This section addresses why the exchange rate plays a more important policy and operational role for emerging economies than for advanced economies. In short, the channels between the exchange rate and economic and financial performance tend to be larger and more uncertain for emerging economies. The following section models the stronger exchange rate channels, and the implications for implementation are addressed in Section V. This section is necessarily qualitative and draws on the research literature, the country and foreign exchange intervention case studies, and descriptive data.
Pass-Through from the Exchange Rate to Inflation
The pass-through from changes in the exchange rate to inflation is important for any relatively open flexible-exchange-rate economy. Pass-through operates directly through the effect of exchange rate movements on prices and indirectly through the impact of exchange rate movements on aggregate demand and prices. Generally, the empirical literature finds that pass-through from exchange rate changes into import prices is less than one (the law of one price does not hold) owing to a variety of factors, particularly transport costs, distribution costs, and price discrimination.
Pass-through has been on a trend decline around the world (Helbling, Jaumotte, and Sommer, 2006). Gagnon and Ihrig (2004) find a threefold decline in pass-through during the 1990s. The reasons for the decline include worldwide disinflation (Taylor, 2000; Choudhri and Hakura, 2001; and Gagnon and Ihrig, 2004), pricing to market, and credibility gains (Mishkin and Savastano, 2001, 2002; and Schmidt-Hebbel and Werner, 2002).
Pass-through is relatively important for emerging economies with shorter track records and other factors that lead to relatively low credibility (Frankel, Parsley, and Wei, 2005; and Ho and McCauley, 2003). Per capita income and distance from trading partners seem to be the most robust determinants of pass-through (Frankel, Parsley, and Wei, 2005). Other determinants include inflation and exchange rate volatility (Gagnon and Ihrig, 2004). Pass-through seems to decrease with the level of development and increase with openness, and emerging economies have higher and more variable inflation, experience greater exchange rate volatility, and are more dollarized (Honohan and Shi, 2003). A regime change to full-fledged inflation targeting has been found to reduce pass-through (Gagnon and Ihrig, 2004; Reyes, 2007; and Nogueira Júnior and León-Ledesma, 2008a, 2008b). This suggests that the emerging economies with other anchors have to deal with larger exchange rate pass-through.
Large, rapid, and uncertain pass-through can lead central banks to put weight directly on the exchange rate to reduce the level and volatility of inflation.5 A change in the exchange rate can rapidly raise inflation and inflation expectations, thus compelling the central bank to quickly take action to influence the exchange rate by changing interest rates or intervening in the foreign exchange market. However, these actions may give the impression that the central bank cares about the exchange rate above and beyond its impact on inflation. Pass-through is modeled explicitly in Section IV, including by simulating the exchange rate in the reaction function. Foreign exchange intervention is covered in Section V.
Output Stability
Many emerging economies aim to manage the exchange rate in order to mitigate the impact on output of relatively short-term (monthly or quarterly) exchange rate movements. A number of IMF staff reports on flexible-exchange-rate emerging economies find that intervention is used to limit exchange rate volatility, although the reasons for such intervention are not always fully articulated. Some empirical studies show that exchange rate volatility has a significant but small negative effect on trade.6
As noted, using the exchange rate to smooth output volatility can create confusion regarding the commitment to an inflation target or objective. Full-fledged inflation-targeting economies with long track records can be relatively transparent and find it easier to credibly explain to the markets the rationale for exchange-rate-smoothing interventions. In contrast, economies with less of a commitment to an inflation target and a shorter track record have a harder time intervening to smooth volatility in a way that is clear to the market. Conceptually, such intervention should be aimed at temporary exchange rate shocks, which raises the challenge of judging the duration of such shocks. The trade-offs posed by managing the exchange rate in order to smooth output volatility are modeled explicitly in Section IV.
Financial and External Stability
Promoting financial and external stability has long been a motivation for active management of the exchange rate, mainly to limit depreciation. In recent years, balance sheets with currency mismatches have been an important channel for large exchange rate depreciations (Allen and others, 2002). Governments are exposed to the extent that their foreign currency debt (foreign and domestic) exceeds their foreign reserves. Banks are exposed directly through their net asset positions and indirectly when borrowers themselves have balance sheet mismatches. Firms and households can have large, negative financial accelerator effects for the economy as a whole.
Emerging economies are typically more exposed to exchange rate fluctuations than advanced economies. The bulk of their international borrowing is denominated in foreign exchange. Currency crises can cause severe recessions (Kaminsky and Reinhart, 2000), especially if coupled with a banking crisis (Hutchison and Noy, 2005). The exchange rate policies of Asia were largely shaped by the crisis of 1997–98. In 2007, the National Bank of Kazakhstan undertook large-scale interventions which helped limit the depreciation prompted by worldwide financial developments.
Dollarization can exacerbate financial stability concerns arising from exchange rate depreciations. Dollarization can raise credit risks when banks lend in foreign currencies while borrowers’ salaries are in local currency. The relatively high level of dollarization of the emerging economies with other anchors may help explain why they have a more active exchange rate policy compared with the advanced economies and the inflation-targeting emerging economies (Table 3.1). In Peru, to prevent risks associated with dollarization, the authorities tend to avoid excess exchange rate volatility, particularly abrupt depreciation, which may have facilitated dedollarization.
Dollar Deposits, Percent of Total
(Selected Countries)
Dollar Deposits, Percent of Total
(Selected Countries)
| Inflation-targeting advanced economies | ||
| Median | 15.0 | |
| Standard deviation | 4.3 | |
| Inflation-targeting emerging economies | ||
| Median | 14.1 | |
| Standard deviation | 20.1 | |
| Flexible-exchange-rate, non-inflation-targeting emerging economies | ||
| Median | 44.6 | |
| Standard deviation | 26.2 | |
| Pegged-exchange-rate emerging economies | ||
| Median | 25.2 | |
| Standard deviation | 17.9 | |
Dollar Deposits, Percent of Total
(Selected Countries)
| Inflation-targeting advanced economies | ||
| Median | 15.0 | |
| Standard deviation | 4.3 | |
| Inflation-targeting emerging economies | ||
| Median | 14.1 | |
| Standard deviation | 20.1 | |
| Flexible-exchange-rate, non-inflation-targeting emerging economies | ||
| Median | 44.6 | |
| Standard deviation | 26.2 | |
| Pegged-exchange-rate emerging economies | ||
| Median | 25.2 | |
| Standard deviation | 17.9 | |
In some countries, systemic financial stress arising from an underdeveloped banking system affects exchange rate policy. Rapid credit growth in Romania and South Africa contributed to current account imbalances, exchange rate appreciation, and inflation pressures. Prudential risks and vulnerabilities in the banking system present challenges when moving toward a flexible-exchange-rate arrangement. There are risks posed by a lack of incentives to manage exchange rate risk, rapid credit growth, and rising exposure of unhedged borrowers. Balance sheet mismatches can lead central banks to have implicit thresholds below which they will not let the exchange rate depreciate. Extended intervention to support such a threshold may run down reserves and reduce the credibility of monetary and exchange rate policy in general.
The threat to external stability of a sudden stop of capital inflows is a special concern for many emerging economies. Many of those emerging economies that have been intervening during the past several years in the face of heavy capital inflows cite concerns that exchange rate overshooting could be followed by a sudden stop and a large contractionary depreciation (for example, Romania). A period of strong capital inflows may cause domestic booms in credit and demand (justifying a tighter monetary policy stance), but it may also cause an appreciation of the currency and a widening of the current account deficit that together make policymakers hesitant to tighten. Prolonged foreign exchange intervention to stabilize the exchange rate can lead the authorities to take on a large share of currency risk, encouraging further (excessive) capital inflows and increasing the risk of a sudden stop. Emerging economies have used blunter tools, such as reserve requirements and even capital controls, to control liquidity, but these are not consistent with inflation targeting and generally are not effective (Roger and Stone, 2005).
There are trade-offs between using exchange rate management to address financial and external stability concerns and using it to promote price and output stability. These trade-offs are modeled explicitly in Section IV.
Underdeveloped Financial Markets
Underdeveloped domestic financial markets reduce the scope for exchange rate flexibility by amplifying exchange rate shocks and constraining policy implementation. The clear differences in the levels of market development across the groups of countries probably help explain the different exchange rate policy approaches (Table 3.2; see also Table 2.1).
Market Development Indicators
(Selected Countries)
(Ratio to GDP)
Market Development Indicators
(Selected Countries)
(Ratio to GDP)
| Stock Market Turnover | Broad Money | Foreign Exchange Market Turnover | ||
|---|---|---|---|---|
| Inflation-targeting advanced economies | ||||
| Median | 79.3 | 85.8 | 1,022.7 | |
| Standard deviation | 52.4 | 37.9 | 597.2 | |
| Inflation-targeting emerging economies | ||||
| Median | 10.8 | 45.6 | 197.9 | |
| Standard deviation | 24.6 | 21.0 | 191.2 | |
| Flexible-exchange-rate non-inflation-targeting | ||||
| economies | ||||
| Median | 2.2 | 34.1 | 112.5 | |
| Standard deviation | 15.3 | 27.1 | 173.2 | |
| Pegged-exchange-rate emerging economies | ||||
| Median | 4.3 | 42.3 | 56.4 | |
| Standard deviation | 56.0 | 51.5 | 17.5 | |
Market Development Indicators
(Selected Countries)
(Ratio to GDP)
| Stock Market Turnover | Broad Money | Foreign Exchange Market Turnover | ||
|---|---|---|---|---|
| Inflation-targeting advanced economies | ||||
| Median | 79.3 | 85.8 | 1,022.7 | |
| Standard deviation | 52.4 | 37.9 | 597.2 | |
| Inflation-targeting emerging economies | ||||
| Median | 10.8 | 45.6 | 197.9 | |
| Standard deviation | 24.6 | 21.0 | 191.2 | |
| Flexible-exchange-rate non-inflation-targeting | ||||
| economies | ||||
| Median | 2.2 | 34.1 | 112.5 | |
| Standard deviation | 15.3 | 27.1 | 173.2 | |
| Pegged-exchange-rate emerging economies | ||||
| Median | 4.3 | 42.3 | 56.4 | |
| Standard deviation | 56.0 | 51.5 | 17.5 | |
Underdeveloped financial markets can lead to more active management of the exchange rate to promote price, output, and financial stability. Uncompetitive markets with thin volume and limited or no derivatives instruments are less able to absorb shocks without wide exchange rate fluctuations. The lack of instruments for managing the exchange rate risk leaves foreign exchange users vulnerable to wider exchange rate movements. These considerations can compel the authorities to choose a more rigid exchange rate arrangement. Extended intervention to stabilize the exchange rate can remove incentives for the private sector to develop currency risk-management tools.
The case studies suggest that a more developed foreign exchange market reduces the need for foreign exchange intervention, provides hedging instruments, and facilitates the signaling channel. In Chile, the foreign exchange market offers a developed market for hedging through foreign exchange forwards and is deep enough to smooth exchange rate volatility, thus alleviating the burden on monetary policy and allowing infrequent and transparent foreign exchange intervention. In Colombia, the developed foreign exchange market allows a sophisticated and transparent approach to foreign exchange intervention. In New Zealand, the very deep foreign exchange market and the central bank’s stringent intervention criteria help limit the need to intervene. In Serbia, the role of the central bank in the foreign exchange market remains important, and intervention objectives are focused on supporting the transition to a fully developed interbank market. In Turkey, the developed foreign exchange market helps the central bank participate only in response to extraordinary events because the availability of risk-management instruments helps market participants manage exchange rate volatility in the interbank market.
Developed money markets and government security markets also provide more policy options. Weak interest rate transmission from underdeveloped money markets can compel a leading policy role for the exchange rate. Furthermore, underdeveloped money and security markets can raise the costs of sterilization and result in large liquidity creation from capital inflows. Finally, the absence of developed money markets can inhibit the adoption of inflation targeting under which a short-term interest rate is used as the operating target.
The implications of underdeveloped financial markets for policy implementation are elaborated in Section VI, and the case studies and examples of good implementation practices are documented in Sections IX and X.
Credibility
A reasonably sized and developed country with a supportive set of economic and structural policies allows for a credible commitment to an inflation target and less reliance on managing the exchange rate. A large dose of credibility is needed for an emerging economy to reap the benefits of a full-fledged inflation-targeting nominal anchor, which frees the exchange rate to float and also facilitates policy implementation. Furthermore, economies with flexible exchange rates that have yet to adopt explicit inflation targets can be considered in transition to a single nominal anchor, and completing this transition requires establishing the groundwork for a credible commitment to the inflation target.
To better understand the differences in credibility across economies with different monetary and exchange rate regimes, credibility is crudely proxied here by the actual inflation outturn and by market ratings of long-term local-currency-denominated government debt. The inflation-targeting economies have much better inflation outturns (Table 3.3). Low inflation signals that a central bank can make a credible commitment to an inflation target. Furthermore, low and positive inflation is supportive of high and stable long-term growth (see, for example, Sarel, 1996), and a monetary policy supportive of long-term growth can be more credible. The lowest inflation rates in recent years have been in the inflation-targeting advanced economies, followed by the inflation-targeting emerging economies; the emerging economies with other anchors have had the highest inflation rates. The inflation-targeting economies have higher ratings of long-term local-currency-denominated government debt.7 This gauge is forward looking and directly captures market perceptions of the degree of long-term market confidence in the stability of a currency, which ultimately is the responsibility of the central bank even though it reflects factors beyond the scope of monetary policy. The inflation-targeting advanced economies have the highest ratings, followed by the inflation-targeting emerging economies and the emerging economies with other anchors. These indicators of credibility, while rough, suggest that higher credibility is associated with a smaller role for the exchange rate.
Indicators of Credibility
(Selected Countries)
Indicators of Credibility
(Selected Countries)
| Average Consumer Price Index Inflation | Standard & Poor’s Rating of Long-Term Local-Currency- Denominated Government Debt | ||||
|---|---|---|---|---|---|
| 1997–2007 | 2002–07 | 2005–07 | |||
| Inflation-targeting advanced economies (11) | |||||
| Median | 2.3 | 1.9 | 1.8 | AAA | |
| Standard deviation | 1.1 | 1.1 | 1.3 | ||
| Inflation-targeting emerging economies (16) | |||||
| Median | 6.6 | 5.1 | 4.9 | BBB+ | |
| Standard deviation | 11.6 | 4.3 | 2.5 | ||
| Non-inflation-targeting emerging economies (15) | |||||
| Median | 9.9 | 11.0 | 9.4 | BB+ | |
| Standard deviation | 29.6 | 10.6 | 4.5 | ||
| Pegged-exchange-rate emerging economies (17) | |||||
| Median | 3.7 | 5.6 | 7.5 | BBB | |
| Standard deviation | 18.6 | 6.4 | 4.2 | ||
Indicators of Credibility
(Selected Countries)
| Average Consumer Price Index Inflation | Standard & Poor’s Rating of Long-Term Local-Currency- Denominated Government Debt | ||||
|---|---|---|---|---|---|
| 1997–2007 | 2002–07 | 2005–07 | |||
| Inflation-targeting advanced economies (11) | |||||
| Median | 2.3 | 1.9 | 1.8 | AAA | |
| Standard deviation | 1.1 | 1.1 | 1.3 | ||
| Inflation-targeting emerging economies (16) | |||||
| Median | 6.6 | 5.1 | 4.9 | BBB+ | |
| Standard deviation | 11.6 | 4.3 | 2.5 | ||
| Non-inflation-targeting emerging economies (15) | |||||
| Median | 9.9 | 11.0 | 9.4 | BB+ | |
| Standard deviation | 29.6 | 10.6 | 4.5 | ||
| Pegged-exchange-rate emerging economies (17) | |||||
| Median | 3.7 | 5.6 | 7.5 | BBB | |
| Standard deviation | 18.6 | 6.4 | 4.2 | ||
The underlying elements of credibility for inflation targeting have been extensively examined and are not directly addressed in this paper.8 All inflation-targeting economies are fairly large and developed, which suggests that inflation targeting requires size and a somewhat advanced economy. An inflation-targeting central bank needs a mandate to pursue the inflation target and sufficient discretion and autonomy to set its monetary instruments accordingly. A strong fiscal position is essential: it is not good enough that monetary policy not be dominated by fiscal priorities, because even suboptimal policies can hurt credibility in a country with high debt and a short history of sound fiscal management (Blanchard, 2005; and Schabert and Van Wijnbergen, 2006).
Other Exchange Rate Policy Channels
This paper does not address the issue of managing the exchange rate to maintain export competitiveness. Recent IMF staff reports have discussed such intervention by emerging economies with flexible exchange rates in order to limit exchange rate appreciation for reasons including competitiveness (Argentina, Azerbaijan, Indonesia, Kazakhstan, Korea, Russia, Tunisia, Ukraine, and Uruguay). In Hungary, the center of the exchange rate band was adjusted as part of other government economic policy measures to support exporters. In general, competitiveness is driven by the real exchange rate which, over the medium and long term, is beyond the control of monetary and exchange rate policy. Of course, there is a thin line between, on one hand, intervention to smooth the impact of monthly or quarterly exchange rate changes on the relative price of exports and on output volatility and, on the other, intervention to maintain competitiveness over the long term. In most circumstances, a depreciated nominal exchange rate will feed into higher inflation and a higher real exchange rate, ultimately undermining competitiveness.
Nor does this paper consider new rationales for maintaining a weak currency that are based on positive growth structural channels. Johnson, Ostry, and Subramanian (2007) find that avoiding prolonged periods of exchange rate overvaluation helps sustain growth, although this may have more to do with the degree of capital account openness and structural policies than with monetary policy. Rodrik (2007) asserts that there is a systematic relationship between growth and undervaluation for emerging and developing economies. In these economies, tradables are different as a result of market failures (information and coordination externalities) and the impact on them of institutional weakness and contracting incompleteness. Levy-Yeyati and Sturzenegger (2007) find that in recent years exchange rate intervention has been asymmetrically aimed at stemming appreciation and that depreciation benefits economic growth not through the traditional channel of increasing net exports but rather by boosting domestic savings and capital accumulation. These channels involve changes in the equilibrium steady state of the economy, which are not modeled in this paper.
During 2007–08, exchange rate policy played some role in containing inflation pressure in a number of inflation-targeting emerging economies, particularly where there was strong exchange rate pass-through to inflation, as discussed in Section VII.
Theoretical work on the impact of exchange rate volatility on trade is ambiguous, reflecting that the nature of shocks that cause exchange rate changes can lead to changes in other macroeconomic variables that offset the impact of the movement in the exchange rate (Clark and others, 2004; Koren and Szeidl, 2003). Aghion and others (2007) find that real exchange rate volatility reduces long-term productivity growth for countries with undeveloped financial markets but has a limited effect on financially developed countries.
It would be preferable to use market-based measures of central bank credibility, but such measures are available for only a few economies. In addition, comparisons of actual versus targeted inflation are precluded by the absence of firm quantitative targets for many economies. Indicators of the stability of inflation expectations in the face of inflationary shocks would be another good gauge of credibility, but cross-country data are not available.