By Judy Stephenson (University College London)
How many days a year did people work in England before the Industrial Revolution? For those who don’t spend their waking hours desperate for sources to inform wages and GDP per capita over seven centuries, this question provokes an agreeable discussion about artisans, agriculture and tradition. Someone will mention EP Thompson and clocks or Saint Mondays. ‘Really that few?’ It’s quaint.
But, for those of us who do spend our waking hours desperate for sources to inform wages and GDP per capita over seven centuries the question has evolved in the last few years into a debate about productivity and when modern economic growth began in an ‘industrious revolution’. A serious body of research in economic history has recently estimated increasing numbers of days that people worked from the late seventeenth century. Current estimates are that people worked about 270 days a year by 1700, rising to about 300 after 1750.
The uninitiated might think that estimates of such important things like the working year would be based on some substantive evidence, but in fact, most estimates of the working year that economic historians have been using for the last two decades don’t come from working records at all. They come from court depositions where witnesses told the courts when they went to and left work, or they come from working out how many days a worker had to toil to afford a basket of consumption goods. This approach, pioneered by Jacob Weisdorf and Bob Allen in 2011, essentially holds welfare as a constant throughout history, and it’s the key assumption made in a new paper on wages forthcoming from Jane Humphries and Jacob Weisdorf. Unsurprisingly for historians familiar with material showing the miserable conditions under which the poor toiled in eighteenth century Britain, this calculation frequently leads to a high number of days worked. It also implies that Londoners, due to higher day wages, may have had slightly more leisure than rural workers. Both implications might appear counterintuitive.
Knowledgeable historians, such as John Hatcher, have pointed out that the idea that anyone had 270 days paid work a year before the industrial revolution is fanciful. But unless there was an industrious revolution, and people did begin to work more days per year in market work – as Jan de Vries posited – the established evidence firmly implies that workers became worse off throughout the eighteenth century, because wage rates as measured by builders wages didn’t increase in line with inflation, and in fact builders earned even less than we thought.
My article, “Working days in a London construction team in the eighteenth century: evidence from St Paul’s Cathedral” forthcoming in the Review, takes a different approach: it uses the actual working records of a team of masons working under William Kempster who constructed the South West tower of St Paul’s Cathedral. For five years in the 1700s, these archives are exceptionally detailed. They show that building was seasonal (it’s not like we didn’t know – it’s just we had sort of forgotten), and building was stage dependent, so not all men could have worked all year. In fact, they didn’t. Surprisingly, for a stable firm at an established and large site, very few men worked for Kempster for more than about 27 weeks. Work was temporal and insecure, and working and employment relationships were casual.
If one was to take a crude average of the days each man worked in any year it would be less than 150 days. To do so is obviously misleading and that’s not what the paper claims, because obviously men worked for other employers too. But, what the working patterns reveal is that unless men seamlessly moved from one employer to another with no search costs or time in between, it would have been impossible for them to have worked 250 days a year. Its more plausible that they were able to work between 200 and 220 days.
Moreover, the data shows that men did not work the full 6 days per week on offer. The average number of days worked per week was only 5.2. This wasn’t because men did not work Saint Mondays (which are almost indiscernible) but because they took idiosyncratic breaks. Only the foremen seem to have been able to sustain six days a week.
However, men that had a longer relationship with Kempster worked more days per year than the rest. This implies that stronger working relationships or consolidation of employers and workers relationships might have led to an increase in the average number of days worked. However, architectural and construction historians generally think that consolidation in the industry did not occur until the 1820s. If there was an industrious revolution in the eighteenth century it might not have happened for builders. If builders’ wages are representative – and that old assumption seems increasingly stretched these days – then the story for wages in the eighteenth century is even more pessimistic than before.
The evidence from working records presented in this article paper are still relatively fragmentary but they do clearly show that holding welfare to be stable by calculating the number of days worked from consumption goods – as the Weisdorf/ Humphries/ Allen approach does not give us the whole story.
But then again, is it really plausible to hold welfare stable? The debate, and scholarship no doubt will continue.
To contact the author: