Volume 4 NOV 2008
Feature articles When a firm is market-oriented
Accounting induced performance
anxiety
Understanding global imbalances
An interview with Robert E. Lucas Jnr.
The use and misuse of intelligent
systems in accounting
Alumni refresher lecture series
Real options analysis and
investment appraisal
New economic geography and
manufacturing
Print
this page
Inflation targeting
A review of the theoretical foundations of inflation targeting and current
research in the area, with special reference to the Australian experience
by Guay C. Lim
Listen to lecture - August 2008 (MP3 11.1MB)
What is inflation targeting?
Inflation targeting is the practice of monetary policy where the central bank aims to keep inflation within a quantitatively-defined band and, just as importantly, where the central bank communicates the policy to the public. To date, inflation targeting is practiced in 26 countries around the world (see Table 1 for a chronological list).
Table 1: Inflation Targeting Countries
Year of adoption of
inflation targeting |
Country* |
|---|---|
| 1989 | New Zealand |
| 1991 | Chile, Canada |
| 1992 | Israel, United Kingdom |
| 1993 | Sweden, Australia |
| 1998 | Czech Republic, South Korea, Poland |
| 1999 | Mexico, Brazil, Columbia |
| 2000 | Switzerland, South Africa, Thailand |
| 2001 | Norway, Iceland, Hungary |
| 2002 | Philippines, Peru |
| 2005 | Indonesia, Romania, Slovakia |
| 2006 | Turkey |
| 2007 | Ghana |
*Various sources: Finland and Spain abandoned inflation targeting when they joined the European Monetary Union in January 1999; Slovakia, Poland, Czech Republic and Hungary are expected to give up inflation targeting when they join the EuroZone.
As shown, inflation targeting is practiced by different types of economies, for example, industrialised countries such as New Zealand, Canada and the UK, and newly industrialised and emerging economies such as Chile, Korea and Mexico. The operational details vary and the practice has been variously described as pure inflation targeting, flexible inflation targeting, full-fledged inflation targeting, forward looking inflation targeting, strict inflation targeting – just to name a few.
Inflation targeting policy frameworks
Table 2 shows the operational details of inflation targeting for Australia, compared to New Zealand, Canada and the United Kingdom.
Table 2: Inflation Targeting Policy Frameworks*
|
Country |
Measure of |
Target band |
Policy timeline |
|---|---|---|---|
Australia |
Consumer Price Index (underlying) |
Average of 2-3% overthe medium |
No explicit timeframe to correct deviations |
New Zealand |
Consumer Price Index |
Average of 1-3% over the medium term |
No explicit timeframe to correct deviations |
Canada |
Consumer Price Index (core) |
Midpoint 2% +–1% band |
6-8 quarters to correct deviations |
United Kingdom |
Consumer Price Index |
Midpoint 2% 1% band |
If deviation > 1%, the Governor must provide a written explanation to the Chancellor |
The consumer price index (excluding volatile elements) is generally used as the measure of inflation. The target band for Australia is between two and three per cent, a range of one per cent compared to the more common range of two per cent adopted by many other countries. Like all central banks that have adopted inflation targeting, the Reserve Bank of Australia communicates its policy stance and releases monthly updates explaining its policy decision. But, unlike other countries’ institutional structures, the Reserve Bank of Australia is not faced with an explicit timeline to correct deviations from the target band. This allows the Reserve Bank the flexibility to pay attention to the state of the economy as measured by, say, the output gap or the employment gap, in its deliberations about the stance of monetary policy. Inflation targeting as practiced in Australia is by no means strict! The Reserve Bank is not an ‘Inflation Nutter’; a term coined by Mervyn King, the Governor of the Bank of England, to describe central banks which focus exclusively on inflation.
A simple graphical way to present inflation targeting with respect to the unemployment gap is shown in Figure 1. The vertical axis shows the inflation gap (deviation of actual inflation from the mid-point of the target band of 2.5 per cent), while the horizontal axis shows the unemployment gap (deviation of actual unemployment from an underlying nine-quarter moving average value). The top left-hand quadrant shows the occasions when the inflation gaps were high and the unemployment gaps low, indicating that tightening monetary policy (positive changes in the cash rate) were warranted. In contrast, the bottom right-hand quadrant shows the occasions when loosening monetary policy (negative changes in the cash rate) were warranted as the inflation gaps were low and the unemployment gaps high. The graphical analysis in Figure 1 provides some indication of the importance of unemployment in the Australian inflation targeting policy framework.
Figure 1: Scatter plot of the inflation gap, and the unemployment gap; positive changes in the cash rate (grey); negative changes in the cash rate (black); 1993:1-2007:4
The quantity theory of money equation (MV=PY) serves as a convenient framework to think about the development of inflation targeting. If output Y is determined by real factors, it follows that price P is determined by the nominal term MV (quantity of money multiplied by the velocity of money). In the past when velocity V was fixed by technology, it followed that P was determined by the supply of money M. Hence maintaining price stability became tantamount to managing the money supply. When the velocity V became more volatile and less predictable due to financial innovations, controlling the money supply to control P was no longer feasible. Instead, the policy strategy became the direct management of P – in other words, inflation targeting!
Australia’s history of monetary policy mirrors this point. In the 60s, monetary policy was about controlling the money supply, in particular bank deposits. During this time, the exchange rate was fixed and it served as the nominal anchor. However, in the 70s, the breakdown of the Bretton Woods system of exchange rates saw many countries abandon fixed exchange rates in favour of flexible rates. The Australian dollar was floated in December 1983. These were also the years of increased globalisation and financial innovations, so much so that it became increasingly difficult to define the money supply. Management of the growth of money, or monetary targeting, was abandoned, and the checklist approach to monetary policy was adopted.
Over the next few years, the Reserve Bank gravitated towards managing inflation directly; and finally, in April 1993, Australia formally adopted inflation targeting as the strategy for monetary policy. Note that a floating exchange rate system is a requirement for a well-functioning inflation targeting system, since in a world of high capital mobility, independent monetary policy cannot co-exist with a pegged exchange rate regime – the so-called impossibility of the holy trinity!
In parallel with globalisation and financial innovations, there were two critical developments in economic theory that promoted monetary policy as a short-run demand management tool focused on inflation. These were the relationship between inflation and unemployment, and the importance of commitment and credibility in anchoring expectations.
Inflation and unemployment
In 1958, A.W. Phillips published his famous article that demonstrated a negative relationship between unemployment and the rate of change of money wages in the UK. Put simply, during periods of high unemployment, employees are unlikely to demand big increases in pay. In 1959, when Phillips was on sabbatical leave in Australia at the University of Melbourne, he estimated his second ‘Phillips Curve’ and once again established the negative relationship between changes in money wages and the unemployment rate, this time for Australia over the period 1947–1958. Since wage inflation and price inflation are highly correlated, research by academic economists became increasingly focused on the negative relationship between price inflation and the unemployment rate. In particular, the burning question became this: if there exists a negative relationship between inflation and unemployment, does it mean that we have to accept high inflation to have low unemployment?
Figure 2 shows the scatter plot of wage and price inflation rates against unemployment rates for the period 1965–2007. It would be difficult indeed to see any empirical evidence to support a negative relationship between inflation and unemployment for Australia.
Figure 2: Scatter plot of wage and price inflation against the unemployment rate 1965–2007
However, there have been many structural changes in the sample period of more than 40 years. Figure 3 illustrates the changing nature of the relationship between inflation and unemployment. We see a positive vertical relationship before the break-down of the Bretton Woods system of exchange rate determination (1964–1974), a negative relationship before the floating of the Australia dollar (1975–1983), a steeper relationship before inflation targeting (1984–1992) and a flatter relation since 1993, after allowing for the introduction of the GST in 2000. Much econometric research has gone into estimating the slope of these relationships as they provide invaluable information about the ‘trade-off’ between inflation and unemployment.
Figure 3: Phillips Curves over time
Alongside the empirical research, developments in economic theory began to question the existence of a long-run tradeoff between inflation and unemployment. Put simply, while attempts by a central bank to change the inflation rate might produce a nominal ‘surprise’ (with real short-run consequences), rational agents would factor the price changes into their decision-making process. Thus, over time, there would be no real effects and, consequently, no long-run trade-off.
More importantly, academic economists began to argue that since unemployment is a real phenomenon, issues about labour supply and productivity were better managed as long-run problems via the fiscal arm of government. In contrast, inflation is a nominal phenomenon and should be managed by monetary policy. So with financial innovations and globalisation, inflation targeting became the preferred monetary policy strategy to manage short-run demand issues.
Anchoring expectations
The second influential strand of economic theory that had a significant bearing on inflation targeting was the body of academic research that showed that commitment to a strategy enhanced credibility. This in turn ‘anchored’ expectations and led to better outcomes. In other words, if a central bank sends a clear signal that inflation control is a priority and the policy is then well-communicated, people soon expect and act on the expectation of stable prices, and actual inflation will remain low and stable.
The left-hand side graph in Figure 4 shows the high correlation between actual underlying inflation and the Melbourne Institute measure of consumer inflationary expectations. More interestingly, look at the behaviour of wage expectations over the recent period of cash rate increases (see the graph on the right). When the Reserve Bank began its period of monetary tightening, wage expectations remained stable and in fact turned down, suggesting a strong belief in the downturn of inflation and the credibility of the inflation targeting policy. Wage expectations only began to creep up when the Reserve Bank kept raising the cash rate because inflation stayed stubbornly high.
Figure 4: Inflation and wage expectations
The practice of inflation targeting
The interest rate is the instrument used to achieve the inflation policy objective. In practice, the Reserve Bank of Australia meets on the first Tuesday of every month to determine the cash rate and its deliberations are announced. The effective transmission of the policy change is then dependent on the banks and changes in private sector behaviour. Much empirical research has been devoted to improving our understanding of how a change in the policy interest rate is transmitted to the rest of the economy.
Australian economic performance pre and post inflation targeting
Figure 5 shows the performance of the Australian economy, pre and post inflation targeting. The level of inflation has certainly come down; GDP growth remains high but is less volatile; the employment to population ratio is trending up; and the unemployment rate is trending down. By all accounts, the economic indicators are more favourable post inflation targeting, but how much this is due to good luck from the resources boom or good management is still debatable.
Figure 5: Some macroeconomic indicators 1984-2007
Criticisms of inflation targeting
The strategy of inflation targeting has been criticised mainly on two fronts. The first criticism is that a narrow inflation-only focus can lead to undesirable outcomes for growth and employment. This criticism is not relevant in Australia, as we have already noted.
The other criticism is that inflation is mainly imported and hence beyond the control of the Reserve Bank. But is this the case for Australia? The graph on the left in Figure 6 shows the recent divergence in the rate of growth in the price index of tradables and non-tradables; while the graph on the right shows the smaller contribution of tradables to overall inflation. It would seem, in recent times, that inflationary pressures in Australia are predominantly homegrown, notwithstanding the contribution – direct and indirect – of the recent hikes in energy and food prices.
Figure 6: Tradables and non-tradables
Concluding remarks
We have looked briefly at the international practice of inflation targeting and noted the importance of the quantitative target and the communication of policy. Developments in support of inflation targeting include: financial innovations and globalisation; and academic research about the nature of the trade-off between inflation and unemployment, the role of monetary policy as a short run demand management tool, and the importance of commitment and credibility. The Australian experience, to date, has been favourable; but how much is due to good luck and good management is something still to be explored. A lot of research, much of which is based on dynamic stochastic general equilibrium models, is currently focused on exploring ways to enhance the strategy of inflation targeting in the face of a range of shocks, especially asset-price shocks. There is still much we can research and learn about the strategy of inflation targeting.