Most emerging market countries have undergone a period of transition prior to adopting a full-fledged inflation targeting framework (Table 4.1). Policymakers have marked the beginning of the transition period by announcing the intention to adopt inflation targeting or by announcing an inflation target in the context of an exchange rate band. The transition period ends when most of the elements of a full-fledged in nation targeting framework are in place. Emerging market countries’ transitions to full-fledged inflation targeting have ranged from slow (Chile and Israel) to fast (Brazil, Czech Republic, Poland, and South Africa).
Timing of Transitions to Full-Fledged Inflation Targeting by Emerging Market Countries
Official announcement of intend on to adopt an inflation targeting framework.
Announcement of first inflation target.
Announcement of details of full-fledged inflation targeting framework.
Abandonment of exchange rate band or broadening of the band to a range that makes policy conflicts between the exchange rate and the inflation target unlikely.
Timing of Transitions to Full-Fledged Inflation Targeting by Emerging Market Countries
Beginning of Transition to Inflation Targeting | Adoption of Full-Fledged Inflation Targeting | ||
---|---|---|---|
Brazil | June 19991 | June 19993 | |
Chile | Sep. 19902 | Sep. 19994 | |
Czech Republic | Dec. 19971 | Dec. 19973 | |
Israel | Dec. 19912 | June 19974 | |
Poland | Oct. 19981 | Mar. 19994 | |
South Africa | Aug. 19991 | Feb. 20003 |
Official announcement of intend on to adopt an inflation targeting framework.
Announcement of first inflation target.
Announcement of details of full-fledged inflation targeting framework.
Abandonment of exchange rate band or broadening of the band to a range that makes policy conflicts between the exchange rate and the inflation target unlikely.
Timing of Transitions to Full-Fledged Inflation Targeting by Emerging Market Countries
Beginning of Transition to Inflation Targeting | Adoption of Full-Fledged Inflation Targeting | ||
---|---|---|---|
Brazil | June 19991 | June 19993 | |
Chile | Sep. 19902 | Sep. 19994 | |
Czech Republic | Dec. 19971 | Dec. 19973 | |
Israel | Dec. 19912 | June 19974 | |
Poland | Oct. 19981 | Mar. 19994 | |
South Africa | Aug. 19991 | Feb. 20003 |
Official announcement of intend on to adopt an inflation targeting framework.
Announcement of first inflation target.
Announcement of details of full-fledged inflation targeting framework.
Abandonment of exchange rate band or broadening of the band to a range that makes policy conflicts between the exchange rate and the inflation target unlikely.
Inflation Trends
Most industrial countries have adopted a full-fledged inflation targeting framework at a time when inflation was declining (Table 4.2). The only exception is Canada, where the move toward inflation targeting was timed to mitigate the potential inflationary impact of the introduction of a new indirect tax. All of the industrial countries adopted inflation targeting during or just before an economic downturn—as gauged by output gaps recorded in the World Economic Outlook database. Moreover, the adoption of inflation targeting by the European countries, notably Sweden and the United Kingdom, occurred in the wake of the breakdown of the Exchange Rate Mechanism of the European Monetary System and the abandonment of exchange rate pegs.
Inflation During Transition to Full-Fledged Inflation Targeting1,2
(Percent)
Monthly CPI year-over-year inflation.
t is time of the adoption of full-fledged inflation targeting framework.
Excluding the estimated impact of increases in indirect taxes.
Inflation During Transition to Full-Fledged Inflation Targeting1,2
(Percent)
t | Inflation at t Less 36 Months | Inflation at t Less 24 Months | Inflation at t Less 12 Months | Inflation at t | |||
---|---|---|---|---|---|---|---|
Emerging Market Countries | |||||||
Brazil | June 1999 | 16.3 | 7.0 | 3.4 | 3.3 | ||
Chile | Sep. 1999 | 6.3 | 6.0 | 4.8 | 2.9 | ||
Czech Republic | Dec. 1997 | 9.9 | 7.9 | 8.6 | 10.0 | ||
Israel | June 1997 | 12.5 | 9.7 | 12.9 | 8.5 | ||
Poland | Mar. 1999 | 20.2 | 16.8 | 13.9 | 6.1 | ||
South Africa | Feb. 2000 | 9.7 | 5.4 | 6.2 | 2.4 | ||
Industrial Countries | |||||||
Australia | Apr. 1993 | 8.3 | 3.4 | 1.4 | 1.7 | ||
Canada | Feb. 1991 | 3.9 | 4.9 | 4.3 | 4.43 | ||
Finland | Feb. 1993 | 7.5 | 5.0 | 2.6 | 2.7 | ||
New Zealand | July 1989 | 8.5 | 12.83 | 5.9 | 4.93 | ||
Spain | Nov. 1994 | 5.7 | 5.1 | 4.7 | 4.4 | ||
Sweden | Jan. 1993 | 8.8 | 10.0 | 5.2 | 4.7 | ||
United Kingdom | Oct. 1992 | 7.5 | 8.9 | 4.1 | 1.8 |
Monthly CPI year-over-year inflation.
t is time of the adoption of full-fledged inflation targeting framework.
Excluding the estimated impact of increases in indirect taxes.
Inflation During Transition to Full-Fledged Inflation Targeting1,2
(Percent)
t | Inflation at t Less 36 Months | Inflation at t Less 24 Months | Inflation at t Less 12 Months | Inflation at t | |||
---|---|---|---|---|---|---|---|
Emerging Market Countries | |||||||
Brazil | June 1999 | 16.3 | 7.0 | 3.4 | 3.3 | ||
Chile | Sep. 1999 | 6.3 | 6.0 | 4.8 | 2.9 | ||
Czech Republic | Dec. 1997 | 9.9 | 7.9 | 8.6 | 10.0 | ||
Israel | June 1997 | 12.5 | 9.7 | 12.9 | 8.5 | ||
Poland | Mar. 1999 | 20.2 | 16.8 | 13.9 | 6.1 | ||
South Africa | Feb. 2000 | 9.7 | 5.4 | 6.2 | 2.4 | ||
Industrial Countries | |||||||
Australia | Apr. 1993 | 8.3 | 3.4 | 1.4 | 1.7 | ||
Canada | Feb. 1991 | 3.9 | 4.9 | 4.3 | 4.43 | ||
Finland | Feb. 1993 | 7.5 | 5.0 | 2.6 | 2.7 | ||
New Zealand | July 1989 | 8.5 | 12.83 | 5.9 | 4.93 | ||
Spain | Nov. 1994 | 5.7 | 5.1 | 4.7 | 4.4 | ||
Sweden | Jan. 1993 | 8.8 | 10.0 | 5.2 | 4.7 | ||
United Kingdom | Oct. 1992 | 7.5 | 8.9 | 4.1 | 1.8 |
Monthly CPI year-over-year inflation.
t is time of the adoption of full-fledged inflation targeting framework.
Excluding the estimated impact of increases in indirect taxes.
The choice between a gradual and fast track transition for emerging market countries reflects the level of inflation at the beginning of the transition (Table 4.3). Chile and Israel decided to follow the riskier course of first announcing inflation targets to bring down stubbornly high inflationary expectations. The other emerging market countries either had short transition periods or moved directly to full-fledged inflation targeting in a setting of lower and declining inflation. In addition, the countries that moved to inflation targeting in the late 1990s may have benefited from what had already been learned about this framework, whereas Chile and Israel had to “learn by doing.”
Inflation Prior to Beginning of Transition to Inflation Targeting1
(Annual overage CPI inflation; percent)
t is time of the start of the transition to full-fledged inflation targeting framework.
Inflation Prior to Beginning of Transition to Inflation Targeting1
(Annual overage CPI inflation; percent)
Country | t | Inflation at t Less 4 Years | Inflation at t Less 3 Years | Inflation at t Less 2 Years | Inflation at t Less 1 Year | Inflation at t |
---|---|---|---|---|---|---|
Brazil | 1999 | 66.0 | 15.8 | 6.9 | 3.2 | 4.9 |
Chile | 1990 | 20.6 | 19.9 | 14.7 | 17.0 | 26.0 |
Czech Republic | 1997 | … | 10.1 | 9.1 | 8.8 | 8.4 |
Israel | 1991 | 19.8 | 16.3 | 20.2 | 17.2 | 19.0 |
Poland | 1998 | 33.3 | 26.8 | 20.2 | 15.9 | 11.7 |
South Africa | 1999 | 8.6 | 7.4 | 8.6 | 6.9 | 5.2 |
t is time of the start of the transition to full-fledged inflation targeting framework.
Inflation Prior to Beginning of Transition to Inflation Targeting1
(Annual overage CPI inflation; percent)
Country | t | Inflation at t Less 4 Years | Inflation at t Less 3 Years | Inflation at t Less 2 Years | Inflation at t Less 1 Year | Inflation at t |
---|---|---|---|---|---|---|
Brazil | 1999 | 66.0 | 15.8 | 6.9 | 3.2 | 4.9 |
Chile | 1990 | 20.6 | 19.9 | 14.7 | 17.0 | 26.0 |
Czech Republic | 1997 | … | 10.1 | 9.1 | 8.8 | 8.4 |
Israel | 1991 | 19.8 | 16.3 | 20.2 | 17.2 | 19.0 |
Poland | 1998 | 33.3 | 26.8 | 20.2 | 15.9 | 11.7 |
South Africa | 1999 | 8.6 | 7.4 | 8.6 | 6.9 | 5.2 |
t is time of the start of the transition to full-fledged inflation targeting framework.
Disinflation Experiences
Most countries that started with higher rates of inflation and crawling exchange rate regimes disinflated over long transition periods to limit disruptions to the real economy. Chile and Israel slowly shifted from a crawling exchange rate regime to an inflation targeting framework. Their experiences are illuminating because they were the first emerging market countries to announce inflation targets and they began their transition to a full-fledged inflation targeting framework when inflation was at double digit levels.
After a decade when inflation fluctuated around 25 percent. Chile began announcing inflation targets in 1990 and adopted full-fledged inflation targeting in 1999. A major reason for the move toward inflation targeting was to bring down stubbornly high inflation expectations by providing the public with an explicit inflation objective (Landerretche and others, 1999). The Chilean central bank relied on a series of annual targets to guide the inflation rate lower in recognition of the high initial inflation rate, potentially large short-term losses in real output, and employment associated with disinflation in the context of widespread indexation of wages and prices, in addition to the risk of triggering turbulence in the foreign exchange markets.
In Israel, inflation targets were introduced in 1991 as an important input into the specification of an upward slope for the crawling exchange rate peg. There was no clear perception at the outset as to the central bank’s degree of commitment to achieving the inflation targets; only gradually, over a number of years, did the inflation targets develop a life of their own as the authorities demonstrated that achieving them was the primary objective of monetary policy (Bemanke and others, 1999; Leiderman and Bar-Or, 2000). The length of the disinflation in Israel reflected the limited degree of public support for an active program to reduce inflation to single digit levels if that program meant incurring real costs in terms of lost output or higher unemployment (Offenbacher, 1996). These costs were potentially significant in Israel given the widespread practice of indexation of prices in local contracts and the country’s history of high inflation. Consequently, monetary policy aimed to encourage more forward looking behavior on the part of wage and price-setters and a move away from indexation.
Chile and Israel distinguished between the long-run inflation goal and interim inflation targets. The real costs of disinflation depend importantly on how fast inflation is brought down, the extent of nominal rigidities in the underlying economy, and on the attainment of credibility. By using annual targets to adjust the speed of disinflation and take advantage of unexpected disinflation opportunities, the real costs of reducing inflation could be moderated, and employment and output maximized, while at the same time maintaining a strong commitment to long-run inflation control. The adoption of inflation targeting can lead to mutually reinforcing structural reforms in other areas, especially fiscal consolidation, which reduce the costs of disinflation (Leiderman and Bar-Or, 2000).
The durable success of the disinflation efforts of Chile and Israel are instructive. Their initial rates of inflation were moderate—less than 25 percent. Moreover, disinflation was made easier by lasting improvements in structural and fiscal policies. Trend improvements in productivity followed from structural reforms in both countries as well as immigration for Israel. Government debt was reduced sharply in Chile, and the financial systems in both countries were strong enough to weather the Mexican and Asian crises of the late 1990s.
Inflation targeting frameworks notably have not been used to engineer major disinflation from a starting point of high inflation, however. More experience with inflation targeting-based disinflation will be needed before it can be compared with exchange rate– or money-based approaches, which have been used with mixed success to bring inflation down from high levels relatively quickly (Calvo and Végh, 1999).
Shifting from an Exchange Rate Regime
Nine of the 13 inflation targeting countries shifted from an exchange rate targeting regime (Table 4.4). Chile and Israel announced their inflation targets in the context of formulating exchange rate target paths. More recently, emerging market countries have adopted inflation targeting rapidly after dropping their previous monetary framework and announcing their commitment to inflation targeting (Brazil, Poland, and South Africa).
Exchange Rate Regime Prior to Adoption of Inflation Targeting
Exchange Rate Regime Prior to Adoption of Inflation Targeting
Emerging Market Countries | ||
Brazil | Crawling exchange rate band | |
Chile | Crawling exchange rate band | |
Czech Republic | Crawling exchange rate band supplemented by a monetary guideline and a disinflation objective | |
Israel | Crawling exchange rate band | |
Poland | Crawling exchange rate band | |
South Africa | Floating exchange rate with informal inflation targeting and announced guidelines for broad money growth | |
Industrial Countries | ||
Australia | Floating exchange rate | |
Canada | Floating exchange rate | |
Finland | Exchange rate peg to European Currency Unit (ECU) | |
New Zealand | Floating exchange rate | |
Spain | Exchange rate band within Exchange Rate Mechanism (ERM) of the European Monetary System | |
Sweden | Exchange rate peg to ECU | |
United Kingdom | Exchange rate band within ERM |
Exchange Rate Regime Prior to Adoption of Inflation Targeting
Emerging Market Countries | ||
Brazil | Crawling exchange rate band | |
Chile | Crawling exchange rate band | |
Czech Republic | Crawling exchange rate band supplemented by a monetary guideline and a disinflation objective | |
Israel | Crawling exchange rate band | |
Poland | Crawling exchange rate band | |
South Africa | Floating exchange rate with informal inflation targeting and announced guidelines for broad money growth | |
Industrial Countries | ||
Australia | Floating exchange rate | |
Canada | Floating exchange rate | |
Finland | Exchange rate peg to European Currency Unit (ECU) | |
New Zealand | Floating exchange rate | |
Spain | Exchange rate band within Exchange Rate Mechanism (ERM) of the European Monetary System | |
Sweden | Exchange rate peg to ECU | |
United Kingdom | Exchange rate band within ERM |
The experiences of Chile and Israel may offer some practical lessons on how to minimize the inherent conflicts between exchange rate and price stability for emerging market countries that want to take the slow track approach. Early on, after the announcement of the new inflation targets, the exchange rate could be viewed as entering independently into the objective function of policymakers owing to its importance for the real sector and for exchange rate stability in these open economies. Many of the features of Chile’s crawling band (width, rate of crawl, reference currency basket, degree of symmetry, and central parity level) were modified in response to changing policy objectives and market conditions (Landerretche and others, 1999; Ugolini, 1996). Over time, conflicts between the inflation target and the exchange rate band were usually solved in favor of the former. In mid-1995, the declining weight of the exchange rate in Israel was marked by the widening of the crawling band in response to appreciation pressures to 14 percent, to 28 percent in 1997, and then to 40 percent in 1999. Chile also continuously widened its exchange rate band and abandoned it in September 1999.
Both countries judiciously used foreign exchange intervention to keep the exchange rate within the pre-announced crawling band. Intervention by the Chilean central bank to keep the exchange rate within the band was frequent and intense. Similarly, heavy sterilized foreign exchange intervention was undertaken by the Bank of Israel to keep the Israeli currency (shekel) from appreciating outside its band in response to capital inflows.
Balancing Risks
The unfavorable economic developments that often motivate the shift to inflation targeting can pose risks for its credibility. Inflation targeting often has been introduced in the wake of exchange rate crises, high inflation, and poor overall economic performance (Table 4.5). In addition, four countries have been motivated by their prospective accession to European Economic and Monetary Union (EMU). Adopting inflation targeting before enough time has passed from the time of the motivating crisis or period of unfavorable economic developments risks a recurrence of these factors, which could compel monetary policy to subordinate the new inflation objective. This risk is especially important for emerging market countries because they are more prone to fiscal imbalances and financial and exchange rate crises (IMF, 1998b).
Motivations for Adopting Inflation Targeting
Instability in monetary aggregates provided a further motive.
Motivations for Adopting Inflation Targeting
Inflation Targeting Countries | Part of Response to Poor Overall Economic Performance | Exchange Rate Instability | Joining EMU | Anchor Inflationary Expectations | |
---|---|---|---|---|---|
Emerging Market Countries | |||||
Brazil | × | × | |||
Chile | × | × | |||
Czech Republic1 | × | × | |||
Israel | × | ||||
Poland | × | × | |||
South Africa | × | × | |||
Industrial Countries | |||||
Australia | × | × | |||
Canada | × | ||||
Finland | × | × | |||
New Zealand | × | × | |||
Spain | × | × | |||
Sweden | × | ||||
United Kingdom | × |
Instability in monetary aggregates provided a further motive.
Motivations for Adopting Inflation Targeting
Inflation Targeting Countries | Part of Response to Poor Overall Economic Performance | Exchange Rate Instability | Joining EMU | Anchor Inflationary Expectations | |
---|---|---|---|---|---|
Emerging Market Countries | |||||
Brazil | × | × | |||
Chile | × | × | |||
Czech Republic1 | × | × | |||
Israel | × | ||||
Poland | × | × | |||
South Africa | × | × | |||
Industrial Countries | |||||
Australia | × | × | |||
Canada | × | ||||
Finland | × | × | |||
New Zealand | × | × | |||
Spain | × | × | |||
Sweden | × | ||||
United Kingdom | × |
Instability in monetary aggregates provided a further motive.
A strong fiscal position is essential for the successful operation of inflation targeting. Monetization of government debt that could jeopardize the credibility of the inflation target is precluded by a sufficiently strong fiscal position (Kumhoff, 2000). A lack of fiscal discipline can also complicate monetary policy by undermining the capacity of the central bank to conduct effective monetary operations with government securities. Moreover, large fiscal imbalances may require the central bank to pursue a tighter-than-desired monetary policy stance to demonstrate its commitment to inflation control. Such a constraint on monetary policy was experienced by the Bank of Canada in the early 1990s, when the high public debt and deficit situation hindered its scope for easing monetary conditions (Clinton and Zelmer, 1997).
The fiscal position of emerging market countries generally strengthened ahead of their transition to inflation targeting, whereas that of industrial countries improved after its adoption.13 Net government debt to GDP declined prior to the beginning of the transition to inflation targeting for the emerging market countries, with the notable exception of Brazil. In contrast, for the industrial countries, government debt rose prior to inflation targeting and structural fiscal balances generally worsened. This discrepancy may reflect the need for a stronger signal on the part of emerging market countries that the new inflation targeting framework will not be threatened by a weak fiscal position, especially as emerging market countries often have higher levels of domestic debt (Appendix I).
Exchange rate stability is also essential for the successful operation of inflation targeting. In this context, exchange rate stability can be defined as a policy framework with an exchange rate value credible enough to convince markets that the inflation target will not be threatened by a currency crisis. The adoption of inflation targeting by most emerging market countries followed several years of sound policies and exchange rate stability. An exception with regard to exchange rate stability is Brazil, which in 1999 weathered pressures on its currency with international support. The European industrial countries adopted inflation targeting shortly after the crisis in the Exchange Rate Mechanism (ERM) of the European Monetary System, suggesting that markets felt that their exchange rates were close enough to equilibrium and policies sufficiently supportive for a credible inflation target.
Financial system stability, as gauged by a lack of financial sector crises for several years, preceded the adoption of inflation targeting by emerging market countries but not by all of the inflation targeting industrial countries. A threat to the financial system may undermine the credibility of an inflation target by raising the possibility that an inflationary liquidity injection might be needed to prop up ailing financial institutions and can limit the scope for a tightening of policy to stabilize the exchange rate. Finland, New Zealand, and Sweden experienced financial sector crises a few years prior to the introduction of inflation targeting (Caprio and Klingebiel, 1996). This pattern suggests that emerging market countries may feel obliged to wait longer after a financial crisis than do industrial countries to ensure that the new inflation targeting framework will not be threatened by financial system instability. If an exchange rate peg is being abandoned, measures might be needed to reduce the vulnerability of the financial sector to exchange rate depreciation. In addition, many of the emerging market countries have taken measures to deepen and broaden capital markets to reduce potentially destabilizing asset price volatility.
Long-Run Inflation Objective
The choice of the long-run inflation objective appears to depend on the extent of upward bias in the target price index and on the degree of rigidities in the economy. The optimal long-term inflation target for inflation targeting countries remains a matter of debate. Akerlof and others (1996) recommend a target rate of around 3 percent on the grounds that there is a significant positive bias in the measurement of consumer price inflation and a large degree of downward wage rigidity. This target also reflects the belief that it is important to allow real interest rates to become negative at troughs in the economic cycle.14 A lower long-run inflation rate could be appropriate if the measurement bias is small and wages are flexible, but there is a broad consensus that a zero target is too low. Thus, in practice the range of long-run inflation rates used in industrial countries (i.e., 0 to 3 percent) is not very wide.
Technical uncertainties about the appropriate long-run target rate of inflation might be more acute for an emerging market country than for an industrialized country. Upward bias in the consumer price data can be accentuated by relatively limited data collection and by the importance of quality enhancement (Clinton, 2000). In addition, the more rapid pace of structural change means that less is known about the productivity growth rate and the degree of wage flexibility that underlie the bias. Chile, the Czech Republic, Israel, and Poland have signaled their aim to achieve a long-run inflation rate similar to that of their trading partners.