^{page 51 note (1)} See, for example, Kalachck, Edward, and Westebbe, Richard, ‘Rates of unemployment in Great Britain and the United States, 1959-60’, Review of Economics and Statistics, November 1961.

^{page 52 note (1)} The unemployment rate multiplied by the average duration has been suggested as a measure of the ‘intensity’ of unemploy ment. See, for example, Geoffrey H. Moore, ‘How full is full employment’, American Enterprize Institute for Public Policy Research, Domestic Affairs Study No. 14, July 1973.

^{page 53 note (1)} The comparatively favourable unemployment experience of British women may be in part due to failure of married women, large numbers of whom are not entitled to benefit, to register. In principle, the ‘standardised’ rates shown in table 3 should correct for this bias, but they may not do so fully.

^{page 53 note (2)} Exact data for such a comparison are laclcing unfortunately. While there is recent information on the numbers of unemployed by age for this country, corresponding information on employment will not become available until the full results of the 1983 Labour Force survey are published.

^{page 53 note (3)} Trends and differentials in earnings by region were the subject of a special article in Employment Gazette, April 1979.

^{page 54 note (1)} Regional unemployment differences in Great Britain are discussed in a paper with that title by P. C. Cheshire published by the National Institute in May 1973.

^{page 54 note (2)} A recent paper by C. A. Pissarides and I. McMaster, ‘Regional migration, wages and unemployment’ (LSE Discussion Paper 204. September 1984), includes a regression equation for regional wage relativities using pooled cross-section and time-series data. This suggests that the short-run association between relative wage levels and relative unemployment rates is negative but the long-run association runs the other way.

^{page 54 note (3)} An article in the Monthly Labor Review, November 1979, by John E. Buckley ‘Do area wages reflect living costs?’ finds it necessary to correct for industry mix before satisfactory regression results can be obtained.

^{page 58 note (1)} Many other explanations have been proposed. A large literature on a controversial subject includes John W. Kendrick, ‘International comparisons of recent productivity trends’, in William Fellow, ed., Essays in contemporary economic problems', 1981-2 edition (American Enterprize Institute. 1981), W. D. Nordhaus, ‘Economic policy and declining productivity growth’, European Economic Review, May/June 1982, and Barry P. Bosworth, ‘Capital formation and economic policy’, Brookings Papers on Economic Activity, 2, 1982.

^{page 59 note (1)} Unpublished; quoted by Kendrick, op. cit.

^{page 59 note (2)} This measure of wages requires a word of warning. It refers not to basic wage rates but to actual payments to the average worker. The latter is likely to vary more with the cycle than the former because of variations in overtime and in payments by results. However it has the drawback for our present purpose of not being a fixed-weight index; it is affected not only by changes in the rate of pay for a particular job, but also by changes in the distribution of wage earners between different jobs. Because we wish to focus on the earnings component of labour costs, no account is taken of taxes on employment. (International comparisons which do take these into account are presented in the note by Mr. Ray in this issue of the Review.)

^{page 61 note (1)} Britton, A. (ed.), Employment, Output and Inflation, Lon don, Heinemann, 1983, chapter 5.

^{page 61 note (1)} These conclusions would seem consistent with the estimated wage equations reported by P. R. G. Layard and S. J. Nickell in ‘Unemployment and real wages in Europe. Japan and the US’ L.S.E. Centre for Labour Economics Working Paper No. 177 (1984).