^{Note (1) page 22} Our estimate of consumers' expenditure in the fourth quarter is rather higher than the provisional official figure.

^{Note (2) page 22} The term ‘exogenous’, here, implies that, to a first approximation, the way in which these demands develop over the forecast period is independent of the evolution of the economy as a whole.

^{Note (1) page 24} See pages 69-71.

^{Note (2) page 24} In terms of the relative export prices of manufactures (dollar values), the loss of competitiveness last year was probably of the order of 4½-5 per cent (see page 12); given the revaluation and expected price developments this year, a further decline in relative price advantage of perhaps 3 per cent is expected for 1972.

^{Note (3) page 24} For some details see the discussion on pages 10-12 above.

^{Note (4) page 24} For example, it is possible that a part of the explanation, at least, lies with long-lagged responses to the 1967 devaluation.

^{Note (5) page 24} It was, though, suggested by more than one method of forecasting. Both the OECD trade model forecast and more informal means of forecasting which, however, embraced an area-commodity breakdown, gave broadly similar results.

^{Note (1) page 25} In the sense that the predicted increase in import unit values, year-on-year, which is put at about 4 per cent, is less than the rise in final consumer prices.

^{Note (2) page 25} Benefits are assumed to be uprated by 10 per cent and contributions pro tanto, with no change in Exchequer con tributions.

^{Note (3) page 25} Before tax, total personal incomes are expected to grow by some 9 per cent between the fourth quarters of 1971 and 1972; after tax (and net transfers abroad), the rise in total personal incomes works out at around 8 per cent, of which price rises are expected to account for some 4½ per cent.

^{Note (1) page 26} Cf. page 8 above.

^{Note (2) page 26} A further possible source of difficulty may lie in the fact that the aggregate consumption forecasting relationship (see M. J. C. Surrey, ‘The analysis and forecasting of the British economy’, CUP, 1971) implicitly weights new credits and repayments equally, though some recent work provisionally suggests that this may not be correct and that new credits should be thought of as more important in expanding con sumer spending than repayments are in reducing it. While the difference would not matter greatly at normal levels of credit flows, it may become important at the exceptionally high levels anticipated in the forecast period, although even then, it would presumably only affect very short-run, quarter-to- quarter movements rather than developments over 18 months.

^{Note (3) page 26} This element in the forecast is obviously likely to be seriously distorted by the restrictions on electricity consumption. (See page 30 below.)

^{Note (1) page 28} See pages 24, and fn. (5) on that page.

^{Note (2) page 28} Complementary procedures are also available—for example, the use of the OECD trade model to generate forecasts for exports of goods.

^{Note (3) page 28} See page 24.

^{Note (4) page 28} The area distribution of exports last year, as well as the commodity pattern, is discussed in chapter I, pages 10-12.

^{Note (5) page 28} At this level of detail, forecasts of exports to the Irish Republic depend on what assumption is made about the diamond trade which was a feature of last year's expansion (see page 10, fn. (4)).

^{Note (6) page 28} See page 26.

^{Note (7) page 28} Average value—this has recently tended to rise faster than the unit value index.

^{Note (1) page 29} This is in sterling terms. In dollar terms, the net effective revaluation implies a further rise up to an additional 2 per cent.

^{Note (2) page 29} This is oversimplified, of course, a significant omitted factor being the effects on primary producers' real incomes and reserves of a, to them, adverse movement in the terms of trade, which ultimately restricts their demand for UK exports. Also, the demand for UK exports will depend inter alia on relative dollar prices, which as indicated above (fn. (1)) are likely to deteriorate faster owing to the revaluation.

^{Note (1) page 30} Strictly, this applies only to the output estimate of GDP; the income estimate is unlikely to be affected to the same extent nor, probably, is the expenditure estimate, because of timing differences and unrecorded de-stocking. Hence, our own ‘compromise’ estimate will probably not show so large a fall as the output figure, but the discrepancy should average out at zero over 2-3 quarters.

^{Note (1) page 35} Or, for the future, on the assumption of no policy change.

^{Note (2) page 35} Reciprocal allowance is again made for the impact of tax changes, the average tax ratio being held constant at the levels reached following the change.

^{Note (3) page 35} For the purpose of this exercise, the forecast was tentatively run on to the end of 1973.

^{Note (1) page 36} By contrast with these sharp movements, our calculations show that there was very little variation in the weighted deficit in the first half of the 1960s.