Precarious Work, ‘Compression’ and Class Struggle ‘Leaps’

Kim Moody, author of In Solidarity, continues the discussion begun by Ian A and continued by Kevin Crane about the nature of work and workplace struggle today

Picture Credit: iStock/MHJ
Picture Credit: iStock/MHJ

In his response to Ian A, Kevin Crane raises a number of serious issues concerning the work of socialists in trade unions and the workplace. The heart of his arguments, I believe, centres on the problems presented by the rise in precarious work and the decline in older forms of ‘regular’ employment. Indeed, more and more jobs have become precarious in one way or another in the last thirty years. It is also true that precarious work has hit the young disproportionately. Yet, precarious work is far from the norm even today in any sector of the UK economy or workforce. Before looking at the extent of precarious work in the last twenty years or so, I want to put the question of the workplace or work and class struggle in historical and theoretical context.

Some history

Historically, the period of relatively stable employment that followed the Second World War and lasted until the late 1970s was the exception in the history of capitalism, not the norm. The relative stability of employment following WWII rested on economic growth and increased real wages as well as on institutional collective bargaining arrangements.

The decades before this brief period, as well as those since, were ones in which precarious forms of employment were widespread, even dominant. The waves of unionisation and mass strikes in the UK, that began in 1888 with the New Unionism, were conducted by a working class in which a huge number of workers had no regular, full-time jobs, subcontracting was widespread, and a majority of employers were typically small outfits.

The major actors in the New Unionism in 1888-91 were highly precarious: match ‘girls’ (‘casuals’, homeworkers, ‘half-time [school] system’), dockers (‘casual hiring’, multiple-employers), and gas workers (seasonal) (Raw 2011, 41-42, 83; Hobsbawm 1964126-178). Building workers, who generally had craft unions until the mergers of more recent times, were (are) also casual and seasonal. In short, if precarious workers were incapable of collective action and unionisation on-the-job there would not have been much of a twentieth century trade union movement beyond the most skilled workers. It was not stable work that brought trade unions, but unions that brought the higher wages, shorter hours, and collective bargaining that encouraged workers to stay on a job in the expectation of improvement. Faced with higher wages after the war, management also found it useful to stabilise employment in hopes of increasing productivity.

Some theory

Marx and Engels were the first revolutionary socialists to see the importance of trade unions and work (the labour process). This was for the obvious reason that it is there, in the ‘social relations of production’, where capital and labour meet and clash, that class struggle begins and recurs day-in and day-out. Of course, this conflict extends well beyond the workplace into almost every crevice of society, but it is in the labour process that workers experience the pressures of capital’s incessant drive for profit most intensely and that trade unions first arise, opening new possibilities for mass struggle and politics.

But the class struggle that encourages workplace organisation is not an even or constant one. It is, as Marx and Engels put it in the Manifesto, ‘now hidden, now open.’ Despite their frustration with and criticisms of trade union leaders, Marx and Engels were particularly emphatic about trade union work even though they lived during the longest trough (‘downturn’) in class conflict (1850 to 1888) prior to our own era. More recently Eric Hobsbawm (1964, passim), Beverly Silver (2003, passim) and others have shown that unions and class conflict do not run an even course or grow in a linear fashion. Rather, class struggle and its organisational outcomes come in ‘leaps’ (Hobsbawm) or ‘waves’ (Silver). Seldom are these waves predicted. As Hobsbawm argues, however, these ‘leaps’ are the result of ‘compression’: intensified work, poor or falling wages, insecure employment, lack or decline in state provision, etc.

The last ‘wave’ or upsurge was in the 1960s and 1970s. This was a rebellion against the new bureaucratic industrial order that began after the war and arose in the 1950s and 1960s as managements sought to tighten control of the shop floor as shop stewards proliferated (Campbell, Fishman and McIlroy 2007, 69-113; McKinlay and Melling 2007, 222-241). Management attacked the autonomy and power of the shop stewards (Cliff and Barker 1966, passim) and attempted to impose ‘working conditions [that] were increasingly brutalised, automated, and deskilled’ (Savage, 2007, 23-42), along with the ‘productivity deals’ that began in the second half of the 1960s (Cliff, 1970, passim).Such waves of struggle are, in my view, usually the result of changes and increased pressures in the nature of work, as in the post-WWII period, not the business cycle as some would argue, though, of course, other factors enter in.

It’s also generally the case that it takes time, often a generation or so, for the activist layer of the working class to figure out effective ways to deal with the new unfolding situation. For example, the rapid growth in number of shops stewards and workplace organization in the 1950s and 1960s leading to the ‘peaks’ in strike activity in 1968-74 and 1977-79 (McIlroy and Campbell 2007, 93-130). This was also the case with the introduction of mass production in the first half of the 20th century. I would argue this is the case with the new forms of work (lean production, just-in-time scheduling, team working, outsourcing, extended supply chains, precarious employment, etc. – and, of course, the neoliberal context) that began to take shape in the 1980s and that have so disoriented trade union leaders and activists. The various recent experiments among precarious workers (fast food, warehouse, Wal-Mart, etc.) we see in the US are efforts to figure out what works in this relatively new (and evolving) context. Kevin Crane is right to point to these efforts, but the fact is they flow from the jobs these precarious workers are employed in and the degrading pay and intensified conditions that characterise this work.

The point of ‘social movement unionism’ is not that it by-passes the workplace or job, where it is rooted, but that it ‘takes it to the street’ as well. The Chicago Teachers Union (CTU) could rally the parents and other ‘community’ groups behind their strike and fight against school closings precisely because they had spent several years building exceptionally strong rank and file organization and power in the workplace, the schools, well before they won union office or went on strike. The centre of power and the unifying force was (is) the CTU, not the ‘community’ (Bradbury et al 2014, 9-81).

One important purpose for engaging in such ‘rank and file’ work today, even though strikes may be a ‘foreign experience’ for many, is to prepare for a time (a ‘leap’ or ‘wave’) when strikes and other forms of mass struggle become a more widespread ‘lived reality.’ If the socialists are not at the heart of such struggles they will not have influence or play a leading role and yet another political opportunity will have been squandered. To be there means to have been fighting the ever recurring smaller more ‘hidden’ fights that precede a ‘leap’ or ‘wave’. In all of this, the union is central whether it is the existing (bureaucratic) union or the union-in-formation now or in the future as with the (precarious) fast food and warehouse workers in the US and hopefully here.

Some facts

That precarious work has grown in the last thirty years is undeniable. Prior to 1992, the Office of National Statistics (ONS) did not even gather figures about what they call ‘temporary employees’, a classification which includes most forms of precarious work other than self-employment. The Workplace Industrial (now Employment) Relations Survey (WERS), which began in 1980, noted in 1984 that ‘peripheral’ workers were ‘perhaps increasingly, being used to make closer connections between output and employment’ (Millward and Stevens, 1986, 209-212). In other words, as the economy and output grew in the 1980s and unions experienced defeats, employers increasingly hired ‘temporary’ workers in the name of ‘flexibility’, rather than permanent full-time or part-time workers. The 1984 WERS survey estimated 174,000 workers on ‘fixed term’ contracts (like fractional lecturers) and 161,000 agency ‘temps’ (Millward and Stevens 1986, 210-211). Comparable ONS figures for 2014 are 715,000 and 331,000 respectively (ONS, EMP07, 2014c). So, indeed, things are different from the 1970s or even the 1980s. But what is the actual extent of precarious work in its various forms?

According to ONS figures, the number of all precarious (‘temporary’) employees grew from 1,220,000 in 1992 to a highpoint of 1,712,000 in 1997 after which their numbers declined to a low of 1,399,00 in 2008 (the crisis) and then rose to 1,682,000 by early 2014. In 1992, ‘temporary employees’ composed 5.5% of the workforce, by 1997 it was 7.4% and by 2014 6.5% of all employees (ONS, 2002, 84; 2010, 88; 2014, EMP01). The number rises as the economy grows, and shrinks as it slows down and employers dump their precarious workers, but it is never more than a small (though significant) percentage of the workforce.

These percentages are higher for younger workers and for jobs in service work such as hotels, restaurants and pubs to be sure. So, there is a concentration effect. But overall more public sector employers (48%) use casual workers of various sorts than private sector employers (32%) (CIPD 2013a, 7). These ‘temporary’ workers are about 88% white and more or less evenly divided between men and women (Forde et al, 2008, 12); that is, not that different from the rest of the workforce.

A relatively new development is the ‘zero hours contract’. ONS says there are about half a million, but the CIPD (2013a, 7) more realistically estimates that as of 2013 there were over a million workers on such contracts. These are spread throughout the temporary, full-time and part-time classifications, not in addition to them. Public sector and non-profit (voluntary) organizations are more likely to use ‘zero hours contracts’, at the rate of 28% and 42% respectively, than private sector employers where only about one fifth use such contracts, according to the CIPD survey (CIPD 2013a, 7). The biggest occupational categories are cleaners, care workers, administrative employees, call centre workers, and teachers.

Outsourcing is a well-known feature of ‘lean’ production methods which now extends far beyond manufacturing into service work including even NHS hospitals. In fact, outsourcing and sub-contracting are far more prominent in the public sector (van Wanrooy 2013, 43). Outsourced jobs often are regular and full-time although more poorly paid in most cases.

Nevertheless, outsourcing obviously contributes to a sense of insecurity for workers and bargaining problems for trade unions. One aspect of this that is now gaining increased attention, however, is that many of these outsourced workplaces form part of production supply chains often linked by just-in-time schedules. On the one hand, this contributes to work intensification, speed-up, stress, and increased surveillance – accumulating ‘compression’. On the other hand, these just-in-time chains are highly vulnerable to disruption by workers at key production points and logistics ‘nodes’ (Slaughter, 2012, 8-10). The growing awareness of this is an example of the class learning how to deal with the changes of the last thirty years – more potential for a ‘leap’.

Overall, according to both ONS and CIPD surveys, 70% of the workforce are full-time permanent employees, 20% are part-time permanent, and 10% are ‘casual workers, and self-employed and agency workers in atypical employment relationships’ (CIPD, 2013a, 9). Permanent does not mean ‘life-time’ and never did. Nevertheless, job tenure has not changed much since the 1970s. The average length of job tenure for the whole labour force is about 8 years for women and 9 for men in 2011, with a majority having been employed for five or more years and 85% for more than two years (CIPD 2013b, 4-5). Thus, while precarious work is more prominent today than thirty or so years ago, it is not the most common type of employment and directly affects only a minority of all employed workers. And, perhaps surprisingly, it is more concentrated in the public sector, where union membership is at its highest.

Estimates of how many workers voluntarily quit their jobs vary widely. According to ONS’s Labour Force Survey, ‘voluntary separations’ fell from about 3% of all employees in the mid-1990s to 1.5% by 2011 (ONS 2011, 2). The WERS survey, on the other hand, shows a much higher rate of ‘voluntary resignations’ which were steady at about 14% in 1998 and 2004, then with the recession fell to 9.1% in 2011, the last survey (van Wanrooy et al 2013, 161 ; Kersley et al 2006, 231-232; Cully et al 1999, 127) The difference in the two surveys may be due to the fact that the ONS/LFS questions workers, who are more likely to play down voluntary quits to the ‘authorities’, while WERS surveyed employers, who are more prone to exaggerate such resignations given their generally low opinion of their employees’ dependability. Nevertheless, the high estimates are probably closer to reality. The decline in voluntary quits in both surveys is explained by the fact that workers are more willing to leave a job in relatively good times when alternatives are more likely and less so once the crisis hit and jobs evaporated. ONS figures show that ‘voluntary separations’ were about twice as high among those aged 16-24 than among older workers (ONS 2011, 5). To put this another way, as age increases, even past 24, voluntary quits decline significantly.

It is true that the composition of the labour force and, hence, the working class has changed dramatically since the 1970s. Where there were 6.3 million manufacturing jobs in 1978 there are now 2.3 million (though producing twice what they did in the mid-1980s) (ONS 2014b, JOBS03). This is still slightly more than those working in accommodation and food service (or any other private sector group), so they still have potential social power. They are, however, less concentrated than in the past. The number of factories employing 1,000 or more workers, for example, has dropped from 324 in 1994 to 90 in 2011 (ONS 2012, 600; 1996, 140). As pointed above, however, many are linked in just-in-time chains with some advantages for direct worker action, although greater disadvantages for conventional collective bargaining. On the other hand, there are more women, ethnic minority, and migrant workers in the labour force today, which makes issues of race and gender more central to the workplace than in the 1970s.

Nevertheless, 90% of Britain’s workers have ‘regular’ jobs that last a number of years. The work of those jobs, like that of the more precarious ones, however, has become more intense and stressful in the wake of lean production, ‘new management methods’, frequent reorganizations, bullying, increased surveillance, and competition. In addition, on average they pay less today than in 1973, with wages rapidly falling even further in the last three years, hitting workers ages 20-29 hardest (Guardian 30/01/2015). Of course, even the ‘permanent’ jobs may disappear due to workforce reductions or workplace closures as the declining number of manufacturing workers shows. When asked by the WERS survey in 2011 if their workplace had been ‘adversely affected’ by the recession only 11% reported ‘no adverse effect’. The incidence of pay freezes or cuts rose from 11% of all workplaces in 2004 to 30% in 2011 (van Wanrooy 2013, 15, 87). You get the picture.

What is more, the welfare state, including the NHS as well as more and more benefits, council housing, etc. is crumbling, the Employment Tribunal System increasingly inaccessible, and the ability of workers to rely on the state-provided ‘social wage’ actually less than in the 1970s. None of the ‘major’ and not-so-major political parties are proposing to reverse this. In other words, we are in a state of ‘compression’, as Hobsbawm put it, and the job is a major pressure-point. Because of this the changing situation is also, now as in the past, a major potential springboard for a ‘leap’ in class struggle somewhere down the road. That is why, I think, Ian A. is on the right track.


Bradbury, Alexandra, et al. How To Jump-Start Your Union: Lessons From The Chicago Teachers. Detroit: A Labor Notes Book, 2014.

Campbell, Alan, Nina Fishman, John McIlroy (2007), ‘The Post-War Compromise: Mapping Industrial Politics, 1945-64’ in Campbell, Alan, Nina Fishman, John McIlroy, The Post-War Compromise: British Trade Unions and Industrial Politics, 1945-64. Monmouth: Merlin Press, 2007, 69-113.

Forde, Chris et al (2008), Agency Working in the UK: what do we know? Centre for Employment Relations, Innovation & Change, Leeds University, 2008.

CIPD (Chartered Institute of Personnel and Development) (2013a), Zero-hours contracts: Myth and reality. London: CIPD, 2013.

CIPD (2013b). Megatrends: Has job turnover slowed down? London: CIPD, 2013.

Cliff, Tony (1970), The Employers’ Offensive: Productivity Deals and How To Fight Them. London: Pluto Press, 1970.

Cliff, Tony and Colin Barker (1966), Incomes policy, legislation and shop stewards, London: London Shop Stewards’ Defence Committee, 1966.

Cully, Mark, et al (1999), Britain at Work: As depicted by the 1998 Workplace Employee Relations Survey. London: Routledge, 1999.

Guardian, 31 January 2014, ‘Real wages have been falling for longest period for at least 50 years, ONS says,’’period-ons-record

Hobsbawm, Eric (1964), Labouring Men: Studies in the History of Labour. London: Weidenfeld and Nicolson, 1964.

Kersely, Barbara, et al (2006) Inside the Workplace: Findings from the 2004 Workplace Employment Survey. London: Routledge, 2006.

McIlroy, John and Alan Campbell (2007), ‘The High Tide of Trade Unionism: Mapping Industrial Politics, 1964-79, in McIlroy, John, Nina Fishman, Alan Campbell, The High Tide of British Trade Unionism: Trade Unions and Industrial Politics, 1964-79. Monmouth: Merlin Press, 2007.

McKinlay, Alan and Joseph Melling (2007), ‘The Shop Floor Politics of Productivity’ in Campbell, Fishman. McIlroy. 2007, 222-241.

Millward, Neil and Mark Stevens (1986), British Workplace Industrial Relations 1980-1984: The DE/ESRC/PSI/ACAS Surveys. Aldershot: Gower Publishing Company Limited,1986.

Office for National Statistics (1996), Annual Abstract of Statistics, 2002 Edition London: The Stationary Office, 1996.

Office for National Statistics (2002), Annual Abstract of Statistics, 2002 Edition London: The Stationary Office, 2002.

Office for National Statistics (2010), Annual Abstract of Statistics, 2010 Edition London: The Stationary Office, 2010.

Office for National Statistics (2012), Annual Abstract of Statistics, 2002 Edition London: The Stationary Office, 1996.

Office for National Statistics (2011), Reasons for Leaving Last Job, 2011, 09 November 2011,

Office for National Statistics (2014a), EMP01 ‘Full-time, part-time and temporary workers’, Aug-Oct 2014,

Office for National Statistics (2014b), JOBS03, ‘Employeejobsindustry’, 1978-2014.xls,

Office for National Statistics (2014c) EMP07 ‘Temporary employees (all and part-time)’Jul-Sep 2014,

Raw, Louise, Striking a Light: The Bryant and May Matchwomen and their Place in History. London: Continuum, 2011

Savage, Mike (2007), ‘Sociology, Class and Male Manual Work Cultures’ in McIlroy, Fishman, Campbell. 2007.

Silver, Beverly J. (2003), Forces of Labor: Workers’ Movements and Globalization since 1870. Cambridge: Cambridge University Press, 2003.

Slaughter, Jane (2012), ‘Supply Chain Workers Test Strength of Links’ Labor Notes 397 (April 2012), 8-10.

Van Wanrooy, Brigid et al (2013), Employment Relations in the Shadow of Recession: Findings from the 2011 Workplace Employment Relations Study Houndsmills: palgrave macmillan, 2013.


  1. Some interesting points in this contribution to the debate:

    I do get irritated when people include shorter hours as part of precarity. Some part-time jobs are insecure, or a result of under-employment. But others are not, and are a result of people juggling paid employment with caring responsibilities, or working past what used to be normal retirement age. I don’t think lumping all these categories together under “precarity” is a particularly illuminating of discussing the subject.

    Is this comment accurate for the USA?
    “It’s not just that the cost of unemployment has skyrocketed. The rate of unemployment has also risen dramatically. Not only has the probability that workers will lose their jobs therefore increased. The length of time it takes for the unemployed to find new jobs has grown longer, while the likelihood that the unemployed will have to accept jobs with worse working conditions has become greater.”
    In the UK, it would be wrong to simply read across from high unemployment to a greater risk of losing your job or longer periods out of work. The size of the population either working or seeking work has grown rapidly – so figures for both employment AND unemployment have gone up. The government is deliberately increasing the potential working population further by driving sick and older people into the labour market. So you can’t simply *assume* that increased unemployment is reflected in either of those factors. However, I agree that high unemployment will increase feelings of precarity.

    Overall, it seems rather mechanical in its view that high unemployment means reduced workers’ power, though it certainly doesn’t help!

  2. this is an excellent article. But where does Kim get the statistics to support the following claim “on average they pay less today than in 1973”? The Office for National Statistics produced a report this year stating “Since 1975 average earnings for full time employees have more than doubled after taking into account inflation” . This is not saying that household real wages have risen (because of dual income families) but that real earnings for each worker have risen. There is clearly a difference between “average earnings” and “earnings of the average person”, but there is some correlation between the two. This isn’t denying the attacks since the crisis of 2007, but I doubt it is the truth that the working class has accepted a 30 year constant decline in real wages. My members haven’t anyway. see—uk-wages-over-the-past-four-decades.html

  3. While this is an important article, I don’t think that the 10% figure that the author cites is useful. There are so many jobs in which workers have no contractual protections, legal rights, protection by enforcement agencies, or employment security, that the 10% figure misleads analysis. Some of these jobs are filled by staffing agencies, some are temporary, some are misclassified as independent contracting, some are so deeply sub-contracted that there is no employer to hold accountable. Some are full-time and some are part-time. Both full time and part-time jobs may come with unstable schedules that make it impossible to accommodate other needs and interests. In addition, there are other jobs where the protections are ambiguous, contested, inconsistent, in flux, or disappearing. Even jobs that were formerly well-protected by law and union contract are losing their security; witness the weakening of pension rights among autoworkers and public sector employees.
    A group of scholars, mainly from France, Brazil and the U.S. are working, under the name of Zogris, on analyzing this “grey zone” in employment relations. Our argument is that this grey zone is now the dominant arena of work in the neo-liberal economy that has developed in the past forty years.

  4. There is much good stuff here and I am grateful to the comrades who commissioned the piece as well as Kim himself. A problem thoughis the sentence to the effect that 90% of workers have “regular” jobs. This sounds like Kim is saying that temporary contracts plus agency worker plus self employment come in at less than 10% of the workforce. But since the self employed alone at 4 million people are already 15% of the workforce ( that would be plainly wrong. I think that in coming up with this 90% figure Kim felt he could discard certain categories of seemingly casualised work as not really casual at all. This is potentially a perfectly sensible approach (eg construction unions have complained for years about “bogus” self employment; if you look closely at agency worker it covers both patches of extraordinary degradation with eg some workers on generous “consultancy”-type terms). But even if that approach was properly thought through and justified rather than just asserted, it would still be up for debate. After all, taking the construction example – isn’t a reason for the very low levels of strikes in the sector, precisely the prevalence of subcontracting and agency contracts? The relationships may be more like “employment” than the employers like to let on, but the barriers to organising caused by subcontracting etc are not trivial, we can’t simply wish them away

  5. Some slightly dubious use of statistics here. By presenting figures in terms of %age of the workforce Kim misses much of significance. For instance, out of 42 million people in the UK ages 16-64 just 19.4 million are in full time employment. About 11 million are not in any form of work, 4 million self employed the balance in part time employment. Similarly, wistful invocations of the power of workers at strategic points in the supply chain miss the very small numbers so located. More importantly the article completely fails to address Kevin Crane’s point about the lessened significance of work for workers. One element here, of central importance but neglected in this type of analysis, is the existence of a mass welfare system which means rallying calls around the match girls strike rather miss their target. That system, on an extended definition, both supports a more precarious workforce and is a central method of discipline. The overall social wage has certainly deteriorated in quality sine the 1970’s but it has been extended massively in scope. There is a reason why the Tories have made ‘welfare’ – on their narrow definition – the centre of their ideological assault, not workplace class struggle. The debate therefore should not be simply about the extent of precarity but about the changing nature of work and its social significance – and about extending our understanding of class struggle


Please enter your comment!
Please enter your name here