⁽¹⁾ These during-the-year (fourth-quarter-on-fourth-quarter) growth rates relate, like the year-on-year changes shown in table 1, to the output measure of GDP. The expenditure and income measures show similar inflections in the growth rate, though with differences of quarterly timing.

⁽¹⁾ When defined to exclude sales of council houses.

⁽²⁾ Statistical Appendix table 14.(3) Statistical Appendix table 15.

⁽¹⁾ These estimates are not very firm. Not all schemes are included. (E.g., the Training Opportunities Scheme, which cost £220 million in 1983/4 and recorded 66,200 completions in the year, is ex cluded, though it could well have a register effect.) Also, as is stated on the official press notice, ‘the figure given for the effect of the measures on unemployment depends on a number of uncer tain assumptions’.

⁽¹⁾ First estimates of profits are liable to heavy revision and should be treated with caution. A large residual error in the National Accounts and a large balancing item in the flow of funds accounts suggest that profits in 1984 may be overestimated.

⁽²⁾ Apart from coverage, the main difference between the two series is that the first includes all (and only) bank borrowing, while the second includes only short-term borrowing from all sources. Whereas in the 1950s and 1960s short-term borrowing and bank borrowing were almost synonymous, since the mid-1970s an increasing share of firms' medium- and long-term borrowing has been provided by banks. This partly accounts for both the lower level and smaller variance in the all-companies' series.

⁽¹⁾ In recent years, initial estimates of net property income (which have become less reliable since the abolition of exchange con trols) have been consistently revised up. (See last August's Review, pp. 17-18.) Even after revisions, the figures show an increase since 1980 that is surprisingly small in relation to the increase in Britain's net stock of overseas assets.

Part of the explanation may be that, after the abolition of con trols, companies increased their portfolio investment abroad and reduced their direct investment. This switch will have tended to reduce net property income (as it is recorded in the accounts), since part of the return from portfolio investment takes the form of market revaluations (which are excluded from the statistics). (See an article entitled, ‘The UK's External Balance Sheet’ in the June 1984 issue of the Bank of England Quarterly Bulletin.)

⁽¹⁾ However the estimated effect in a full year was a reduction in revenue of £1.7 billion. This large difference resulted mainly from the decision to withdraw lagged VAT accounting on imports, tem porarily increasing revenue by £1.2 billion.

⁽²⁾ Sales of council houses (treated as negative investment in the National Accounts), sales of British Telecom shares, the rebate from the European Community, and the acceleration of VAT pay ments on imports all have weights of zero. Oil revenues have a low weight, since a high proportion of North Sea profits is remitted abroad. Details of the weighting scheme are given in the February 1982 Review, no. 99, p.95.

⁽¹⁾ In a speech at the Lord Mayor's dinner to the bankers and merchants of the City of London, 18 October 1984, printed in the Bank of England Quarterly Bulletin for December.

⁽¹⁾ The estimates assume that domestic production of energy besides coal (oil and gas) and domestic consumption of energy have not been affected by the strike, so that the loss in coal production is matched exactly by the sum of the reduction in energy stocks and the increase in net imports of energy. It is possible, however, that North Sea production was somewhat higher than it would otherwise have been, and domestic energy consumption somewhat lower.

⁽²⁾ The Association of British Mining Equipment Companies has estimated that 10,000 workers have lost their jobs in auxiliary industries as a direct result of the strike and a further 20,000 have lost jobs in associated industries. (Oral answer by the Secretary of State for Employment, Mr Tom King, to a Parliamentary Question on 20 November 1984.)

⁽¹⁾ National Institute Economic Review, no. 108, May 1984, p.6.

⁽¹⁾ Over the period 1967-83, Treasury forecasts made at Budget time have an average (absolute) error of 1½ per cent of GDP; for 1983/4, this is equivalent to £4½ billion (Financial Statement and Budget Report 1984/5, table 3.8). Over the much shorter period 1976-83, NIESR forecasts made in May have an average error of a little less than 1 per cent of GDP, equivalent to about £2½ billion at today's values.

⁽¹⁾ We assume no change in the special employment and training programmes.

⁽²⁾ By ‘real’ what is meant here is volume (i.e. sp-ending at current prices deflated by own price indices), not cost i.e. spending at current prices deflated by the GDP deflator). ‘Real’ is used in the latter (unusual) sense in the White Paper.

⁽³⁾ The definition of fixed investment in table 15 differs from the National Accounts definition (shown in table 19) in that it excludes sales of land and existing buildings (including council houses) and expenditure by public corporations that have been, or are about to be, denationalised.

⁽¹⁾ In addition to the doubts expressed earlier about the seasonal adjustment, exports in the previous quarter may have been reduced by dock strikes in July and September.