skip to primary navigation skip to content
 

Other sources of occupational data

Other sources of occupational data

In addition to sources with a relatively wide temporal and geographic range - Anglican parish registers, testamentary documents, court records, coroners' inquests - sources of high quality occupational information (although usually on men only) are sometimes available for specific years and/or areas. These have been systematically explored and digitised as part of the Occupational Structure Project's data collection effort. Two major examples are discussed below

Poll tax returns

Carolyn Fenwick's efforts have made the returns of the 1377, 1379, and 1381 poll tax available to historians by organising and transcribing these sets of documents, many of which are in poor condition and very difficult to read.1 These returns provide information on a significant number of English men and women, and have been used by historians before to examine the occupational composition of the late medieval labour force before, notably by Clark, for his analysis of the long-term development of the agrarian labour share, and by Broadberry et al in their recent reconstruction of Britain's historical national accounts.2 The poll tax of 1381 has survived most completely and has more information on female occupations than the other two. It was aimed at all individuals of fifteen years and older, except for the poor and those in holy orders. Many of the returns have survived, although often in a poor state, and some contain occupations. Altogether, they provide the occupations of '30,292 individuals (approximately 2 per cent of all adults)' living in 892 vills from many parts of England, with the fact that many of these individuals are women being 'a particular strength'.3

Richard Smith, a leading member of the Cambridge Group, has, in a number of conference papers, critically assessed existing estimates based on the 1377-81 poll taxes.4 Much of the problematic nature of these estimates is the direct result of the similarly problematic nature of the occupational information provided in the poll tax returns. It is, for example, clear that women are underrepresented, but establishing an integral, male-and-female occupational structure is difficult as it requires assessing female labour force participation in full-time equivalent terms, for which no firm data are available. Other problems with the existing estimates are: the allocation of servants to domestic service when the vast majority, in reality, were farm servants; the unrepresentativeness of many of the vills in the sample with regards to the county estimates derived from them. Taking such issues into account, Smith estimates differ considerably from the existing ones; a paper presenting these new estimates is in progress.

Although the most geographically extensive and important, the 1377/81/83 returns are not the only surviving poll tax returns which contain occupational information. An example of a later return of this kind is the one for the Cheshire hundred of Northwich from 1660., which was transcribed for the Record Society of Cheshire and Lancashire by G.O. Lawton, and which contains occupations for most men.5 All except those on poor relief were expected to pay this tax. An occupation was recorded for 73 per cent of the three thousand men listed in the returns. Table 1 provides a comparison of the male occupational structure according to the poll tax returns with those derived from (calibrated) probate data. As the table shows, the two sources provide a very good match.

The Northwich poll tax compared to calbirated probate data

Table 1. Comparison of occupational structure according to poll tax and calibrated probate data (Northwich hundred, 1660)
Source: Keibek, S.A.J., The male occupational structure of England and Wales, 1600-1850 (PhD dissertation, University of Cambridge, 2017), p. 52.
Note: the probate data have been calibrated, using a methodology described in the above PhD and in a recently published working paper.6 Labourers were allocated using the approach outlined in another working paper.7

Militia ballot lists and similar sources

In order to determine the military capacity of the country, central and local authorities, from time to time, requested that lists of men be drawn up. Some of these lists survive, and some contain occupations. An early example are the muster rolls for 1522, which survive for several (parts of) counties. Some of these rolls – for the county of Rutland, the Suffolk hundred of Babergh, and the city of Coventry – list occupations. Combined, these three rolls list 5,083 men of which 3,918 have been given a meaningful, clear occupation. Another, more famous example owing to the landmark paper which the Tawney's based on it, is the 1608 muster list for Gloucestershire.8

Using these early militia ballot lists as a source of occupational information is not straightforward, as an analysis of the Gloucestershire source illustrates. Firstly, the share of men provided with an occupation in the muster list varies from place to place, and this variation is not random: the more agricultural the area, the lower the share of men for whom an occupation was listed, as Figure 1 shows. If the muster list data would be taken at face value, the agricultural sector would be underrepresented.

1608 muster list - evidence of rural underrepresentation

Figure 1. The agricultural share of the male labour force in the uncorrected probate record as a function of the percentage of men of unspecified occupation in the muster list (Gloucestershire, 1608)
Source: Keibek, The male occupational structure of England and Wales, 1600-1850 (PhD dissertation, University of Cambridge, 2017), p. 53.

Why the occupations of a relatively high proportion of men were not specified in farming-dominated areas is a matter for speculation. It may simply have been caused by farming being so dominant in these area that it was considered the 'default' activity, with occupational specification (mostly) reserved for the few men who were not involved in agriculture. It is also possible that in the hamlets and small villages in which these men lived, an occupational denominator was simply not required to distinguish one man from another, as the likelihood of the same first name/surname combination occurring more than once was very low in these small settlements.

A second problem with the muster list data is the sizeable number of servants. The Tawneys allocated these to occupational sectors based on the occupation of their masters. They considered servants of masters specified by status only – gentlemen, knights, esquires, etcetera – to have been domestic servants. The Tawneys' approach leads to 39 per cent of servants being allocated to agriculture, 24 per cent to the secondary sector, and 37 per cent to the tertiary sector, almost all as domestic servants in the households of gentlemen and other men of status. This distribution seems rather light on farm servants. The Tawneys admit that the number of domestic servants is overstated as 'some of those classed as household servants … were undoubtedly farm servants'; indeed, they note that in a number of stately households, the servants are specified as domestic or agricultural, with the latter making up around forty per cent of the total. But their assumption that servants of secondary sector workers were employed in their masters' trade is also debatable. It is possible that many of them, perhaps even the overwhelming majority, were farm servants.

Table 2 presents Gloucestershire's estimated occupational structure in 1608 as derived from the muster list, taking the above two issues into account. The agricultural sector has been upwardly corrected for under-specification in farming-dominated parishes. To express the uncertainties in the allocation of servants to sectors, two scenarios were calculated. One copies the Tawneys' approach, allocating servants in line with their masters' occupation. In the alternative approach, forty per cent of the servants of gentlemen and other men of status were allocated to farming rather than considered domestics, and all servants of secondary-sector workers were considered farm servants too. The result is a (manageable) range rather than a single set of values. Table 2 also includes the occupational structure derived from contemporary (but calibrated) probate data; the muster list and calibrated probate estimates match well – with the exception of the textiles sub-sector.

Comparison between 1608 muster list and calibrated probate data

Table 2. A comparison of male occupational structures as derived from muster list and calibrated probate data (Gloucestershire, c.1608)
Source: Keibek, The male occupational structure of England and Wales, 1600-1850 (PhD dissertation, University of Cambridge, 2017), p. 54.
Note: the probate data have been calibrated, using a methodology described in the above PhD and in a recently published working paper.6 Labourers were allocated using the approach outlined in another working paper.7

Later examples of militia ballot lists also exist, in particular for the second half of the eighteenth century when, after the Militia Act of 1758, local authorities were required to make 'Fair and true lists, in writing, of the names of all men usually and at that time dwelling within their respective parishes, tythings and places, between the ages of eighteen and fifty years distinguishing their respective ranks and occupations ... and which of the persons so returned labour under any infirmities, incapacitating them from serving as militia men.'9 Although we have other sources at our disposal to study male occupational developments in this time period – such as parish registers and probate data – these militia ballot lists provide valuable checks on the results derived from these other sources. In a remarkably thorough and complete survey, Paul Glennie has shown how powerful a source of occupational information they are for late eighteenth-century Hertfordshire and other counties, and has made a systematic analysis of their availability by location and year.10


1 Fenwick, The poll taxes of 1377, 1379 and 1381. Part 1, Bedfordshire-Lincolnshire / edited by Carolyn C. Fenwick (Oxford: Oxford University Press for the British Academy, 1998); Fenwick, The poll taxes of 1377, 1379 and 1381. Part 2, Lincolnshire - Westmorland / edited by Carolyn C. Fenwick (Oxford: Oxford University Press for the British Academy, 2001); Fenwick, The poll taxes of 1377, 1379 and 1381. Part 3, Wiltshire-Yorkshire, unidentified documents and additional data / edited by Carolyn C. Fenwick (Oxford: Oxford University Press for the British Academy, 2005).

2 Clark, '1381 and the Malthus delusion', Explorations in Economic History, 50:1 (2013), pp. 4-15; Broadberry et al, 'When did Britain industrialise?', Explorations in Economic History, 50:1 (2013), pp. 17-8; Broadberry et al, British economic growth, 1270-1870 (Cambridge: Cambridge University Press, 2015), pp. 346-50.

3 Broadberry et al, British economic growth, p. 346.

4 These conference papers are not currently publicly available, but are being converted into a paper intended for publication.

5 Transcribed in Lawton, Northwich hundred: poll tax 1660 and hearth tax 1664 (Bath: Pitman Press, 1979).

6 Keibek, 'Using probate data to determine historical male occupational structures', Cambridge Working Papers in Economic and Social History, 26 (2017), http://www.econsoc.hist.cam.ac.uk/docs/CWPESH_number_26_March_2017.pdf.

7 Keibek, 'Allocating labourers to occupational (sub-)sectors using regression techniques', Cambridge Working Papers in Economic and Social History, 27 (2017), http://www.econsoc.hist.cam.ac.uk/docs/CWPESH_number_27_March_2017.pdf.

8 Tawney and Tawney, 'An occupational census of the seventeenth century', The Economic History Review, a5:1 (1934); Smyth, The names and surnames of all the able and sufficient men in body fitt for his Majestie's service in the warrs, within the City of Gloucester and the inshire of the same, manuscript (1608).

9 As quoted in Glennie, 'Distinguishing men's trades': occupational sources and debates for pre-census England (Bristol: Historical Geography Research Group, 1990), p. 46.

10 Glennie, 'Distinguishing men's trades', pp. 46-128.