Summary [Top]
The most recent business cycle began in 1997Q2. We base this assertion on our new measure of potential output.
We use this new measure of potential output to highlight developments in UK macroeconomic policy. We show that, during the course of 1990s, monetary policy can be characterised by a simple policy response function, featuring the deviation of the inflation rate from the Bank of England’s target and the output gap. We also estimate a fiscal policy rule and find that, between the late 1980s and the mid-1990s, the budget surplus moved closely with the economic cycle. There was, however, a significant structural deficit.
In the second half of the 1990s, there appears to be a break in the relationship between fiscal policy and the business cycle. Fiscal tightening between 1997 and 2001 facilitated the removal of the pre-existing structural imbalance.
Fiscal policy has been loosened since 2001. Substantial tightening in fiscal policy would be required to remove the newly created structural deficit.
Output Gap [Top]
We introduce a measure of potential output based on a Cobb-Douglas production function, akin to that used by the CBO (2001) for the US . We believe this more fundamental approach represents an improvement over other popular ways to calculate the output gap based on a linear-quadratic or Hodrick-Prescott filtered estimation of the trend in real GDP. As well as being theoretically more appealing, it largely eliminates the end-of-sample bias implicit in these other purely statistical methods.
Figure 1 plots the estimates of the output gap obtained using residuals from a 1955Q1 to 2005Q1 linear-quadratic regression on log real GDP (LQ), from an HP-filtering exercise using the same sample period (HPF), and using the log difference between actual real GDP and CDMA’s estimate of potential real GDP (CDMA). The solid line is our measure of the output gap. Towards the end of the sample, the line becomes dashed in order to indicate that this is actually a forecast of the output gap based on extrapolation of underlying variables, since capital stock data are only available until 2003.
CDMA’s output gap estimate indicates that the previous full cycle lasted for approximately a decade, starting in 1986Q2 and finishing in 1997Q2. The current cycle is estimated to have been running since 1997Q2.
Monetary Policy [Top]
Monetary policy has been conducted under different regimes between 1987 and 2003. Nelson (2001) provides a good summary of this period.
To characterise policy a little more formally, we investigate the existence of Taylor-type rules, following Taylor (1995, 1997), which might explain interest rate behaviour in the period analysed. Based on Chadha and Nolan (2007), we can show that the rule
with i=0.06, δ1=0.25, Φπ=1.53, Φy=0.08 and infl*=0.025 reasonably well explains the variability in the base interest rate during the course of the 1990s.
ir_base t is the Bank of England’s base interest rate. infl tis the annual change in the seasonally-adjusted quarterly RPIX index. infl* stands for the Bank of England’s inflation target, which was 2.5 percent for RPIX inflation up until 2003Q4. Φπ and Φy are the long-run response coefficients for inflation and the output gap. A value of 1 for Φπ indicates that a one percentage point upward deviation in inflation from the target eventually results in a one percentage point increase in the interest rate. The constant i=0.06 implies an average 3.5 percent real natural rate of interest. Figure 2 displays the actual and calibrated values of the base interest rate.
Fiscal Policy [Top]
There have been four governments in power during the period analysed. We show below in our Figures and estimation that the overall fiscal balance, as well as current revenues and spending, were tracking the cycle up until 1997. In the period between 1997 and 2001, the budget was delivering “excess” surpluses. In other words, the tight control over government expenditure under the first Labour government of Tony Blair, led to a surplus which was well above what would have been required by tracking of the economic cycle as had gone before. This policy played a key role in curbing the structural deficit which on average stood at an estimated 2.5 percent of GDP between 1987 and 1996. On the other hand, the post-2001 period is characterised by “excess” deficits, in the same sense. This implies a worsening of the structural balance in recent years.
To check formally our hypotheses, we estimated a rule for UK fiscal policy similar to Taylor (1995, 1997):
st is the current surplus as a percentage of GDP. s can then be interpreted as the structural surplus/GDP ratio.
Hence, equation (2) represents a simple way of separating the structural and the cyclical component of the total surplus. If the primary source of changes in the structural balance is discretionary action, then – according to Taylor (1995) - the output gap term in (2) represents the full effect of automatic stabilisers. The estimates of the coefficients using dummy variables to obtain estimates for the sub-periods are reported in Table 1. We use seasonally-adjusted HMT data for the current budget surplus in our estimations.
We find that the budget was running a structural deficit of around 2.5 percent of GDP between 1987 and 1996. At the same time, the output gap very well explains fluctuations in the fiscal balance around the structural level. This close correlation can be seen from Figures 3 and 6.
An important result from our estimation is that the output gap becomes statistically insignificant once the sample period is reduced to the post-1997 years. In other words, we find evidence for the claim that there is little systematic co-movement between the economic cycle and the current surplus (as a percentage of GDP) after 1997. Figures 4 and 5 show the breakdown in the correlation with total current revenues and current government expenditure, respectively.
The budget is estimated to have run a structural surplus of around 2.3 percent of GDP when only a short sample period from the late 1990s is used. This provides some support for our remark that the fiscal tightening administered by the first government of Tony Blair addressed the structural deficit run by the preceding government.
When observations only from the most recent period are used, a relationship with the cycle seems to re-appear. It is too early to make a strong claim on this since the sample period used is too short. Nevertheless, it looks as though the deficit has become more responsive to the cycle, producing larger swings in response to changes in the output gap. At the same time, however, our estimation reveals signs of a re-emerging structural deficit of around 3.1 percent of GDP. This would imply that the fiscal authority would have needed a one-off improvement in the balance by around £9.3bn as of 2005Q1 to get rid of this structural deficit.
We also conducted the following simple thought experiment. We enquire what the development in the surplus would have been, had the government not changed its policy after 1997. Of course, this is not to make any judgement regarding the optimality of the pre-1997 policy from the viewpoint of economic theory. So, we took the parameters from the period when budget balance was closely – though not necessarily optimally - tracking the economic cycle, and apply this formula to the post-1997 period.
Figure 6 displays the result of our exercise. It is evident that the fiscal authority was accumulating “excess” surpluses in the period between 1997 and 2001. As pointed out before, this policy addressed the negative structural balance estimated for the preceding years.
We estimate that the balance is broadly where it would have been had the government not changed its policy in 1997. That policy was, however, characterised by a structural deficit of 2.5 percent of GDP.
Conclusions [Top]
Our newly introduced output gap measure implies that the most recent cycle began in 1997Q2. We find that Taylor rules are helpful in explaining some recent episodes from UK monetary and fiscal policy history. On the fiscal side, the late 1990s brought a period of fiscal conservatism, whereas the post-2001 period is marked by relative fiscal laxity. We show in our estimations and a simple thought exercise that there is some tentative evidence that a structural deficit is emerging. A fairly substantial change in fiscal policy would be required to restore structural balance.
Details on the methodology used here and a more detailed presentation of the estimation results will be published in Horvath and Nolan (2005).
References [Top]
Chadha, J.S. and Nolan, C. (2007): Optimal Simple Rules for the Conduct of Monetary and Fiscal Policy, Journal of Macroeconomics, forthcoming.
CBO (2001): CBO’s Method for Estimating Potential Output - An Update, Congressional Budget office, Washington D.C., August 2001
Horvath, M. and Nolan, C. (2005): Monetary and Fiscal Interactions: Evidence from the UK and the US , CDMA Working Paper, forthcoming.
Nelson, E. (2001): UK Monetary Policy 1972-97: A Guide Using Taylor Rules, CEPR Discussion Paper No. 2931
Taylor, J. B. (1995): Reassessing Discretionary Fiscal Policy, Stanford University , mimeo.
Taylor, J. B. (1997): The Policy Rule Mix: A Macroeconomic Policy Evaluation, Stanford University , mimeo.
Abbreviations used [Top]
BoE – Bank of England
CDMA – Centre for Dynamic Macroeconomic Analysis
HMT – Her Majesty’s Treasury
inv. lhs – inverted left-hand scale
lhs – left-hand scale
ONS – Office for National Statistics
rhs – right-hand scale
sa – seasonally adjusted
Disclaimer [Top]
The views expressed here accurately reflect the personal views of the authors and not necessarily of the University of St Andrews and its governing bodies. While utmost care has been taken to ensure that the information contained herein is correct, CDMA makes no representation that it is accurate or complete. The information contained herein is subject to change without notice. It may be reproduced, distributed or published by any person for any purpose only subject to prior consent of the CDMA.
