FISCAL POLICY:
LESSONS FROM THE LAST ECONOMIC CYCLE
HM Treasury November 1997 |
|
Summary | |||||||||||||||||||||||||
|
The Treasury estimates that the economy has probably just moved above its trend level as it did in late 1986. The present fiscal position too bears some similarities to that of the mid-1980s, with government borrowing projected to come down following a number of years of substantial borrowing. | |||||||||||||||||||||||||
|
As the 1980s progressed, the PSBR declined and surpluses were reached. On the basis of unchanged policies, surpluses were projected to continue. But by the early 1990s the public finances had deteriorated markedly with the deficit reaching 7 per cent of GDP in 1993-94. The cost of this deterioration has been high. This episode highlights the following lessons: | |||||||||||||||||||||||||
|
Lesson 1: Take a prudent approach |
Adjust for the economic cycle and build in a margin for uncertainty.
At the end of the 1980s, when the economy was at the peak of the cycle, the fiscal surpluses gave a misleading picture of the health of the public finances. However, fiscal policy was relaxed. Output, revenue and spending projections, and, in particular, the estimate of the level of output relative to its trend, were subject to large errors. The fiscal position turned out much worse than the forecasts and illustrative projections suggested. The costs of reversing a deterioration of the fiscal position suggests building in caution. | ||||||||||||||||||||||||
|
Lesson 2: Be open and transparent |
Set stable fiscal rules and explain clearly fiscal policy decisions. Throughout the last cycle the stated fiscal objectives changed on a number of occasions as the fiscal position evolved. Estimates of the trend level of output and cyclically-adjusted fiscal indicators were not published. As such, evaluation of policy against its objectives by outside observers was not straightforward. | ||||||||||||||||||||||||
|
The Government has taken on board these lessons in the design of the new fiscal framework:
| |||||||||||||||||||||||||
| Introduction | |||||||||||||||||||||||||
|
Start of a new cycle |
Although measures of trend output are uncertain, the Treasury estimates that the economy has probably just moved above trend, following a period of below trend output since the turn of the decade. Chart 1 shows that the present position of the economy in relation to its trend is probably similar to that at the end of 1986. | ||||||||||||||||||||||||
|
| |||||||||||||||||||||||||
|
This paper first examines how the fiscal position developed during the last cycle, and then considers what went wrong. Finally, the paper draws out the lessons from that period for fiscal policy today. | |||||||||||||||||||||||||
|
What happened over the last cycle? | |||||||||||||||||||||||||
|
Like the cyclical position of the economy, the present situation with respect to public borrowing bears many similarities to that of the mid-1980s. Following a number of years of high deficits, the forecast published in the July 1997 FSBR showed the PSBR falling below 1 per cent of GDP next year. | |||||||||||||||||||||||||
|
Borrowing came down |
For the first part of the 1980s, the fiscal policy objective was to reduce the PSBR as a proportion of GDP over the medium term to support monetary policy. The PSBR remained above 3 per cent of GDP until 1984-85. However, consistently outperforming forecasts, it fell gradually to just below 1 per cent of GDP in 1986-87. The FSBR in March 1987 set a medium-term PSBR objective of 1 per cent of GDP, which was described as 'the modern equivalent of a balanced budget'. It stated that 'the buoyancy of revenues has enabled the Government to achieve its long term objective for the PSBR sooner rather than later' and that 'the PSBR is projected to remain at 1 per cent of GDP thereafter.'
| ||||||||||||||||||||||||
| Surpluses at the end of the decade | In 1987-88, the UK recorded a budget surplus for the first time since 1969-70. Assuming unchanged policies, surpluses were forecast to continue. The March 1988 Budget introduced the balanced budget fiscal objective and what was described as a 'cautious relaxation in fiscal policy', cutting taxes by over £6 billion. The FSBR stated that 'the PSBR is now assumed to be zero over the medium term: a balanced budget'. The surplus peaked that year. | ||||||||||||||||||||||||
|
The following Budget in March 1989 cut taxes by a further £3½ billion, to achieve 'a gradual return to a balanced budget ...over the medium term, with both taxes and public expenditure declining as a share of GDP'. The FSBR reaffirmed that the 'long term objective of fiscal policy remains a balanced budget.' This was reflected in the published illustrative projections which showed a debt repayment of £3¼ billion in 1992-93. | |||||||||||||||||||||||||
|
Early 1990s: fiscal deterioration |
However, the surplus fell faster than forecast. It became apparent that the underlying fiscal position had deteriorated markedly. As the economy entered a drawn-out recession, annual outturns were successively worse than forecasts and illustrative projections. The deterioration continued steadily before the PSBR peaked in 1993-94 at just over 7 per cent of GDP (£46 billion), representing a full ten percentage point swing over just five years, and an error between outturn and the 1989-90 illustrative projection of nearly £50 billion. | ||||||||||||||||||||||||
| |||||||||||||||||||||||||
|
From 1993 on: attempted consolidation |
Both Budgets in March and November 1993 tried to repair the damage, in particular introducing significant revenue raising measures amounting to about 1 per cent of GDP in 1994-95, and more than twice that in subsequent years. The fiscal policy objective was recast as 'to bring the PSBR back towards balance over the medium term'. Subsequent Budgets focussed on consolidation via Control Total expenditure restraint. Much of the spending restraint was achieved only by cutting back public investment and, significantly, receipts continually came in under forecast. Five years of stringent fiscal control have been needed to retrieve the situation. The deficit reduction plan introduced in the July 1997 Budget took a major step forward towards securing sound public finances. | ||||||||||||||||||||||||
| |||||||||||||||||||||||||
| So what went wrong? | |||||||||||||||||||||||||
|
Relaxation at the end of the 1980s |
Some of the improvement in the public finances over the second half of the 1980s was recognised at the time to be due to cyclical factors. But most of the improvement was seen as structural, representing a step change in the performance of the economy following supply-side reforms throughout the decade. The debt repayment of above 3 per cent of GDP in 1988-89 taken at face value appeared to justify some relaxation of fiscal policy. | ||||||||||||||||||||||||
| |||||||||||||||||||||||||
|
The underlying trend rate of growth of the economy was overestimated and actual recorded growth was initially underestimated. No quantitative estimate of the economy's cyclical position, or of its relationship to the public finances, was published by the Treasury until the July 1997 FSBR. However, current Treasury estimates now suggest that, despite the large surplus in 1988-89, the debt repayment was due to cyclical factors: the budget was approximately at balance in structural (ie cyclically-adjusted) terms. Therefore, with hindsight, there was no scope for a significant relaxation in policy. As Chart 5 shows, the March 1988 Budget contributed to an estimated increase in the cyclically-adjusted deficit of about 2 per cent of GDP between 1988-89 and 1989-90. | |||||||||||||||||||||||||
|
Forecast errors |
Therefore, the relaxation of the fiscal stance at the end of the decade can now be seen to have led to adverse consequences and to have been inconsistent with the stated fiscal policy objective. However, it might not have led to such a dramatic deterioration in the public finances. If the extent of the output gap at its peak had been recognised early enough there might have been more pressure for fiscal policy to have been tightened significantly before the PSBR reached 7 per cent of GDP in 1993-94. However, successive forecasts and illustrative projections published by the Treasury (and by most independent forecasters) from 1990 onwards failed to predict the magnitude of the deterioration in the public finances. | ||||||||||||||||||||||||
|
A distinction should be made between the short-term forecasts, which are based upon a fully considered view of economic prospects, and the medium-term illustrative projections for later years which were a conditional view based on much more stylised assumptions for growth and inflation. The most significant errors occurred between outturns and the illustrative projections. | |||||||||||||||||||||||||
|
Errors in the illustrative projections of output occurred largely because, although there was increasing evidence of overheating, the extent to which the economy was running above its trend level at the end of the decade was underestimated. Hence the cyclical position was mis-read. The Treasury now estimates that the output gap had closed by the end of 1986, and that output was running at 5 per cent above trend in 1988-89. So, with hindsight, a downward correction to output was probably inevitable. | |||||||||||||||||||||||||
That the economy was entering recession did not become clear until the autumn
of 1990 and, even then, it proved far longer and deeper than anticipated. The
Treasury now estimates that the output gap became negative in 1991-92 and
widened to over 4 per cent of GDP by 1992-93. In 1993-94 the actual level of
output was over 9 per cent lower than the illustrative projections for that year
published in the March 1990 FSBR. | |||||||||||||||||||||||||
|
In turn, this meant that estimates of future revenues and expenditures, and hence of the underlying state of the public finances, were excessively optimistic. There was an error of nearly 8 per cent of GDP on the illustrative projection for the 1993-94 PSBR published in the FSBR in March 1990. The Treasury's latest published estimates of the cyclicality of the public finances suggest that errors in the output projection could explain more than half of this error. Table 1 shows the average errors in Budget projections of the PSBR for different time horizons and the extent to which these errors could be explained by errors in projections of output. | |||||||||||||||||||||||||
| |||||||||||||||||||||||||
|
Discretionary spending increases |
Increases in spending, resulting from the 1990 and 1991 Public Expenditure Surveys (without significant offsetting increases in revenue), also meant that the PSBR turned out higher than illustrative projections. Total spending (GGE) rose from under 38 per cent of GDP in 1988-89 to over 43 per cent of GDP in 1993-94. The increase was over 6 per cent in real terms in 1992-93 alone. | ||||||||||||||||||||||||
|
| |||||||||||||||||||||||||
| Much of this increase was cyclical. But a significant proportion of the increase was discretionary. Current Treasury estimates now suggest that, after adjusting for the effects of the cycle, expenditure rose by over 1 per cent of GDP between 1989-90 and 1992-93. | |||||||||||||||||||||||||
|
Misreading the position of the economy relative to the trend level of output was partly to blame. With the economy entering recession, and monetary policy constrained by membership of the ERM, some increase in the structural deficit might have been expected on demand management grounds. Nevertheless, because the structural deficit had increased significantly following the March 1988 Budget, any further relaxation of the fiscal stance would inevitably lead to a further rise in the PSBR in a downturn. | |||||||||||||||||||||||||
| Lessons from this episode | |||||||||||||||||||||||||
|
There are a number of lessons to be learnt from the fiscal policy experience of the last cycle. The following stand out. | |||||||||||||||||||||||||
|
Lesson 1: Take a prudent approach |
Adjust for the cycle and build in a margin for uncertainty.
At the end of the 1980s the fiscal surpluses were taken largely at face value and so gave a misleading picture of the health of the public finances. The UK was only able to repay part of its public debt because output was significantly above its trend level. With hindsight, had policy been set on the basis of cautious estimates of the cyclically-adjusted position there was no case for relaxing policy. | ||||||||||||||||||||||||
|
However, it needs to be recognised that forming accurate figures of the cyclically-adjusted position is difficult. Cyclical adjustment requires estimates of the current level of output and of the trend level of output, both of which are uncertain. Overoptimism about the trend level of output at the end of the 1980s was a critical factor for the deterioration of the public finances over the last cycle. In the early stages of recovery it is possible to combine above trend growth without an immediate increase in inflation. Thus it can be tempting to conclude that the underlying growth which the economy can sustain without higher inflation is higher than is actually the case. This points to taking a cautious view of the underlying trend rate of growth. | |||||||||||||||||||||||||
|
Moreover, the Treasury's estimates of the cyclicality of the public finances are based only on a stylised average cycle, so can give only a broad indication of the underlying fiscal position rather than a precise estimate(1). So although forecasts of the cyclically-adjusted deficits are key indicators for assessing the appropriate fiscal stance, they are themselves bound to be subject to some margins of error over and above the uncertainty over the output gap. | |||||||||||||||||||||||||
|
Finally, the two components of the fiscal balance - revenues and expenditures - are in themselves difficult to forecast with accuracy. For example, outturns for VAT receipts were consistently below the forecasts published in successive FSBRs since 1989-90(2). So although forecasts of the cyclically-adjusted deficits are key indicators for assessing the appropriate fiscal stance, they are bound to be subject to wide margins of error. Therefore, policy should recognise the significant uncertainty inherent in the forecasting of the public finances. | |||||||||||||||||||||||||
|
The experience of fiscal policy over the last economic cycle demonstrates that it is easier to reduce revenue and increase expenditure than to do the reverse. So when things go wrong with the public finances it is difficult to put them right, particularly since it can take a long time even to recognise the problem. Moreover, when public borrowing turns out higher than expected, debt interest costs also rise intensifying the problem. | |||||||||||||||||||||||||
|
All of this underlines the value of a deliberately prudent approach. If, in the event, the prudence turns out to have been unnecessary - because events turned out as expected, or even better than expected - it is not difficult to deal with the resulting bonus. Therefore a lesson from past mistakes is that policy makers should err on the side of caution. | |||||||||||||||||||||||||
|
Lesson 2: Be open and transparent |
Set stable fiscal rules and explain clearly fiscal policy.
Over the last economic cycle, the stated fiscal policy objectives changed on a number of occasions. However, with a lack of constancy of stated objectives, policy is never obliged to compensate for past slippages since the objective can be changed according to circumstances. Further, objectives tended to be set over 'the medium term', an imprecise time horizon. Estimates of trend output and cyclically-adjusted fiscal indicators were not published. | ||||||||||||||||||||||||
|
Objective evaluation of policy over the last economic cycle was thus made more difficult. Clear fiscal rules or objectives, which remain stable over the economic cycle, allow objective ex post evaluation, particularly when relevant information is published to explain fiscal policy decisions. With this approach, a commitment to meeting stated fiscal objectives can be strengthened. A more credible and predictable fiscal policy can contribute to producing a stable macroeconomic environment. | |||||||||||||||||||||||||
| The new fiscal framework: learning from the past | |||||||||||||||||||||||||
| The Government is determined to avoid the mistakes of the past. The lessons have been taken on board in the design of the new fiscal framework. | |||||||||||||||||||||||||
|
Margin for uncertainty |
The deficit reduction plan set out in the FSBR in July explicitly errs on the side of caution with 'a margin to guard against future uncertainties'. Alternative fiscal projections are presented based upon an assumption for the level of trend output of 1½ per cent of GDP less than the Treasury's central estimate. | ||||||||||||||||||||||||
|
Cautious assumption of trend growth |
Further, the assumption for the trend rate of growth of output used in the projections of the public finances is deliberately designed to be cautious. The NAO independently scrutinised and endorsed this, and other key assumptions, in July. This approach recognises specifically the errors in the estimates of the level of trend output which contributed to the overall mistakes at the end of the 1980s. | ||||||||||||||||||||||||
|
Two clear fiscal rules |
The Government has set the following two fiscal rules which deliberately take account of the significant effect of cyclical variations on the public finances: | ||||||||||||||||||||||||
|
|
These rules are fundamental. For the purposes of evaluating fiscal policy against these rules, outturns for 1997-98 will represent the starting-point of the cycle. | ||||||||||||||||||||||||
|
By setting the two fiscal rules over the economic cycle, the aim is to enhance stability by allowing borrowing to fluctuate due to cyclical changes in output (ie by allowing the automatic stabilisers to operate). However, the exclusion of cyclical expenditure from the Control Total should prevent cyclical changes in social security spending being transferred into permanent changes in other programmes. | |||||||||||||||||||||||||
|
Greater role for cyclically-adjusted indicators |
In the July FSBR, the Government published Treasury estimates of trend output and cyclically-adjusted fiscal indicators for the first time. The Treasury's cyclically-adjusted estimates of key fiscal indicators now play a greater role in setting the stance of policy. Policy will be set by looking at projections of these estimates, as well as at past and projected progress against the fiscal rules, so that any slippage or over-achievement can be counterbalanced by future policy. | ||||||||||||||||||||||||
|
Commitment to transparency and stability |
The Government's commitment to transparency and stability, and to achieving its fiscal rules, will form key features of the proposed Code for Fiscal Stability. | ||||||||||||||||||||||||
| |||||||||||||||||||||||||
1. For more detail of the Treasury's present approach to adjusting the public finances for the effect of the economic cycle, see 'Public Finances and the Cycle' , Treasury Occasional Paper No.4, (September 1995). Earlier Treasury research on this issue was published in 'Fiscal Developments and the Cycle' , Treasury Bulletin, Volume 2 Issue 1 (Winter 1990-91).
2. For more detail about the reasons for the VAT shortfall, see Treasury Occasional Paper No. 9, 'The VAT Shortfall: Report of the Working Group on VAT Receipts and Forecasts.' (September 1997)