Working time: yet another study says…

There has been a lot of coverage on the BBC yesterday and today, both via the website and on the news bulletins, of the publication of the results of the shorter working hours trial in Iceland. Judging by the number of comments – upwards of 2,500 at the time of writing – there is a fair amount of UK interest in this (I did break a rule and peek below the line) and, while some comments are clearly misanthropic in tone, there is also a fair amount of genuine objective curiosity. The existence of the trial has been well-publicised and its results have been keenly awaited.

There is, of course, no reason why such an experiment should not work in the UK; and neither is there much to be lost by at least instituting a trial to see what lessons can be learned about the application of shorter hours in this country.

The Iceland trials (there were actually two) come on top of several others worldwide in recent times: a pilot project in Microsoft found that going to a four-day workweek (not necessarily associated with a reduction in working time) in 2019 in Japan boosted productivity by 40 per cent; Perpetual Guardian, a New Zealand trust management company, introduced in 2018 a successful trial of a four-day, 32-hour, working week, later made permanent for all staff, which saw sizable increases in productivity and in worker well-being; while TalkTalk found last year that ‘five days work could be done in four’ (actually a quote from the Chief Executive rather than a quantified research result) as a result of productivity gains reported by a clear majority of workers working from home during the pandemic (NB TalkTalk and Microsoft do have products to sell you as a way of working differently).

Some of the UK interest will have been sparked by the involvement in a joint analysis of the trial by a UK organisation, Autonomy, geared towards analysing the future of work not least in the area of working time. This might give it an specific interest in a successful outcome of the trial but Autonomy has, by the look of its funders, some interesting connections, including on this specific project and incorporating the past and present trade union movement, and therefore what it has to say is of interest. Autonomy’s partner in the analysis is the Icelandic organisation, ALDA (Association for Sustainability and Democracy), a think tank also engaged in analysis of the changing reality of work which openly advocates cuts in working hours. Shorter working hours with no loss of pay also of course featured in the 2019 Labour Party election manifesto, whose section on work included a whole sub-section on working time; and influenced quite heavily by the New Economics Foundation’s identification of the decoupling between leisure time and increases in productivity. Party policy here was driven by the Communication Workers Union, behind one of the major union campaigns to reduce working time in recent years (in the Royal Mail), and itself a backer of Autonomy.

What the trials (firstly in Reykjavík City Council between 2014 and 2019; and secondly in the Icelandic Government between 2017 and 2021 – together entailing nearly 3,000 workers out of a working population of around 200,000) concluded was that a drop in weekly working hours from 40 to 35 or 36 hours per week, with no loss of pay, resulted in productivity and service provision remaining the same or improving across the majority of trial workplaces (a substantial number were involved both in trials as well as in acting as control groups in which there had been no changes); while worker well-being dramatically increased including in terms of perceptions of stress and burn-out, and in health and work-life balance. Furthermore, the trials also remained revenue neutral for both the Reykjavík City Council and the government. While underway, the trials’ evident effects led either to collective agreements being signed between 2019 and 2021 for permanent reductions in working time or for the right to negotiate shorter hours covering a total of some 86 per cent of the country’s working population.

The trials were established following concerns that Iceland’s low productivity, long working hours, poor work-life balance and poor well-being – items which have no little connection with the UK – could be addressed by shorter working time on the basis of the correlation between shorter working hours and increased productivity both in wealthy nations and in individual workplaces. Indeed, the trials were set up to explore the veracity of these links within Iceland. This might be thought to give the trials something of a head-start since there is a degree of vested interest in their success: the trials were set up to prove on the ground something thought of as likely. However, this does not negate the full data gathering exercises, both qualitative and quantitative, that accompanied the trials – and we should also not forget that the trials were lengthy enough for workers to have established and embedded new routines by the end of them. We should also not forget that the trials did not just drop working hours – the intention to retain service provision levels required the trials to set about the reduction intelligently, by re-thinking some tasks and working practices while dropping others to re-organise work around the most efficient activities. As with Microsoft in Japan, this frequently entailed cuts in meetings.

Working hours of full-time workers in the UK are lower than in Iceland – 39 is the average (according to the 2019 Annual Survey on Hours and Earnings, picked deliberately to omit the effects of Covid-19 on the 2020 provisional ASHE data), and this has inched incrementally downwards from 40 in 1997. The equivalent in the UK of the Icelandic trials would thus be not to move to a four-day week, as much of the debate has recently envisaged, but to a lunchtime finish on Fridays. Thus, the Iceland trials were actually quite relatively conservative in their ambitions (and in this respect we should also note that the trials were, at the same time, also more radical in terms of the hours cuts than the agreements which have subsequently introduced shorter working time in Iceland which are, for the most part, considerably more modest).

More does indeed need to be done in this direction but, as a trial, the Iceland experiment successfully points to the direction of change; with lessons that are common to both the public and the private sectors. And there is plenty of arguments as to why shorter working time needs to happen, including the slowing down of the historic trend towards reductions in working time which would see workers in the UK on target for a 30-hour week by 2040 had pre-1980 trends continued; automation and the difficulties with implementing a robot tax; the failure of the share of national income going to workers to keep pace with productivity improvements across Europe (and the US and Japan) in recent decades; and the need to ‘build back better’, in the workplace as much as elsewhere, after the pandemic. It’s not as though workers are looking for something for nothing in this area as survey evidence, such as from Kronos Incorporated, has noted: there is real appetite among workers across the globe for a four-day week in which good employers will be ahead of the curve.

The main lessons from Iceland would seem to be these:

  • the requirement for revenue neutrality is not essential, but it was an important component in these particular trials. In some places – either on a country-wide basis or in individual workplaces – there might be a desire to invest in reducing working hours, recognising the disconnect between productivity and working time to which the NEF has pointed, and redressing the existing imbalance between productivity and the labour share
  • there is, otherwise, indeed a link between reducing working hours and productivity increases where the attempt is made strategically to re-design or re-organise work around the more productive activities
  • working time reductions need to be actively introduced if there is to be radical, rather than incremental, change in working time in the future
  • staff need to be actively engaged in the design of the programmes to achieve the aims of retaining service provision; and monitoring committees need to encompass trade unions not least from the perspective that organisational changes may have a damaging effect on some staff even where working time is less. Organisational change, even when implemented to achieve reductions in working time, is rarely a painless experience
  • a replacement of meetings with e-mails may well be a productivity solution which has a somewhat slimmer chance of working in the UK than it apparently did in Iceland – hence the importance of locally-negotiated solutions in which staff are engaged in identifying what will work best
  • take-up of options across a range of negotiated settings may well vary from sector to sector, recognising different job loading and peak gearings
  • managers need be involved in the programme too, not least in terms of setting examples to and acting as role models for those they manage. This means, in the UK, addressing unpaid overtime activities – and it also means active policy engagement with the ‘right to disconnect’ for which Prospect is currently campaigning: the same tools that facilitate improvements can also be used to depreciate working conditions and we need specifically to ensure that workers’ own practices start from a protected right to switch off.

As the UK’s Covid-19 lockdowns ease and calls are made from the usual Luddites for a return to an office-based way of working, either as a result of a desire for managerial control or else to ‘stop the city from crumbling‘, and as concerns rise over the climate impact of travel costs and interest grows in shorter working hours in this domain, too,* the publication of the Iceland trial data is a timely reminder that, not least under this current UK government, whose concern over working time is a well-established point of debate, a post-pandemic future that is not ‘the same as before’ won’t just fall into our laps: it has to be won, which means articulating it and organising around it.

As always: your quickest and best route to getting organised is to join a union.

* A report also funded by Alex Ferry Foundation while Autonomy is a supporter of the 4 Day Week campaign which produced it.

4 thoughts on “Working time: yet another study says…

  1. That’s a useful commentary Calvin (and I’ll probably share the link on LinkedIn).

    With regard to your statement “Working hours of full-time workers in the UK are lower than in Iceland – 39 is the average”, is that referring to contract hours or hours worked?

    Unless it’s the former, it seems difficult to reconcile this figure with the stats quoted around Work Your Proper Hours Day (i.e. is the report only looking at working time that is paid?).


    1. Thanks Malcolm: after a bit of digging, I think WYPHD uses the data from ONS’s Labour Force Survey, not ASHE. (Actually, I think they get the figures via Eurostat which quotes an annual figure for ‘hours worked per week of FT employment‘ rather than pick one of the quarterly Labour Force Surveys ) There are major differences between the two – ASHE being a survey of 1 per cent of employees drawn from company records; the LFS being based on responses submitted by individual respondents in 15,000 households. There are advantages and drawbacks to each method.

      In hours terms, ASHE ought to provide a precise estimate of the hours worked and paid for within that 1 per cent sample (of those paying NI); the LFS will be based on responses to a series of questions about usual hours worked, including (and excluding) overtime, paid hours, paid and unpaid overtime hours, etc. Accuracy (or recall) will vary between the two and, while ASHE ought to be more authoritative; the sheer range of questions on hours within the LFS does give greater comparability between the different components of survey respondents’ working hours. ASHE won’t include unpaid overtime. Its figures are based on a mean basic working week of 38.1 hours and a mean overtime of 0.9 hours. So, the difference between the two will be based on the level of unpaid overtime.

      The full LFS dataset is available and, while not all the variables may be publicly available, this does raise some interesting questions about what it says about the relationships between those different components. I might well have a look at the dataset and see what I can find out…

      But, the answer to your question essentially is that I am using ASHE; whereas WYPHD uses the LFS.


  2. Hi Calvin, great article and I love the use of misanthropic. A much underused term especially with todays social media!


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s