Historical Statistics and British Economic History: The British Historical Statistics Project (BHSP)

Abstract

The project now underway to produce a new, multi-volume print and online edition of Brian Mitchell's British Historical Statistics is an opportunity to examine the history of the celebrated work. An account of its conception, reception, and evolution provides a window through which to examine a central aspect of economic history's disciplinary history and methodenstreit; certain fundamentals relating to the demand for, and supply of, historical statistics; and to introduce the British Historical Statistics Project, which is now underway but still in its infancy, during which it is closely examining its antecedents in order to learn the correct lessons from that history.

Keywords:

• Britain
• historical statistics
• economic history
• quantitative history
• cliometrics

Acknowledgments

This article was written in my role as a General Editor (with N. Goose and M. E. Turner), of the British Historical Statistics Project (BHSP; http://www.bristol.ac.uk/history/bhsp). Financial support for BHSP is gratefully acknowledged from the University of Bristol alumni, Cambridge University Press (CUP), and the Economic History Society (EHS). The primary acknowledgement goes to Brian Mitchell for allowing BHSP to go forward in the best possible manner, and then to Susan Carter and Richard Sutch, two of the editors-in-chief of the millennial edition of Historical Statistics of the United States (2006; HSUSME), for their great assistance to the BHSP general editors, and particularly at the project launch at EHS2010. Additionally, I thank Roger Backhouse and Mary Morgan for sharing their knowledge of the history of Cambridge's Department of Applied Economics (DAE); Nick Crafts for insights into the production of Deane and Cole (1962); Tony Wrigley for information on CAMPOP; participants at the “Lure of the Aggregates” conference, University of Reading, March 30, 2011; and the two anonymous referees and editor of this journal for their assistance in greatly improving earlier versions of this article. The usual disclaimers apply about remaining errors.

Notes

1. Details of the contributions by Cole, Deane, Feinstein, Matthews, Mitchell, and colleagues of the 1950s and 1960s DAE can be found in Lyons et al. (2008); see also, on Feinstein, Offer's (2008) British Academy memoir; and, on Reddaway, Tribe (1997, ch. 5).

2. Offer slightly mis-references Lundberg (1971), written for the award of Kuznets’ Nobel prize, who in turn (p. 447) cites Deane (1967), a review of Kuznets (1966a), who displayed a regard for Kuznets’ methodology which so clearly was inspirational for her.

3. We lack a biography of Kuznets, but see Lundberg (1971), Abramovitz (1986), Kapuria–Foreman and Perlman (1995), and Fogel (2000) for assessments which make clear how the national accounts research program matured naturally into the growth program. Kuznets received the Nobel prize in 1971 for “his empirically founded interpretation of economic growth which has led to new and deepened insight into the economic and social structure and process of development.”

4. http://www.roiw.org/1/10.pdf; accessed November 29, 2011.

5. Kuznets’ role is mentioned in Easterlin (2001) which cites Kuznets (1949), of which ch. 1 details the project proposal (“Suggestions for an inquiry into the economic growth of nations”) and ch. 4 (“Notes on the quantitative approach to economic growth”) is an early version of the series of ten papers first published in Economic Development and Cultural Change (1956–67) which formed the basis for his Modern Economic Growth (1966b).

6. Nonetheless, intermediate outputs followed quickly, beginning with Deane (1955), which in table 1 presented Gregory King's social accounts for England and Wales c.1688 in the same format as that which opens chapter 1 of DC62.

7. Comim (2000) is the closest we come to a historical analysis of the DAE. Whilst he does not mention MD62 or indeed Mitchell, DC62 is acknowledged in the context of, under Reddaway's directorship, the DAE's important work in the quantification of economic history that paralleled the ongoing Stone growth project (p. 165); see also Feinstein (1998); Offer (2008) on Feinstein; and Pesaran and Harcourt (2000, F149–50) on Stone.

8. See Deane's preface to MD62, vii. From Mitchell (2009, 80), it seems that the manuscript for MD62 was completed in 1960–1, and he notes that there was then a 12-month delay in publication because Cambridge University Press were “busy with the New English Bible.”

9. Importantly, and again with Kuznets’ heavy involvement, the International Economic Association conference of 1960 was devoted to Rostow's work, with Rostow (1963) the published proceedings and papers. These included Habakkuk and Deane (1963), in which early results from what would become DC62 (at this time entitled The Course of Economic Growth) were included; there were also further extensive comments by Deane on other papers. In his introduction (p. xiv), Rostow made clear Kuznets’ influence on his own research, whilst Kuznets’ (1963) paper made clear his very considerable conceptual and empirical reservation about Rostow's thesis. See also Deane in Lyons et al. (2008, 138) who, some thirty years later, when interviewed by Crafts, agreed that the Rostovian program was “misguided but had been fruitful,” and “was fruitful partly because it was flamboyant.”

10. DC62 makes a brief appearance in both of Coleman's historical studies (1987, 122; 1995, 640), but Mitchell receives no mention, and Feinstein only appears in the former (pp.130–1) in relation to MFOS82, though at least this work does merit a more extended treatment than is the fate of many other canonical postwar works. DC62 receives mention in Greasley and Oxley (2010, 758), whose focus is cliometrics, as does Feinstein (1972), but there is no mention of MD62. As we have seen, for Drukker, DC62 is a canonical volume, while MD62 and successive editions were cited as evidence of a “not unimportant tradition in historical statistics” (p. 114, n. 35).

11. This omission was rectified in M88 as tables II. 30–31. In introducing them, Mitchell (1988, 97) provides a thoroughly even-handed assessment of their quality and significance, but does not enlighten as to why there had been a quarter century delay; see, however, Falkus (1996, 163–4).

12. http://www.hpss.geog.cam.ac.uk/research/; accessed December 1, 2011.

13. Coleman (1995, 639–40), in his strictures on the “numbers game,” has the DAE and DC62 as key only for the cliometricians, though it could not be said that they feature prominently in his analysis.

14. In addition to the reviews here discussed, see also, on DC62/MD62, Crouzet (1965); on MD62, Olsen (1963) and Thirring (1963); on MJ71, Nicholson (1972) and Neveux (1974); and on Mitchell 1975, Floud (1977).

15. As befits a review in the Agricultural History Review, Dyos (1963, 166) focused on the agricultural series, but additionally he commented that “yet, with all its riches, there is more room for disappointment than ought perhaps to be there.”

16. For a broader European perspective, see Cipolla (1991, table 3) which had MD62 in advance significantly of a comparable German collection but lagging Italy and Sweden.

17. Platt (1989) followed close on the heels of his 1986 book on the same theme, but with the narrower subject of British overseas investment in the century to 1914; see also the posthumous volume, Platt, Michie, and Latham (1993). His concerns (in print) about Britain's foreign trade statistics stretch back to Platt (1971).

18. The earlier trade data, back to 1697, drew upon Schumpeter (1960) and this remains the source reported in Mitchell (1988, ch. IX).

19. The potential of M88 was immediately recognized by one educational software company (SECOS) which produced a customized selection of BHS series as a dataset for a quantitative methods workbook aimed at sixth-formers and undergraduates (Garner 1993; reviewed in Middleton and Wardley 1994, 389, 391).

20. The most recent ONS annual report, for 2007/8, makes no mention of historical data in the sense we would understand the concept and practice, whilst the latest strategy document, ‘ONS Business Plan, 2011–2015’, June 2011 (available from ons.gov.uk), combined with current fiscal retrenchment, makes it unlikely that ONS will follow its predecessor organisations in recognising the importance of high-quality historical series and of resourcing their production.

21. The papers for the launch session are available at http://www.bristol.ac.uk/history/bhsp

22. This Riksbank project provides a portal for other national historical statistics resources (http://www.historicalstatistics.org/).

23. Bob Allen, Richard Britnell, Nick Crafts, Cormac Ó Gráda, Knick Harley, Rab Houston, Jane Humphries, Larry Neal, Richard Smith, and Keith Wrightson.

24. Freeman (2010) produced by a historian, for historians acknowledge the broader environment to which BHSP will connect; see also Stratton (2011).

25. Using the lists published between 1976–2011, a 5% sample of pages has been conducted to produce a series for 1975–2010. Whilst Harte's original exercise distinguished between books and articles, it is encouraging that on the basis of this sample, the 1975 list (of 1974 publications) is within 0.5% of Harte's estimate.

26. Searches respectively “Mitchell historical statistics” and “Mitchell British historical statistics,” conducted December 21, 2011.