Worker’s Status

A worker who has not been paid their wages has the right to bring a claim in an Employment Tribunal (ET) for an unlawful deduction from wages.

But who is a worker?

The law states a worker is someone who personally performs any work or services under a contract of employment or any other contract, provided that the worker is not in business on their own account (i.e is genuinely self-employed).

The legal definition of a worker is important because workers have some employment rights (see Box on Employment Rights) including a right to be paid the national minimum wage and the right to pursue a claim for unlawful deduction from wages in an Employment Tribunal if they have not been paid the wages they are owed including, for example, if they have not been paid holiday pay.

This compares with those who are genuinely self-employed (i.e. in business on their own account), who do not have those same employment rights.

Employers in the new ‘gig economy’ often categorise those working for them as self-employed.  For example, taxi drivers, couriers and cleaners are often given contracts or “written terms” stipulating that they are “independent contractors” or are “self-employed”.  In other cases, there may be nothing in writing at all other than the fact that the company classifies them as self-employed.

With reports of couriers having pay deducted because they have not made a delivery within the allotted hour or not being paid for the last shift because they did not accept a job in the last minute before their shift ended, those working in this sector are often unclear as to what action they can or cannot take to recover pay that is rightly due to them.

The correct categorisation of the working relationship is all-important in determining whether someone is a worker and therefore is able to take action to recover unpaid wages.

Case law has held that there are two key elements to determining whether someone is a worker.  The person must:

  • Provide personal service; and
  • There must be mutuality of obligations between the parties.

What is personal service?

At one level establishing that someone provides personal service seems relatively easy.  Surely, if you are asked to do the work and you do it there is no problem? However, some employers claim that those that work for them have the freedom to get others to do the work for them – during periods of sickness or holidays, for example.  This, the employers claim, means that the person can provide a substitute who can do the work for them, so they are not required to carry out personal service.  In Autoclenz Ltd v Belcher, the company provided car valeters with a written contract which stipulated that the car valeters were “entitled to engage one or more individuals to carry out the valeting” on their behalf.  However, in practice, it is difficult, if not nigh on impossible, to find anyone else to do the work or at least anyone whom the company is willing to accept.  If that is the reality then there is every chance they are in fact providing personal service and so satisfy the first step in establishing that they are a worker.

What is mutuality of obligation?

This is the obligation on the employer to provide work and a correlating obligation on the person to accept work when it is offered.

Many of those working in the new “gig economy” are  sold the idea of self-employment on the basis that they can decide when they want to work and that they can be their own boss.  However, when there is rent to pay, food and clothes to buy the ability to choose not to work is often no choice at all.

The reality is that if work is not accepted when it is offered not only will there be no pay but the person is unlikely to be offered any further work in the future.  So, in practice, people rarely refuse work when it is offered.

The profits companies stand to make from the self-employed similarly means that in most cases the company is obliged to offer work to stay in business.  An article in the Guardian dated 18 July 2016 on delivery firm, Hermes, revealed that pre-tax profits of £36 billion in the year February 2015 was the product of a workforce of couriers, 84% of whom were self-employed!

The question then is, can a company write into a contract terms which would prevent someone from claiming they are a worker?

Generally a written contract is seen as sacrosanct when determining the working relationship.

In Stevedoring and Haulage Services Limited v Fuller and Ors, (albeit a case which concerned employee status rather than worker status), the contractual terms expressly stated that work would be provided on an ad hoc basis with “no obligation on the part of the company to provide work nor for [you] to accept any work so offered”.  The Court said that there was no scope for implying a positive obligation on the parties to offer work and for work to be accepted where to do so would contradict the express terms.

But that was back in 2001.  Today the Courts take a dim view of employers relying on contractual terms to deny workers employment rights where the terms do not reflect the reality of the working relationship.

Contracts which do not represent the reality of the situation are now recognised to be sham contracts.  A sham contracts as expressed in the case of Consistent Group Limited v Kalwak is a contract where “the reality of the situation is that no one seriously expects that a worker will seek to provide a substitute or refuse the work offered, the fact that the contract expressly provides for these unrealistic possibilities will not alter the true nature of the relationship”.

In many cases for workers in the new “gig economy” the written terms or the written contract will not represent the reality. Anyone unsure of the status of their working relationship and associated employment rights should join and seek the advice of a trade union who can advise on this hugely important issue.

 Employment Rights Box

Statutory Rights Employee Worker Self Employment
Dismissal Y
Redundancy Y
Notice Y
Maternity Leave Y
Parental Leave Y
Fixed Term Employment Y
National Minimum Wage Y Y
Protected Disclosure Y Y
Working Time Y Y
Part time work Y Y
Right to be Accompanied Y Y
Unlawful deduction from wages Y Y
Protection from discrimination because of a protected characteristic Y Y Y

Impact of Employment Tribunal fees on selected employers

We have reported before on our trips to the Employment Tribunal (ET) judgement registry, hidden away in the heart of East Anglia (Bury St Edmunds). We are back there now, and as before concentrating on the cases which are key to our research namely those featuring “unauthorised deductions from wages” (generally known, for historical reasons, as Wages Act claims), and those relating to failure to provide holiday pay.

As we plough through scanning hundreds of judgements, we see some respondent (that is to y, employer) names recurring. This makes us wonder if it would be easy to see how frequently these company names come up as respondents in cases involving claims of unpaid wages. So we chose a few of the larger outsourcing and facilities management companies, based on their tendency to have subsidiaries beginning with the parent’s name. I should say that this method is far from fool proof – not all of the subsidiary names will follow this model, and it of course leaves out parent companies which do not follow this model at all.

So our leaving out some large employers is not intended as a slight – we were simply using the rudimentary search strategy which is available on the database of ET judgements, namely putting the wildcard “*” after the first few characters of the company name. So the five parent companies selected almost but not quite at random, were Capita, ISS, Mitie, Serco & Sodexo. We looked to see how many cases which included “wages act” claims had reached a hearing, by the year in which the claims had been accepted. We looked at each year from 2010 to 2015. In the first year, these five companies accounted for 148 such cases, but by 2015, this had fallen to 41.

Perhaps this is good news for their collective workforces. Perhaps they are just much less disgruntled, and more confident that they have been paid correctly for their labour power. Or then again, perhaps the introduction of ET fees has been of great benefit to these companies, many of them delivering outsourced and privatised public services. Because there does seem to be a dramatic fall after 2013, the year fees were imposed. Have a look at the graph, and see what you think.

ET cases includin WA claims reaching hearing

Looking for unpaid wages, finding Delinquent Employers

Sometimes, when a researcher asks for comments on a piece of work, they are really asking for validation. I’m not saying I was when I proposed a typology for unpaid wages to a workshop that Unpaid Britain organised in May. But if I had been, I would have been disappointed.Group C picture

We brought together a pretty expert group of practitioners and academics. Then we asked them to try and categorise real life cases of unpaid wages by whether or not they were deliberate, and whether workers might have given some form of consent. The response could be summarised as “yes, very interesting, but you can’t make it work”. Instead, we were directed back to the specifics of employers’ strategies in different economic sectors. And if this did not endorse my proposed typology, it certainly validated the use of the participative workshop.


The research team have spent the intervening months considering how this might be achieved. The result is a method for identifying the sectors most likely to exhibit non-payment. To do this, we had to confront the lack of data specifically measuring non-payment. Key Informant interviews, workshop participants, our Project Advisory Group, and our examination of Employment Tribunal judgements all suggested that unpaid wages may present in several forms simultaneously, as well as being accompanied by other abuses of workers’ rights.

So, using two sets of official statistics we have tried to find data which might indicate the abuse of workers’ rights. From the Labour Force Survey (LFS) we chose reported unpaid overtime, which may reflect a general level of pressure in the workplace, but also breaches of at least some workers’ contractual entitlements. We also selected reporting of zero hours contracts (ZHCs). While these are not automatically unlawful, by not committing the employer to provide a certain minimum level of payment per pay period they blur other entitlements such as that to holiday pay, and the variability of hours renders under-recording (and therefore underpaying) of hours more feasible. The LFS may undercount this – overall, only 1.8% of those in employment reported having a ZHC, but this ranged from 0% in mining support services, up to 9.2% in security and investigation activities. Finally we looked at paid holidays – about 5% of respondents in employment report that they have no entitlement to paid holidays, and this is not evenly distributed across sectors (it goes from 0% in manufacture of electrical equipment up to 18.4% in sports, amusements and recreation). We can be certain that workers not receiving any paid holidays have had their rights breached.

A second survey, the Family Resources Survey (FRS), includes two sets of data which could be indicative of abusive practices. One relates to the provision of payslips, which is a legal requirement under the 1996 Employment Rights Act. 8% report that their employer does not provide a payslip, ranging from 0% in extraction of crude petroleum and natural gas up to 21.3% in domestic households. The other is self-employment. Operating on the assumption that work falsely classified as self-employment by employers wishing to reduce labour costs (as opposed to genuine self-employment) is likely to be low paid, the FRS permits the identification of the extent of low-paid self-employment by sector, indicating the likely presence of “bogus” self-employment.

These five indicators can shed some light onto the variety of practices which might be deployed to separate the worker from what they have rightfully earned. What they cannot do is give us a definite count of abuse, so we opted for a series of inter- sectoral comparisons, allowing us to identify those sectors most likely to display forms of abuse, and therefore the non-payment of wages.

Excluding sectors with too few cases, we scored each sector from 1-10 for each factor, with 10 representing the most abusive. Then we added them together to arrive at an overall “Index of Employer Delinquency”. The top (or perhaps bottom) ten for London[1] are shown below.

Top Sectors in London Index of Employer Delinquency
1 Creative, arts and entertainment activities
2 Food and beverage service activities
3 Other personal service activities
4 Sports activities and amusement and recreation activities
5 Libraries, archives, museums and other cultural activities
6 Other professional, scientific and technical activities
7 Education
8 Advertising and market research
9 Repair of computers and personal and household goods
10 Security and investigation activities


In one sense the overall outcome is probably unsurprising for those who are familiar with employment practices in Britain today. The Index shows service sectors predominating amongst those showing the highest tendency to abuse employment rights.

Pride of place, if that term can be applied in this context went to “Creative, arts and entertainment activities”, where substantial low paid self-employment exists alongside frequent failure to provide paid holidays (these two categories are unlikely to occur in the same individual cases, since the holiday question is posed only to those identifying themselves as employees).

Next, and perhaps most significantly in numerical terms, is “Food and beverage service activities” – restaurants, cafes and bars in other words. Here we see the use of ZHCs assume greater significance than low paid self-employment. Then we have “other personal services” (such as hairdressing and nail salon) and another cultural area – sports and entertainments, followed closely by museums and libraries, which has been the scene of several long-running industrial disputes over attacks on conditions and cuts in services (for example at the National Gallery and Barnet libraries).

The case studies for our next phase of research will be selected from this list, and we will be looking for a balance between large and small employers, and for types of worker (by gender, migration status and age). Suggestions are always welcome.

A more detailed description of and justification for our index can be found on the full paper here: Unpaid Britain project – developing a selection grid for case studies.




[1] Agriculture and mining and quarrying, which both score relatively highly in national data are not significant in the London labour market, and have been left off the list for the purposes of this project.

Unpaid wages to unpaid benefits

When does a job count as ‘work’? This question is pressing across the EU (which at time of writing includes the UK and will do so whatever the result on Thursday 23rd for some time to come)[1]. While free movement across EU member states is the right of most EU citizens, the right of residence for longer than three months is for self-sufficient (i.e. wealthy) people and for workers and their family members. The EU citizen is paradigmatically a worker citizen. According to the EU a person must be in ‘genuine and effective’ work, but the question of what constitutes genuine and effective work is largely settled at a national level. In the UK from 1 March 2014 one of the tests for work being genuine and effective was that earnings exceed the national insurance threshold for 3 months, or £153 a week. Notably this was, at the time, considerably more than the wages of apprentices, raising the question of whether their work is considered ‘genuine and effective’. Given European Economic Area  (EEA) nationals’ concentration in low-waged jobs in sectors such as hotel and catering, cleaning, agriculture and social and childcare, where zero-hours contracts, agency and temporary working are rife, this kind of threshold is likely to prove challenging. It also means that unpaid wages become unpaid benefits. EEA migrants who lose their jobs can claim Job Seekers Allowance (JSA) as a worker only if they have been in ‘genuine and effective’ work for the three months before making their claim. If they have not been paid enough money to count as workers then they will not be eligible.

The welfare consequences of this definition are perverse even following the government’s logic.  An EEA national who passes the test to count as a worker for the purposes of claiming JSA can claim for a maximum of 6 months before being required to pass a Genuine Prospect of Work Test. To pass this, they must provide ‘compelling evidence’ that that they will receive an offer ‘imminently’ for work of a type that is considered ‘genuine and effective’. To be eligible for benefit an EEA national must have a genuine chance of finding genuine and effective work in the labour market sector that is relevant to their previous experiences. In contrast, as the guidance for JSA decision makers on sanctions makes clear, resident UK nationals must be prepared to do any work, irrespective of whether it is genuine and effective, or whether it is in a sector that is of interest to them. Indeed citizens are finding welfare rights they took for granted increasingly circumscribed. There has been an increase in the use of sanctions for failure to comply with work-related conditions. In written evidence submitted to the House of Commons Work and Pensions committee inquiry into benefit sanctions policy, David Webster found that the total number of sanctions issued to people claiming JSA and Employment Support Allowance (ESA) before reconsiderations/appeals in the year to June 2014 was an estimated 1,030,000. That is not a typo. In the summary of his evidence, Webster described the system as ‘a huge secret penal system, rivalling in its severity the mainstream judicial system but without the latter’s safeguards’. For someone who works on immigration this is eerily familiar territory.

Prime Minister Cameron has often referred to immigration and welfare benefits as ‘two sides of the same coin. It is difficult to see both sides of a coin at the same time, but as the EU referendum has demonstrated, we badly need to connect exclusions from formal citizenship (migration) with exclusions within formal citizenship (unemployment, exploitation, and other forms of marginalisation).  The urgency of this political project is palpable.





[1] This article was written on the 21/06/2016 this was before the EU referendum vote took place.

The delinquent employers’ dilemma: settle or dissolve?

Employment rights in the news

Last week two very interesting (from Unpaid Britain’s perspective) stories hit the headlines. In the first, Mike Ashley, owner of Sports Direct, admitted that 15 minutes of pay was deducted from workers who arrived at work even a minute late. This is a high-profile example of the low-level “gaming” of employment contracts which Unpaid Britain is studying.  The second story saw former Chelsea FC doctor Eva Carneiro  accept a settlement from Chelsea shortly before she was due to testify to an Employment Tribunal (ET). The latter story illustrates a particular phenomenon which we have been examining, as this blog will attempt to explain.

The Employment Tribunal

Since March 2016 we have been analysing Employment Tribunal cases brought before London and Watford tribunal offices for unauthorised deductions from wages. This has so far involved a two week stint at the Bury St Edmunds employment tribunal registry. Even though employment tribunal judgements are public records they can only be accessed via terminals located in the basement of Bury St Edmunds county court. There, surrounded by disintegrating boxes containing paper copies of ET judgements, we went about selecting a random sample of judgments from London for the years 2012 and 2014 (before and after the introduction of fees). Details (parties, jurisdictions, general outcome) of the judgements had to be copied from an old computer screen onto our laptops (no digital downloads!). Then we had to find and scan the paper copies of the judgements. This was a lengthy task; many copies of the judgement are mis-filed and some are absent.

ET computers  ET roling Stacks

Companies House

Back at Middlesex, we reviewed the Companies House databases for the respondent companies named in the judgements. We sought to identify the company’s status, sector and identity of their directors. Companies House has two database systems, of which Beta is the more detailed. However many of the companies don’t appear on it especially if they had been dissolved a few years previously, so a constant transition between Beta and the second (WebCheck) is required.

Half way through this process we decided to examine our findings so far, and they suggest something very interesting about employers who settle.

Preliminary Findings[1]

What is interesting is the marked difference between the status of companies for the different outcomes of claims. Where the case has been “dismissed on withdrawal”, dismissed on settlement (as in the Carneiro case) or fails at a hearing the company is most likely to be still active. Where there is a default judgement (usually occurring when the respondent i.e. the company doesn’t respond to the claim or appear at the Tribunal) or the workers’ claim was successful the company is most likely to be dissolved through insolvency or other means.  What does this reveal regarding these companies intention, or the prospects of workers recovering their wages?

Dismissal on Withdrawal

21% of outcomes out of our preliminary sample were “dismissed on withdrawal”, meaning the “claimant informs the Tribunal through writing or in the course of a hearing that the claim or part of the claim is withdrawn”, following which the Tribunal issues a judgement ruling that the claim is dismissed. This leaves the actual outcome in terms of whether the worker was successful in getting their money unknown.  A case might be settled informally, or in cases such as that of Eva Carneiro a settlement was reached during proceedings and be classified as dismissed on withdrawal. Members from the Unpaid Britain project’s Advisory group call this type of settlement “napkin cases”, when a settlement is agreed and in some cases scribbled onto a napkin just before or during a hearing. However a dismissal on withdrawal could also occur where the claimant no longer wanted to pursue the case, (for reasons such as stress or feeling that the claim might not succeed).


Where settlements take place, it may be assumed that the worker received  at least some of their money, but judgements stating that a settlement has occurred are the smallest group amongst our sample. We had many more cases classified as “dismissed on withdrawal”, and  wanted to investigate the likelihood that these had  been withdrawn due to a settlement occurring (this had been suggested to us by legal advisors). To do this we chose to compare the status of respondent companies where we know a settlement has occurred with that of those in cases withdrawn on dismissal. We found the Companies House status of respondents in cases dismissed on settlement to be very similar to those dismissed on withdrawal, with around 80% of companies still being active.  This is in contrast to the profile of other groups of respondents.

Default Judgements and successes

Looking at cases that were Default Judgements the majority of companies appear to have been dissolved, with only a small minority still active. This may not be that surprising: if they are about to be or are already insolvent, employers may be unable or reluctant to defend the case. However where the claim was heard and succeeded (the most common outcome) the majority of respondent companies are now either dissolved or insolvent – only around 30% are still active. Research done by the Department for Business Innovation and Skills  in 2013 found that only 32% of claimants whose unpaid wage claims were successful were paid in full. Larger companies were more likely than small ones to pay awards, and this may be explained by widespread use of insolvency and limited liability to evade payment.

Failed claims

In cases which were heard, but the claim failed, our preliminary findings show that 94% of respondent companies are still active.

What the findings suggest

These findings suggest to us that the majority of cases that are “dismissed on withdrawal” are actually settled before or during the hearing.

One further hypothesis is that if an employer expects to pay the money to their workers they are likely to settle before the hearing, only going ahead with a hearing where they are confident of winning. Settling reduces litigation costs for employers, especially if the unpaid wages claim is linked to other claims such unfair dismissal or discrimination.  Cases classified as being dismissed on withdrawal leave no evidence of guilt of non-payment. This enables the employer to maintain their image both in terms of brand as well as to other employees. Thus systematic non-payment of wages can carry on unnoticed/unchallenged.

Where employers have no intention of paying, however, they may simply fail to defend, leading to a default judgement, or go to a hearing and use insolvency and limited liability to protect directors from any personal cost. It could also be that losing the ET case pushes the company into insolvency, but given that most unpaid wages claims are for few hundred pounds, this could only apply in a minority of cases.

Although we still have a lot of work to do to understand the phenomenon of non-payment and employment tribunal claims, these preliminary findings are consistent with there being a group of employers who “game” the system with apparent impunity.

For the time being we will keep you all updated. Any comments on this blog or our research are as always greatly appreciated.





[1] The preliminary findings are based on 205 cases including unpaid wage claims, predominantly made in 2012, and for which the respondent company’s identity has been located at  Companies House. The most common outcome in these cases is that the case has been “dismissed upon withdrawal”.

Unpaid Britain Project Towards a typology of non-payment

Towards a typology of non-payment summary 

A paper proposing a typology of non-payment of wages has been prepared for discussion at a workshop on 9th May.  This typology (following revision as a result of the workshop) will be used to help us pick case studies for the next stage of our research.

The full paper outlines categories which have been produced by other authors (including state regulators), and discusses factors considered to have potential relevance to for the construction of the typology. It also includes a brief summary of the Unpaid Britain project, followed by a discussion of the existing data on non-payment.

Click for full paper

The proposed typology of non-payment

Taking into account the various discussions and caveats that are outlined in the above working paper, we find there are four key dimensions to unpaid wages:

Consent; Intent; Means & Magnitude

Means (the way the wages have not been paid) and Magnitude (the size of the sum involved relative to normal earnings) are clearly important. They might determine whether restitution by the worker is pursued, the route by which this occurs, whether there is a role for regulators, and the impact of non-payment. But they are descriptive and determined by the circumstances and intent of the employer, and the degree to which workers’ consent may be given.

typology of non-payment

Non-payment might occur when imposed by circumstances (absence of funds, or catastrophic failure of payment system for example), or as a result of error (data recording mistake), or misinterpretation (misinterpretation of minimum wage regulations etc.)

These are distinct from circumstances where there has either been a decision not to pay, or one (even by default) not to rectify systemic problems which result in non-payment. Interviews suggested strongly that plausible deniability was common, but difficult to detect in official and administrative statistics, because it is feasible to rectify any loss in individual cases without disrupting the model or incurring penalties. This differs from the construction industry “knocking” (complete failure to pay outstanding wages) or deliberate insolvency associated with phoenix companies, in that those are detectable, and (in theory at least) subject to penalties and thus less sustainable as a business model. This could also apply to extreme examples such as forced labour or criminal extortion. These are evident non-payments.

Consent is not a relevant consideration when non-payment is imposed by circumstances, or where there is an error or misunderstanding, but is of significance for purposeful non-payment. The proposed typology separates informed consent, which includes some unlawful, but voluntary contracts and agreed delays to payment, from without consent which would include all other categories (such as not receiving entitlements of which the worker was unaware, as well as the more blatant unauthorised deductions of which they were fully aware).

Unpaid Britain Project Towards a typology of non-payment

Unpaid Britain Typology Workshop

Have you ever had wages withheld or unpaid? Or helped someone who has? As a result of late cancellations, the Unpaid Britain project has two spare places going at our workshop on 9th May, and we would like to fill them with people with direct experience of unpaid wages.

The workshop is intended to help us with our research into this phenomenon. Our next phase will be the selection of case studies to look in detail at the causes, experiences and remedies for unpaid wages. However we feel that this cannot be done effectively without developing a typology of non-payment, to help us choose case studies that illustrate each type. We have developed a draft typology but the time has come for it to be scrutinised, and tested to ensure its robustness.  The Typology workshop will consist of 30 experts from a variety of backgrounds ranging from trade union, employment lawyers, academics, citizens advice, conciliation organisations, think tanks, charitable trusts, monitoring organisations and employers, all of whom have an interest in the subject of non-payment.

Held at Middlesex University’s Hendon Campus, the workshop will start at 12:30 with a light lunch reception, continue until 4.30 and end with a drink and a discussion. There will be short contributions from Anna Kyprianou Dean and Pro-vice chancellor of the Middlesex University Business School, Michael Reed, Principal Legal Officer (Employment) at the Free Representation Unit, Oxford University lecturer Jenny Chan (expert in Labour Rights in China), and project leader Nick Clark. However most of the afternoon will consist of discussions in working groups, as they scrutinise, critique and think of cases that fit or do not fit in to the Typology.

We would love to fill the two places now vacant. If you or any one you know has experience or expertise in the field of non-payment and is interested in attending on the 9th of May please contact us by emailing:

The creeping notion of working for free

Yesterday’s Guardian carried the story of a recruitment agency advertising for warehouse staff who would have to complete a three-day unpaid “induction” (yes, I know it went online on Sunday, but I am old school and still read a print version). The agency (GB Recruitment) was recruiting for a logistics company (Wincanton) who in turn was running the warehouse for DIY retailers B&Q.  Quite apart from being a vivid illustration of the chains of labour supply now at work in the economy, it highlights what according to Unpaid Britain’s initial (if largely, at this stage, anecdotal) findings suggest has become a common practice, seen by many young workers in particular as normal practice.

This seems to be combining two factors. One is the gradual creep of the notion that some workers – as a working category I am calling them new entrants (or re-entrants) to the labour market[1] – have labour power that is worth little or nothing. The other is the ingenuity of employers in labour intensive sectors in finding ways of trimming labour costs so that they fall below the National Minimum Wage.

Unpaid training had already been identified as a problem area in the care sector, by HMRC . They found unpaid training to be the most common type of unpaid working time contributing to non-compliance with the National Minimum Wage. However, they seemed to suggest some circumstances in which this might be lawful:

We found instances where prospective workers attend pre-employment induction events to assess their suitability for employment or as part of the job application process. In the circumstance where this activity is not undertaken as part of their terms of employment or not for any form of remuneration paid to or benefit received by those attending, the time spent is not working time for NMW purposes” (p. 5).

In conducting research on internships in an earlier job, I was struck by the many ways in which the work of the young had been down-valued to the extent that even some of them accepted that their labour power had no market value, beginning with work experience at school, job placements at college or university, advice to “volunteer” to improve c.v.s and culminating with full-blown unpaid internships.  Wage-free labour has become embedded in some sectors of the labour market, such as the cultural industries where unpaid internships still abound, and entertainment trade unions such as Equity and the Musicians Union are being forced to campaign for young people to be paid at all.

Interviews conducted during the early stages of the Unpaid Britain project revealed an initial unpaid training (also known as induction, or trial shifts) has almost become a rite of passage for young workers. Staff from one advice organisation told me that they were aware of one employer who seemed to have a constantly replenished supply of young workers working “training weeks” for which they would never be paid, nor would they progress on to paid contracts, thus guaranteeing the company an almost free workforce. Of course abuse on that scale can hardly be a long term strategy, but the more limited unpaid induction, trial shifts and unpaid training can be found with relatively little research.

So for example last autumn one health and social care recruiter was advertising care worker posts in London for which there was an unpaid induction (of unspecified length), while another agency  was recruiting parcel sorters in the Midlands who would also be required to “complete an unpaid induction during unsociable hours”. But agencies do not appear to be alone in playing fast and loose with the notion of a wage/work bargain. A hairdressing apprenticeship scheme in North London, which claims to be sponsored (among others) by a London borough, the Skills Funding Agency, City & Guilds and the EU Social Fund, advertised vacancies for apprentices who might be “subject to a 6 week to a 6 month unpaid trial period”. This perhaps reflects the 2014 NMW (Amendment) Regulations which introduced new rules for traineeships for young people aged 16-24, excluding them from NMW rights – again suggesting that such workers have little value.

But then JobCentre Plus has for decades been organising work trials of up to 6 weeks, for which the worker receives no wage from the employer – only benefits (plus travel expenses) from the state. It is said to be “voluntary”, but it is work, and it is unpaid. Unfortunately, “…the TUC has accepted Work Trials. We are sympathetic to the plight of long-term unemployed people, unemployment is a threat to all workers and the people who need to be helped are our friends, neighbours and relatives.” (TUC 2008)

B&Q and Wincanton both told the Guardian that unpaid induction contradicted their policies, and that the advertisements had been withdrawn, but the idea of them did not appear out of thin air.  Economist Joan Robinson’s statement that “The misery of being exploited by capitalists is nothing compared to the misery of not being exploited at all”[2] is often used to justify all manner of abuses. But it is based on the notion that while the system denies (for those who have only their labour to sell) the means of sustenance to those who do not work, for those who do, work is the route out of poverty. But this notion is turned on its head if, in order to have the chance to be exploited, one must provide labour power for free. This is the lesson which is apparently being passed on to young workers, and is no doubt music to the ears of the abusive employer.

[1] This would include young workers (including working students), recent migrants, ex-offenders, those returning to work from long-term illness, and the long-term jobless, for example.

[2] Robinson, J. (1962) Economic Philosophy p.45

The Risk of the Wage-Work Bargain

According to Secretary of State for Business Sajid Javid, “the idea of employers paying a fair day’s wage in return for a fair day’s work has been the basis of our economic system for generations(BIS & Home Office 2016), no doubt expressing a widely held view. Similarly, the contract of employment has been said to channel “the risks of economic security in such a way as to protect the individual worker…” (Deakin & Wilkinson, 2005, p. 109). And while considerable attention has been given by scholars to the circumstances in which the right to be paid for work might be qualified or defeated (such as absence, partial or defective performance, for example), Napier’s (1984) conclusion that the “residual rule continues to be that the actual performance of services is a condition precedent of the payment of wages”, in general, is still held to be true.

And yet millions of workers find that having carried out the work, their wage is not forthcoming, either in its entirety, or at all. It seems that far from being fair and secure, the exchange of labour power for wages may prove to be a risky enterprise.

For example, China Labour Bulletin reported mass protests in China in the run up to the lunar New Year on the 8th of February, caused in the main by “chronic wage arrears”. It revealed that of 1,050 strikes recorded between December and February, 90% were related to wage arrears, and no longer confined predominantly to the construction industry, were spreading across other sectors.

Meanwhile, Human Rights Watch (2015) reported that two years after Rana Plaza disaster in Bangladesh, and despite legislation being put in place many garment workers (alongside other abuses) fail to receive their full wage, or are paid late.

For those tempted to think that this is a phenomenon associated with labour markets in newly industrialised economies, consider the substantial body of scholarship and campaigning over “wage theft” in the United States. A 2008 study of low paid workers in three cities found 26% of them to be paid below the minimum wage, and over three quarters who worked overtime to be paid below the legally required overtime rate. Researchers estimated that the workers lost over $2600 per year to pay violations (Bernhardt et al, 2009). Kim Bobo, meanwhile in her book ‘Wage Theft’ (2011), described the phenomenon occurring across different industries all around the US, featuring abuses such as total non-payment, payments to supervisors and bouncing pay-cheques.

Similar experiences can be found in Britain. The BBC’s Inside Out programme recently reported on security staff employed at the Olympic Games who remained unpaid four years later. It revealed that 11 stewards were owed between £1,000 and £1,800 each, but because of subcontracting arrangements and the dissolution of their employer Britannia Security Services Ltd have been unable to secure their outstanding wages.  While the UK labour market is arguably less regulated than that of the USA (perhaps surprisingly), and so there are fewer rights to violate, the limited official sources of data reveal tens of thousands of cases of non-payment per year.  What the BBC saw as remarkable, may turn out to be commonplace.

Take those who are unpaid when their employer becomes insolvent. Freedom of Information (FOI) requests by Unpaid Britain to the Insolvency Service have determined that about 40,000 workers per year are paid out from the National Insurance Fund for arrears in wages or unpaid holiday pay (frequently for both). At the height of the recession the numbers approached 100,000 per year.

insolvency payment of arrears

Another indicator is claims made to Employment Tribunals (ETs). Figures published by HM Courts and Tribunals Service (HMCTS) show that in 2014/15, 28,000 claims were lodged for “unauthorised deductions from wages” – which generally equates to unpaid wages. There were a further 31,000 under the working time regulations, most of which will relate to unpaid holiday entitlement. While some of these figures (10,000 according to HMCTS) relate to group submissions which may reflect a litigation strategy rather than specific individual cases of non- payment), the data will under-count claims including unpaid wages, by counting each claim only once. So an unfair dismissal claim will be counted under that “jurisdiction” only, even if there are also claims for unpaid wages and holiday pay (this appears quite common according to our preliminary examination of Employment Tribunal judgements). It is almost certain that taking into account claims with more than one issue at dispute, unpaid wages is the issue most frequently raised in ETs.

However, these figures report on the picture after the 2013 introduction of fees for ET claims. This has led to an enormous drop in claims. In the HMCTS London Region, for example, there were 18,700 claims classified as “unauthorised deductions” (also known as Wages Act claims) in 2013 – the year in which fees were introduced. The following year there were only 2,200. Given that the success/failure ratio of claims has not significantly altered since fees were introduced, it is likely that many legitimate claimants have been deterred by fees[1]. Citizens Advice data on enquiries to their bureaux nationwide shows that the number related to problems with payment of wages is on fact rising year on year, with 9,000 cases of unauthorised deductions being dealt with in 2015 (up 84% on the previous year).

Even if these figures for non-payments due to insolvency, and wage and holiday pay claims to ETs accurately represented the extent of unpaid wages, it would suggest a significant problem. The total of about 100,000 puts the offence on the same level as “robbery”, as reported in the Crime Survey for England and Wales, which registered 106,000 cases in 2014. They are likely to represent only a minority of cases, however (as the Unpaid Britain project hopes to reveal) due to legal barriers, low levels of enforcement by regulators, workers’ reluctance to pursue some cases and ignorance of their rights (particularly regarding holiday pay).

Over the course of the next two years, we will be shining a light on this neglected, but fundamental, breach of employment rights.  We would welcome readers’ experience of this – whether it is of employers going bust, unpaid periods of “training”, holiday pay not being paid on leaving a job, or just a few hours work here and there remaining unpaid. On the other hand, there are reports of misunderstandings by workers as to what they should be paid, an what deductions are legitimate, and we want to hear of them too.

[1]Another possibility is that more cases are being resolved through ACAS – a hypothesis that we will address in a later blog.



Bernhardt, A., Milkman, R., Theodore, N., Heckathorn, D.,  Auer, M., DeFilippis, J., González, A. L.,  Narro, V., Perelshteyn, J., Polson, D. & Spiller, M. (2009) Broken Laws, Unprotected Workers

BIS & Home Office (2016) Tackling Exploitation in the Labour Market Government Response, Department for Business Innovation and Skills (January 2016)

Bobo, K (2011) Wage Theft in America: Why Millions of Americans are not being paid-and what can we do about it, New York: New Press.

Deakin, S. & Wilkinson, F. (2005) The Law of the Labour Market, Oxford Monographs on Labour Law, Oxford University Press

Napier, B. (1984) Aspects of the Wage-Work Bargain Cambridge Law Journal, 43:2 pp 337-348