^{note (1) in page 62} A. J. H. Dean, ‘Earnings in the public and private sectors 1950-1975’, National Institute Economic Review, no. 74, November 1975.

^{note (2) in page 62} A full description of the way in which the data were processed for the purposes of deriving series for public and private sectors, plus a listing of the weights used, is given in appendices 1-4 of the original article.

^{note (3) in page 62} Ibid., page 62.

^{note (4) in page 62} Ibid., page 67.

^{note (1) in page 63} Department of Employment, New Earnings Survey; Part B: analyses by agreement, HMSO, annually.

^{note (2) in page 63} The relation of the median to the mean will stay constant if everyone receives equi-proportionate increases. If, however, the income distribution becomes narrower (as has happened during the period in question), this is likely to lead to the gap between the median and the mean becoming smaller. Per centage increases in the median will then be larger than increases in the mean. However one can probably assume that this narrowing is likely to have taken place in roughly the same way in the public and private sectors. In this case, when examining movements in relative pay between the two sectors, it should make little difference whether one looks at the median or the mean.

^{note (1) in page 64} A similar study, also employing NES data, has been conducted by Hawkesworth (see R. I. Hawkesworth, ‘Research note: private and public sector pay’, British Journal of Industrial Relations, Vol. XIV, no. 2, July 1976). He used a different sample (26 private sector and 25 public sector agree ments) and weighted by employment rather than by sample numbers. He found that ‘ … for the period as a whole (1968-75), there has been an improvement in the relative position of public sector manual employees of about 15 per cent’.

^{note (2) in page 64} Although it is possible to calculate figures for female employees the sample size is so much smaller that far fewer national agreements can be covered.

^{note (3) in page 64} A four year average is chosen because the sample numbers tend to fluctuate considerably from one survey to another; these weights are shown in Appendix 1.

^{note (4) in page 64} In particular we should note that the private sector coverage of the ‘old’ series is far broader than the 31 agree ments covered by the NES series presented here. There has also been one allocational change; the British Steel Corporation national agreement is included in the public sector in the new NES series whereas the iron and steel industry was included in the private sector in the old series, the steel companies being in private ownership for the larger part of the period covered by the ‘old’ series (see original article, loc. cit., Appendix 2).

^{note (1) in page 65} The figure of £6.7 for the private sector does not indicate that the pay limits were broken since various increases apart from the basic supplements were allowed; in particular overtime payments, which are more important in the private sector, increased strongly during this period.

^{note (2) in page 65} De Zoete and Bevan, Economic Bulletin 2/77, July 1977.

^{note (3) in page 65} The 21 private companies covered are four banks (Barclays, Lloyds, Midland and National Westminster), two breweries (Bass and Whitbread), two building firms (APCM and Taylor Woodrow), four engineering companies (GEC, GKN, Hawker Siddeley and Tube Investments), two food firms (Associated British Foods and Cadbury Schweppes) and British Leyland, Lucas, Courtaulds, ICI, Beecham, BP and Marks and Spencer. The seven public sector organisations are the National Coal Board, the British Steel Corporation, the Electricity Boards, British Gas, British Rail, London Transport and the Post Office.

^{note (1) in page 66} However an unpublished study of interconnections in public sector pay movements between 1957 and 1971 by John Fyfe suggests that the links between movements in rates and earnings have to be explained in more than statistical terms. The study suggested that there are complexities underlying the bargaining process which need to be more fully understood, not only to ‘explain’ the relationships between rates and earnings, but also to improve our under standing of the factors influencing the wage determination process in the public sector.