THE RELIABILITY OF VACANCY STATISTICS

AuthorDEREK ROBINSON,BRIDGET ROSEWELL
Publication Date01 Feb 1980
DOIhttp://doi.org/10.1111/j.1468-0084.1980.mp42001001.x
OXFORD BULLETIN
of
ECONOMICS and STATISTICS
THE RELIABILITY OF VACANCY STATISTICS
By BRIDGET ROSEWELL and DEREK ROBINSON
INTRODUCTION
The Employment Services Division (ESD) of the Manpower Services Commis-
sion (MSC) collect and publish monthly in the Department of Employment Gazette
figures of the number of vacancies notified to the ESD that are still unfilled on a
certain date each month. There is no compulsion on employers to notify their
vacancies. The published figures are often used in economic analysis, frequently in
association with statistics relating to the numbers registered as unemployed, to
comment on, or express some quantitative relationship about, the labour market.1
Some post-Phillips analysts have sought to use the series of notified vacancies and
registered unemployed in more sophisticated and refined ways.2 However, there
is a major difficulty in using the published series of notified vacancies. It is
generally accepted that the series seriously understates the number of actual
vacancies, and that the proportion of vacancies notified varies according to skill
and locality.3 There is little, if any, evidence to indicate whether the understatement
is constant or whether it varies over time, or by industry or occupation, or according
to the type of services provided by the Manpower Services Commission.4
The relationship between actual and notified vacancies is of interest and
relevance to policy-makers as well as to economists. If officially recorded figures
'See for example,J. C. R. Dow and L. A. Dicks-Mireaux, 'The Excess Demand of Labour, A Study of
Conditions in Great Britain 1946-56', Oxford Economic Papers, February 1958; D. Gujarati, 'The
Behaviour of Unemployment and Unfilled Vacancies', Economic Journal, March 1972.
2 K. W. Rothschild, 'The Phillips Curve and all That', Scottish Journal of Political Economy, November
1971 ;J. Bowers, P. Cheshire and A. Webb, 'The Change in the Relationship between Unemployment and
Wage Increases: A Review of some Possible Explanations', National Institute Economic Review,
November 1970; M. Parkin, 'Incomes Policy: Some Further Results on the Determination of the Rate
of Change of Money Wages', Economica, November 1970; J. I. Foster, 'The Relationship between
Unemployment and Vacancies in Great Britain (1958-72): Some further evidence', in D. Laidler and
D. E. Purdy (eds.), Inflation and Labour Markets, Manchester, 1974.
See 'Skill shortages in British industry', Department of Employment Gazette, May 1979, and D. I.
Mackay, D. Boddy,J. Brack,J. A. Diack, N.Jones, Labour Markets under Different Employment Conditions,
London, 1971, particularly Ch. 13.
F. T. Blackaby in G. D. N. Worswick, (ed), The Concept and Measurement of Involuntary
Unemployment, London, 1976. '.. . there is plenty of scope for the relationship between unemployment
and vacancies to shift as employers make more use of the Department of Employment's facilities.' See
also Department of Employment Gazette, July 1978, and discussion below indicating the Department's
view that Jobcentres increase the proportion of vacancies which are notified. See also B. Thomas and
D. Deaton, Labour Shortage and Economic Analysis, Oxford, 1977, pp. 2-3, who raise some of the
difficulties of using unemployment and vacancies statistics.
1
Volume 42 February 1980 No. 1
2BULLETIN
of vacancies and unemployment do not accurately reflect the true underlying
relationships it will not be possible to form accurate assessments of the state of the
labour market. Search theories and related micro-economic approaches to macro-
economic issues5 place a heavy reliance on vacancies and information about them,
and attempts to test these theories require better knowledge than we currently
have of notified vacancies. Appreciation of the problems that arise from using
figures of registered unemployment only as an indicator of the state of the labour
market have led some to use both employment and vacancies.6 Indeed it has been
suggested by one informed commentator that, given the uncertainties of interpreting
the unemployment statistics, complicated by temporary distortions bróught about
by job creation measures and seasonal adjustment factors, and the problems of
using adjusted employment in production industries which tends to be a lagging
indicator, 'a more reliable indicator (from the labour market) is the behaviour of
vacancies over a reasonably long period.'7
Labour market adjustment presumably responds in some way to employers'
demand for labour, and that demand is expressed in part by their notifying
vacancies to the ESD. The statistics of notified vacancies are the only regular
comprehensive series available, and it is not surprising that they have been used to
examine and discuss the economic situation. It was noticed in the late 1 960s that
there has been a shift in the relationship between registered unemployment and
vacancies for males particularly, and the role of unemployment in this has been
analysed in some depth.8
Macro-economic analysis requires reliable statistical series and it is important
to assess how far there have been changes in the vacancy figures as a result of
modifications in employer practices regarding either the use of labour or their
propensity to notify vacancies to the ESD. There may have been important
structural changes in the composition of vacancies and of the unemployed. For
example, the increase in notified vacancies in 1972-3 might have been a result of
changed employers' practices whereby they held less internal labour reserves, i.e.
had previously dis-hoarded labour and needed to turn to the external labour
market to meet the upsurge in demand. The sudden upsurge may have generated
the increase in the flow of vacancies.
In addition, manpower training policies require some estimate to be made of
supply and demand positions unless the training provided is to be decided in
isolation from market requirements. The use of published notified vacancy
See E. S. Phelps et. al., Microeconoinic Foundations of Employment Theory, London, 1971; there is an
excellent bibliography in A. M. Santomero andJ.J. Seater, 'The Inflation-Unemployment Trade Off: A
Critique of the Literature', Journal of Economic Literature,June 1978.
6See, for example, M. T. Sumner, 'Aggregate demand, price expectations and the Phillips curve', in
M. Parkin and M. T. Sumner (eds.), Incomes Policy and Inflation, Manchester, 1972; and J. K. Bowers, P.C.
Cheshire, k E. Webb and R. Weeden, 'Some Aspects of Unemployment and the Labour Market, 1966-
1971', National Institute Economic Review, November 1972.
'S. Brittan, 'Labour Markets and the Budget', Financial Times, 30 January 1978.
8See, for example, Department of Employment Working Party of the Changed Relationship
between Unemployment and Vacancies Report, which was summarized in the Department of Employment
Gazette, October 1976. The findings and some implications are considered in M. Scott, Can We Gei Back
to Full Employment, London, 1978.

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT