In this blog post, Professor Sara Horrell of London School of Economics and Political Science introduces her latest article for the Economic History Review, about “Household consumption and the consumer price index, England, 1260-1869”.
The long-run level of real wages forms the cornerstone of recent explanations for England’s early economic growth (Allen, 2001). High wages arguably underpinned north-west Europe’s adoption of the European Marriage Pattern after the Black Death, incentivised investment in human capital, and assured long-run growth (van Zanden, 2009). Continued high wages led entrepreneurs to economise on labour costs by substituting technology, so creating the conditions for industrialisation (Allen, 2009). Elsewhere lower levels of real wages resulted in the Little Divergence within Europe and the Great Divergence when north-west Europe is compared with Asia. Real wages are central to these accounts, leading to the accuracy of some of these wage series to be questioned.
While considerable attention has been given to the validity of the nominal wages collected, far less has been paid to the deflator, the consumer price index. Indeed, Allen’s pricing of the quantities of those goods necessary to assure an individual a ‘respectable’ standard of living has proved highly tractable over time and across space. However, history and theory concur that expenditure on an unchanging, pre-modern basket of commodities will not accurately capture longer-run changes in the price level. Consumption shifted from a medieval diet of grains, pottage, beef, and beer to a fare of wheaten bread, tea, sugar, and bacon during industrialisation. Domestic comfort improved and fashion-conscious consumers eschewed coarse woollens in favour of brightly printed cottons. Theory too requires the consumer to be able to revise the bundle they purchase in response to changes in income and in relative prices if the consumer price index is to reflect the resultant alteration in welfare. Pricing a fixed basket of goods that assures basic needs are met can be expected to become increasingly inaccurate as a measure of living costs as consumption patterns evolve.
Here we investigate the effect of allowing the expenditure weights in the cost-of-living index computation to change. Detailed and varied sources are used to construct the household budget of a labouring family of five people at eight points through the six centuries under study. The sources range from medieval accounts of small holders’ activities, information on peasant housing and dress, and maintenance agreements drawn up for retired property owners; through early modern probate inventories, records of diets fed to inmates of institutions, and the value given to items of clothing that appeared in court records having been stolen and the theft then prosecuted; to household budget data, reports of family possessions, and estimates of self-provisioning activities in the industrialisation era. The resultant expenditure shares illustrate the changing importance of broad commodity categories, such as bread, beer, household goods, and linen, through time (see Figure 1). More detailed shifts in consumption are evident once the analysis is expanded to allow 33 categories of expenditure, including potatoes, sugar, tea, tobacco, household furniture, and services.
Existing price series (Allen, ‘Data’. Clark (2005); ‘Macroeconomic’) are used to construct a chained-Laspeyres cost-of-living index. This uses these detailed expenditure accounts as the basis for creating a price index for the surrounding years, described as a link, then joins these links into a chain to construct the consumer price index from 1260 to 1869 (see Figure 2).
Comparison with Allen’s cost of the respectability basket from 1260 to 1830 and Clark’s expanded index 1260 to 1869 reveal surprisingly few differences. While the tendency for a fixed-weight index to overstate price rises in inflationary times is evident, significant divergence between the different estimates of the price level only occurs from the later eighteenth century. The close correspondence earlier is attributed to the strong dependence of most items in ordinary households’ consumption on agricultural output. The prices of these goods tended to move in tandem, so offered few opportunities for substitution into alternative products to combat the effect of price rises. Eighteenth-century trade introduced exotic groceries (such as sugar) and raw material inputs that loosened the ties to the fortunes of domestic agriculture. The advent of industrial manufacturing further reduced the reliance of consumption on organic outputs by offering alternative fuels, new materials, and a range of novel products. The industrial era brought with it a changed dynamic in household consumption choices. Industrialization, economic growth, and improved incomes drove a wedge between the ‘respectability’ basket and actual consumption. The true cost-of-living could no longer be measured using a fixed basket of consumables, instead changing consumption patterns had to be acknowledged and captured by using the chained-Laspeyres method to construct the consumer price index.
Allen, R. C., ‘The Great Divergence in European wages and prices from the Middle Ages to the First World War’, Explorations in Economic History, 38 (2001), pp.411-47.
Allen, R. C., The British Industrial Revolution in global perspective (Cambridge, 2009).
Clark, G., ‘Condition of the working-class in England, 1209-2004’, Journal of Political Economy, 113 (2005), pp.1307-40.
Van Zanden, J. L., The long road to the industrial revolution: the European economy in global perspective, 1000-1800 (Leiden, 2009).
Allen, R.C., www.nuffield.oc.ac.uk/people/sites/allen-research-pages/ Data – Age and Price History, London.xls.
Clark, G., ‘The macroeconomic aggregates for England, 1209-2008’, Revised version October 2009, UC Davis working paper 09-19. www.core.ac.uk/download/pdf/6832298.pdf.
Contact the author: