Abstract
The Selected Issues paper examines the dynamics of the inflation process in the United Kingdom, particularly the influence of external shocks. The paper provides a brief summary of existing empirical work. It presents the expectations-augmented Phillips curve model used in the analysis. The basic conclusion is that low inflation in recent years can be explained reasonably well by a combination of increased competitive pressures. The paper also presents some estimates of the degree of possible overvaluation in housing prices and outlines the links between housing prices and the macroeconomy.
I. Why Has Inflation Been Low?1
A. Overview and Introduction
1. Despite higher oil prices and limited slack in the economy, the U.K. CPI inflation has remained surprisingly low in recent years. Inflation fell to below 2 percent during 1993–99, similarly to developments in the United States and the euro area. Since 1999, while inflation edged up in both the United States and the euro area, it has remained low and stable in the United Kingdom, averaging about 1¼ percent. This impressive inflation performance did not appear to come at the expense of lower growth, with the U.K. economy weathering the global slowdown better than both the United States and the euro area. In fact, U.K. growth has remained not only robust but unemployment has also declined from about 6 percent in 1999 to 4¾ percent in 2004.
Comparison of Inflation Performance
(YoY in percent)
Citation: IMF Staff Country Reports 2005, 081; 10.5089/9781451814286.002.A001
2. It is widely recognized that the adoption of inflation targeting in 1992 and the subsequent decision to give the Bank of England operational independence have been key factors contributing to the U.K.’s stable and low inflation environment (Box I.1). Since 1997 when the target for inflation was changed to a point target, long-term inflation expectations (10-year-ahead), measured by the difference in the yields on indexed and non-indexed government bonds, have been firmly anchored to the target. Between 1997 and 2003, RPIX inflation remained on average close to the target. However, over the past year CPI inflation has remained below the 2 percent target, undershooting prior projections. Why has inflation been so low? In part, it may be due to a series of favorable external shocks.
3. These favorable external shocks have been reflected in the striking differences between goods inflation and services inflation. Prices for services, which now account for about half of the U.K. CPI basket, have been relatively stable since 1995, increasing at an average annual rate of about 3½ percent. In contrast, goods prices have declined since 1999, falling at an average annual rate of ¾ percentage points, offsetting inflationary pressures from services.
CPI inflation (YoY) and REER
Citation: IMF Staff Country Reports 2005, 081; 10.5089/9781451814286.002.A001
Inflation Measures and Targets in the United Kingdom1
The United Kingdom adopted inflation targeting in October 1992 in the aftermath of its exit from the ERM of the European Monetary System. Originally, the target was set in terms of the traditional Retail Price Index (RPI) excluding mortgage interest payments—referred to as RPIX.2 RPIX inflation was to be maintained in the lower half of the range of 1–4 percent. The present inflation targeting framework was introduced in 1997, when responsibility for making interest rate decisions was transferred to the Monetary Policy Committee (MPC) of the Bank of England. The target was redefined as a point target of 2½ percent as measured RPIX.
Inflation Performance
(12-month changes, percent)
Citation: IMF Staff Country Reports 2005, 081; 10.5089/9781451814286.002.A001
In December 2003, with a view to facilitating convergence with the euro area, the Chancellor changed the MPC’s remit to targeting 2 percent inflation as measured by the harmonized index of consumer prices, called the CPI in the United Kingdom. Given the differences in composition, coverage, and aggregation between the RPIX and the CPI, the new 2 percent target for CPI inflation is roughly consistent with RPIX inflation of 2¾ percent; similarly the previous 2½ percent target for RPIX inflation was consistent with CPI inflation of 1¾ percent.
1/ For more details about the experience of the United Kingdom with inflation targeting, see for example Bean (2003). 2/ RPI includes various costs associated with owner-occupied housing costs including council tax, mortgage interest payments and housing depreciation.4. The decline in goods prices is largely attributable to competitive pressures. Increased globalization of production following the conclusion of the Uruguay Round and strong productivity growth in the IT industry were contributing factors. The emergence of China as a major exporter of textiles and other small manufactured goods also increased competition in world markets, and other Asian countries became significantly more price competitive after their currencies depreciated during the 1997–98 Asian crisis. Since 1995, U.K. imports of goods from the emerging Asian economies2 have increased from about 8 percent (1 percent from China) to nearly 12 percent (4¼ percent from China) of total imports. In addition, the United Kingdom experienced a more than 30 percent real appreciation of sterling in 1996–97, which served over time to limit increases in domestically produced goods prices. Both the United States and the euro area have also benefited from declining goods inflation in the 1990s, as a similar tendency for rising imports, especially from Asia, reflected increased competitive pressures. However, in contrast to the United Kingdom, the tendency for goods inflation to decline has dissipated in recent years in these economies.
Goods Inflation
(YoY in percent)
Citation: IMF Staff Country Reports 2005, 081; 10.5089/9781451814286.002.A001
5. This paper attempts to shed light on the dynamics of the inflation process, particularly the influence of external shocks, to help understand the recent low inflation environment in the United Kingdom. It is organized as follows. Section B provides a brief summary of existing empirical work. Section C presents the expectations-augmented Phillips curve model used in the analysis. Section D presents the results, followed by concluding remarks in Section E. The basic conclusion is that low inflation in recent years can be explained reasonably well by a combination of increased competitive pressures, as proxied by sterling real appreciation, declining import prices, and an important degree of persistence in inflation. Looking ahead, the model suggests that, based on the assumptions and forecasts in the most recent World Economic Outlook, inflation in the United Kingdom will rise gradually towards the 2 percent target over the next 2–3 years. Three factors will contribute to push inflation higher: rising import prices; the closing of the small output gap in the United Kingdom; and well-anchored inflation expectations.
B. Summary of Existing Empirical Work
6. Expectations-augmented Phillips curves have been widely used in analyzing price adjustments to changes in some measure of real disequilibrium in goods or labor markets. A common specification of the traditional Phillips curve takes the following form:
where πt is inflation,
7. Based on the specification for inflation expectations suggested by Rudebusch-Svensson (1999), the following equation was estimated over the period 1981:Q1 to 2004:Q3 using four lags of inflation as a proxy for adaptive or backward-looking expectations (standard errors are shown in parenthesis):3
where πt is seasonally adjusted quarterly CPI inflation at an annual rate and gapt is excess demand—calculated based on detrended output using a Hodrick-Prescott (HP) filter. The coefficients on lagged inflation sum up close to unity, which suggests considerable persistence in inflation.4 The output gap term enters significantly with a positive sign.
8. This simple backward-looking Phillips curve model does a reasonably good job in fitting the historical data. However, it persistently overpredicts inflation during the period 1997–2002, following the sharp appreciation of sterling. The overprediction also reflects the limitation of backward-looking expectations, as agents’ expectations adjust only slowly to declining inflation.
CPI Inflation: Backward-looking Phillips
(YoY in percent)
Citation: IMF Staff Country Reports 2005, 081; 10.5089/9781451814286.002.A001
9. To better characterize the inflation dynamics in the United Kingdom, a recent strand of the literature has adopted broadly four different approaches: (i) the introduction of external shocks in a variety of Phillips curve models (e.g., Kara and Nelson (2002)); (ii) the New Keynesian Phillips Curve derived from micro-foundations with an explicit forward-looking component in inflation expectations formation (e.g., Balakrishnan and López-Salido (2002)); (iii) a Kalman filter, which jointly estimates the Phillips curve and a time-varying NAIRU to capture the dynamics underlying structural changes in the United Kingdom (e.g., Greenslade et al. (2003) and Driver et. al. (2003)); and (iv) an analysis focusing on credibility and inflation expectations in inflation targeting regime (e.g., Laxton and N’Diaye (2002)). These approaches have pointed to the importance of external shocks and inflation expectations in explaining the dynamics of U.K. inflation. However, as shown by Balakrishnan and López-Salido (2002), there is little empirical evidence to support the proposition that forward-looking expectations play a dominant role in inflation dynamics in the United Kingdom. Similarly, Batini and Nelson (2002) show that there is a considerable sluggishness in the response of inflation to changes in monetary policy in the United Kingdom, and that this response lag has persisted over different monetary policy regimes.
10. This paper uses a model based on an expectations-augmented Phillips curve, taking into account the influence of external shocks in the context of inflation targeting. Despite wide use, the traditional Phillips curve model has been criticized for being subject to the Lucas (1976) critique, which suggests that the underlying structural relationships contributing to the inflationary process cannot be identified with a reduced-form specification, as the parameters are unlikely to be stable over different monetary policy regimes. Notwithstanding their limitations, the focus of Phillips curve models on the relationship between inflation and the output gap makes them useful for policy simulation exercises.5 This paper aims to contribute to the understanding of U.K. inflation dynamics by including a measure of external shocks to take account of the long-term effects of competitiveness changes. It also uses a hybrid formulation for inflation expectations, which allows for both forward-looking and lagged inflation following Chadha, Masson, and Meredith (1992) and Laxton and N’Diaye (2002).
C. Model Specification
11. With exports and imports accounting for roughly a third of GDP, the United Kingdom is a highly open economy, which suggests that the economy is sensitive to external shocks. Inflation in an open economy, relative to a closed economy, is affected by an additional transmission channel arising from changes in competitiveness and other import price shocks. To capture both current and lagged effects of external shocks, the model includes contemporaneous and lagged changes in import prices, and the effect of increased competitiveness pressures, proxied by changes in the real exchange rate based on unit labor costs. The competitiveness pressures are specified as a moving average process to summarize long-term effects.6 The augmented model of the Phillips curve is given by:
In this specification, inflation depends on inflation expectations
12. Inflation expectations consist of a forward-looking and a backward-looking component. The forward-looking component is based on the monetary authorities’ inflation objective (π*). Consistent with the timing of the adoption of inflation targeting in the United Kingdom, this study assumes π*=1.7 from 1993 until the change in end-2003, and π*=2 in the subsequent period. The backward-looking component reflects the degree of persistence in inflation. It reflects people’s expectations as to how fast the monetary authorities will adjust the policy rate of interest to steer inflation toward the target. More specifically, it reflects people’s view about the frictions in price adjustment, such as the presence of uncertainty about competitors’ price decision, and lags between cost changes and price adjustments (Blinder, Canetti, Lebow and Rudd (1988)). It may also reflect that some fraction of firms may use a strategy of backward-looking rule-of thumb pricing. Following the specification suggested by Laxton and N’Diaye (2002), inflation expectations are defined as a linear combination of the target and observed lagged inflation.8
D. Results
13. Using a general to specific modeling strategy, the following equation for the alternative model of inflation was estimated over the period 1981:Q4 to 2004:Q3:
Virtually all estimated coefficients have the anticipated signs. Both effects of the output gap enter significantly with positive signs. The level of the output gap a year ago affects inflation but also the speed at which the gap is closed during the succeeding three quarters. In estimating this equation, a 16-quarter moving average of the real exchange rate was chosen as best fitting the historical data, as well as minimizing the forecast errors in and out of sample. d91Q2 is a dummy variable for the second quarter of 1991. A test for the imposed restriction on the sum of coefficients cannot be rejected, consistent with a vertical long-run Phillips curve.
CPI Inflation
YoY in percent)
Citation: IMF Staff Country Reports 2005, 081; 10.5089/9781451814286.002.A001
14. The equation based on the alternative inflation model fits the data better than the simple model, especially during the period after 1996. The equation suggests that the low inflation environment in the United Kingdom during this period can be explained reasonably well by downward pressures from external shocks coupled with an important degree of persistence in inflation (see chart below). The favorable impact on inflation of increases in competitive pressures, as proxied by the sharp real appreciation of sterling in 1996-97, peaked in 2000. In particular, it helped to sustain downward pressures on inflation by offsetting pressures from a rebound in import prices during 1999 and 2000. Since then, import prices have declined, playing an increasing role in holding inflation down. The backward-looking component of inflation expectations, in turn, has spread the favorable effects on inflation of the external shocks over time.9
Inflation: Contributions
(QoQ annualized, in percentage points)
Citation: IMF Staff Country Reports 2005, 081; 10.5089/9781451814286.002.A001
15. The importance of inflation persistence does not conflict with the presence of credible inflation targeting and expectations anchored close to the target. Lags in monetary transmission support a degree of stability in inflation on one hand, while credible inflation targeting would ensure that inflation converges toward the target. In addition, as inflation settles down near the target, it becomes increasingly difficult to identify whether the backward- or the forward-looking component of expectations is playing a relatively more important role in explaining inflation.
16. Forecasting with the estimated inflation equation suggests that U.K. inflation will rise gradually towards the 2 percent target over the next 2–3 years (see charts below). The upper and lower bounds in the chart represent the projected mean squared error of the forecast. Under the staff’s baseline projections, the small negative output gap in the U.K. economy is expected to be eliminated in 2005, and there would be no output gap over the medium term as output would grow at potential. At the same time, import prices are projected to rise—based on the assumptions and projections in the most recent World Economic Outlook (WEO).10 These two factors and well-anchored forward-looking expectations would all work to drive inflation to target. But the persistence in inflation would work to slow the convergence. Accordingly, the equation predicts that CPI inflation will rise to 1¾ percent by end-2006 and reach 2 percent in three years time. However, these projections need to be interpreted with some caution as forecast errors are generally large as shown in the chart, and this is only a partial-equilibrium model.
CPI Inflation
(YoY in percent)
Citation: IMF Staff Country Reports 2005, 081; 10.5089/9781451814286.002.A001
Inflation Projections: Contributions
(QoQ annualized, in percentage points)
Citation: IMF Staff Country Reports 2005, 081; 10.5089/9781451814286.002.A001
17. There are some significant uncertainties associated with the projection. An important downside risk stems from estimates of the current output gap. If there is actually more slack in the economy at present, as estimated by the U.K. Treasury, for example, then inflation would pick up more slowly than forecast.11 There are also uncertainties regarding the projection of rising import prices going forward. In particular, with competition in world clothing and textile markets expected to stiffen with the phase out of MFA quotas at the end of 2004 and continuing strong competition provided by Asian economies, it is not clear whether import prices will rise as quickly as projected. Indeed, over the past several years, the WEO has consistently projected a pick-up in import prices 2-3 years ahead, and the actual data have continued to surprise on the downside. Important upside risks center on the pace of increase in the prices of oil and other raw materials, and how much of such price increases is likely to be passed on to consumer prices. Finally, relatively high services inflation in the United Kingdom poses another uncertainty in the forecast. One would expect that sustained competitive pressures on the tradable sector would contribute in part to contain pressures on the non-tradable sector. However, it is not clear how this will play out.
Services Inflation
(YoY in percent)
Citation: IMF Staff Country Reports 2005, 081; 10.5089/9781451814286.002.A001
E. Conclusions
18. The analysis presented in this paper suggests that the recent low inflation in the United Kingdom can be explained reasonably well by a combination of downward pressure from external shocks and persistence in inflation. Specifically, the favorable impact of increased competitive pressures, as proxied by sterling real appreciation and declining import prices has helped to contain inflationary pressures. Persistence in inflation, in turn, has spread the favorable effects on inflation of the external shocks over time. A forecast based on the assumptions in the most recent WEO suggests that inflation in the United Kingdom will rise gradually towards the 2 percent target over the next 2–3 years.
19. Looking ahead, the model suggests that key uncertainties for monetary policy will be associated with prospects for import prices and the size of the output gap. In particular, difficulties in forecasting import prices, which are affected by a wide range of forces outside of the U.K. monetary authority’s control, could pose a significant policy challenge. If import prices do not pick up as projected and goods prices remain subdued reflecting competitive pressures, a question arises as to whether the Bank of England should attempt to achieve higher services inflation to steer overall inflation up to 2 percent. The decision may entail the risk of having to bring down services inflation in the future when import prices do come back, which could be costly in terms of foregone output. Alternatively, should the Bank just take a wait-and-see approach until import prices eventually come back? However, such an approach also might entail costs, as it is not clear how long inflation can remain below the target without jeopardizing monetary policy credibility.
APPENDIX I.1 Data Sources and Definitions
This appendix describes data sources and definitions used in the model of inflation in the United Kingdom for 1981:Q1-2004:Q3.
Inflation (π): Quarterly inflation on annual basis, calculated from CPI index, seasonally adjusted. Data prior to 1988 have been extrapolated using the historical estimates published in the Office for National Statistics (ONS) Economic Trends No. 541 (December 1998). Data source: ONS.
Output gap
Import prices (Pm): Quarterly changes on annual basis calculated from import price index for total trade in goods; not seasonally adjusted as there are no statistically significant seasonal factors using a standard X12 technique.
Real exchange rate (REER): Based on unit labor costs. The moving average process is calculated using year-on-year changes over the lag period.
Inflation target (π*): 1.7 percent from 1993 to 2003, following the adoption of inflation targeting in 1992, and 2 percent in 2004.
References
Balakrishnan R., and J.D. López-Salido (2002), “Understanding U.K. Inflation: the Role of Openness,” Bank of England Working Paper No. 164.
Bank of England (1999), “Economic Models at the Bank of England.”
Batini, N. and E. Nelson (2002), “The Lag from Monetary Policy Actions to Inflation: Friedman Revised,” External MPC Unit Discussion Paper No. 6.
Bean, C. (2003), “Inflation Targeting: The U.K. Experience,” Bank of England, October 2003.
Benati, L. and G. Kapetanios, “Structural Breaks in Inflation Dynamics,” December 2002.
Campa, J. M., and L. S. Goldberg (2002), “Exchange Rate Pass-through into Import Prices: A Macro or Micro Phenomenon?” National Bureau of Economic Research Working Paper 8934.
Chadha, B., P. R. Masson and G. Meredith (1992), “Models of Inflation and the Costs of Disinflation,” IMF Staff Paper Vol. 39, No. 2.
Debelle, G., and J. Wilkinson (2002), “Inflation Targeting and the Inflation Process: Some Lessons from an Open Economy,” Reserve Bank of Australia Research Discussion Paper 2002-01.
Driver, R. L., J. V. Greenslade, and R. G. Pierse (2003), “The Role of Expectations in Estimates of the NAIRU in the United States and the United Kingdom,” Bank of England Working Paper No. 180.
Ericsson, N., and J. S. Irons (1994), “The Lucas Critique in Practice: Theory Without Empirical Basis,” Macroeconometrics: Development, Tensions, and Prospects, Ed. By Kevin D. Hoover.
Fuhler, J.C. (1997), “The (Un)Importance of Forward-Looking Behavior in Price Specifications,” The Journal of Money Credit and Banking, Vol. 29, pp. 338–50.
Herzberg, V., G. Kapetanios, and S. Price. (2003), “Import Prices and Exchange Rate Pass-through: theory and Evidence from the United Kingdom,” Bank of England Working Paper No. 182.
Kara, A. and E. Nelson (2002), “The Exchange Rate and Inflation in the U.K.,” External MPC Unit Discussion Paper No. 11.
King, M., (2002), “The Inflation Targeting 10 Years On,” Speech delivered at London School of Economics on November 19, 2002.
King, M. , (2005), Speech delivered at the CBI on January 20, 2005.
Laxton, D., and P. N’Diaye (2002), “Monetary Policy Credibility and the Unemployment-Inflation Trade-Off: Some Evidence from 17 Industrial Countries,” IMF Working Paper, WP/02/220.
Lucas, R., (1976), “Econometric Policy Evaluation: A Critique,” Carnegie-Rochester Conference Series on Public Policy, Vol. 1, pp. 19–46.
Rotemberg, J., and M. Woodford, (1999), “The Cyclical Behavior of Prices and Costs,” Handbook of Macroeconomics (ed. Taylor, J.B., and M. Woodford) Vol. 1, pp. 1,050–1,135.
Rudebusch, G.D., and L. Svensson, (1999), “Policy Rules for Inflation Targeting,” Monetary Policy Rules, University of Chicago Press, pp. 203–246.
Svensson, L., (2000), “Open-Economy Inflation Targeting,” Journal of International Economics, Vol. 50(1), pp. 155–183.
Prepared by Keiko Honjo.
Including China, Hong Kong SAR, Malaysia, Philippines, South Korea, Singapore, Taiwan, and Thailand.
See the Appendix for an explanation of the data set used.
An alternative estimate of inflation persistence (or inertia), derived from an estimated autoregressive model of inflation, suggests a gradual decline in persistence over time; however, it remains considerable.
Ericsson and Irons (1994) show that the lack of significant empirical evidence supporting the Lucas critique indicates that it may still be valid to use reduced-from models for empirical estimation.
Most other studies looking at the long-term effects of competitiveness use the real exchange rate with a four-quarter lag.
The operator Δ denotes a one-quarter change and Δ4 four quarter changes. All quarterly changes are on annual basis.
The coefficient Φ is related to the average time required by the monetary authorities to eliminate any deviation between observed inflation and the inflation target. In general, it takes the monetary authorities around six to eight quarters to achieve the inflation objective after a shock. This study assumes Φ =0.5, which is consistent with the range suggested by Laxton and N’Diaye (2002) and Batini and Nelson (2002).
Standard tests for a structural break in the CPI equation suggest that such a break might exist in the early 1990s, which would be consistent with a prior expectation of a change in inflation expectations with a shift from a deflationary to a stable inflation environment. A CPI equation estimated over the period from 1992 to 2004 indicates that the explanatory power of the forward-looking component of inflation expectations increases significantly, particularly in gearing inflation towards the target; inflation persistence, however, continues to play an important role in slowing the rise in inflation.
The WEO projects 12-month change import prices to rise gradually from just above zero in 2004Q3 to about 2 percent by the end of 2007.
While there are large uncertainties associated with estimates of the output gap, the widely accepted view in the United Kingdom is that the economy is operating in the neighborhood of full capacity.