GETTING AND SPENDING

A recession in the UK is unlikely. But low output growth, defined as being less than 1% (for any quarter), has a forecast probability of 35% for the period up to the end of the first quarter of 2003. And the probability of inflation falling within its target range in 2002 has fallen to around 40% as the likelihood of the rate going below 1.5% increases. These are some of the predictions of macroeconomic modelling by Anthony Garratt, Kevin Lee, Hashem Pesaran and Yongchoel Shin, which were presented at the Royal Economic Society’s Annual Conference on Tuesday 26 March.

The research outlines a simple and transparent method for forecasting the probability of macroeconomic events. Using UK data up to the first quarter of 2001 (2001q1), forecasts are computed for the probability of the following three events:

· First, the occurrence of inflation lying within the Bank of England’s target range 1.5-3.5%, an event realised in 2001, which was forecast with high probability estimates of 87% for 2001q2, 70% for 2001q3 and 54% for 2001q4. But this is predicted to be less likely in 2002 with the probability estimate falling to around 40%.

· Second, recession, defined as the occurrence of two consecutive quarters of negative output growth, is forecast with a probability of 10% for the period 2001q2-2003q1.

· Third, the joint event of the inflation target being met and recession avoided (the preferred policy outcome in relation to the trade-off between low inflation but at the potential cost of low output growth or recession). The probability of this event, which occurred, was forecast to be high in the short run, 87% in 2001q2, 63% in 2001q3 and 50% in 2001q4, but to fall to around 35% in 2002.

One of the many problems economic forecasters and policy-makers face is conveying to the public the degree of uncertainty associated with point forecasts. The fan charts produced by the Bank of England are an important step towards acknowledging the significance of forecast uncertainties in the decision-making process and it is clearly a welcome innovation. But the approach suffers from two major shortcomings:

· First, it seems unlikely that independent researchers can replicate the fan charts. This is largely due to the subjective manner in which uncertainty is taken into account by the Bank, which does not readily lend itself to independent analysis.

· Second, the use of fan charts is limited for the analysis of uncertainty associated with joint events. Currently, the Bank provides separate fan charts for inflation and output growth forecasts, but in reality one may also be interested in joint events involving both inflation and output growth.

This research addresses both of these issues. It uses a small but comprehensive model of the UK macroeconomy, the precise nature of which is specified such that the sources of forecast uncertainty are clear. The researchers argue for the use of probability forecasts as a method of characterising the uncertainties that surround forecasts from a macroeconomic model, believing this to be superior to the conventional way of trying to deal with this problem through the use of confidence intervals.

They then provide a general simulation framework, where each step is made explicit, for the computation of probability forecasts and the characterisations of different forms of uncertainty that surrounds them. The analysis of joint events is straightforward in this multivariate setting, where the presentation of the forecast uncertainty associated with such events is conveyed through the use of a single probability estimate, exactly as in the single event case.

The various probability forecasts presented in the study are encouraging from the point of view of the government's inflation objectives. The probability of inflation falling within the target range is quite high in the short run, accompanied with only a small probability of a recession. The relatively low probability of the inflation target being met at the slightly longer horizon does not however represent a poor forecast from the point of view of the Bank of England. Here, the probability of inflation falling within the target range starts to decline, primarily due to a predicted rise in the probability of inflation falling below 1.5%, the lower end of the target range. But while recession is unlikely, low output growth, defined as being less than 1% (for any quarter), has a forecast probability of 35% for the period 2001q2-2003q1.

Notes for Editors: ‘Forecast Uncertainties in Macroeconometric Modelling: An Application to the UK Economy’ by Anthony Garratt, Kevin Lee, M. Hashem Pesaran and Yongchoel Shin was presented at the Royal Economic Society’s 2002 Annual Conference at the University of Warwick.

The paper was delivered by Tony Garratt, who is at the Department of Applied Economics, University of Cambridge, Sidgwick Avenue, Cambridge, CB3 9DE.

For Further Information: contact Tony Garratt on 01223-335270 (fax: 01223-335299; email: A.Garratt@econ.cam.ac.uk); or RES Media Consultant Romesh Vaitilingam on 0117-983-9770 or 07768-661095 (email: romesh@compuserve.com).

The Bank of England’s Monetary Policy Committee (MPC) has been overestimating uncertainty and the risks of higher inflation, according to new research by Professor Ken Wallis of the University of Warwick, presented at the Royal Economic Society’s Annual Conference on Tuesday 26 March. His analysis of the ‘fan charts’ used to represent probability forecasts of future inflation lead him to the following conclusion:

‘The demons that the MPC thought they saw did not materialise. These are early days yet – the numbers are small, and the MPC is still learning. But these mistakes have costs, and more accurate probabilities would have made earlier reductions in interest rates more likely.’

Our economic prospects are uncertain. In many areas of economic activity, it is important to have forecasts of future developments, but these too are uncertain. Forecasters often overlook this, and present forecasts as if they are exact – ‘inflation next year will be 2.4%’, for example – with no guidance as to their likely accuracy.

In 1996, the National Institute of Economic and Social Research (NIESR) and the Bank of England began to publish information about the uncertainty surrounding their forecasts. This was a welcome development. The Bank presented estimates of the complete probability distribution of possible outcomes for future inflation up to two years ahead. Forecast intervals into which inflation might fall with probabilities 10%, 20%, … , 90% were plotted, and since uncertainty increases and the intervals ‘fan out’ as the forecasts get further into the future, the plot has become known as the fan chart.

The Monetary Policy Committee (MPC) has adopted this practice since its inauguration in 1997. The fan charts give a full account of the MPC’s subjective assessment of inflationary pressures, recognising its imprecision.

How good the forecasts are is a question that is asked wherever forecasts are made. It applies to probability forecasts too. If it turned out that it rained on half the days Michael Fish said there was a 50% chance of rain, he would be doing a good job. Ken Wallis’s study develops techniques for answering this question for the fan chart forecasts, and applies them to the first three years of the MPC’s forecasts.

He finds that the average forecast of inflation one year ahead overestimates inflation by 0.2%. But the novelty of the fan charts is the dispersion, not the average. Here he finds that the fan charts overestimate uncertainty, and overestimate the upside risks to the future outcome. Taking a central interval into which inflation is forecast to fall half the time, he finds that inflation actually fell there two-thirds of the time – the MPC overestimated uncertainty. And inflation never fell above this range, whereas in an accurate probability forecast this would have occurred one quarter of the time.

Notes for Editors: ‘Chi-squared Tests of Interval and Density Forecasts, and the Bank of England’s Fan Charts’ by Kenneth F. Wallis was presented at the Royal Economic Society’s 2002 Annual Conference at the University of Warwick.

Wallis is Emeritus Professor of Econometrics, Department of Economics, University of Warwick, Coventry CV4 7AL.

Economic forecasters face two sources of uncertainty. The first is where the probabilities involved are understood from past analyses. For example, this source is reflected in the width of the Bank of England's fan charts. The second is due to new unpredictable events - such as the oil crisis and the terrible events of September 11 - and lies outside the usual calculations.

Although the events in the second class are unpredictable, their consequences are not, and so must be allowed for in theories of economic forecasting. Speaking at the Royal Economic Society’s Annual Conference on Monday 25 March, Professor David Hendry showed how recent developments have achieved precisely that, namely, allowing for both these sources. The resulting framework can successfully explain many well-known phenomena in economic forecasting, hitherto regarded as anomalies, including:

· why some forms of such unexpected sudden shifts are not as pernicious as others;

Surprisingly, poor methods, inappropriate models and inaccurate data are not the main factors in explaining earlier episodes of bad forecasting.

Most of these findings were anomalous in previous theories. The new framework thus provides a general theory against which future proposals can be judged. It also has a number of important practical implications for forecasters:

· different types of model may be preferable for forecasting and for analysing economic policy;

· some widely used methods of economic policy analysis are not robust to unanticipated events;

· and some popular econometric models fail systematically after major shocks, whereas others are more robust.

Hendry also discussed ten areas where further research is needed to clarify other empirical findings that currently lack theoretical explanations, proposing his framework as the vehicle for doing so given its success to date.

Notes for Editors: ‘Economic Forecasting: Some Lessons from Recent Research’ by David F. Hendry and Michael P. Clements was presented at the Royal Economic Society’s 2002 Annual Conference at the University of Warwick.

Hendry is in the Economics Department at Oxford University; Clements is in the Economics Department at the University of Warwick. Their research was financed by the Economic and Social Research Council (ESRC) under grant L138251009.

For Further Information: contact Professor David Hendry on 01865-278554 (email via: maureen.baker@nuf.ox.ac.uk); or RES Media Consultant Romesh Vaitilingam on 0117-983-9770 or 07768-661095 (email: romesh@compuserve.com).

MONETARY POLICY SHOULD RESPOND TO MOVEMENTS IN ASSET PRICES, ESPECIALLY PROPERTY PRICES

Monetary policy-makers should pay more attention to asset prices, especially property prices, in their conduct of monetary policy. Disregarding changes in equity and property prices means that the macroeconomic outcomes are not as desirable as they could be in terms of the stability of both prices and output. That is the conclusion of new research by Charles Goodhart and Boris Hofmann, presented at the Royal Economic Society’s Annual Conference on Tuesday 26 March.

Goodhart and Hofmann find that asset prices have a significant effect on future excess demand conditions in the G7 countries and that property prices appear to have a substantially larger effect on output than share prices. Property prices and equity prices may affect future aggregate demand conditions via their effect on private sector wealth and the borrowing capacity of households and firms, so that a change in these asset prices may call for an offsetting response by monetary policy.

But the analysis of optimal monetary policy for inflation targeting is still mainly based on models where property and share prices are not considered. In these models, optimal monetary policy is given by a Taylor-type interest rate rule, with the interest rate as a function of current and lagged inflation rates and current and lagged output gaps. In addition, in open economies, monetary policy reacts to the real exchange rate in order to offset its effect on future demand conditions. Monetary policy does not respond to property or share prices.

There are certainly good reasons why asset prices have been ignored in the conduct of monetary policy. The main argument is that asset price movements are inherently hard to interpret and that the central bank should not claim to have better knowledge of whether a change in asset prices reflects economic fundamentals or not. These researchers believe that these concerns should not lead to a complete disregard of asset prices in the analysis of monetary policy, but rather to more empirical effort to assess the information content of asset prices movements for future excess demand conditions.

Based on quarterly data over the period 1972-98, Goodhart and Hofmann estimate a simple, small structural model for the G7 countries. They find that the future output gap and future CPI inflation in the G7 countries are significantly affected by the real interest rate and the real exchange rate, and also by real property and real share prices. Real property prices have a substantially larger effect on the output gap than share prices and the exchange rate.

Taking the UK as an example, the researchers discuss the implications of the significance of asset prices for the conduct of monetary policy. They show that disregarding asset price movements would lead to a sub-optimal outcome for the economy in terms of inflation and output gap variability. This result not only obtains because the information contained in asset prices about future demand conditions is be ignored, but also because their omission from the model introduces considerable biases, so that monetary policy is based on a mis-specified model of the economy.

The research also shows how a Financial Conditions Index (FCI), a weighted average of the short-term real interest rate, the real exchange rate, real property and real share prices can be derived based on the estimated model for the UK. The derived FCI appears to be a useful predictor of future CPI inflation.

Notes for Editors: ‘Asset Prices and the Conduct of Monetary Policy by Charles Goodhart and Boris Hofmann was presented at the Royal Economic Society’s 2002 Annual Conference at the University of Warwick. It was delivered by Hofmann, who is at the Zentrum fuer Europaeische Integrationsforschung, University of Bonn, Walter-Flex-Strasse 3, 53113 Bonn, Germany.

The study was presented at a special session of the conference on ‘Monetary Policy and Asset Prices‘, organised by Jagjit Chadha of the University of Cambridge.

The other studies presented were ‘The Role of Asset Prices in Transmitting Monetary and Other Shocks’ by Stephen Millard and Simon Wells (Bank of England), ‘Optimal Monetary Policy, Asset Prices and Financial Frictions’ by Jagjit Chadha (University of Cambridge) and Charles Nolan (University of Durham) and ‘The Collapse of the Argentine Peso: Economic Fundamentals or Self-fulfilling Panic?’ by Javier Fronti, Marcus Miller and Lei Zhang (University of Warwick). The papers were discussed by Sumru Altug of the University of York.

For Further Information: contact Boris Hofmann on +49-228-73-1732 (fax: +49-228-73-1809; website: http://www.zei.de); or RES Media Consultant Romesh Vaitilingam on 0117-983-9770 or 07768-661095 (email: romesh@compuserve.com).

TEN YEARS OF INFLATION TARGETS: WHAT IMPACT ON UK MONETARY POLICY?

How has UK monetary policy changed since the introduction of inflation targets nearly a decade ago? Speaking at the Royal Economic Society’s Annual Conference on Monday 25 March, Christopher Martin and Costas Milas described how UK interest rates have been set since inflation targets were introduced in October 1992 and since the Bank of England gained operational independence in May 1997. Their research shows that:

· Policy-makers are not trying to hit the inflation target: they are attempting to keep inflation within the range of 1.4%-2.6% rather than pursuing a precise target of 2.5%.

· Monetary policy is more concerned with high inflation than with low inflation, suggesting there is a deflationary bias.

· Output and unemployment do not affect interest rates; policy-makers are only concerned with the inflation rate.

Inflation targets have proved be a successful basis for monetary policy. Over the past decade, countries with inflation targets have largely succeeded in maintaining low inflation while also experiencing less output volatility, lower unemployment and more predictable monetary policy. To quote Mervyn King: ‘inflation targets form a clear and transparent framework for monetary policy …I think they are here to stay’.

But there are questions and concerns about inflation targets. Are policy-makers trying to hit the inflation target or simply to keep inflation low? Is low inflation seen as being as dangerous as high inflation? Do policy-makers consider output growth or employment or do they focus entirely on inflation? Have inflation targets made any difference to how interest rates are set?

Notes for Editors: ‘Modelling Monetary Policy: Inflation Targeting in Practice; by Christopher Martin and Costas Milas was presented at the Royal Economic Society’s 2002 Annual Conference at the University of Warwick.

The authors are in the Department of Economics and Finance, Brunel University, Uxbridge, Middlesex UB8 3PH.

For Further Information: contact Costas Milas via email: costas.milas@brunel.ac.uk; or RES Media Consultant Romesh Vaitilingam on 0117-983-9770 or 07768-661095 (email: romesh@compuserve.com).

‘KEEP IT REAL!’: USING REAL-TIME DATA FOR ECONOMIC POLICY EVALUATION

Evaluation of past economic policy performance often uses data that were unavailable to the policy-maker at the time. Speaking at the Royal Economic Society’s Annual Conference on Tuesday 26 March, Don Egginton, Andreas Pick and Shaun Vahey argued that it makes far more sense to use ‘real-time’ data, those faced by the policy-maker at the time. Their application of this approach to monetary policy in the 1980s shows that data mismeasurements did not in fact obscure the signs of high inflation as has been subsequently claimed by the Chancellor of the time, Nigel Lawson.

Macroeconomic data are often subject to revisions as the statistical offices update their estimates. These updates occur over long periods and sometimes lead to drastic changes in the economic data. But policy evaluation should use the data faced by policy-makers at the time, which are termed ‘real-time' data.

Until now, real-time data sets have only existed for the United States, made available by the Federal Reserve Bank of Philadelphia. These researchers present a complementary real-time macro data set for the UK, containing real-time data for GDP, consumers' expenditure, unemployment, industrial production, retail sales, monetary aggregates, budget deficit and average earnings.

Using the newly developed real-time data, the authors investigate the claim that downward bias in the measurement of demand side variables disguised inflationary pressures in the late 1980s. It has been claimed by, among others, the then-Chancellor Nigel Lawson that inaccurate data contributed to the lack of policy responsiveness to inflationary pressures. These researchers find that the reality of the policy errors of that time is more apparent using real-time data than with current vintage data.

Notes for Editors: ‘Keep It Real!: A Real-time UK Macro Data Set’ by Don M. Egginton, Andreas Pick and Shaun P. Vahey was presented at the Royal Economic Society’s 2002 Annual Conference at the University of Warwick.

Egginton is at Daiwa Institute of Research Europe; Pick and Vahey are in the Faculty of Economics and Politics, Austin Robinson Building, University of Cambridge, Cambridge CD3 9DD.

For Further Information: contact Andreas Pick via email: andreas.pick@econ.cam.ac.uk; or RES Media Consultant Romesh Vaitilingam on 0117-983-9770 or 07768-661095 (email: romesh@compuserve.com).

THE ‘TECHNOLOGICAL GAP’: WHY US ‘NEW ECONOMY’ PRODUCTIVITY GAINS ARE SUSTAINABLE

It is reasonable to project the recent productivity gains made by the US economy into the future, according to Jason Cummins and Gianluca Violante. Their research, presented at the Royal Economic Society’s Annual Conference on Tuesday 26 March, finds that quality improvement in equipment and software (E&S) explains much of the resurgence of US growth in the 1990s. What’s more, the speed of adoption of new technologies was driven by the ‘technological gap’ - how much more productive new machines are compared to the average machine – which grew dramatically. Since this gap continues to increase, the prospects for future growth are bright.

Since a substantial increase in the quality of E&S was responsible for much of the 1990s growth resurgence, it may be suspected that such gains are unsustainable. But these results show that there is a great deal of potential productivity improvement that remains to be done. The technological gap for E&S between the productivity of the best technology and the productivity of the average practice in the economy was 15% in 1975. In 2000, the figure had jumped to 40%. The technological gap actually increased by five percentage points in the 1990s, despite the boom in capital spending.

According to many macroeconomic models, the improvement of the average productivity of capital depends on the technological gap between the best technology and average practice and on ‘adaptable labour’, which defines human capital. Cummins and Violante find that the growth rate of the average practice moves nearly one-for-one with the technological gap and is correlated with measures of adaptable labour (such as the shares in the labour force of college graduates and young workers). Since the technological gap continues to grow, the analysis suggests that the prospects for future growth remain bright so long as labour market conditions are not unfavourable.

Quality improvement in investment goods is pervasive, especially in high-tech categories. For example, a new PC may have the same price today as a new PC had five years ago, but if it provides ten times as much computing power as before, in effect the constant-quality price of the new PC is one-tenth the price of the old PC. The opportunity cost of innovating - whether it is in producing PCs or tractors—is forgone consumption. Intuition therefore suggests that a comparison of constant-quality investment prices with a constant-quality consumption price is an informative measure of technical change.

The table breaks down the sources of GDP growth, distinguishing between the contribution made by the quantity of capital and by the quality of capital. The quantity of capital is measured in terms of constant-quality consumption units. The quality of capital is the difference between capital measured in terms of constant-quality investment prices and capital measured in terms of constant-quality consumption units. Hence, the quality of capital isolates the contribution to growth from the quality-adjusted investment price indexes they construct.

Improvement in the quality of capital (line 3) explained no more than 0.82 percentage points of growth until the 1990s, when the contribution jumped to 1.1 percentage points. Moreover, even in the 1990s, quality improvement in high-tech categories (line 4), which includes computers, software, communications equipment, and instruments, was less important than quality improvement outside of high-tech categories (line 5). Since the contribution of the quantity of capital (line 6) falls in the 1990s and the contribution of labour (line 9) only edges up, Cummins and Violante’s findings indicate that the increase in the quality of capital in the 1990s explains as much of the resurgence in GDP growth as the increase in multifactor productivity (MFP, line 10).

1960-69

1970-79

1980-89

1990-99

1. GDP Growth

Contribution of:

2. Capital

3. Quality of capital

4. High-tech

5. Other

6. Quantity of capital

7. High-tech

8. Other

9. Labour

10. MFP

4.82

2.11

0.79

0.12

0.67

1.33

0.17

1.16

1.26

1.45

3.89

2.17

0.61

0.23

0.39

1.56

0.18

1.38

1.24

0.47

3.60

2.04

0.82

0.31

0.51

1.23

0.21

1.02

1.40

0.14

4.15

2.27

1.10

0.44

0.66

1.17

0.17

1.00

1.48

0.40

Notes for Editors: ‘Investment-specific Technical Change in the US (1947-2000): Measurement and Macroeconomic Consequences’ by Jason Cummins and Gianluca Violante was presented at the Royal Economic Society’s 2002 Annual Conference at the University of Warwick. Cummins is at the Federal Reserve; Violante is at University College London and is currently visiting the London School of Economics, Houghton Street, London WC2A 2AE.

TACKLING UNEMPLOYMENT: COMPETITION POLICY IS MORE IMPORTANT THAN LABOUR MARKET FLEXIBILITY

Addressing imperfections in product markets through competition policy may be a better way to tackle unemployment than reducing unemployment benefits and rendering labour markets more flexible. That is the conclusion of new research by Jakob Madsen of Brunel University, presented at the Royal Economic Society’s Annual Conference on Tuesday 26 March.

His study challenges the view held by many economists that unemployment in the OECD is a result of wages that have been pushed in excess of their full employment level by easy access to generous unemployment benefits, and by increasing taxes and pushy unions. Rather, he blames the increasing concentration of firms, which can have permanent adverse effects on unemployment.

The study tests the extent to which the level of unemployment exerts downward pressure on real wage growth, using half a century of data for 18 OECD countries. The estimates show that the level of unemployment has a strong effect on wage growth, which suggests that most of the unemployment effects of adverse supply shocks are eliminated after some years and almost fully after a decade. Furthermore, the results indicate that the wage growth response is remarkably similar across countries.

This result challenges the widely held view among economists that variations in the unemployment experiences of OECD economies are due to institutional differences that result in different labour market adjustments. More difficult access to and lower unemployment benefits combined with relatively low taxes is often stressed as the key to the relatively low unemployment in the UK, the United States, Japan and Ireland. By contrast, inflexible labour market institutions have often been blamed for the high and persistent unemployment in continental Europe. Madsen challenges this view.

An adverse supply shock, such as an increase in oil prices, taxes, unemployment benefits or union pushiness, may lead to higher real wages and therefore higher unemployment. Although this scenario may follow such shocks, the key point of Madsen’s analysis is that the effects of the supply shocks on wages will be short-lived.

An adverse supply shock that pushes wages in excess of their full employment level will lead to higher unemployment, which will in turn put downward pressure on wage growth. The downward pressure on real wage growth will prevail until unemployment is eliminated or down to a level at which structural forces make it difficult for it to be reduced further.

The analytical results of the study show that increasing concentration of firms can have permanent adverse effects on unemployment. This trend in firm concentration leads to persistent ‘disequilibrium’ in markets and suggests that addressing goods market imperfections is a more important policy issue than reducing unemployment benefits and rendering labour markets more flexible.

Although several initiatives have already made labour markets more flexible in most OECD countries since the 1960s and 1970s, unemployment remains high in many OECD countries despite the recent economic boom, which has been the largest and longest post-war economic upturn. Madsen suggests that alternative initiatives, such as competition policy, may be better measures to reduce unemployment.

Notes for Editors: ‘The Dynamics of Income Shares and the Wage Curve-Phillips Curve Controversy’ by Jakob B Madsen was presented at the Royal Economic Society’s 2002 Annual Conference at the University of Warwick.

Madsen is in the Department of Economics and Finance, Brunel University, Uxbridge, Middlesex, UB8 3PH.

For Further Information: contact Jakob Madsen on 01895-203168 (email: Jakob.madsen@brunel.ac.uk); or RES Media Consultant Romesh Vaitilingam on 0117-983-9770 or 07768-661095 (email: romesh@compuserve.com).