If in a year’s time a Chancellor …
I have heard people say that the Budget was a bore. This may be true for those who had to listen to it or for those who are interested in minutiae. But as one interested primarily in economic strategy, I cannot remember a more intriguing situation. How can it possibly be right to propose a huge tax increase for next year, and an even larger one for the year after that, when by general consent unemployment is going to be in the region of three million and rising? I am going to argue that this Budget only makes sense on very special assumptions about the performance of the economy; and that if, a year from now, the situation is broadly unchanged – if, that is, output has not changed much, with unemployment still around three million and the balance of payments not in serious deficit – it will be every bit as wrong to raise taxes then as it would now.
Vol. 15 No. 9 · 13 May 1993
In his analysis of the Budget, with much of which I agree, my friend Professor Godley (LRB, 8 April) says that when private savings and private investment return to normal it will be absolutely essential to increase taxation by whatever amount is necessary to get personal disposable income and consumption back to their normal relationship to GDP; and he goes on to say that the tax increase required to do this ‘is fairly conveyed by the extent to which the share of personal consumption is abnormally high: that is, a sum in the region of £30-40 billion, two or three times as large as what the Chancellor is proposing for the year after next’.
This estimate is most startling. It implies that taxation must be increased by about £1500-£2000 per household. I wonder if in arriving at it he has taken into account the following points: 1. In a recession, retained company profits, and the company investment they finance, are squeezed much more severely than personal income, with the result that the share of personal income in GDP rises in a manner that is reversed automatically in a recovery. 2. In a recession, personal disposable income rises relative to pre-tax personal income as payments of benefits are swollen by the increase in numbers unemployed. This, too, will reverse itself in a recovery. In so far as these two factors have been at work, a tax increase will be needed only to the extent that a return of savings and investment towards normal causes aggregate demand to rise faster than can be met by our capacity to increase output and pay for increased imports – a point which Professor Godley recognises in the first part of his article but seems to forget when, in the second half, he estimates the needed tax increase directly from the increase in the share of personal disposable income and consumption in GDP. After these cyclical effects have been allowed for, there may of course still remain a ‘structural’ budget deficit of some size that will need to be corrected once recovery has reached the feasible limits.
Trinity College, Cambridge
Wynne Godley writes: The effect of the business cycle should have been taken care of by making comparisons between 1992 and 1980-82, two periods of roughly equally severe recession. Between these two periods, personal disposable income (after deduction of direct, indirect and poll taxes to allow for the effects of switches between them) rose by a remarkable six percentage points of GDP – an increase which creates a strong presumption that a substantial rise in taxation or cut in public expenditure will be required at some stage. Having said that, I now think that I may have exaggerated the scale of the fiscal restriction which will be needed and was probably, in any case, wrong to put a figure on it without presenting a great many figures to support the argument.
I cannot here do full justice to the complexity of the issues involved. But there are two relatively simple and important things to be pointed out. First, at the trough of the last recession, the Government was receiving sums in the form of oil taxation and royalties equal to just over 3 percent of GDP. These receipts have dwindled to almost nothing and, other things being equal, will have to be replaced with something else. (Three per cent of GDP is about £15 billion.) Second, there was a rise in social security payments (net of receipts) of about 1 percent of GDP between 1980-82 and 1992 and by a larger amount if comparison is made with years in which unemployment was at more acceptable levels. The extent to which the rise in net social security payments must be considered ‘structural’ depends mainly on the extent to which Britain’s foreign trade performance permits unemployment to fall below current levels. If unemployment has to stay for several years in the region of three million, the social security benefits received by the unemployed will become a ‘structural’ burden that has to be met out of general taxation and this will be magnified because the unemployed do not pay taxes.