Jump to ContentJump to Main Navigation
British Economic Growth 1856-1973The Post-War Period in Historical Perspective$

R. C. O. Matthews, C. H. Feinstein, and J. Odling-Smee

Print publication date: 1982

Print ISBN-13: 9780198284536

Published to Oxford Scholarship Online: November 2003

DOI: 10.1093/0198284535.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 21 February 2017

(p.612) Appendix N Basic Statistics: Sources and Methods

(p.612) Appendix N Basic Statistics: Sources and Methods

British Economic Growth 1856-1973
Oxford University Press

The following statistical series are considered in this appendix:

  1. (1) Gross domestic product and the main national income accounting subaggregates (e.g. income from employment, consumers' expenditure)

  2. (2) Income, output, and factor shares by industry

  3. (3) The total in employment and hours worked by industry, and the total working population

  4. (4) The gross fixed capital stock; inventories and gross fixed capital formation by industry, sector, and asset; and net overseas assets and the net fixed capital stock in the whole economy

  5. (5) Imports and exports of goods and services

Except where otherwise indicated, all data refer to the contemporary area of the United Kingdom.

With one major exception, the industrial classifications used are identical to those in Feinstein 1972, to which reference should be made for further details. The exception is that the rent from non‐trading property owned and occupied by the central government and local authorities is treated as the income of the public administration and defense sector, rather than as the income of insurance, banking, and finance. In general, the classification for 1920–48 follows the 1948 Standard Industrial Classification, that for 1948–64 the 1958 Standard Industrial Classification, and that for 1964–74 the 1968 Standard Industrial Classification. There are therefore breaks in classifications in 1948 (in the capital stock series, for which the classification is anyway not very precise, the break occurs in 1938) and 1964. The classification for 1855–1920 corresponds approximately to the 1948 Standard Industrial Classification.

Gross Domestic Product and Subaggregates

The two sources for nearly all the data are Feinstein 1972, for 1855–1964, and National Income and Expenditure, 1964–74, UK[21], for 1964–73. Additional detail was obtained from the working papers and sources used for Feinstein 1972 and, for 1948–64, National Income and Expenditure, 1968.

The estimates in Feinstein 1972 are consistent with those in National Income and Expenditure, 1968. But the latter are not necessarily consistent with the (p.613) estimates in National Income and Expenditure, 1964–74, because of data revisions. In a number of series, adjustments were made to either the 1948–64 or 1964–73 series to enable comparisons to be made.

The reliability of the historical estimates is discussed in Feinstein 1972: Chapter 1. The following reliability gradings are attached to the current‐ and constant‐price series of GDP:













where the corresponding margins of error are A, ± less than 5 percent; B, ± 5–15 percent; and C ± 15–25 percent.

The differences between the alternative estimates of real GDP over our usual periods are discussed in Appendix A.

Income, Output, and Factor Shares by Industry

The main sources are Feinstein 1972 and working papers, and National Income and Expenditure, 1964–74.

Additional sources for income by industry are as follows:

W. A. Lewis 1978: Table A.3, for 1907

A. L. Chapman 1952, for additional detail of income from employment in 1920–38

Inland Revenue Statistics, UK [57], for PAYE (Pay as you earn) statistics used to allocate income from employment between professional services and miscellaneous services in postwar years

Input‐output Tables for the postwar years (UK [9], [16], [17])

A variety of sources for estimates of end‐year inventories at current prices, from which stock appreciation estimates were obtained, including Feinstein 1965; Economic Trends (e.g. October 1968, Table M); Sources and Methods, 1956, UK [22]: 321, and 1968, UK (20): 400; and Board of Trade Journal (e.g. issue of Feb. 23, 1968, for motor repairers and catering)

Forestry Commission, Annual Reports; Trading Accounts and Balance Sheets, UK [96]; Inland Revenue Statistics, UK [57] (to isolate agriculture from forestry and fishing in the postwar period)

National Income and Expenditure, 1958 and 1959 issues, for estimates of the effects of the change in Standard Industrial Classification

Sources and Methods, 1968, UK [20]: 91–93, for the weights of industries in GDP in 1958

All real output indicators were derived from the same sources as (and for individual industries and sectors they agree with) those in Feinstein 1972: Tables 8, 51–53, and National Income and Expenditure, 1964–74, UK [21]: Table 15. For aggregates (e.g. manufacturing and GDP) the agreement is not exact for periods after 1913 because the individual industry indicators were combined using different base years, as follows: (p.614)


Base year





1951–55, 1951–64, 1951–73






1964–68, 1964–73




This system of first‐year weighting (except for 1913–24) was adopted to permit some of the calculations in Chapter 9. In principle, it is not so accurate as that employed in Feinstein 1972 (see his Chapter 10, especially Table 10.1), where the base year was changed more often. But the growth rates of the aggregate series do not differ significantly from those of the Feinstein series, except across the wars, as the estimates of the growth of GDP in Table N.1 show.

Labor income is the sum of income from employment and an estimate of labor income from self‐employment, and property income is the residual from total income. The method of estimating labor income from self‐employment is explained in Chapter 6. In addition to the main sources, those cited above for estimating total income, and those in Chapter 6, note 7, estimates of the numbers of self‐employed and of employees in employment are required. They are discussed in the next section.

The Total in Employment and Hours Worked by Industry, and the Total Working Population

The main sources for the total in employment are Feinstein 1972: Table 59, for 1920–64; and the Department of Employment Gazette, March and October 1975 and December 1976, for 1964–73.

A minor adjustment to Feinstein's estimates for agriculture and hence all industries in 1920–38 was made to exclude some casual and part‐time workers, mostly female. On pre‐World War I and post‐World War II definitions, these workers would not have been included in the labor force, and so they are excluded to permit transwar comparisons. Even so, the interwar and postwar seires are not strictly comparable because of a further discrepancy in the coverage of casual workers in agriculture, and a difference in coverage of seamen abroad. A further break in the concept of the total in employment occurs in 1964, after which the new census of employment concept is used. These breaks are summarized below in the paragraph on the working population. In tables and figures the series have been spliced at 1951 and 1964 where it would have been misleading not to do so. To do this, a separate estimate for 1951 had to be made on a basis exactly comparable with the interwar estimates. This required an adjustment to agriculture (because the interwar concept of the total in employment was still broader than the postwar one, even after the minor adjustment of Feinstein's data noted above), an adjustment to shipping (because again the interwar concept was broader than the postwar one), and a general reclassification (because the interwar statistics are classified according to the 1948 Standard Industrial Classification, and the postwar statistics according to the 1958 Standard Industrial Classification). (p.615)

Table N.1 Alternative Estimates of the Growth of Gross Domestic Product, 1913–1973

(Annual percentage growth rates)


Present estimatesa

Alternative estimatesb


Present estimatesa

Alternative estimatesb




























(a) From Table A.1, column 3.

(b) Calculated from Feinstein 1972: Table 6, column 1; National Income and Expenditure, 1964–74, UK [21]: Table 15.

The source for the total in employment in the whole economy in 1856–1920 is Feinstein 1972: Table 57. Estimates by industry for benchmark years before 1914 were obtained by interpolation between the decennial estimates derived from Census of Population data in Feinstein 1972, Table 60, and a comparable estimate made for 1851. Uniform rates of unemployment (as in Feinstein 1972: Table 57) were applied to all industries.

The totals in employment by industry in 1920–73 were disaggregated by sex (postwar only) and employment status (employees and self‐employed) on the basis of data in the Gazettes cited above and the working papers and sources used for Feinstein 1972: Tables 57–59. The total self‐employed in 1911 in Feinstein 1972: Table 7.17 was disaggregated into seven sectors on the basis of rough backward extrapolations from 1921 (for Great Britain and Northern Ireland) and 1926 (for Southern Ireland, from Saorstát Éireann, Census of Population, 1926). For 1861–1901 it was assumed that the self‐employed were the same proportion of working population in each of the seven sectors as in 1911.

New estimates of hours worked in 11 benchmark years were prepared for this study. They refer to the hours actually worked after allowing for overtime, short‐time, and part‐time working; holidays; sickness; and strikes. Their sources and methods are summarized in Appendix D.

The sources for the total working population are Feinstein 1972: Table 57, for 1855–1964, and the Department of Employment Gazette, March and October 1975 and December 1976, for 1964–73. The minor adjustment described above for the total in employment in 1920–38 was also made for the total working population. Our definition of the working population changes twice in the whole period. For years from 1951 a slightly narrower concept is employed than for earlier years. Some casual workers in agriculture and some seamen abroad are excluded: they numbered 115,000 and 44,000, respectively, in 1951. The census of employment that replaced the National Insurance card estimates in 1971 excludes employees who worked for part of the year but who had jobs during census week, and private domestic workers, and includes the secondary jobs of multiple job holders. In practice we switch onto the census of employment concept in 1964 or 1966, depending on the context. In most of the tables, (p.616) especially those showing growth rates, the series on different bases have been spliced according to their ratios in overlapping years.


The main sources for the fixed capital stock, gross and net, are Feinstein (forthcoming) for 1856–1938; Feinstein 1965 for individual manufacturing industries in 1920–38; Feinstein 1972 and associated working papers, as amended by Chapter 5, note 1, for 1938–64; and National Income and Expenditure, 1964–74, UK [21], for 1964–73, with additional detail kindly supplied by the Central Statistical Office. The same sources were used for gross fixed capital formation except that Feinstein (forthcoming) was used for 1860–1920 and Feinstein 1965 and 1972 for 1920–64.

Gross capital stock by sector (i.e. unincorporated, other never‐public, and sometime‐public sectors) was estimated by allocating industries to sectors. First, the sometime‐public sector was defined to comprise mining and quarrying, iron and steel, gas, electricity, and water, and transport and communications (excluding shipping). Then capital in manufacturing, construction, and commerce was divided between the unincorporated and corporate sectors on the basis of the two sectors' shares of depreciation allowances given in Inland Revenue Statistics (UK [57]). The unincorporated sector's capital stock consists of capital in the unincorporated parts, together with capital in professional services. Other never‐public sector capital consists of shipping capital, and capital in the corporate parts of manufacturing (other than iron and steel), construction, and commerce.

The main sources for inventories are for 1856–1920, Feinstein (forthcoming), whole economy and agriculture only; for 1920–38, Feinstein 1965; for 1938–51, Agricultural Statistics, UK [60], for agriculture; for 1951–73, a variety of sources including National Income and Expenditure, 1968 and 1964–74 and the works cited in the list on p. 613.

The estimates of the growth in constant prices for aggregates (e.g. manufacturing and the whole economy) were obtained by summing the series for individual industries at the prices of the base years shown in the tabulation on p. 615. As with output, this system of first‐year weighting is necessary for some of the calculations in Chapter 9.

The fixed capital stock estimates for 1920–73 might be graded C and D (margin of error ± more than 25 percent) for earlier years. The fixed capital formation estimates might be graded as follows: 1856–90, C; 1891–1919, B/C; 1920–73, B.

The estimates of net overseas assets were built up from a variety of sources, of which the most important are these:

1856–1913, Morgan 1952: 331–34, 342–43; Imlah 1958: 68–80

1924–37, Kindersley 1929–39; Finance Accounts, UK [93]; London and Cambridge Economic Service 1972: Table N; Reserves and Liabilities, UK [94]


1951–73, Conan 1952: Appendix 1; UK Balance of Payments, UK [25]; Radcliffe Committee, 1959, UK [32]: 264; Bank of England, Quarterly Bulletin and Statistical Abstract 1970, 1975; Midland Bank Review; London and Cambridge Economic Service 1972

Imports and Exports of Goods and Services

The main sources are Feinstein 1972 and working papers; National Income and Expenditure, 1968 and 1964–74, UK [21]; and UK Balance of Payments, 1968 and 1964–74, UK [25]. For 1855–70, the sources used for Feinstein 1972, especially Imlah 1958, were used. For re‐exports, the following sources were also used: Imlah 1958; Board of Trade Journal; Annual Abstract, UK [10]; London and Cambridge Economic Service 1972.

A variety of sources were used for estimates of imports by main commodity group. The most important, in addition to those above, are Schlote 1952; Scott 1963: Appendix II; Statistical Abstract, UK [7]; Annual Abstract, UK [10]; Annual Statement of Trade, UK [2]; Trade and Navigation Returns, UK [8]; Report on Overseas Trade, UK [6]; Board of Trade Journal; and Mitchell & Deane 1962. The classification into the four commodity groups is, as far as possible, that of Scott 1963: Appendix II, pp. 210–13. (p.618)