note (1) page 16 The figures are given in table 3 of the Statistical Appen dix. The index is compiled from the figures of unfilled vacancies and unemployment and roughly corresponds to the difference between them, expressed as a percentage of employees: see J. C. R. Dow and L. A. Dicks-Mireaux, ‘Excess demand for Labour’, Oxford Economic Papers, February 1958.

note (1) page 17 One of these studies was made at the National Institute by L. A. Dicks-Mireaux and J. C. R. Dow: ‘The Determinants of Wage Inflation’, read to the Royal Statistical Society in December 1958 and to be published in the Journal of the Society. The other was made at the Oxford Institute of Statistics by L. R. Klein, R. J. Ball and A. Hazlewood as part of a general econometric study of the United Kingdom, not yet published in full (for a summary see their article ‘Econometric Forecasts for 1959’ in the Bulletin of the Oxford Institute of Statistics for February 1959). Both studies used multiple regression techniques: the second attempted to explain changes in wage rates, and wage earnings, as part of the more general process of price formation. The difference in procedure may account for the somewhat differing estimates for the price and demand effect: the differences do not affect the broad conclusions discussed here. The first study refers to the years 1946-56, the second to 1948-56.

note (2) page 17 Actual earnings are generally greater than standard wage rates but there is reason to suppose (see page 20 below) that the gap between the two is not very sensitive to changes in demand. Changes in negotiated wage rates explain most of the increase in hourly earnings.

note (1) page 18 A. W. Phillips, ‘The Relation between Unemployment and the Rate of Change of Money Wage Rates in the United Kingdom, 1861-1957’, Economica, November 1958.

note (2) page 18 Taken at face value, Professor Phillips' curve implies that if ‘demand were kept at a value which would maintain stable wage rates the associated level of unemployment would be about 5 1/2 per cent’. The curve is based on data for 1861-1913. The upper end of Professor Phillips' curve fits surprisingly well to post-war data. But for a variety of reasons, including the fact that neither the wage rates index nor the unemployment statistics used to calculate it have the same coverage as post-war statistics, it is doubtful how far the conclusions can be applied to present conditions.

note (1) page 20 E. Devons and R. C. Ogley, ‘An Index of Wage Rates by Industries’, The Manchester School of Economic and Social Studies, May 1958. Figures for years after 1956 have kindly been made available to us by Mr. J. R. Crossley of Manchester University.

note (1) page 21 Accounting profits roughly measure the ‘historical profit margin’, that is, the margin between selling prices and the costs actually incurred in producing the goods currently being sold. When costs are rising, current costs will be higher than these historical costs; and the margin over current costs, which is roughly measured by profits net of stock appreciation, will be lower than accounting profits.

note (1) page 25 Being based on a comparison of calendar years, it shows broad trends but not all the short period changes: for instance, some of the rise in prices shown between 1957 and 1958 occurred in 1957, and during 1958 itself prices on average rose only 1 to 1 1/2 per cent.

note (2) page 25 The retail price index is based on the pattern of expen diture of working-class and medium salary households. Chart 5 also shows an index of overall consumer prices, an average value index based on consumers' total expenditure.

note (1) page 26 It is hoped to give a more detailed analysis of the rise in food prices in a future number of this Review.

note (2) page 26 Though movements in the components of the retail and wholesale price series for manufactures can be compared in some detail this comparison can hardly be exact.