Substitutability – the ability to substitute another person to do the work.

  • Economic dependence – the degree to which individuals rely on employers for subsistence.

  • Mutuality of obligation – the extent to which each party agrees to provide/accept work on an ongoing basis.

None is conclusive on its own and cases turn on their own facts, which makes it hard for employers or would-be litigants to gauge whether precedents are likely to apply to them.

Legislators have provided a modicum of clarity when it comes to some forms of atypical working. For instance, there is bespoke legislation to protect fixed-term employees (but not workers), which means they should be treated no less favourably than comparable permanent employees – although with caveats and potential defences.

There are also separate laws for agency workers, who are given basic day-one rights and another layer of protection after 12 weeks’ service, again stemming from EU-level protections negotiated by Labour in government. But it is not always easy for individuals and employers to know when working arrangements are captured by these carve-outs. And in any event, many atypical working arrangements remain outside their purview.

As the new technologies and business models enable ever-more diverse and complex working arrangements, determining employment status is becoming increasingly hard. As discussed above, it has instead fallen to the courts to grapple with the nuances and clarify how these workers should be treated. Even then, there is only so much the courts can do: while new judgements can build on previous case law to refine the tests used to determine worker status, each case ultimately turns on its facts and decisions can take time, which leaves plenty of scope for distinguishing other working arrangements from legal precedents. In this context, it is hard for the law to keep pace with developments on the ground, which adds to the broader uncertainty about workers’ rights.

Why Uncertainty is a Problem

Broadly speaking, there are three main reasons why uncertainty about employment status undermines good working practices: 1) Individuals do not know what rights they have so are unable to fight for them; 2) businesses can exploit ambiguity to undercut competitors through regulatory arbitrage rather than providing better goods or services; and 3) businesses are wary of the legal risks of providing individuals with more rights.

1) Individuals Who Do Not Know Their Rights Cannot Defend Them

In our categorisation of non-traditional work, self-employed individuals exist in the hustler, gigger and impermanent groups. The dispersion and varieties of non-traditional work make it hard for individuals to know what their employment status and associated rights should be or whether they have changed over time. In turn, this makes it more difficult to enforce those rights. This is because employers first decide which category applies to any given worker, leaving the burden on workers to challenge any erroneous categorisations, either directly with their employers or through the courts. For them to do this effectively, they need to reliably ascertain where they stand (as discussed earlier).

The difficulty in determining employment status for tax purposes has compounded this uncertainty for businesses and individuals. Recent reforms have put the burden of determining employment status for tax purposes on the hirer and have created perverse incentives for UK companies to work with overseas freelancers, and for UK freelancers to work with overseas hirers, as the IR35 rules do not govern these relationships.[_]

Given the complexity of determining an individual’s employment rights, it is perhaps unsurprising that some individuals do not readily understand their legal position. It is hard to know exactly how pronounced this level of uncertainty is, but there are clear indications that many lack a reliable impression of their status.

2) Businesses Exploiting Ambiguity for Competitive Advantage

The second reason why uncertainty is problematic is that it enables employers to exploit that ambiguity for unfair competitive advantage. While most employers do not seek to actively circumvent the rules, the sheer diversity of non-traditional work means that our system of determining and enforcing workers’ rights creates incentives for employers to seek commercial advantage by exploiting gaps, uncertainties and complexities in employment law.

The rise of umbrella companies is a case in point. Undefined in legislation, these companies assume responsibility for workers’ employment rights from other parties in exchange for fees – typically from agencies which continue to broker work for those individuals – but do so unbridled from the legal responsibility to uphold their employment rights. Evasive practices like these are damaging to the individuals concerned because they increase precarity.

For example, ostensibly self-employed individuals who are really “workers” miss out on key rights, including the national minimum wage, paid holiday and rest breaks. Meanwhile, individuals working under umbrella companies often have lower pay than they would have otherwise commanded because intermediaries skim off their cut and some of these workers do not even know who their employer is.[_]

But evasive practices also hurt business: by avoiding regulation and associated costs, employers gain an illegitimate commercial advantage over their competitors; conversely, businesses who play by the rules are unnecessarily penalised, which risks precipitating a race to the bottom on wages and working standards to remain competitive.

3) Businesses Wary of Legal Risk

The third reason why uncertainty undermines workers’ rights is that it puts employers off from offering non-traditional workers more support. Whereas some employers seek to exploit uncertainty to derive commercial advantage, many others want to give the people they engage better rights. For instance, both small and medium-sized enterprises (SMEs) and large platforms[_] have gone on record to say they want to give more benefits, including sick pay, insurance, maternity pay, pension contributions and training. However, the same businesses also say that uncertainty around worker status can often be an unmoveable obstacle to such endeavours. This is because extending rights raises the legal risk of having to reclassify an individual’s status.

By deterring employers from offering these types of benefits, the uncertainty that currently exists around worker status further entrenches precarity among atypical workers, who would otherwise have enjoyed more financial security in their roles. But it also undermines productivity by stymying benefits like training, which could otherwise help to improve human capital by supporting non-traditional workers to improve the quality of their goods and services, increase their pay and potentially get a better job in the future.[_]

Chapter 6

A Drag on Investment

Non-traditional work offers individuals a flexible source of income and gives businesses a workforce more capable of adapting and flexing to shifting customer needs. Surveys of non-traditional workers suggest this is a mutually beneficial and efficient arrangement, especially over shorter time frames. Yet the uncertainty over employment status makes it less efficient over time, as businesses may be wary of offering non-traditional workers training because of the inherently temporary nature of their relationship, leaving these individuals less able to develop their skills. This risks exacerbating one of the central productivity problems of the UK economy: inadequate skills within the workforce. Addressing skills deficits, particularly at the lower end, could boost the economy by £20 billion a year according to some estimates.[_]

As shown in Figure 13, our analysis suggests that traditional workers are one-and-a-half to three times more likely than non-traditional workers to receive employer-funded training. While inadequate skills development is a problem for the entire economy, it appears to be a greater problem for non-traditional workers.

Skills development and training either in or outside work not only raise a worker’s productivity but are also important in enabling people to benefit from new job opportunities or better pay.

Overall, we find that there is not only a difference in the proportion of traditional and non-traditional workers engaged in training, but also a qualitative difference in the type, frequency and purpose of training workers across the different categories engage in.

Figure 13

Traditional workers are one-and-a-half to three times more likely to get employer-funded training

Source: TBI calculations using Understanding Society

Among traditional workers, 27 per cent reported they had received some training in 2019–2020, and were more likely to receive more, shorter, employer-funded training sessions than any of the non-traditional workers. This training is also far more likely to focus on aspects of the current role, including improving their skills for their job or for promotion, or health and safety.

In contrast, 30 per cent of impermanents were engaged in some form of training. On average, they are receiving 47 hours of training per year, compared with 37 hours per year for traditional workers, but are more likely to be studying at university, getting training to help them start a job or training for hobbies and leisure.

One-in-four giggers were engaged in training for, on average, about 39 hours per year. The training giggers were engaged in appears to be split: the bulk of it was paid for by an employer and so was more likely to be related to their current job but, like the impermanents, giggers were also receiving training at a university. Hustlers are the least likely to report they are engaged in training and are more likely to report that the training they are undertaking is to maintain their current professional qualifications. This fits with hustlers’ average profile of being slightly older workers striking out on their own.

While non-traditional workers may be less likely to get employer-funded training and typically have fewer spells of training, they are engaged in training and development outside work. This is especially true for younger impermanents and giggers, who appear to be taking non-traditional jobs in addition to attending university or other government-funded training. For these individuals the lack of training is less of an issue because the work is being done precisely to support their own personally funded training or education. Lastly, all categories of non-traditional workers are three times more likely than their traditional counterparts to do some form of training related to leisure or hobbies, supporting the idea that non-traditional work is a lifestyle choice for some individuals.

Figure 14

Traditional workers more likely to get training related to their job and progression

Source: TBI calculations using Understanding Society

Transitions Into and Out of Non-Traditional Work

The longer an individual spends in non-traditional work and does not receive training, the more likely they are to see a stagnation in their skills development and career prospects. To explore the degree to which non-traditional work is transitory, we looked at panel data gathered by Understanding Society to follow the same people over time to understand the flows into and out of traditional and non-traditional work between 2016 and 2019. Just 6 per cent of traditional workers in 2016 were in non-traditional roles by 2019.

Below, we plot the individuals who had changed economic status or employment type in 2019 compared to their status in 2016. Overall, 31 per cent of non-traditional workers transitioned from non-traditional roles to traditional roles between 2016 and 2019. Only 13 per cent of hustlers in 2016 transitioned into traditional roles by 2019. Giggers and impermanents were far more likely to move: 43 per cent of giggers in 2016 that were in traditional roles by 2019 and 56 per cent of impermanents shifted into traditional roles during this period.

Hustlers were also far more likely to stay in this category between 2016 and 2019, with almost 70 per cent remaining in this category of work. In contrast, only 35 per cent of giggers in 2016 were still in similar roles in 2019 and only 16 per cent of impermanents were still in similar roles after three years.

Figure 15

Flows between traditional and non-traditional work

Source: TBI calculations using Understanding Society

Lack of training appears to be an issue in the non-traditional labour market, and it is one that the UK needs to address given the significant skills challenges facing the economy and its impact on productivity. What accounts for this failure to train? The conventional reason is that, in very flexible, deregulated labour markets, a “free rider” problem exists. Firms are discouraged from investing in training their workers as they fear they will be poached by non-training rivals who will put some of the money saved into offering higher wages instead. Most, perhaps all, will opt not to train to avoid this possibility even if the industry (not to mention the country) might benefit from a better-trained workforce.

Our analysis of the legal uncertainties surrounding employee status points to an additional factor which compounds this problem. Many firms are likely to be wary of providing additional benefits such as training and sick pay to self-employed staff as they worry the courts could see this is as more reminiscent of an employment relationship and reclassify them accordingly. Start-ups and SMEs in particular could be wary of offering more rights and benefits because they fear the legal risk of being taken to court to reclassify employment status.

While the free-rider problem will be difficult and complex to solve, clearing up the legal uncertainties surrounding employment status could be a more straightforward way of getting firms to train more.

Chapter 7


Non-traditional jobs are growing in significance in terms of their relative share of the labour market and in their importance to individuals at various stages in their careers. As they do so, clear patterns around power dynamics are emerging in non-traditional work, which we define as “hustlers”, “giggers” and “impermanents”.

The various types of work that conform to these descriptions are not only different from traditional work, but there is also a great deal of variety within these categories. Moreover, the growing complexity of work patterns in non-traditional labour markets is reflected in the difficulties encountered in identifying the appropriate mechanisms for specifying rights and legal protections that match the realities of doing these jobs.

With many workers’ rights linked to an individual worker’s employment status, the issue is no longer just what rights individuals in different jobs should be entitled to but also what rights an individual should have when they engage in work of any kind.

The UK is not the only country facing these challenges, and governments, businesses, unions and individuals are taking a range of approaches to strengthen workers’ rights (see Annex C). This analysis, together with discussions that we have had with businesses, workers and unions, suggests that the UK needs to address these issues on several different levels:

Maintain the Existing Three-Tier System of Employment Rights

Our discussions with business, unions and individuals, coupled with our analysis, suggest that the UK’s three-tier system offers business and workers flexibility and is a competitive advantage for the country globally.

The alternative of a shift to a binary system of “employees” and the “genuinely self-employed” would face two problems. First, it would simply hurt individuals by restricting their flexibility, worsening their working conditions or pushing them out of the labour market. In Spain, for example, a new law that presumed all couriers on digital platforms were “employees” led to 8,000 job losses[_], lower wages and longer hours for gig workers,[_] while tech platforms have been forced to leave the market or shift their business models in order to continue operating sustainably.[_] Similarly, in Geneva, a court-ordered reclassification of delivery drivers led to two-thirds of drivers (some 2,000) losing their jobs, with almost 80 per cent of those still unemployed six months after the reform.[_]

Second, any reforms that seek to define the “genuinely self-employed” – such as those recommended by the Taylor Review of Good Work[_] – will struggle to draw hard boundaries around the term. This uncertainty risks businesses resorting to using outsourcing[_] tricks to exploit definitions (as has happened with “umbrella” companies and agency workers) or undermining the sustainability of business models. Ultimately, the impact on individuals is reduced flexibility, lower wages[_] or unemployment.[_]

Introduce a Set of Core Minimum Employment Rights That Apply to Anyone Engaged in Work

As discussed above, the push to get non-traditional workers additional employment rights through “employee” status has led to a rigid trade-off between additional rights for individuals and a loss of flexibility for both individuals and workers.

Rather than encouraging legislative and court battles over status, creating a set of minimum employment rights would put a floor of rights underneath all individuals engaged in work regardless of how they access work or their actual employment status. Doing so could also help enhance individuals’ agency over their work and their bargaining power.[_]

These core rights would be focused on five specific areas, some of which would be new rights for all three categories of employment status and some of which extend existing rights of “employees” and “workers” to the “self-employed”:

  • The right to a floor on pay. New ways of working make it more difficult for the self-employed to ensure they are adequately paid for their work. This core right to a floor on pay would extend to the self-employed the following rights: access to the national minimum wage (which currently covers only “employees” and “workers”), protection against unlawful pay reductions, and the right to request sick pay and parental pay.

  • New digital rights. Matching shifts in technology, all individuals should have a core set of digital rights, including the right to access their data at work, the right to ask for a human review of AI management or work practices, and the right to take their data from one platform to another.

  • The right to representation. While the “self-employed” already have the right to join a trade union, a union’s scope to support the “self-employed” is more restricted than it is for “employees” or “workers”. Policymakers should consider extending some of the negotiating rights associated with statutory recognition to bargaining units that comprise self-employed individuals who operate under the strict restrictions of focusing only on pay and working conditions.

  • The right to health, safety and wellbeing in the workplace, including new rights to switch off for remote/anywhere jobs, the right to adequate safety training for new roles and the right to a safe working environment (including maximum temperature thresholds).

  • The right to request benefits. Individuals would have the right to request a limited set of benefits, including training and pension contributions.

Figure 16

A core set of minimum worker rights

Individuals engaged in work should expect a certain set of core rights, regardless of their status. This would give both businesses and individuals greater clarity, certainty and flexibility, and would help support a more dynamic labour market and economy.

Digital Rights
  • Right to access personal data related to work and anonymised collective data

  • Right to the data/information about how algorithmic management affects access to work, pay, performance and hiring/termination

  • Right to data portability

  • Right to request a human review

Right to a floor on pay
  • Right to the national minimum wage

  • Protection from unlawful pay deductions

  • Right to request sick pay and parental pay

Right to health, safety, wellbeing in the workplace
  • Remote worker/freelancer digital rights, including the right to switch off

  • The right to a safe working environment, including maximum temperature thresholds

Right to representation
  • Right to form a recognised bargaining unit with the aim of negotiating certain core terms

Right to request benefits

Right to request benefits, specifically:

  • Training

  • Pension contributions

Create a Single Enforcement Body

The UK needs a single enforcement body (SEB) with extensive powers to assess and advise on employment status, monitor the minimum set of rights and recommend updates, inspect workplaces and bring prosecutions and civil proceedings on behalf of workers relating to health and safety, minimum wage, worker exploitation and discriminatory practices.

This would rationalise the currently fragmented and confusing system, in which responsibility for enforcement is split among at least four different organisations, and would enable information to be pooled more effectively. The goal of the enforcement body should be to support businesses intent on complying with labour regulations, while being tough on those which do not, and supporting workers in raising complaints. It should also identify pervasive instances of non-compliance more broadly and publicise persistent breaches.

The SEB would also be responsible for providing non-binding opinions on the employment status of workforces. It should also be responsible for exploring options to hold businesses accountable for the pay and working conditions in their supply chains. Developing a supply-chain requirement would help remove incentives for firms to seek a competitive advantage by ignoring the details of the employment practices of their suppliers.

Improve Predictability for Businesses and Individuals Over Employment Status

Our analysis has shown that there is a great deal of uncertainty when it comes to deciphering employment status. This matters because employment status largely shapes the rights workers can draw on. Distinguishing between categories is not easy – the courts have applied various tests to make sense of them and cases turn on their own facts, which means businesses and workers find it hard to know where they stand. And the growth of atypical work has made this task even harder because it has brought more diverse, complex working arrangements to the market.

Uncertainty, as we have explained, undermines good working practices. It makes it harder for individuals to know and therefore enforce their rights. It invites employers to seek commercial advantage by exploiting ambiguities and complexities in the law. And it puts employers off from giving atypical workers more benefits because they worry about precipitating a reclassification of worker status. All of this reinforces precarity for workers and is bad for business because it creates an unlevel playing field. Policymakers should, therefore, urgently seek to improve the clarity of employment status by undertaking the following measures:

  • Introduce an early-determination service for employment status. Currently the Advisory, Conciliation and Arbitration Service (ACAS), a non-departmental public body, provides employment advice for businesses and individuals on its website and through its helpline. It also offers dispute-resolution services. However, as it makes clear in its guidance, it “cannot give an opinion” on employment status when giving employment advice, and its dispute-resolution service attempts to resolve relationships only once they have broken down, before parties go to court.[_]Policymakers should introduce a more proactive approach to determining worker status. To do this, the new single enforcement body should be given the powers and resources it needs to provide individuals and employers with considered opinions on the likely status of any given worker. These non-binding opinions would not carry the weight of tribunal or court judgements, although they would be strongly indicative and could be used as evidence in any subsequent proceedings.

  • Recommend that businesses and sectors assess the employment status of their workforce. The single enforcement body should recommend that businesses which may be in doubt about individuals’ employment status voluntarily seek an opinion on their status, similar to the voluntary declarations HMRC requires if an individual uses or promotes specific tax-avoidance schemes.In addition, policymakers should introduce a requirement for certain businesses to obtain an opinion from a single enforcement body (as per our recommendation above) regarding the rest of their workforce and make this publicly available. Such circumstances might arise, for instance, if a company has been judged by a tribunal to have miscategorised the employment status of one or more individuals, and the judgement in question is not in the process of being appealed.Lastly, if a business has been assessed to need to change the employment status of members of their workforce (either by the single enforcement body, through a tribunal or via the courts), this should trigger a notice to other businesses in the sector to obtain an opinion from the SEB on the employment status of their workforce.

  • Introduce rapid, low-cost tribunal and court processes on employment status. Where disputes exist, the government should seek to create fast-track tribunal and court processes (that speed up existing processes rather than diluting due process). This likely means more resources for the tribunal and court processes but would also be facilitated by requiring the SEB to provide an opinion on employment status prior to moving to dispute resolution.

  • Help businesses provide the self-employed with limited benefits without affecting employment status. Policymakers could do this by explicitly allowing businesses to offer certain benefits (including, for instance, training, accident insurance, income protection, pension contributions and/or sick pay) without these factors being considered when determining employment status in law. In so doing, policymakers could improve individual security for self-employed individuals while also preserving flexibility for them and the businesses that engage them.

Address Interactions With Other Systems

While we believe the measures above are the starting point for improving the agency and rights of non-traditional workers, they do not address all the intricacies of how the UK’s employment-rights system interacts with the skills, welfare and tax systems. While the complexities of these systems mean they are outside the scope of this work, addressing them would further strengthen the UK’s employment regime. The areas to tackle are as follows:

  • Expanding access to human and social capital. Access to capital used to mean access to finance – and specifically your house – as potentially both your retirement fund and collateral for entrepreneurship. Now, it also needs to include access to human capital – for example, through life-long learning and individual-learning accounts – and social capital, including the support systems and communities[_] that make flexible work possible.[_]

  • A welfare system that is not just a safety net, but a support system. Research suggests that the stronger the social safety net, the greater the market power of workers relative to digital platforms. The UK’s social-insurance system needs to be redesigned to fit a more flexible world of work[_] and to help people be resilient to risk,[_] cope with challenges[_] and improve human capital.[_]

  • Simplifying the tax and employment regimes. The government needs to align the definitions of employment in the tax and employment-status systems: currently some  “workers” are classified as “self-employed” for tax purposes, and some “self-employed” are classified as “employees” for tax purposes. Aligning the definitions of employment, as recommended in the Taylor Review of Good Work, would help remove such anomalies and help address the wider uncertainty.

The UK’s existing legal and institutional system of protecting workers’ rights has failed to keep pace with profound technological, economic and social changes. As a result, individuals and businesses face more uncertainty. In turn, this is leading to greater precarity for individuals and an uneven playing field for business, both of which undermine prosperity and productivity.

Our proposals would restore agency and opportunity to individuals and give businesses more predictability, a level playing field and the chance to be proactive on workers’ rights without fear of consequence. They also amount to the simplest approach, avoiding complex legal battles and the economic damage that would stem from wholesale reclassification of employment statuses while also putting a floor under a deregulatory race to the bottom on wages and standards that most businesses do not want.

The UK government needs to modernise the legal and institutional frameworks underpinning workers’ rights without sacrificing the dynamism of either individuals or businesses.

Chapter 8

Annex Employment Status and Employment Rights

Figure 17

How rights tally with employment status *

*Subject to various qualifying criteria – for instance, in relation to duration of time worked.

Source: TBI

Methodological Appendix

For our analysis we use the UK-household longitudinal survey: Understanding Society, Waves 1-11, spanning 2009 to 2021.[_]

To analyse the characteristics and experiences of atypical workers we use the latest wave available, Wave 11, corresponding to household interviews carried out between 2019 and 2021. The analysis is weighted using the survey’s cross-sectional weights and sampling variables.

We define our population of interest as individuals that are 16 years of age or older and report they worked some hours in the week up to the time of the interview or, if they did not, they report being away from their position due to circumstances such as holiday, sick or maternity leave.

Since atypical forms of work may extend the lifespan of an individual’s economic activity, we do not limit the analysis to individuals aged under 64, traditionally thought of as the “working-age” population.

Of those who report being aged over 16 and in work, individuals are classified as traditional or atypical workers according to the following criteria.

Figure 18

How traditional workers, giggers, impermanents and hustlers described what they did

Traditional work
Traditional employee

The individual reports being a permanent employee, paid by the organisation they are working for.

Traditional business owner

The individual reports being self-employed, describes the nature of their self-employment as running a business or being a partner in a professional practice in which they employ others

Agency workers

The individual reports their work as impermanent and when asked in what way is the work impermanent the individual reports that they work through an agency

Platform-based work

The individual says that in the past month they made money using a website, a platform or an app to:

  1. Carry passengers in their vehicle (e.g. taxi rides), to deliver food and drink from restaurants and food outlets to people.

  2. Provide courier services (e.g. package and postal deliveries, messenger services).

  3. Perform manual tasks (e.g. cleaning, decorating, building, home fixtures and repairs, pet-sitting).

  4. Perform non-manual tasks (e.g. web and software development, writing and translation, accounting, legal and admin services, marketing and media, audio and visual services).

Tripartite employment

The individual reports being a permanent employee who is paid by an organisation different to the one they are working for.

Seasonal work

The individual reports their work being impermanent and when asked in what way, they describe the work as seasonal.

On-demand work

The individual reports that their work is impermanent and describes it as casual, or reports the work as impermanent and says their work has zero-hours or on-demand arrangements.The question on zero-hours and on-demand arrangements at work (jbflex) is asked only in even years so we merge information for individuals from Wave 10 and assume that if they are on non-zero-hours and on-demand arrangements in 2019–2021.

Fixed term work

The individual reports their work is impermanent and when asked in what way, they describe the work as done under a fixed-term contract.


The individual reports being self-employed, has no employees and describes the nature of their self-employment as working for themselves, freelancing or being self-employed in some other way.


The individual reports being self-employed, has no employees and describes the nature of their self-employment as being sub-contractor.

Solo-business owner

The individual reports being self-employed, has no employees and describes the nature of their self-employment as running a business or a professional practice or being a partner in one.

Mutually Exclusive, Collectively Exhaustive

Individuals are originally categorised by what their main employment is. In Understanding Society, someone’s “main job” is that with the highest earnings. In the case of equal earnings, the main job is that with the longest hours.

When individuals fall into multiple categories after they have been classified, we assert an order according to the overriding factor in distributing worker power and agency over their work, hours or pay.

Since gig-work questions are asked on the basis of whether an individual has engaged in gig work in the last month as opposed to the last week, it is possible an individual both has engaged in gig work and has another job that is considered their main job. If an individual has been working more than 50 per cent of their total working hours in the gig economy, then they are considered a gig worker over other types of work they may be engaged in. We use an hours instead of an earnings criterion because gig workers can choose to report their earnings by day, as using earnings as a benchmark requires additional assumptions based on using the hours worked in order to scale their reported daily earnings to make the data comparable to weekly pay reported in other roles.

If an individual reports being both self-employed and working through an agency, we consider the tripartite nature of the relationship in an agency setting to be more important for the way the individual accesses work than the fact they are self-employed, so we classify that individual as an agency worker. The same holds if the individual says they are a fixed term, on-demand or seasonal worker but also report working through an agency.

If an individual reports being self-employed, on a fixed-term contract, working seasonally and on-demand, we assume the on-demand nature of the working relationship overrides the impermanence and non-subordinate features the other categories imply.

We apply these criteria iteratively until all workers belong to only one category.


Individuals who do not respond to any of the questions used in the criteria outlined in Figure 18 will not be assigned to any category. As a result, there is a small share of individuals who report working some hours last week but are not assigned to a category. We assume non-responders are equally likely to have been traditional or non-traditional workers, so we scale the total shares of traditional and non-traditional employment to add to 100.

Atypical Work Over Time

To understand the growth in atypical work, we employ the remaining ten waves of the survey. Not all variables we use to identify atypical employment are available for previous waves. Questions identifying gig workers appear for the first time in Wave 11, individuals are asked whether they are paid by the same organisation they work for from Wave 10 onwards and the question about on-demand and zero-hours contracts appears from Wave 8 onwards but is only asked in even waves of the survey, amounting to two waves by the time of our analysis.

We carry out the analysis first by keeping the definitions of atypical workers tied to what can be identified in the first wave of the survey (2009 to 2011) and second by also incorporating new questions on atypical work as they are introduced in the Understanding Society survey. The new questions are likely to capture a mix of individuals whose circumstances could not be identified as “atypical” prior to their being introduced, as well as people who identified as atypical but would have put themselves in a different category.

We decided to allow the categories captured in atypical employment to increase with the new questions that become available. This is for several reasons. New questions are not just a matter of more accurately classifying what is already happening in the labour market but also capturing new trends that may not have been there in 2009, such as the gig economy. As people are given more options to classify their work more accurately, keeping the classifications constant means the share of atypical work that is captured is falling. Figure 19 shows the difference between holding the definition of atypical consistent or adapting it to reflect new questions. The small fall in the total share between Wave 8 and 9 and Wave 10 and 11 is likely due to the fact that the on-demand and zero-hours question is asked biannually. The merging of individuals’ answers from two waves for Waves 9 and 11 results in attrition.

Figure 19

Figure 19 – How data-capture varies depending on whether atypical work is classified consistently or inconsistently over time

Source: TBI calculation using Understanding Society

Atypical Work Over the Lifecycle

Finally, we exploit the panel dimension of the survey linking individuals across time to map the lifecycle role of atypical employment. We follow individuals between Wave 8 and Wave 11 and collapse observations between bilateral pairs of transitions between worker categories. For Figure 15 in the report, we illustrate the flows of all pairs where workers transition into a different group from the one they belonged to in 2016.

Lead Image: Getty Images


  1. 1. per cent20Glossary.pdf

  2. 2.

    Part-Time Workers (Prevention of Less Favourable Treatment) Regulations 2000. It is true that a part-time worker must not be treated less favourably than a full-time comparator, and that this covers the big stuff (pay rates, pension benefits, holidays, training, selection for promotion, selection for redundancy, career breaks, to name a few). There are, however, some nuances here. For example, the need to identify a full-time comparator to demonstrate less equal treatment (like-for-like not always easy to find). And the fact that there is a potential defence if the employer can justify an omission on objective grounds - an example would be only offering full-time workers private health insurance/asking part-time workers to contribute something on the grounds that the cost of providing this is disproportionately high, particularly given the size of a company.

  3. 3.

  4. 4.

  5. 5.

  6. 6.

  7. 7.

  8. 8.

    The Amazonian era. How algorithmic systems are eroding good work

    . Institute for the Future of Work. May 2021.

  9. 9.

    The experiences of individuals in the gig economy

    . UK Government. February 2018.

  10. 10.

    My boss the algorithm: an ethical look at algorithms in the workplace

    . ACAS Policy Brief. March 2020.

  11. 11.

    Good work. The Taylor review of modern working practices

    . July 2017.

  12. 12.

  13. 13.

    From rights to reality. Enforcement of labour market laws in the UK

    ; Resolution Foundation. 2019.

  14. 14.

    ‘Jobs paid below the minimum wage by category.’ ONS 2021.

  15. 15.

    ‘We need health and safety at work now more than ever.’ O’Connor, S. FT, 26 May 2020.

  16. 16.

    Uber BV and others v Aslam and others [available at:

    Uber BV and others (Appellants) v Aslam and others (Respondents) - The Supreme Court]

  17. 17.

    Supreme Court rules tribunal fees unlawful | Croner-i (

  18. 18.

    GMB and Deliveroo sign historic recognition deal | GMB

  19. 19.

    Trade unions protecting self-employed workers_EN.pdf (

  20. 20.

    Fabian-Society_Future-Unions-report-1.pdf (

  21. 21.

    Do Trade Unions play a positive or negative role in Britain today? (

  22. 22.

    Fabian-Society_Future-Unions-report-1.pdf (

  23. 23.

  24. 24.

  25. 25.

  26. 26.

    Call for Evidence: umbrella company market - GOV.UK (;

    Government plans let ‘dodgy umbrella companies off the hook’, warns TUC | Employment law | The Guardian

  27. 27.

    110322 Gig Economy Report V2 (

  28. 28.

    110322 Gig Economy Report V2 (

  29. 29.

    Time for action. Skills for economic growth and social justice

    . Learning and Work Institute. March 2019.

  30. 30.

  31. 31.

  32. 32.

  33. 33.

  34. 34.

  35. 35.

  36. 36.

  37. 37.

  38. 38.

  39. 39.

    Types of employment status: Checking your employment rights - Acas

  40. 40.

  41. 41.

  42. 42.

  43. 43.

  44. 44.

  45. 45.

  46. 46.


Practical Solutions
Radical Ideas
Practical Solutions
Radical Ideas
Practical Solutions
Radical Ideas
Practical Solutions
Radical Ideas
Radical Ideas
Practical Solutions
Radical Ideas
Practical Solutions
Radical Ideas
Practical Solutions
Radical Ideas
Practical Solutions