Judy Stephenson

The concept of an ‘industrious revolution’—a period when household productivity and consumer demand increased before industrialization, generating surplus for investment in new technology—has been influential since the late 1990s. For economic historians, the measure of industriousness is the number of days people worked per year. For anybody who was paid by the day, annual income was a function of the portion of the day rate that they received, and the number of days that they received it for. How many days people worked per year is therefore of profound importance to understanding preindustrial living standards, as well as economic growth.  

How many days did people work per year? 

We have surprisingly little information about the number of days people worked in the past. What economic historians know about what people earned in the past has been taken from ‘day wages’, collected since the mid 19th century, from the archives of big institutions. These are mostly for masons, carpenters, bricklayers and their labourers for building work. (It turns out that many of those day wages are labour costs rather than wages.)  

Detail from ‘Political electricity; or, an historical & prophetical print in the year 1770‘.

Standard practise in calculating annual wage income is to assume that most people were working nearly all the time; like today’s full-time jobs. For instance, Bob Allen in his seminal work on real wages assumes that the average building and agricultural labourer was working 250 days a year. Greg Clark assumes that the average was 300 days. 

These are high numbers. In the latter case for instance, they imply that more than half of the workforce were working 300 days or more. Since we know that the working week was six days long, a 300-day working year implies that the average worker only missed two weeks of work all year. Just Christmas and Easter? No time off at midsummer? No ‘saint Mondays’ (the early modern equivalent of a duvet day)? No searching for work or odd jobs? 

Marc Chagall, Reclining on the bed.

Even at 250 days a year, the implied level of labour intensity assumes that the average worker was pretty much fully employed. Those retained on annual contracts and apprentices were expected to work all days in the year it must be said, but that was not the predominant proportion of the workforce.  

The most recent research accepts that these assumptions need some qualification. Humphries and Weisdorf (2019) use the price of goods to calculate the number of days work needed to afford a basket of goods as the key to industriousness. However, since wages were sticky or stable in many periods, and prices were not, unsurprisingly this research finds that ordinary workers in the late 17th century and late 18th century (two periods when prices were high) worked over 250 days annually. The implication of this approach is that the standard of living was constant – an assumption which is firmly contradicted by history.   

Although historians understand that the regularity of work was somehow different before the industrial revolution, there has been surprisingly little progress on studying actual work patterns from actual workplaces. There are lots of good reasons for this, the best being that there are not many records. Until production in factories or large workplaces became the norm, most employers did not record the number of working days of their employees, because they paid workers for their output not their labour input. 

Jean-Francois Millet, A stonemason.

Most manufacturing workers and many service workers were paid by the task or by the piece. In this way, the early modern economy was very much like a gig-economy. Only people in construction, shipbuilding, or casual day labouring would have been regularly paid by the day, and it is only at exceptional sites that you can get a full picture of how many people were on site, because the nature of the contracting system was that not all work was accounted for by the day.  

Seasonal patterns

When you do look at actual working records, though, two things become apparent. The first is that working life and production before the industrial revolution was profoundly seasonal. Not much happened in Britain during January and February. Construction site records indicate significant industriousness and high labour intensity on projects until the end of December, followed by a period of quiet in January and early February, where particularly on building projects only the most essential tasks were done, largely for lack of light.  

Many other trades also show a seasonal slowing of trade and business at this time, as transport by roads and waterways was made more difficult due to colder weather conditions. Hence the proverb “Candlemas day, put beans in the clay; put candles and candlesticks away”, as people urged a fresh start after a dark January to prepare to sow beans and work in the new light.   

James Patrick, Sunset on Snow. East Dunbartonshire Council.

Building records from the 18th century often indicate that it was early March before you might see a full team back on site. Some of the best records we have are from the rebuilding of St Paul’s Cathedral. Over the decade 1700-1710 at St Paul’s, one key mason contractor’s team worked on average about 45 percent fewer days in January, February and March than in June, July and August. The cathedral’s own labourers worked half the days in January they did in July.  

In agricultural work, too, the important rest and maintenance time of the dark months gave way to high work intensity in spring. At the docks there was less to unload and pass on, the seasonality of most trade’s shipping being profound. In shipbuilding the numbers of men on site would increase by a third or more from one end of the year to another.  

Precarity

The second observation is that most preindustrial workers did not have a ‘steady’ job. Work was often precarious – much like zero hours contracts today. Building workers moved from site to site, and were not permanently retained by employers. It could take time to find the next job. In textiles, spinners, weavers and knitters worked to order and had to allow time to get materials and supplies from their employers, and time to bring in their work and have it measured and weighed and accounted for. The amounts they worked for orders varied.  

Mark Senior, The Flax Spinners, Rotterdam. The Hepworth Wakefield.

In mining it seems to have been expected that hewers and heavers did not work every day. In potteries and foundries, the working week included much preparation time worked less intensely (on Mondays and Tuesdays) with hours adapted to the demand of customers, with shorter stints worked in times of depression and overtime in time of high demand.  

Of course, the existence of overtime indicates that sometimes people worked more than six days a week. The Royal dockyards are a classic example of this. Overtime and double stints paid high wages when times were good – or war was declared. But at other times, servants and non-regular men were laid off for months at a time.

It is also notable that even when a workplace operated six days a week it was unusual for all workers to work all six days, or get paid for them. At the potteries, the average was between five and six days, and analysis of building site records show that often many people were on site for three to four days a week.  

Conclusion

Only if a full working week was sustained by everyone, without any search costs of looking for other work or moving between sites and employers, could the average number of days worked in a year be above 250 in a workplace. And of course, the early factories, like those of Richard Arkwright, or Benjamin Gott, offered working conditions that encouraged and supported such full-time work – and, in fact, required it. That a labour force unused to such time discipline was not always willing to work to this regularity, is well established.  

Men at work inside a plate glass factory, 1747. Wellcome Collection.

The case for an industrious revolution before regularised factory work is therefore a bit shaky. It certainly seems to be the case that in order to afford new goods or to maintain a standard of living, workers had to work more, but records tell us that precarious employment is nothing new, and work was uncertain and badly paid for many. 

Further reading

• Allen, R. C., ‘Prices and wages in SE England’. https://www.nuffield.ox.ac.uk/people/sites/allen-research-pages/ 
• Allen, R., and Weisdorf, J., ‘Was there an ‘industrious revolution’ before the industrial revolution? An empirical exercise for England, c. 1300—1830’. The Economic History Review 64:3 (2011), 715-729. 
• Baugh, D. A., British naval administration in the age of Walpole (Princeton University Press, 2015). 
• Burnette, J., ‘Seasonal Patterns of Agricultural Day-Labour at Eight English Farms, 1835–1844’, in Hatcher and Stephenson (eds.) Seven Centuries of Unreal Wages (London, Palgrave, 2018).  
• Boulton, J., ‘Economy of time? Wedding days and the working week in the past’, Local Population Studies 43 (1989), 28-46 
• De Vries, J., The industrious revolution: Consumer behavior and the household economy, 1650 to the present (Cambridge, 2008). 
• Humphries, J., and Weisdorf, J., ‘Unreal wages? Real income and economic growth in England, 1260–1850’. The Economic Journal 129:623 (2019), 2867-2887. 
• Lipson, E., The history of the woollen and worsted industries (Psychology Press, 1965). 
• Riley, J., ‘Sickness in an early modern workplace’. Continuity and Change 2:3 (1987), 363-385.  
• Stephenson, J.Z., ‘Real’ wages? Contractors, workers, and pay in London building trades, 1650–1800’. Economic History Review 71:1 (2018), 106-132. 
• Stephenson, J. Z., ‘Working days in a London construction team in the eighteenth century: evidence from St Paul’s Cathedral’. The Economic History Review 73:2 (2020), 409-430. 
• Thompson, E., ‘Time, Work-Discipline, And Industrial Capitalism’. Past and Present 38:1 (1967), 56-97.