FISCAL REPORT.

Young, Garry

The macroeconomic effects of fiscal policy since 1997

One of the intriguing features of the UK economy's recent behaviour is that it has combined a strong exchange rate with an apparently tight public sector budgetary position. This is unusual because a strong and uncompetitive exchange rate is more commonly associated with a loose fiscal position where the public sector is using more resources than it has raised by way of taxation. In that situation, upward pressure on interest rates and the exchange rate crowd out other activity and thereby free up resources for use by the public sector. It is largely for this reason that trade groups representing firms in the manufacturing sector have been lobbying for continued restraint on public sector spending as a means of preventing further upward pressure on the exchange rate.

It might therefore appear somewhat puzzling that the tightening of fiscal policy in the second half of the 1990s did not lead to a softening of the exchange rate. There are a number of possible explanations for this that do not require any change in understanding of the link between the exchange rate and fiscal policy. Most obviously, some of the many other influences on the exchange rate could have been operating in the opposite direction more than offsetting the impact of tighter fiscal policy. One contender here is policy changes in other countries. Since most of the apparent overvaluation of sterling has been against the euro, it could be that the tightening of fiscal policy in the Euro Area has been more vigorous than in the UK, causing the euro to depreciate against sterling. Another possible explanation is that the tightening of fiscal policy has been undone by a loosening of private sector spending which has caused the exchange rate to rise.

While these potential explanations have a good deal of weight, it is not possible here to carry out an exhaustive study of all the many influences on the exchange rate. Instead, the focus will be on the stance of fiscal policy in recent years and whether it has been as tight as headline figures suggest. This Fiscal Report examines the change in public borrowing over the past three years and shows how the improvement has been brought about. It discusses the impact of the identified changes on demand in the overall economy and through this the exchange rate. It suggests that one reason that the tightening in fiscal policy has not had a bigger effect on the exchange rate is that conventional measures of the fiscal stance have overestimated the size of the policy change. This comes about because some tax measures have affected saving rather than spending and so have had a smaller effect on aggregate demand than changes in government spending. Nevertheless, even when a weighted measure of the fiscal stance is used , it still appears that there has been a significant tightening in recent years. It is argued that this has not had a bigger effect on the exchange rate because much of the fiscal tightening was already anticipated and that which was not is likely to be temporary.

Moving on from this, we go on to discuss the outlook for the public finances over the next five years. This takes account of the spending announcements made by the government on 18 July in the Comprehensive Spending Review. [1]

The falling deficit

The government deficit has fallen sharply since it peaked at 7.8 per cent of GDP in 1993-4. By the time the Labour government came into office in May 1997 it had already fallen to 3.6 per cent of GDP in 1996-7. One of the first acts of that government was to introduce a budget in July 1997 which set out its fiscal objectives and introduced a number of tax measures. The development of fiscal policy from the end of 1996-7 therefore coincides fairly closely with the period of office so far of the Labour government.

In the three year period from 1996-7 to 1999-2000, the general government deficit to GDP ratio has been reduced by almost 6 percentage points, from a deficit of [pound]30 billion to a surplus of [pound]18 billion. This has been brought about both by reducing government spending and by increasing the government's income through higher taxation.

The overall improvement can be explained in terms of a few key components of the Government's budget. The main factors are a reduction in public consumption of 0.8 per cent of GDP, a fall in interest payments of 0.9 per cent of GDP, a fall in social benefit payments of 1.4 per cent of GDP, a rise in taxes on households of 1.1 per cent of GDP and a rise in indirect taxes of 0.8 per cent of GDP. These together account for a fall in the deficit of 5 per cent of GDP. Other components of the budget, including for example company taxes, have been less significant in their effect in reducing the deficit.

Looking at the changes in these different key components of the budget, it is clear that the overall improvement is not due just to discretionary policy changes but is also a response to other changes in the economic environment. For example, the reduction in interest payments has come about because of lower interest rates in the world economy. Because most British government debt is issued for long periods at fixed interest rates it can take many years for lower rates to feed fully into lower interest payments. As such, much of the beneficial effect of lower rates on debt interest payments is yet to have its full effect. Similarly, the further improvement in labour market conditions since 1997, which has reduced unemployment by about 500,000 and raised employment by a million, has had a substantial impact in reducing social benefit payments by the government and raising income and indirect taxes.

Together these two changes, falling interest rates and rising employment, account for a large proportion of the improvement in the fiscal position. According to our forecast, there is no reason to suppose that these factors cannot last, although this cannot be taken for granted. We discuss further below their impact on the outlook for the public finances. The key areas where the government acted to improve the fiscal position were in raising certain taxes and in holding down public spending.

On spending, the government initially adopted the previous, Conservative administration's spending plans. Its own policy on spending was set out in the first Comprehensive Spending Review (CSR) in 1998 which outlined spending for the fiscal years from 1999-2000 to 2001-2. The last of these overlaps with the first year covered by this year's CSR. These spending plans were widely acknowledged to be very restrictive and real spending growth was much lower than the growth of the economy as a whole. As it turned out, real current spending grew by only 0.2 per cent between 1996-7 and 1999-2000; its share of GDP fell from 38.7 per cent to 35.8 per cent over the same period. Total managed expenditure (TME), which includes investment, fell by 0.7 per cent in real terms over this period. To give some idea of the importance of this in reducing the deficit, it is notable that TME would have been about [pound]30 billion higher in 1999-2000 if it had not fallen in relation to national income from 1996-7.

On taxation, the government has introduced a range of measures in its four budgets that have had some overall restrictive effect. The major changes are listed in Table 1, which also reports estimates of the yield or cost of these measures as given by the government at the time they were announced. These are provided only for the first three years but, with the obvious exception of the Windfall tax, in most cases the yield or cost continues after this.

This table shows that the direction of tax policy has changed as the public finances have been put on a more sustainable footing. In its first budget the Government ratified some tax measures announced by the previous Conservative administration that had not yet had an effect. These included the abolition of tax relief for profit related pay (PRP) and the effect of the tobacco and road fuel tax escalators. These measures alone were set to raise [pound]4.8 billion in 1999-2000. It also introduced new measures of its own that would raise [pound]5.2 billion in 1999-2000. If these estimates are accurate, then [pound]10 billion of the fiscal tightening to 1999-2000 was due to the policy measures ratified or introduced in the Government's first budget. But since that time, the main thrust of tax policy has been much less restrictive and, since March 1999, tax measures have had a net cost to the exchequer.

One factor that should be borne in mind in considering recent fiscal changes is a possible asymmetric effect of tax changes on aggregate spending in the economy. It is likely that the restrictive effect of the taxes that have been increased is smaller than the expansionary effect of the taxes that have been cut. A number of examples illustrate this point. The tax changes that have raised the most revenue are the additional indirect taxes on fuel and tobacco and the abolition of tax credits on company dividends. Despite their possible environmental and health justifications, it is unlikely that extra indirect taxes on tobacco and petrol have much effect on the amounts consumed (although tobacco duties have clearly affected the amount smuggled). In this case, real incomes are reduced and the taxes are paid for either by reducing current consumption of other goods or by reducing future consumption generally by lowering the rate of saving. For many people, the initial impact will have been on saving. Similarly, t he reduction in tax credits reduces the disposable income of pension funds in the short term and the income of the pension beneficiaries in the long term. Again, the effect on current consumption is likely to be small.

Against this, it is likely that some of the tax reductions will have a more expansionary effect because they are more obvious to the recipients. The redistributory nature of these changes also means that the recipients are more likely to spend their gains. Examples here include the reduction in the basic rate of income tax and the introduction of a new 10p rate announced in Budget 1999 which have an obvious effect on the pay packet. Similarly the working families tax credit and various child tax measures are going mainly to low income families who are likely to spend a high proportion of their extra disposable income.

In addition to these tax measures there have been other changes in the tax system such as the introduction of self assessment for income tax in. For reasons which are not entirely clear this appears to have resulted in permanently higher tax receipts. This, together with the various tax measures, has led to a substantial increase in income tax payments in recent years. Taxes on income paid by households were steady at just under [pound]75 billion from 1995 to 1997, but rose to [pound]91 billion in 1998 and have continued rising since. Household gross saving was also flat between 1995 and 1997 at around [pound]52 billion, but fell in 1998 to [pound]34 billion and has remained very low since. The coincidence of the size of the change in income tax payments and saving, both around [pound]17 billion, suggests clearly that much of the increase in income taxation has so far been paid for out of saving rather than spending.

It is considerations such as these that point to the inadequacy of some straightforward measure of the change in public sector net borrowing as an indicator of the impact of the changing fiscal position on demand in the economy as a whole. But just as some weighting measure is needed to take account of different tax changes, a similar argument applies to other aspects of the budget. For example, while a reduction in public consumption affects aggregate demand directly, a reduction in interest payments only does this to the extent that bondholders reduce their spending. In the short term, much of the reduction in bond-holders' income is likely to lead to lower saving rather than lower spending.

In previous issues of the Fiscal Report we have described a weighting scheme that allocates different weights to every component of the budget according to an estimate of its contribution to national savings or consumption. (This was outlined in detail in the July 1998 issue of the Review.) This can also be used to calculate a weighted measure of the contribution of the budget to aggregate demand. Here public spending on consumption and investment goods are given a weight of one, while spending on social benefits and debt interest are given weights of 0.68 and 0.46 respectively. On the receipts side of the budget, taxes on non-property income and social contributions are given a weight of 0.68, indirect taxes have a weight of 0.76, taxes on property income have a weight of 0.46 and company taxes a weight of 0.23.

According to this 'macroeconomic budget' the changes since 1996-97 have reduced the contribution of fiscal policy to aggregate demand by 3 percentage points of GDP. A different way of putting this is to say that changes in the different components of the budget have an effect equivalent to a reduction in government consumption of 3 percentage points of GDP.

This is only half of the size of the change in the unweighted deficit over the same period, suggesting that the unweighted deficit has exaggerated the scale of recent fiscal tightening. Indeed, our weighted measure is rather crude and could be said to miss some of the subtleties of recent tax changes. For example, there has been relatively little change in company tax receipts since 1996-7, but there have been substantial changes in the company tax system. These have consisted of increases in dividend taxes, through the abolition of dividend tax credits, advance corporation tax and offsetting cuts in the company tax rate. It is likely that these changes have increased investment incentives and raised aggregate spending even though they have had only a modest impact on tax receipts. Similarly our weighting makes no allowance for the various changes in household taxes which have redistributed income to the less well-off.

Thus it is possible that our estimate of a fiscal tightening worth 3 per cent of GDP is an overestimate of the true effect. Nevertheless, while there may be doubts over the magnitude of the effect, it does seem fairly clear that fiscal policy has been restrictive over the past three years. A better estimate might be that the tightening is worth 2 per cent of GDP. Accepting this, we can then ask what scale of effect this is likely to have on the exchange rate. For this purpose, we can use the National Institute's domestic model to outline the effect of such a tightening in fiscal policy.

In Chart 3 we show the response of the exchange rate to a previously unanticipated tightening in fiscal policy worth 2 per cent of GDP. This is assumed to come about through reductions in public consumption and investment of [pounds]10 billion per annum each. As the chart shows, this would be sufficient to reduce the exchange rate immediately by about 7 per cent; that is, to reduce the sterling exchange rate from [epsilon]1.65 to [epsilon]1.54.

Given the very large movements in the exchange rate over the past few years, where sterling's rate against the euro appreciated by 18 per cent between May 1997 and April 2000, it is difficult to argue that the fiscal policy tightening has not had some downward effect on the exchange rate of this magnitude.

But more than this, it is clear that the fiscal tightening seen over the past three years is not the permanent, unanticipated change that is shown in the simulation. Looking back to our Fiscal Report written in April 1997, based on the assumption that a Labour government would be elected in May, we were forecasting that the public sector borrowing requirement (PSBR) would fall from 3.5 per cent of GDP in 1996-7 to 0.7 per cent of GDP in 1999-2000. This was very close to the latest Treasury view at the time of a fall from 3.1 per cent to 0.5 per cent of GDP. So it can be fairly argued that much of the fiscal tightening over the past three years was already anticipated. As a consequence we would expect it to have had some of its effect on financial markets prices, such as interest and exchange rates, at the time that the policy of fiscal consolidation was first believed. Chart 3 also shows the effect on the exchange rate of an anticipated change in fiscal policy five years into the future.

It is also the case that the fiscal tightening seen over the past three years is not expected to be permanent. The government's main fiscal target is the Golden Rule, that it will borrow no more than is necessary to finance its net investment. But this target was overachieved in 1999-2000 by about 2 per cent of GDP. Although this particular rule is asymmetric and does not oblige the government to aim for a zero current balance, it is clear that the government has no intention of maintaining a surplus of this magnitude. Its Budget 2000 projections show the current balance returning to a little over 0.5 per cent of GDP by 2003-4, with the surplus maintained mainly for prudential reasons. Chart 3 also shows the effect on the exchange rate of a tightening in fiscal policy of 2 per cent of GDP that is believed to be temporary. Here the exchange rate depreciation is a much smaller 4 per cent.

It should be stressed that the reason that temporary fiscal changes are thought to have a smaller impact on the exchange rate and the economy more generally than permanent changes of a similar size is because expectations of the future development of fiscal policy are believed to be important. When spending is believed to be permanently lower, markets will expect downward pressure on interest rates to be more intense for longer and this will exert more downward pressure on the exchange rate. When spending is only temporarily lower, the effect is more muted.

Similarly, there may also be expectational effects on other aspects of behaviour. For example, people might expect that the savings built up by the government when the budgetary surplus is temporary will be returned to them through lower taxation in the future as the fiscal position is returned to balance. Such an expectation of lower future taxation encourages desired private saving to be lower at the same time that public saving is higher. This type of behaviour, where private savings partially offset the effect of temporary changes in public saving, is a feature of our model. While this feature may be looked at with some scepticism, the evidence is such that there have been counteracting movements in public and domestic private saving in recent years. These movements are not just a reflection of national accounting identities since the current account of the balance of payments could also change to accommodate them.

These arguments indicate that the impact of budgetary changes on aggregate demand and the exchange rate are much more muted when fiscal policy is set within a clear medium-term framework. This provides the background for expectations and enables people to anticipate future movements in tax and spending. On the one hand it means that countercyclical fiscal policy is less effective than it would otherwise be; on the other hand it leaves less for it to do.

Thus to put together the various points of our discussion so far, the observed fiscal tightening between 1996-7 and 1999-2000 of 6 per cent of GDP is probably only worth about 2 per cent of GDP once the different components of the change are weighted together to take account of their effect on aggregate demand. A permanent, unanticipated change of this magnitude would reduce the exchange rate by about 7 per cent. But with much of the change already anticipated and the unanticipated component thought to be temporary, the effect on the exchange rate over the period from 1996-7 is likely to have been quite small, say at most 3 per cent.

The Comprehensive Spending Review

The results of the second CSR were announced on 19 July. This contained relatively little news on the overall levels of public spending for the period from 2001-2 to 2003-4, since these plans were largely unchanged from those announced in Budget 2000. The forecast described in the UK chapter of this Review incorporates the decisions announced in the CSR.

The key aspects of this are that current spending is set to rise by 2 1/2 per cent per annum in real terms, with net investment rising substantially to 1.8 per cent of GDP by 2003-4. Total Managed Expenditure (TME) is planned to rise at an average annual rate of 4 per cent in real terms. On the basis of the Treasury forecast, this means that TME will rise from 37.7 per cent of GDP in 1999-2000 to 40.5 per cent of GDP in 2003-4.

The key questions that arise from the CSR are whether it is prudent in terms of the needs of both the public finances and the economy as a whole.

As far as the public finances are concerned, there is generally a danger in fixing spending plans for a period stretching almost four years into the future. As the experience of the early 1990s showed, economic circumstances can change suddenly so that what appeared prudent at one time might seem ridiculously extravagant at another. But the risks of a very bad outcome on this occasion seem relatively small. There are three main reasons for this. First, the public finances now appear to be on a sound footing with the balance sheet on an improving trend. On the basis of Treasury forecasts, underpinned by a range of cautious asumptions, the current balance will be in surplus by a little over half a per cent of GDP by 2003-4 and public sector net borrowing will be about 1 per cent of GDP. This means that it should be possible to absorb even large shocks from such a strong position. Second, our own forecast for the economy is much more buoyant than the Treasury's and on this basis, discussed further below, the Gov ernment's spending plans are very easily affordable. Third, the departmental spending plans themselves are fixed in cash terms. This means that the main risk to which the budget is exposed is of a deflationary shock which reduces money GDP relative to the Treasury's forecast. Such a shock would allow real expenditure to rise more rapidly than the Government is currently anticipating. Other shocks, such as faster money GDP growth, would tend to increase government receipts so that the actual surplus would turn out to be bigger than that projected. Given our forecast, this appears to be the most likely direction in which the government might be 'surprised'.

Table 2 outlines our forecasts for the public finances on the basis of the government's spending plans and our assessment of the likely development of the economy over the period of the CSR, to 2003-4, and the two years following to 2005-6.

Because our forecast is quite buoyant, partly due to the expansionary effect of future government spending, and money GDP rises faster than the government is assuming, we see TME staying broadly constant as a share of GDP. With the investment share rising to 1.8 per cent of GDP, current spending is actually projected to fall slightly in relation to national income. With receipts also projected to fall slightly as a share of GDP (see Table 3), the current balance is set to remain constant at around [pound]20 billion per annum. The surplus on the public sector financial balance is set to decline from [pound]16 billion in 1999-2000 to around balance by 2003-4. The public sector net cash requirement will have a similar profile except for in the current year when it is reduced by receipts from the sale of spectrum licences which raised [pound]22 billion. The national accounts treatment of this windfall is described in Box 1.

Thus, we are fairly confident that the CSR spending plans can be afforded. However, it is less obvious that the increases in spending are appropriate for the economy as a whole when it is operating so close to full employment. The most complete way of discussing this is within the context of our overall forecast described in the UK chapter of this Review. That shows that interest rates will have to rise to about 61/2 per cent and the exchange rate remain at a high level for there not to be a spillover into faster inflation. The forecast also predicts that there will be some increase in private sector saving to make way for the fiscal expansion and that a pick-up in productivity growth will allow output to grow more quickly for a period broadly coincident with that of the CSR.

Table 4 shows that the contribution of fiscal policy to aggregate demand up to 2003--4 is increasing by about 1 per cent of GDP. This reverses about half of the fiscal tightening from 1996-7 discussed above. As that discussion suggested, the fiscal expansion is likely to have a different effect on the economy because it is anticipated and is correcting a period of excess saving by the government. By conducting fiscal policy within a predictable framework, movements in the budgetary position are known to be temporary and so have less effect on prices in financial markets. Thus we do not expect a loosening in the fiscal position to lead to a significant rise in the exchange rate for the same reason that the rate did not fall far over the period when policy was tightened. In effect we are arguing that changes in private sector behaviour appeared to offset much of the fiscal policy change when policy was tightened and are expecting that it will do so again in the opposite direction when it is loosened.

Summary and conclusion

In this Fiscal Report we have looked at changes in budgetary policy over the past three years and looked forward to the period covered by the Comprehensive Spending Review. We have tried to throw some light on the question of why an apparent substantial tightening of policy has not led to a more competitive exchange rate as conventional economic analysis would suggest. Accepting that other influences on the exchange rate, particularly changes in the mix of policy in the Euro Area, might have moved in the opposite direction, we find two broad reasons why the effect might have been small.

First, the effect of the fiscal policy changes on aggregate demand is much smaller than their effect on the fiscal deficit. This is because some of the reduction in public spending, particularly that on debt interest, tends to have relatively little effect on aggregate spending in the short term since those whose income is reduced will save less rather than spend less. Similarly, where taxes have gone up, say on company dividends, the recipients will have reduced their saving rather than their spending. Putting this together we estimate that the tightening of fiscal policy of about 6 per cent of GDP is worth about 2 per cent of GDP in terms of its effect on aggregate demand.

Second, while this leaves a significant fiscal tightening, we believe that its impact has been reduced because much of it was already anticipated and that which was unexpected is thought to be temporary Because fiscal policy is set within a medium-term framework which sets expectations of how it will evolve over time, short-term movements in the public sector budget have a smaller effect on the overall economy than if the same changes occurred unexpectedly. Thus, the behaviour of both the financial markets and the private sector more generally will be such that private sector spending will move partially to offset the effects of changes in public spending. Certainly, we have observed such changes in recent years.

The Comprehensive Spending Review can be analysed from the same perspective. We estimate that the relaxation of fiscal policy over the next four years is increasing the government's contribution to aggregate demand by about 1 per cent of GDP. But given the government's fiscal rules, the expansion in spending and general easing of the fiscal stance is not surprising. Just as the tightening of policy within a medium-term fiscal framework had relatively little effect in reducing aggregate demand or the exchange rate over recent years, so its easing is unlikely to add much upward pressure.

Although there are many risks, we are reasonably confident that the public finances are not unduly threatened by the spending outlined in the CSR. This is mainly because it is being introduced from a strong budgetary position. Our central forecast suggests that the current balance will remain in surplus over the coming years after allowing for the increased spending in the CSR. While such forecasts have a large margin of error a very adverse outcome is seen as remote. This is partly because the spending is fixed in cash terms so that inflationary shocks would actually improve the public finances relative to our forecast. The main threat comes from deflationary shocks that reduce money income and hence tax receipts. But were such shocks to occur it is likely that interest rates would be cut and this would help the budgetary position by reducing debt interest further. The other type of shock that poses a risk is one where the economy behaves well but tax receipts disappear as mysteriously as they have arrived i n recent years. In such a circumstance, if spending cannot be cut other taxes would have to be raised instead. This may not be popular but it is the implication of the government's medium-term fiscal framework.

NOTE

(1.) Details are contained in Spending Review 2000, Cm 4087, The Stationary Office.

Box I. The accounting treatment of the broad band spectrum M.R. Weale

The government has auctioned rights to use the broad band spectrum for 20 years for a price of [pound]22bn. The question is how this should be treated in the government's accounts. The government's view is that it would be wrong to show the [pound]22bn as income in the current year and it is more sensible to 'take in' a constant amount in each of the 20 years, so that the revenue accrues over a 20-year period. [1] EUROSTAT, however, argues that the whole of the amount should appear in the government's accounts when the money is paid.

Our own view is a synthesis of the two alternatives, as the following makes clear. The general position is that the government owns both financial and real assets. The broad band spectrum is a real public asset which was believed to be worthless. It has now been revalued by [pound]22bn and the government has sold a 20-year lease on it for this amount. Our analysis assumes it buys financial assets with the proceeds. This is exactly equivalent to debt reduction.

Neither the revaluation, nor the sale of the lease affect the governments current income. However, an asset has been sold and the public sector net cash requirement is plainly reduced by [pound]22bn over the period in which the cash is received. The use of the proceeds to acquire financial assets means that the permanent benefit of the sale appears as interest received. This results in more money being available for expenditure on goods and services within the overall expenditure total.

However, the recurrent benefit to the public purse is slightly greater than these calculations suggest because the government has sold not a freehold but a 20-year lease. Any prudent leaseholder shows payments into a sinking fund as current expenditure so as to be able to renew the lease on expiry. So too the vendor of the lease has slightly more to spend than the interest on the proceeds because the lease will come up for renewal in 20 years. This is analogous to negative amortisation. It is current income just as much as ordinary depreciation is current expenditure. To put it another way the government could allow the stock of financial assets bought with the proceeds of the sale to run down gradually.

The overall annual income is thus equal to the receipt by a lender on a mortgage with a 20-year term, taking interest and capital together and the stock of financial assets purchased initially is gradually reduced. The chart shows the interest and amortisation components this calculation generates, assuming an interest rate of 5 per cent per annum, It also shows the outstanding value of the stock of financial assets. With the assumption that interest payments are received half-yearly, the initial interest saving on the proceeds of the sale is [pound]1.09bn while the total value of interest plus amortisation is [pound]1.75bn. Thus the fact that the spectrum licences last for only 20 years increases the estimate of the benefit from the sale by 60 per cent compared with the interest alone. Plainly, if it were believed that the spectrum would have no value when the licences come up for renewal, then the amortisation amount should be ignored, and the only benefit is the interest saving. But otherwise the effect of the sale as a capital transaction has to be shown when the cash is received, while the recurrent benefit is shown as income and amortisation accruing during the life of the licence.

The tables below illustrate how the broad band revenues should be treated in their first year, assuming that the only consumption is equal to interest plus amortisation; in subsequent years there is no revaluation and the mix between interest and amortisation changes as shown in the chart.

This analysis is probably similar to the government's plans in that the amounts described here as amortisation are probably the same as those which the government refers to as rent. However, the government plans to treat the cash received on sale of the licence as an amount receivable. It is, to say the least, odd to treat an amount which has already been paid as an amount receivable. Our analysis suggests an alternative way of linking the payment of the capital sum at the start of the licence period to the income which is taken in during the period. In that sense it can be argued that it is also consistent with the view that the receipt of the money should show up immediately in the government's capital account.

(1.) See "Classification of Spectrum Licence Receipts", Office for National Statistics, 13 June 2000.