Spinning little stories: Why cotton in the Industrial Revolution was not what you think

Judy Stephenson reviews some of the developments in debates about causes of the Industrial Revolution from this year. 

When Nick Crafts reviewed competing ‘meta-narrative’ explanations of the Industrial Revolution by Joel Mokyr and Robert C. Allen in 2010 he noted that explanations of the cause of the IR were a bit like the Holy Grail (1). He was expressing the feeling of a generation of economists who believed that economics could ‘explain’ the modern world, and so must explain the IR. When Deirdre McCloskey’s critique (2) of every meta narrative ever generated about the Industrial Revolution declared that economics could do no such thing it was indicative of a shift generally away from that kind of thinking, or as John Kay said, there are no meta-narratives, only little stories. But in economic history, Allen’s meta narrative has largely prevailed. A recently published little story has questioned it.

 Allen ‘explained’ the Industrial Revolution by claiming that because Britain was a ‘high wage’ economy in the eighteenth century, the high cost of labour and the relative cheapness of coal and capital incentivised labour saving mechanisation, and this is why Britain industrialised before other countries. The theory has met with challenges already. In 2013 Humphries undermined the assumptions about household conditions, and since she has also produced new wage data with Jacob Weisdorf that is fundamentally at odds with the day wages Allen used. (3) (I have shown to these be too high, 4). Humphries and Ben Schneider have also shown that spinning was a very low wage activity (5). Whilst all this undermines Allen’s theory, a well told little story by John Styles in the latest edition of The East Asian Journal of British Studies, is notable because it challenges much of what we understand about the innovations at the core of the IR.

http://www.history.ac.uk/sites/history.ac.uk/files/eajbhvol5.pdf see p.161 onwards.

 Allen’s favourite and oft used case of the effect of British factor prices has always been that of the spinning jenny. In ‘The Industrial Revolution in Miniature’ (6) he sets out high British spinning and labouring wages, and low French wages to show that the jenny was only economical in England. Styles brings some other facts to the case. It’s enjoyable reading so I won’t attempt to reproduce it here but suffice to say he also shows in the first part of the paper that French wages were higher, British lower, and there were more jennies in France than thought.

 Whilst every economic historian knows that whoever says the IR says ‘cotton’, what many probably didn’t know is that ‘cotton’, in England, for most of the eighteenth century meant cotton weft spun on linen warp. The inability of English spinners to create cotton warp strong enough to go on larger frames needed for calicoes meant that English ‘cottons’ were a cotton linen mix, which, although popular and cheap, was not a match for the colour and fineness of proper cotton calico. The burgeoning American market wanted calicoes above all else, and to provide it, and tap into that valuable demand, properly spun cotton warp was the only answer. The spinning jenny did not provide such warp, and whilst the calico acts protected or sheltered the home market it was not until Arkwright’s water frame that English cotton could conquer the profitable American market.

In the story of the spinning jenny, high wages (nor cheap coal) had no part to play. It was Arkwright’ invention which fundamentally changed the production of cotton and which met the demand for fine new cotton fashions, and the incentives are far less clear in this story. Styles makes the important point that Arkwright’s macro-invention was the “outcome of a long history of applying capital-intensive, mechanical solutions to quality and supply problems in luxury textile manufacturing” (see particularly pp.186-7). This is not all bad news for Allen. Styles is clear that the wage for those who could produce good warp was very high, but Arkwright, nor others, could not produce enough of it at the volumes needed at any wage. As many readers will understand this little story has implications for our understanding of jennies as ‘macro’ or ‘micro’ inventions, and so for Mokyr too. Bear in mind that Wallis, Colson, and Chilosi and separately Keibek have both shown this year that industrialisation was as much 17th century phenomenon as 18th, so we may need some new models anyway (7). 

 Styles contribution highlights how a strong empirical basis for economic analysis is essential if meta narratives or ‘big theories’ are to explain economic developments of the past. Allen has always stressed the comparative level of wages in Britain, and there is work to be done on current sources here. (There is a developing Twitter conversation between myself, @pseudoeramus @VincentGeloso @MarkKoyama @benmschneider @ulyssecolonna regarding this subject and market size for instance). But in a year where the notion of economic rationality itself has been shaken to its foundations perhaps it’s not surprising that the Industrial Revolution is moving back towards being ‘unexplained’, although we should await reviews of Mokyr’s latest contribution, (which seems to chime with McCloskey’s ‘cultural’ one) before returning to the view of a couple of decades ago that there really wasn’t one at all.

In the meantime research on the coal tax in early modern England is long over due….

Notes:
(1) http://wrap.warwick.ac.uk/44710/1/WRAP_Carfts_10.2010_craftsindustrial.pdf

R..C Allen, The British industrial revolution in global perspective, CUP, 2009.

J. Mokyr, The enlightened economy: an economic history of Britain, 1700-1850, Yale, 2009.

 (2) McCloskey, Deirdre N. Bourgeois Dignity : Why Economics Can’t Explain the Modern World. Chicago: University of Chicago Press, 2010.

http://www.johnkay.com/2011/10/04/the-map-is-not-the-territory-an-essay-on-the-state-of-economics/ http://www.johnkay.com/product/the-truth-about-markets/

(3) http://www.ehs.org.uk/app/journal/article/10.1111/j.1468-0289.2012.00663.x/abstract

http://www.economics.ox.ac.uk/Oxford-Economic-and-Social-History-Working-Papers/unreal-wages-a-new-empirical-foundation-for-the-study-of-living-standards-and-economic-growth-1260-1860

(4)http://www.lse.ac.uk/economicHistory/workingPapers/2016/wp231.pdf

(5)http://www.economics.ox.ac.uk/Oxford-Economic-and-Social-History-Working-Papers/spinning-the-industrial-revolution

(6) Allen, Robert C. “The Industrial Revolution in Miniature: The Spinning Jenny in Britain, France, and India.” The Journal of Economic History 69, no. 04 (2009): 901-27.

(7)http://www.lse.ac.uk/economicHistory/workingPapers/2016/WP240.pdf , for Keibek see this blog 

  J.Z. Stephenson, Wadham College, Oxford. judy.stephenson@wadh.ox.ac.uk, tw:@judyzara

10 thoughts on “Spinning little stories: Why cotton in the Industrial Revolution was not what you think

  1. Dear Judy, I’m writing a blog post for NEP-HIS about Humphries and Schneider’s paper, but it will come out only on December 19th. So, because of the debate on twitter today, I decided to make some earlier comments. For me, the main problem with Humphries and Schneider’s paper (and the other paper with Weisdorf) is, in the words of Robert Allen, that “Humphries never analyses the British labor supply from an international perspective.” John Styles argument about France is interesting, but we need better data on France’s wages before claiming that “the route to mechanization and factory production was a response to low not high wages.”

    About the weft and warp, Arkwright’s invention improved warp production, but it was not suitable to produce high count yarns. The Mule allowed this and the reduction in short-stapled cotton imports (Smyrna and most of West Indies) is significant after 1780. With the mule, demand for longer and more resistant raw cotton was a consequence of an increase in production of better fabrics (as the debated between Harley and Cuenca Esteban showed). We can see this on reports from Manchester producers that relate how different cotton staples were used in different fabrics.

    Like

    1. I agree. The comparative issue is what it all rests on and the current sources on wages and prices for France mean that we cannot be accurate about that at all at the moment.
      J

      Like

    2. Yes but the other thing shown by the Cuenca-Esteban/Harley debate (especially Harley 1998) is that although the high-count yarn prices fell the most in the classic IR period, they were not economically very significant. Most cotton products used relatively low-count yarn. So while Crompton’s invention was important, I think Styles is correct to put the emphasis on Arkwright. The binding technological constraint on the British ‘cotton’ industry was its inability to produce cotton warp at manageable prices.

      Like

      1. I think we first need to define what is a “high-count” yarn. Harley argues (1998 EHR paper) that fine yarn represented counts of 100 or more. With this definition, I agree with you, their share was small. In the 1780-1800 period, however, counts above 29 were already out of reach for the raw cotton used in fustian production, the old cotton sector.
        What I was thinking when I made the comment is that a large part of the calicoes and muslin production that expanded with Crompton’s mule could be classified as higher count yarns between 1780-1800 because they were the new industry. Counts above 41 (in 1788) used only very high quality raw cotton, and represented 16% of all raw cotton used in textiles production. Since the prices of these fabrics (counts above 41) were much higher than the low-count fabrics, they overall importance was significant.
        Harley writes that “Before the spinning inventions, most yarns were below 20 count;
        towards the end of the century the average count had increased to the
        high twenties; by the 1830s it reached about 40.” (1998 p.55). So the definition of a high-count yarn could change over time. In the spinning debate, I think the end of the eighteenth century is the most important period to look at and, therefore, classified high-count yarns as 100 or above is not so useful.

        Liked by 1 person

      2. I guess we must clarify what is being argued.

        If the question is what incentivised inventions, then…

        Ex ante, it’s quite plausible that inventions were motivated by the highest markup sector (in this case, calicoes and muslins) rather than by the cheaper stuff.

        But that leaves the question of whether the mule was quasi-inevitable given the previous inventions (the jenny, the frame).

        If the question is economic consequence of inventions, ex post, then I think the role of calicoes and muslins is usually a bit exaggerated. It’s the cheap stuff that ultimately made Lancashire.

        Harley 1998 points out that in 1818 that printed calicoes accounted for 15% of the yarn spun. The % of value added is not given but was surely higher. Muslins, a subset of this, were smaller.

        Like

  2. Agree –> “But that leaves the question of whether the mule was quasi-inevitable given the previous inventions (the jenny, the frame).”

    “ex post, then I think the role of calicoes and muslins is usually a bit exaggerated. It’s the cheap stuff that ultimately made Lancashire.” Ok, the machine made calicoes a cheap stuff over time, but as Baines argued in 1835, the “cheap textiles” only became important in the 1830s. I think this is something that we need to research a little more, but just let me point out that in 1818, “printed calicoes” represented 49.2 percent of British total “manufactured cotton” export value (based on declared, not “official” prices). Plain calicoes 23 percent, and plain muslins 16 percent.

    Also, James A. Mann (The Cotton Trade of Great Britain: Its Rise, Progress and Present Extent) shows that in 1779, the profit from a 80 count yarn was almost 3 times higher than a 40 count yarn (which entered the super fine calico category).

    Anyways, my remarks are more to help me think about this issue than to create an endless debate.

    Like

    1. Thales: Not endless bickering, just endless IR geekery !

      1- About profit levels in 1779, that makes perfect sense, and I think those are the right incentives to talk about for inventors. But 1779 markups were almost certainly supernormal profits. After the Napoleonic Wars the ratio of profit margins between the counts of yarn almost certainly diminished.

      2-Speaking of something that needs further research…. We should never again infer total production just from exports (i.e. making assumptions about the size of domestic consumption from exports à la Deane & Cole or was it Hoffmann?). Surely it’s high time someone looked at Excise rather than Customs to estimate the value of domestic consumption. (Ashworth doesn’t count.)

      Like

  3. It’s been fascinating seeing the responses to this post on Twitter and in blogs. Thank you Judy for setting off such a fruitful discussion. I’m keen to offer my own responses, but I don’t have a Twitter account and I’ve more to say than will fit into a blog reply box. So I’ve posted a set of comments on the blog page of my own spinning-wheel.org website, under the title ‘Spinning little stories about the High Wage Economy’ (www.spinniing-wheel.org/blog). It addresses three issues that have been raised in this debate:-
    1. Robert Allen, induced innovation and the High Wage Economy, and in particular the pitfalls of Franco-British comparisons.
    2. Spinning earnings and work intensity.
    3. Yarn counts and quality.

    Like

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s