^{page 9 note (1)} Apart from the exceptionally bad winter of 1947.

^{page 9 note (2)} Weekly Hansard, 21 March 1972, col. 1354.

^{page 10 note (1)} Speech at the Lord Mayor's Dinner to the bankers and merchants of the City of London, 19 October 1972.

^{page 12 note (1)} The London and Cambridge Economic Bulletin (The Times, 8 January 1973) drew attention to the ‘lopsided’ nature of developments in 1972 compared with the first years of the three preceding upturns. Consumers' expenditure contributed 4 per cent to output growth compared with 3 per cent in the past, while the deterioration of the foreign balance, in 1972 1 3/4 per cent, was only 0.4 per cent on average before.

^{page 12 note (1)} For an earlier version of the calculations see National Institute Economic Review no. 59, February 1972 pages 35-7; also ‘Fiscal policy for stabilisation’ by M. J. Artis in ‘The Labour Government's Economic Record: 1964-1970’ ed. W. Beckerman, Duckworth 1972.

^{page 13 note (1)} The tendency of the progressive tax system to take an increasing proportion of money income as the latter rises.

^{page 13 note (1)} Similar to that used by M. J. Artis op. cit.

^{page 13 note (2)} Partly as a result of the switch to investment grants which are included here in expenditure.

^{page 13 note (3)} At first sight the 1971 observation looks very expansionary on the revenue side—no less than in 1972. This is due in part to a large reduction in tax accruals in the company sector following cuts in corporation tax in both 1970 and 1971 as well as the switch back to investment allowances. In addition, national insurance contributions rose only modestly—there being no uprating until towards the end of the year.

^{page 14 note (1)} The CSO has suggested, however, that the output measure may be the most accurate at present. On this basis, the GDP forecasts made by the London Business School and the National Institute were apparently very accurate indeed, though this still came about as the happy result of offsetting errors among the forecasts of the various components.

^{page 14 note (2)} The Treasury were probably the earliest, and their fore casters may not have had so much evidence of the sharp deterioration in export performance early in the year as was available to LBS and NIESR.

^{page 17 note (1)} Bank of England Quarterly Bulletin, December 1972, pages 459-60.

^{page 17 note (1)} Comparisons in the text are given in terms of the table. Thus they allow for effects of recording errors in 1971.

^{page 17 note (2)} See F. Meyer-zu-Schlochtern and A Yajima, ‘OECD trade model, 1970 version’, OECD Economic Outlook, Occasional Studies, December 1970.

^{page 18 note (1)} See also page 51 below.

^{page 21 note (1)} It might be objected that the budget must have been worth more than 1/2 per cent on the volume of final demand in 1972, and therefore that our error (adjusted for policy changes) must have been larger than £900 million. In the circum stances of significant supply difficulties however it is not clear that one can assume that satisfied final demand increased by the full amount of the budget stimulus. In addition if one regards the devaluation of the pound as a policy change too (rather than an error of forecast) this would work to reduce the adjusted error.

^{page 23 note (1)} National Institute Economic Review, no. 55, February 1971, pages 38-9.

^{page 24 note (1)} This may have been due in part to the small revaluation of sterling in the Smithsonian realignment of December 1971.

^{page 24 note (2)} The import unit value index rose by 8 1/2 per cent between May and November.

^{page 25 note (1)} See W. A. H. Godley and J. R. Shepherd, ‘Long-term growth and short-term policy’, National Institute Economic Review, August 1964; J. R. Shepherd, ‘Productive potential and the demand for labour’, Economic Trends, August 1968.

^{page 25 note (2)} The sum of Civil Employment (including Employers and Self Employed), Armed Forces and the Wholly Unemployed.

^{page 26 note (1)} Op. cit.

^{page 26 note (2)} Assuming that the net contribution of demographic changes and long-term changes in participation rates was zero.

^{page 26 note (3)} This assumes that the official 1966 projection of the labour force at constant demand pressure (see ref. in footnote to table 13) was correct, i.e. little change 1966-1971.

^{page 27 note (1)} Unemployment statistics, Report of an Inter-Depart mental Working Party, Cmnd 5157, HMSO 1972, page 22.

^{page 27 note (1)} Though perhaps the demand for female labour did not fall so dramatically (see table 12). The Department of Employ ment also estimated, somewhat surprisingly, that changes in demand pressure do not have a very large effect on activity rates of married women over twenty—five—see Department of Employment Gazette, August 1971, page 719.

^{page 27 note (2)} Their productivity, in terms of output per man-year, will of course be very low compared with that of full-time workers.

^{page 28 note (1)} The process was summed up in April 1970 by Mr John Silkin (then Minister of Public Buildings and Works) as follows: ‘Current pressures to adopt self-employed status are mainly of a financial kind. Charges avoided by an employer if his men assume self-employed status include SET, the employer's share of national insurance and National Health Service contributions, redundancy fund contributions, and payment for annual and public holidays. He is also relieved of the necessity of obtaining employers' liability insurance cover, of contributing to redundancy payments if redundancy occurs, of making wages up to a guaranteed weekly minimum under the national working rules, or paying sick pay. The self- employed man's own national insurance contribution is, of course, lower than that of an employee. In addition to these legitimate avoidance factors, there is under the self-employed system an opportunity for the unscrupulous to evade income tax’.

House of Commons Weekly Hansard, 28 April 1970, col. 893; quoted in J. B. Wood, ‘How much Unemployment?’, Institute of Economic Affairs, Research Monograph 28, 1972.

^{page 28 note (2)} ‘The fall in the working population since 1966’, Depart ment of Employment Gazette, June 1970.

^{page 28 note (1)} Even after allowing that the net migration figures may have been affected by the low pressure of demand—in other words, that a return to high pressure might encourage some further inflow of foreign workers, particularly in the context of the EEC.

^{page 29 note (1)} Productivity is measured here as output per employee in employment. The movement into recorded and unrecorded self-employment means of course that these figures overstate the productivity performance, particularly in construction. On the other hand, there has probably also been an increase in unrecorded output—again largely in construction.

^{page 29 note (2)} See J. D. Whitley and G. D. N. Worswick, ‘The pro ductivity effects of Selective Employment Tax’, National Institute Economic Review, no. 56, May 1971, and no. 58, November 1971.

^{page 29 note (3)} See J. K. Bowers et al., ‘Some aspects of unemployment and the labour market 1966-71’, National Institute Economic Review, no. 62, November 1972.

^{page 30 note (1)} The proportion of part-time workers in all employees in manufacturing did fall between 1966 and 1967—see British Labour Statistics Historical Abstract, tables 142, 143.

^{page 30 note (2)} See Whitely and Worswick, op. cit.

^{page 30 note (3)} See T. S. Barker and V. H. Woodward, ‘Inflation, growth, and economic policy in the medium-term’, National Institute Economic Review, no. 60, May 1972, pages 47-8.

^{page 31 note (1)} Recognised hours fixed in national collective agreements, etc. beyond which overtime rates normally become payable.

^{page 31 note (1)} There is some evidence for this in the British Labour Statistics Year Book 1970, table 68. We calculate that the average length of paid annual holidays for manual workers, which rose from 1.7 to 2.0 weeks between 1951 and 1962, had risen to 2.5 weeks by the end of 1970.

^{page 33 note (1)} We have for example ignored the possibility that the rate of obsolescence of the existing stock has accelerated.

^{page 33 note (2)} See National Institute Economic Review, no. 51, February 1970, pages 17-21; 36-8, no. 55, February 1971, pages 18-21; also Bowers et al., ‘Some aspects of unemployment and the labour market 1966-71’, NIER, no. 62, November 1972, and references cited there.

^{page 33 note (1)} See ‘People and Jobs’, Department of Employment: ‘In (one) self-service experiment, vacancies notified by employers increased in the first year of operation by about 30 per cent’.

^{page 33 note (2)} See ‘Plan for modern employment service’, Department of Employment Gazette, December 1972.