New Estimates of British Unemployment, 1870–1913

Abstract

We present new estimates of the British industrial unemployment rate for , which improve on the Board of Trade's prior estimates. We use similar sources, but our series includes additional industrial sectors, allows for short-time working, and aggregates the various sectors using appropriate labor-force weights from the census. The resulting index suggests a rate of industrial unemployment that was generally higher, but less volatile, than the board's index. We then adjust our series to an economywide basis, and construct a consistent time series of overall unemployment for 1870-1999. T he Board of Trade's unemployment series for the period 1860-1913 has been widely used by economists and economic historians to evaluate the labor-market implications of economic fluctuations in the half-century before the First World War. However, many contemporaries and historians have noted that the index has serious shortcomings that limit its usefulness as a measure of unemployment at any point in time. The Board of Trade index was constructed from data reported by trade unions that administered benefit schemes for their unemployed members. It was based on a relatively small, nonrandom sample of industrial workers, and it excluded those in sectors of the economy that were not unionized or in which unions did not offer unemployment benefits. Moreover, in constructing an aggregate unemployment rate the Board of Trade weighted the individual unions included 644Boyer and Hatton in the index by their membership rather than by the size of the labor force in the industries they represented.In this article we provide a new index of unemployment. Our index relies chiefly on trade-union records, but it also incorporates other information where possible, in order to include sectors of the economy for which tradeunion unemployment data are not available. It reweights the component trades with appropriate labor-force weights obtained from the decennial census. We construct versions of the index that include a measure of unemployment for unskilled general laborers, and also a measure of the loss of employment through short-time working, which was common in certain major industries. Finally, we use post-1919 data to adjust our unemployment series, which covers only the industrial sector, to an economywide basis. On this basis we derive a consistent unemployment series for the entire period 1870-1999. Our results support the views of critics who maintained that the Board of Trade series underestimated the level of unemployment in industry. ] The results also support another criticism of the board's index: that it exaggerates the extent of fluctuations in unemployment. In this respect our results parallel those of Christina Romer and David Weir, who have found that the American labor market was more stable in this era than previous estimates suggested. 2 THE EXISTING INDEX OF UNEMPLOYMENTIn 1888 the Labour Bureau of the Board of Trade began reporting a monthly (and annual) unemployment index, calculated from information supplied...

Can Productivity Growth Explain the NAIRU? Long‐Run Evidence from Britain, 1871–1999

“…Labour force : own calculations based on Feinstein (1972, table 57, pp. T125–6) for 1892–1913 and Boyer and Hatton (2002) for 1920–99. Coverage of national agreements: calculated (as midpoint of range, benchmark estimates) from the discussion paper underlying Milner (1995, p. 82); pre‐1974 figures are from changes‐in‐pay‐rates data, post‐1973 from survey data.…”

Unemployment and the UK labour market before, during and after the Golden Age

“…Number unemployed divided by the labour force, based on postwar definitions from Boyer and Hatton (2002), Appendix Table A2 , p. 31. 1920-1999: number unemployed based on unemployment insurance and claimant count measures, adjusted for changes in eligibility conditions in the 1980s, and divided by the labour force (below); 1870-1914, based on trade union data and other sources, adjusted to the insurance-based estimates for unionised sectors for the interwar period and then adjusted to the whole economy using the relationship between unemployment in the sectors covered for the nineteenth century and the aggregate unemployment rate estimated for 1923-39 and 1948-71.…”

“…Unemployed plus employed wage and salary earners. 1920-1999: unemployed plus employed wage earners (workforce jobs) derived in Boyer and Hatton (2002);1870-1913 working population net of the armed forces from Feinstein (1972), Table 57, pp. T125-6, adjusted to exclude employers and the self-employed.…”

Asylum nursing as a career in the United Kingdom, 1890–1910

“…With well over half of the nurses not coming from the locality, even as broadly defined within this study, it is clear that there was considerable geographical mobility in those entering asylum nursing. Nationally, industrial employment reached high levels, over 8%, for three periods during the 20‐year study period (Boyer & Hatton 2002), potentially providing impetus for travel of significant distances to seek employment.…”