Chapter

Chapter 3 A Taxonomy of Intervention

Author(s):
Marcos Chamon, David Hofman, Nicolas Magud, and Alejandro Werner
Published Date:
February 2019
Share
  • ShareShare
Show Summary Details
Author(s)
Oscar A. Hendrick, Nicolás E. Magud and Asad Qureshi 

The opinions expressed in this chapter are the sole responsibility of the authors.

This chapter presents different frameworks for implementing foreign exchange interventions, including whether they are transparently communicated to the market or kept secret, and whether they are rules-based or discretionary. The chapter also delves into the motives for the various types of foreign exchange intervention, including whether interventions take place in the spot market or in the derivatives market—and the rationales for each type. The chapter also summarizes recent interventions in some Latin American countries.

Objectives and Transparency

The issue of transparency in foreign exchange intervention has been the subject of much debate among policymakers and economists. A review of the literature shows that secret intervention by central banks has been supported by several authors. For instance, Vitale (1999) argued that secret sterilized intervention can be used to influence agents’ expectations and exchange rates, given the access to private information on these fundamentals. Sarno and Taylor (2001) found evidence of the effectiveness of secret official intervention, through either the portfolio channel or the signaling channel. They also argued that coordination among central bankers, and some degree of transparency may enhance the foreign exchange intervention. The empirical work undertaken by Dominguez and Frankel (1993), with the use of data on intervention and exchange rate expectations, was instrumental to overcome two major handicaps characterizing the empirical studies of the 1980s, which largely rejected the effectiveness of intervention. Based on the comprehensive IMF’s 2001 Survey of Foreign Exchange Market Organization, Canales-Kriljenko (2003) found that on some occasions the central bank would benefit from keeping its foreign exchange intervention secret. In these cases, the informational advantage to the central bank may protect it to some degree from speculative attacks and falling into speculative trading games from large traders in the market. On the other hand, in the context of a meeting of deputy governors of central banks from major emerging market economies to discuss foreign exchange intervention, Archer (2005) found that many policymakers are more in favor of transparency regarding the intervention ex ante, and transparency about actual intervention operations ex post. However, secret interventions were still supported by some, under the rationale that the market has no target to attack, or to avoid the perception that the central bank has failed. More recently, central banks with full-fledged inflation-targeting frameworks and with strong credibility rely more on transparency ex ante and ex post regarding the frequency and amount of intervention.

Yet, even when transparency is chosen, several issues surround its degree and type. It is important to distinguish between policy transparency and operational transparency. In the first case, central banks can decide to disclose the rules of foreign exchange intervention on a general level (such as smoothing out excess volatility), or on a specific level (such as the triggers for intervention and how they work). Regardless of the level of transparency, some central banks prefer not to disclose trading tactics, because, in some cases, those tactics can give some market participants an undue advantage to bet against the central bank and undermine the objectives of the foreign exchange intervention.1

It can be argued that “tactical ambiguity” about the exact timing and amount of intervention will heighten prospects for achieving the intervention objective efficiently, that is, with the least amount of intervention. Transparency can be in “real time,” when the central bank explains its actions as they happen, ex ante (before they happen), or ex post (after they happen) (Enoch 1998).

A central bank’s credibility is also relevant when deciding the level of transparency. On the one hand, some experts suggest that if a central bank is credible, and market participants understand its underlying rationale for intervention, transparency could be reduced. On the other hand, a central bank with strong credibility may want to reveal its actions so that the market can benefit from the central bank’s signaling effect.

Transparency policy (or the degree of its transparency) may also vary with the specific objectives of intervention, the tools available to the central bank, the number of players in the foreign exchange market, and the depth and liquidity of money and capital markets in the country. In some cases, market participants may speculate against a central bank in which operations are bound by excessively strict rules. Ex post transparency could also be effective if the central bank’s signaling, after the intervention, influences market expectations by transmitting information on the fundamentals or on future policy actions.

Transparency is also related to the motives of intervention (see Chapter 2). As argued in previous chapters, spot interventions help address liquidity imbalances in the foreign exchange market: buying if there is excessive foreign exchange in the market (such as owing to capital inflows) or selling if there are sudden capital outflows or seasonal liquidity shortages of foreign exchange. Against a foreign exchange hedging demand, swap interventions could be more useful for easing the hedging needs of participants and avoiding excessive and unnecessary pressure on the spot market. This point is subsequently elaborated.

The transparency of intervention practices varies across countries. For example, Group of Three (G3) countries began enhancing intervention transparency in the mid-1990s. The Federal Reserve started to report its intervention activity on its webpage quarterly, and it released daily intervention figures with a one-year lag. Hung (1997) estimates that about 40 percent of the Federal Reserve’s foreign exchange interventions during 1985–89 were not announced. The Bank of Japan did not announce its interventions either, but it reported the amounts of exchange rates ex post. The European Central Bank (ECB) announced some of its interventions, although the information contained in the announcements was limited and did not include amounts and timing (Canales-Kriljenko, Karacadag, and Guimarães 2003). In emerging markets, according to a survey on intervention practices in 90 countries, about half of the central banks in these economies announce their presence in the market, while the evidence on how much central bank practices are secret is mixed (Canales-Kriljenko 2003).

In Latin America, some countries have had episodes of secret ex ante foreign exchange interventions in the past, although in most cases, the intervention was made public ex post and its rationale explained. Recently, a move to more transparency and more to rules than discretion, has been observed, in tandem with central banks’ shift from monetary aggregate anchors to full-fledged inflation-targeting frameworks. Also, the degree of intervention has been reduced or eliminated altogether in some countries. Yet, during the global financial crisis of 2008–09, some monetary authorities reinstated foreign exchange intervention as a temporary recourse, either to rebuild the level of international reserves, as in Chile in 2011, or to smooth out volatility, like Colombia did in 2012. Some countries, such as Peru, have continued to use discretion over rules (see Chapter 12).

Debate is ongoing about the costs and benefits of either strategy, but empirical observation, as illustrated in the country chapters, makes it clear that there is no one rule that fits all. Yet, a case can be made that transparency may be the best approach to foreign exchange intervention, as Chapter 6 discusses.

A few central banks disclose or publish data pertaining to foreign exchange interventions. Some provide initial guidance close to intervention time, while others publish the data with a lag. Rare interveners, such as the Bank of Canada or the Bank of England, disclose volumes close to the intervention time; others prefer to publish intervention volumes with a lag spanning one to six months. For example, Japan discloses daily volumes with a six-month lag, while Australia discloses monthly volumes with a one-month lag. Brazil used to publish intervention data with a one-week lag but now prefers to release market operations data through monthly press releases on its website. The Hong Kong Special Administrative Region and Paraguay publish foreign exchange intervention volumes the same day. In general, most central banks in Asia and Africa do not publicly disclose intervention data. Table 3.1, from the IMF’s Annual Report on Exchange Arrangements and Exchange Restrictions, reveals several central banks’ disclosure practices.

Table 3.1.Current Disclosure Practice: AREAER Survey Results 2016
Economy or RegionImmediate Disclosure and Other Disclosure Channels and Time Lags
ArmeniaThe Central Bank of Armenia publishes intervention data on its website weekly on Mondays.
AustraliaMonthly purchases and sales of foreign exchange are published on the Reserve Bank of Australia website with a one-month lag. Daily data on foreign exchange market interventions are published on the website annually, with the release of the central bank’s annual report.
AzerbaijanThe Central Bank of Azerbaijan publishes foreign exchange intervention data quarterly.
BoliviaThe Central Bank of Bolivia publishes exchange market intervention and foreign exchange purchases and sales data in “Weekly Statistics” on its website.
BrazilThe Central Bank of Brazil publishes intervention data monthly on its website on currency flows and open market operations.
CanadaInterventions are announced on the Bank of Canada’s website, and the amount of intervention is published in the government’s monthly official press release on international reserves.
ChileThe Central Bank of Chile usually announces the amount of foreign exchange it intends to buy or sell in its monthly bulletin. It publishes auction results daily and reserves weekly.
ColombiaThe central bank of Colombia, the Banco de la República, publishes daily and monthly intervention results in the foreign exchange market and information regarding each auction on its website.
European UnionThe European Central Bank publishes information on interventions. When it intervenes, the ECB intervenes at the market prices’ quotes.
GuatemalaThe Bank of Guatemala, effective November 2012, publishes the foreign exchange intervention data on its website.
Hong Kong SARThe Hong Kong Monetary Authority makes immediate announcements of the impact on the aggregate balance from purchases or sales of US dollars through various information outlets, including Reuters and Bloomberg Finance L.P., and on its website.
IcelandThe Central Bank of Iceland publishes monthly data on foreign exchange interventions in the foreign exchange market.
IndiaThe Reserve Bank of India publishes monthly data on its interventions (foreign exchange purchases and sales) in its monthly bulletin, with a six-week lag.
IndonesiaBank Indonesia does not disclose data on its interventions.
JamaicaThe Bank of Jamaica publishes information on its interventions in its Annual Report and Quarterly Monetary Policy Report.
JapanInterventions in Japan fall within the mandate of the ministry of finance, which publishes daily purchases and sales amounts as well as intervention currencies on its website with a lag.
KoreaKorean interventions in the spot market or through derivatives in the forward market are not announced, and intervention data are not published; there are no regular channels. Figures on foreign exchange reserves are published twice a month, but their movements are affected by several factors, in addition to foreign exchange interventions.
MexicoThe Bank of Mexico publishes results of all foreign exchange interventions on its website, including the new foreign exchange hedging program of nondeliverable forwards announced by the Foreign Exchange Commission on February 21, 2017. All results of past rules-based and regular auctions are also publicly available.
PeruThe Central Reserve Bank of Peru publishes daily information about its intervention operations on its website.
Russian FederationThe Bank of Russia publishes information about the frequency and volume of foreign currency interventions on its website under the “Liquidity of the banking sector and monetary policy instruments” subsection of the “Statistics” section.
SingaporeThe Monetary Authority of Singapore intervenes through agents and does not publish information on its interventions.
SwedenSveriges Riksbank announces each intervention in a press communiqué, explains the motive, and discloses the framework of intervention and amount data.
TurkeyThe Central Bank of the Republic of Turkey publishes results of auctions and the volume of its direct interventions on its website with a lag.
UgandaThe Bank of Uganda publishes information on interventions in its monthly, quarterly, and annual reports, including breakdowns for reserve buildup, interventions, and targeted transactions.
United KingdomThe Bank of England has a separate pool of foreign exchange reserves, which it uses at its discretion to intervene in support of its monetary policy objectives. A monthly press release issued by Her Majesty’s Treasury reports the treasury and central bank interventions.
United StatesThe Federal Reserve Bank of New York acts as the operating arm of the Federal Reserve System. Interventions are announced when they occur, and the size of the interventions is reported in the Federal Reserve Foreign Exchange Operations quarterly bulletin.
Source: IMF AREAER 2016.
Source: IMF AREAER 2016.

It is important to note that a central bank’s credibility in its commitment to the inflation target is crucial for all inflation-targeting countries. Clear rules of engagement for foreign exchange intervention, and clear messages that the main objective is the inflation target, and not a specific exchange rate level, are instrumental for building and maintaining a central bank’s reputation.

Rules Versus Discretion

Regardless of how transparent it is, foreign exchange intervention can be rules-based or discretionary. In some cases, monetary authorities clearly establish the conditions for an intervention to take place. Some rules are specific and state the amount or the nature of the purchases or sales of foreign currency. Some even clarify the objectives for such interventions. Those objectives could include mitigating exchange rate volatility to rein in financial stability, alleviating hedging needs for exporters, or supplying cash for those needing to fulfill external debt obligations or for imports. It can also be related to competitiveness in a growth-led strategy. In other cases, the central bank follows no rules. When perceptions of problems in the foreign exchange market arise—including financial stability, exchange rate level, liquidity issues, or others—the central bank decides to arbitrarily intervene. That decision is an example of discretionary foreign exchange intervention. Discretionary interventions are not only ad hoc about what triggers the operation but also about the amount of sales and purchases, and the modalities of the sales and purchases. By definition, discretionary foreign exchange interventions are much less predictable than rules-based interventions.

Beyond preestablished policy response functions based on specific rules, central banks can announce programs of purchases or sales of foreign currency. For example, central banks can aim for certain levels of international reserves as a share of GDP, of imports, or of short-term debt obligations. Typically, the goal is to be on par with countries of similar characteristics. The goal could also be to preempt exchange rate levels perceived as persistently deviating from fundamental values as a way to minimize resource misallocations that could be detrimental to economic activity.

Regardless of the ultimate objective, central banks preannounce programs for purchasing or selling foreign exchange for a preestablished period. Typically, the program also specifies regularly timed auctions of foreign exchange as well as the volume in each auction (often at a constant rate). This mechanism allows the central bank to predictably convey the message to the market of an intervention to limit any disruption. It may also implicitly point to the expected path of monetary policy that is consistent with the foreign exchange intervention in some cases (Mussa 1981)—thus also revealing the central bank’s projections to the market and, thereby, its price stability. In Latin America recently, Chile aimed to build reserves and mitigate the effect of a persistent appreciation in 2011, and Mexico implemented a program for purchasing foreign exchange in 2017 to counterbalance the instability arising out of the North American Free Trade Agreement renegotiations.

In theory, the discussion between rules and discretion for monetary policy is well established. Going back as early as Kydland and Prescott (1977) and Barro and Gordon (1983), monetary policy debates have focused on problems of time inconsistency. In that literature, when discretionarily choosing monetary policy to achieve an inflation target jointly with an output gap target, the welfare-optimizing equilibrium results in an inflation bias, owing to a conflict of interest (Drazen 2003). In Calvo (1978), the central bank faces the problem that the optimal inflation target today may not be the optimal target come next period for maximizing some fiscal objective. Agents internalize this. Yet, the outcome is not first best. The central bank can then have rules that solve the maximization problem and thus avoid the cost of discretionary policy. However, rules also involve economic costs. The trade-offs related to each of the mentioned costs result in the optimality of rules or discretion. Rogoff (1985) extends this literature, showing that having a more hawkish central banker reduces the costs associated with the time-inconsistency problem.

For foreign exchange intervention, the same logic need not necessarily apply. Rather, rules give the authorities predictability instead of tying the central bank’s hands. Presumably, such predictability would reduce financial instability. For example, Montoro and Ortiz (2016) show that in a general equilibrium model, the amount of foreign exchange intervention needed to stabilize the exchange rate under rules is much smaller than under discretion. However, rules also limit the ability to respond differently when needed; in some circumstances, discretion to intervene in foreign exchange markets could be more effective precisely because of its lack of predictability. Not surprisingly, then, it is observed that in Latin America and in many other regions, central banks sometimes prefer rules-based foreign exchange intervention, but many other times they choose to stick to discretionary policies for foreign exchange intervention.

Moreover, for inflation-targeting countries, foreign exchange intervention needs to be clearly subordinated to the inflation target, the main objective of the central bank. Otherwise, it could trigger another variety of the time-consistency problem raised earlier; for example, if the central bank is perceived to pursue an exchange rate objective rather than an inflation target objective. For inflation-targeting central banks, then, it is key that the market understands that regardless of whether foreign exchange intervention aims at mitigating the exchange rate pass-through to domestic prices of rapid and large changes in the exchange rate or at reducing the effects of financial stability resulting from excessive exchange rate volatility, the ultimate goal of intervention is to keep inflation in check. The connection between foreign exchange intervention and the inflation target need not be direct, however. An indirect channel would be financial stability concerns resulting from currency mismatches that could lead to higher inflation if they triggered financial instability. In any case, foreign exchange intervention should always be subordinated to achieving the inflation target.

The central bank can announce the general rules of intervention without specifically indicating limits on the amounts and frequency of interventions (see the earlier discussion on transparency of intervention). Rules of intervention could be very specific; yet they must be designed to preclude the main players in the foreign exchange market from taking unfair advantage of their position in the market. In general, inflation-targeting central banks would be better served by announcing the rules of engagement for foreign exchange intervention. These rules could be specific or more qualitative, such as leaning against the wind or reducing excess exchange rate volatility to limit the negative effects of large exchange rate fluctuations not supported by changes in the fundamentals.

Connected to these rules, and the transparency of interventions raised in the first section, are interventions related to paragovernmental institutions. For example, in Mexico, the cash flows of the state-run oil company PEMEX demand and supply substantial amounts of US dollars each year because of the company’s crude oil exports and gasoline and other petroleum-related imports. Low-capacity utilization resulted in a negative balance contribution to international reserve accumulation in the first half of 2017. By law, PEMEX must sell to the central bank all foreign exchange that results from exporting crude oil. However, when the proceeds from crude exports are not enough, or when foreign debt payments are due, PEMEX buys US dollars from the central bank. Given the size of PEMEX—which until 2017 had been the largest contributor to international reserve accumulation—its purchases and sales of foreign currency need to be properly coordinated with the central bank. In the past, especially before the 1990s, it was common for Latin American state-owned firms to borrow abroad. Lack of coordinated sales and purchases of these government agencies worked in practice as very volatile and unpredictable foreign exchange interventions and increased financial instability. In Chile, the state-owned copper company Codelco distributes part of its sales to the government. Whether that foreign exchange ends up in the government’s account in the central bank or a commercial bank (in particular, the state-owned bank Banco Estado), is not clear. Notwithstanding that, Codelco also purchases and sells foreign exchange in the market—including foreign exchange hedging—thus affecting market exchange rates.

The frequency of exchange rate interventions depends partly on the nature of the interventions. Preannounced programs are the most predictable, not only in volume, but also in frequency. Rules-based intervention can be anticipated by the market. Although frequency cannot be perfectly estimated, market conditions on the back of a transparent rule enable the anticipation of when thresholds could trigger an intervention. Anticipating discretionary intervention is, by definition, more difficult. Thus, its frequency is less homogenous over time.

In Latin America, foreign exchange intervention has been varied. Different countries have resorted to rules and discretion. Moreover, several countries have switched from rules to discretion and back over time. For example, as mentioned earlier, Chile intervened in 2011 with a preannounced program to purchase international reserves to match similar countries’ reserves-to-GDP ratios (see Chapter 8). Colombia and Mexico have, at times, used rules that specified that when the daily volatility of the exchange rate over a specific number of days’ moving average exceeded a preestablished threshold (20 days), intervention was triggered. Colombia discontinued this rule in May 2016 (Chapter 9). Mexico stopped that program in February 2016 (Chapter 11).

At other times, intervention in Colombia and Mexico has been more ad hoc, that is, discretionary. Brazil also implemented rules-based interventions for some time in response to the so-called 2013 taper tantrum (Chapter 7). This involved daily auctions of $3 billion per week from August 2013 focused on preannounced swap and repo operations. The program was originally scheduled to stop at the end of 2013, but it was extended several times, before ending in March 2015. More recently, Mexico introduced a program of up to $20 billion of nondeliverable forwards (NDFs) settled in pesos in late February 2017; and $1 billion short-dollar contracts were auctioned in March 2017. In October of the same year, this program increased to $5 billion. The additional $4 billion was allocated as follows: $1 billion the day after the announcement (October 25, 2017), with maturities of one month ($400 million), two months ($300 million), and three months ($300 million), followed by weekly auctions of $500 million every Wednesday until early December 2017.

In which Market—Spot or Derivatives?

As Chapter 2 notes, foreign exchange intervention typically has several objectives, which may include price stability, financial stability, buffer building, and in a few instances, market development. To achieve these objectives, authorities consider foreign exchange intervention complementary to interest rate policy to contain inflationary pressures from exchange rate pass-through and preserve financial stability by mitigating risks from currency mismatches. It therefore supports economic growth in episodes of financial instability, at least indirectly. Accumulating international reserves to build stronger buffers that can respond to external shocks is typically also referred to as a motive for intervention.

Several factors drive a central bank’s decision to intervene to achieve its intended objectives. These then lead to the modalities and instruments used to achieve the desired results. Inflation-targeting central banks in Latin America predominantly use foreign exchange intervention to smooth out excessive exchange rate volatility and to create strong external buffers against unexpected shocks by accumulating a high level of international reserves. Regardless of the motives within the region, a key question is which market and instruments are most effective in attaining the objectives. Table 3.2 lists some key motives.

Table 3.2.Importance of Various Motives for Intervention, 2005–06 and 2011–12
Importance in 2005–06Importance in 2011–12
MotiveHighModerateLowHighModerateLow
To curb excessive exchange market speculation8401140
To maintain monetary stability7221022
To discourage sharp capital inflows or outflows431551
To build or reduce foreign exchange reserves702622
To smooth the impact of commodity price fluctuations313413
To maintain or enhance competitiveness223413
To alleviate foreign exchange funding shortages of banks and corporations420520
Source: Bank for International Settlements Questionnaire, February 2013.Note: The data show the number of central banks, out of 19 that were surveyed, that responded to the importance of various motives, rating on a scale of 1 (most important) to 7 (least important); “high” indicates a response of 1 or 2; “moderate” indicates a response of 3 through 5, and “low” indicates a response of 6 or 7.
Source: Bank for International Settlements Questionnaire, February 2013.Note: The data show the number of central banks, out of 19 that were surveyed, that responded to the importance of various motives, rating on a scale of 1 (most important) to 7 (least important); “high” indicates a response of 1 or 2; “moderate” indicates a response of 3 through 5, and “low” indicates a response of 6 or 7.

This section aims to present a taxonomy of intervention practices in Latin America. Specifically, it highlights practices typically used by countries in the region that are conditional on the type of shock they are trying to mitigate.

Sterilized intervention in spot markets is one instrument at central banks’ disposal. It remains the market choice, but by no means the only one. Latin American central banks use a diverse range of instruments to intervene in the foreign exchange market. These include forwards, swaps, repos, NDFs and options, as well as US dollar–linked debt. Most central banks in the region have used both spot and derivatives markets extensively. Table 3.3 presents some of these instruments, as well as the mechanism through which each operates.

Table 3.3.Widely Used Foreign Exchange Intervention Instruments
InstrumentMechanism of Central Bank
Foreign exchange spot transactionsBuys and sells foreign exchange spot
Foreign exchange forwardsBuys and sells foreign exchange at an agreed rate and date in the future
Foreign exchange swaps or reposBuys and sells foreign exchange spot and purchases foreign exchange forwards on a predetermined date
Forwards, nondeliverable forwards, futuresPays domestic currency equivalent of change in foreign exchange value on a predetermined date
Foreign exchange optionsSells option to buy foreign exchange if the currency exceeds threshold
Source: IMF staff.
Source: IMF staff.

The rationales that central banks give for experimenting with a wide variety of instruments besides the spot market include the structure and growing sophistication of the markets and the multiple objectives central banks are trying to achieve. In addition to the level of international reserves, central bank foreign exchange interventions in the spot market affect interest rates in monetary and capital markets. Hence, a few of the region’s central banks have introduced instruments to intervene in the foreign exchange market that would reduce not only the pressure on the spot foreign exchange rate but also lessen the distortions of foreign exchange forward market transactions on interest rates in money and fixed income markets, and at the same time, prevent higher hedging costs. Table 3.4 summarizes the various instruments used in Latin America. These practices are then elaborated, and their utilization and frequency are explained.

Table 3.4.Foreign Exchange Instruments of the Latin American Central Banks
Modality of Foreign Exchange Instruments
InstrumentBilateralAuctionWindow2TenorsOther Characteristics
Foreign exchange spotToday/tomorrow/spotDiscretion/rules-based
Foreign exchange forwards/nondeliverable forwardsStandard tenorsDiscretion/rules-based Exchange traded
Foreign exchange swaps1Standard tenorsBilateral/discretionary Auctions/rules-based Window/rules-based
Foreign exchange reposStandard tenorsIrregular/discretionary
Foreign exchange optionsMostly in one monthDiscretion/rules-based
OthersDiscretion/rules-based
Source: IMF staff.

Usage includes for hedging with underlying exposure.

Window operations include exchange-traded instruments.

Source: IMF staff.

Usage includes for hedging with underlying exposure.

Window operations include exchange-traded instruments.

As noted, some countries use a rules-based foreign exchange intervention framework and others use a discretionary-based one. In some instances, countries have even switched from one to the other. The choice of instruments in Table 3.4 in most instances is dictated by the objective and depth of the market, while the choice of foreign exchange intervention framework is mostly influenced by the frequency of interventions.

Regardless of the choice of framework or instrument, most foreign exchange interventions in Latin America have focused on financial stability or on building buffers. In some episodes of large depreciation pressures, price stability has been the objective. Table 3.5 shows the variety of frameworks and instruments Latin American central banks use.

Table 3.5.Foreign Exchange Intervention Framework and Instrumens, by Country
Foreign Exchange Intervention FrameworkMain Instruments of Foreign Exchange Intervention
CountryRulesDiscretionSpotSwapsNondeliverable Forwards1OptionsOthers2
Brazil
Chile
Colombia
Costa Rica
Mexico
Paraguay
Peru
Uruguay
Sources: Central banks; and IMF staff.

Nondeliverable forwards are settled in local currency; they are also referred to as currency swaps.

“Others” includes instruments such as repo and certificates of deposit linked to the exchange rate.

Sources: Central banks; and IMF staff.

Nondeliverable forwards are settled in local currency; they are also referred to as currency swaps.

“Others” includes instruments such as repo and certificates of deposit linked to the exchange rate.

Brief descriptions of country practices follow, with further details in Chapters 7 through 13. See Figure 3.1 for the interventions by instrument and exchange rate.

Figure 3.1.Foreign Exchange Intervention in Latin American Countries, by Instrument and Exchange Rate, 2006–17

(Billions of US dollars, left scale)

Sources: Central banks; and IMF staff.

Note: Data are by quarter for each year.

Country Practices in Foreign Exchange Intervention

Brazil

The Central Bank of Brazil has intervened in foreign exchange markets since the adoption of the floating exchange regime in 1999. Sterilized intervention in spot markets is one instrument at the central bank’s disposal, but by no means the only one. The central bank uses a variety of instruments to intervene in the foreign exchange markets, including outright US dollar sales and foreign exchange repos and swaps. In addition, up to 2002, the central bank was able to use US dollar– linked debt instruments (either issued by the national treasury or by the Central Bank of Brazil) in support of the currency, which directly affected the country’s gross debt.

In 2001–02, pressures on the Brazilian real intensified amid three major shocks—spillover from the Argentine debt crisis, the 9/11 market jitters in the United States, and public debt solvency concerns following the Brazilian presidential election. The real depreciated by up to 44 percent in 2001, and 71 percent in 2002, as international reserves fell to less than $40 billion. The central bank responded by using US dollar–linked debt instruments. However, the Fiscal Responsibility Law, passed in 2002, prohibited the central bank from issuing its own securities beginning in May 2002. Hence, starting in March 2002, the central bank replaced US dollar–linked instruments with so called “Brazilian FX swaps.”

In Brazil, spot market transactions are no longer the dominant intervention instrument, mainly because its derivatives markets are among the largest in the world. Trading volumes in derivatives are four times larger than that of the spot market; derivatives markets also lead spot markets into price discovery (see Figure 3.2). The large variety of interventions in Brazil reflects the different central bank objectives, which in turn depend on the period under analysis. During several episodes of market turbulence, the central bank had to counter depreciation pressures and sharp movements of the exchange rate (such as after the election of Lula da Silva in the 2002 election, the 2008 global financial crisis, or the US taper tantrum), while in other circumstances, it seemed to have been weighing against an appreciation of the currency. During 2008–09, the Central Bank of Brazil intervened through a range of tools simultaneously, including spot dollar sales, auctions of foreign exchange swaps and repos, and even indirect US dollar loans to Brazilian firms. Although the central bank has also used repos to provide temporary liquidity to the market, this type of intervention has been limited.

Figure 3.2.Ratio of Daily Turnover in Derivatives Market to Spot Market

(Net-net basis, as of April 2013)

Sources: Bank for International Settlements Triennial Survey 2016; and IMF staff calculations.

Until recently, the predominantly used instrument has been Brazilian foreign exchange swaps, technically a nondeliverable forward, which factors in the exchange rate risk but is settled in local currency.2 The instrument requires a high degree of substitutability—that is, a well-developed derivatives market. From the investors perspective, the swap is a good substitute for spot US dollars, to the extent that the real is fully convertible to US dollars at the date of settlement, thus meeting the demand for hedging. The instrument is structured such that, at maturity, the Central Bank of Brazil pays the international interest rate, it* , plus the actual rate of depreciation, Δet+1, while it receives the overnight domestic interest rate it.

Brazilian foreign exchange swaps provide hedging for investors with open positions, thus directly bidding down the forward exchange rate. At settlement, the Central Bank of Brazil pays its counterparty the additional amount of reais necessary to keep the dollar value of the initial open position unchanged. The central bank announces the details of each foreign exchange swap auction one business day before receiving market participants’ bids through the Sistema Especial de Liquidacao e Custo system (known as SELIC). On special occasions, the central bank announces the foreign exchange swap auction for the same day without previous announcement. Until the end of 2015, the notional balance of outstanding foreign exchange swaps amounted to close to $110 billion.

Chile

Chile does not intervene regularly in the foreign exchange market, and its intervention policy has been modest; in past decades, it has intervened sporadically. During 1998–99, interventions were not preannounced and were discretionary. However, in 2001 and 2002, interventions were conducted after a formal policy announcement, under a rules-based framework. These interventions were transparent, including explicit definitions of periods and amounts involved, while clearly explaining the reasons for the interventions. The first of these types of intervention started in August 2001 when the central bank communicated that spot market interventions could occur for up to a maximum of $2 billion over the following four months. Additional sales of $2 billion of dollar-denominated central bank bills were also announced. During that period, spot market interventions totaled $803 million, less than half the maximum announced.

Foreign exchange interventions were again used in 2008 and 2011, but mostly to accumulate reserves. The Central Bank of Chile conducted weekly announced competitive buy auctions of $50 million during the 2011 accumulation program, but it has not intervened in the foreign exchange market since then. Exchange rate movements in Chile allowed for significant current account adjustment, while diluting exchange rate volatility. The central bank usually announces the amount of foreign exchange it intends to buy or sell.

Colombia

The flexible exchange rate regime plays an important role in helping the Colombian economy adapt to changing global conditions. The freely floating framework has been complemented since October 2015 by a rules-based-contingent foreign exchange auction program aimed at mitigating excess volatility. Colombia experienced a sharp and fast real depreciation of its exchange rate, of about 34 percent, during 2015. This triggered the introduction of a rules-based foreign exchange intervention program through a competitive auction mechanism. The program was introduced in October 2015 (Table 3.6). The rules-based foreign exchange auction program, discontinued on May 31, 2016, was an effective mechanism to prevent disorderly depreciations and was only effectively triggered on May 20, 2016.

Table 3.6.Colombian Central Bank’s US Dollar Auction Program
Start dateOctober 30, 2015, with subsequent adjustments.
ObjectiveModerate disorderly increases in the exchange rate, which might contribute to an unanchoring of inflation expectations, as well as provide liquidity to the foreign exchange market.
ModalityAuction of call US dollar options for about $500 million, which is about one-third of the daily turnover. The option’s strike price is the average spot exchange rate on the previous day. Options are sold through Dutch auctions.
Rule basedTrigger for the sale (as well as for the exercise) requires the daily exchange rate movement to exceed a threshold, usually a given percentage relative to the 20-day moving average.
TriggerOn October 30, 2015, a trigger of 7 percent (depreciation in the 20-day moving average) was established. As foreign exchange volatility subsided, the threshold was lowered to 5 percent on December 23, 2015, and to 3 percent on February 19, 2016.
ExecutionThe auction program has not been triggered yet, which is likely because of the relative stability in oil prices and the program’s success in reducing foreign exchange uncertainty.
Sources: Central Bank of Colombia; and IMF staff.
Sources: Central Bank of Colombia; and IMF staff.

Guatemala

Guatemala has a long-standing, rules-based intervention policy that aims to stabilize excessive exchange rate volatility, while not affecting its trend. Intervention is triggered when the weighted average exchange rate of the sell (buy) transactions is less (more) than the five-day moving average reference exchange rate minus (plus) 0.75 percent. If triggered, the central bank offers up to a maximum of five daily auctions of $8 million each.

Mexico

Mexico has a long history of intervention in foreign exchange markets. The modalities of intervention have evolved. During 1996–2001, interventions were predominantly put options, where the central bank bought US dollars mainly to build up reserves. Thereafter, Mexico moved to rules-based intervention to moderate exchange rate volatility and to build international reserves. Table 3.7 presents a recent intervention program, although not the ongoing one.

Table 3.7.Bank of Mexico’s Foreign Exchange Intervention Program
Start dateNovember 29, 2011, with subsequent adjustments.
ObjectiveModerate disorderly increases in the exchange rate, which might contribute to an un-anchoring of inflation expectations, as well as provide liquidity to the foreign exchange market.
ModalityRules-based foreign exchange Dutch auction program provided $400 million call dollar options. The option’s minimum price equaled the previous day’s benchmark exchange rate plus 2 percent. This rule remained in place until April 2013. In December 2014, the mechanism was reintroduced, but with a reduced amount of $200 million and the previous day’s depreciation of 1.5 percent.
Rules basedTrigger for the sale requires the daily exchange rate movement to exceed a threshold, usually a given percentage relative to the previous day.
TriggerUntil 2013, a trigger of 2 percent had been established. As foreign exchange volatility subsided, the threshold was lowered to 1.5 percent starting in December 2014, and to 1 percent starting in July 2015.
ExecutionBecause the minimum price is set at a fairly large threshold for a one-day depreciation, it was triggered only on a few occasions.
Sources: National authorities; and IMF staff estimates.
Sources: National authorities; and IMF staff estimates.

In February 2017, they announced a new framework for interventions using NDFs settled in pesos for a maximum $20 billion; $1 billion worth of notional principal in short-dollar contracts were auctioned in March, and a new series of auctions were announced in October 2017 to sell NDFs for $4 billion in seven weekly auctions.

Peru

Peru has had a successful inflation-targeting framework since 2002. The constitution gives the Central Reserve Bank of Peru the mandate to preserve monetary stability, a goal achieved during the last 2 ½ decades. The central bank has had an active policy to moderate foreign exchange volatility to limit the negative effects of large exchange rate fluctuations. Interventions are made frequently using spot interventions, NDFs (also referred to as currency swaps), and certificates of deposit indexed to the exchange rate. Spot interventions are invariably performed during a fixed, preannounced, 2-hour window at the end of each trading day.

Peru’s foreign exchange market is composed of spot and derivatives, but the latter is shallow and relatively illiquid, even though most traded instruments in the derivatives market are NDFs. As such, commercial banks are used to selling (buying) US dollars in the foreign exchange forward market, hedging their positions by buying (selling) dollars in the spot market, and thus affecting the spot rate. The forward price (spot rate and forward rate differential) may therefore vary considerably according to the demand and supply of US dollars in the forward market; they typically deviate from the covered interest rate parity. These deviations used to produce profitable arbitrage opportunities. However, the central bank introduced a new instrument to reduce pressure on the spot exchange rate, and at the same time, lessen the distortions of the foreign exchange forward market transactions on interest rates in money and fixed income markets, while pre-venting the rise of hedging costs. So, in September 2014, the central bank added the Central Reserve Bank FX Swaps (SC BCRPs) as an instrument of foreign exchange intervention. SC BCRPs are basically NDFs settled in local currency.

SC BCRPs are derivatives instruments. One party commits to pay a variable interest rate in the local currency, calculated by using an overnight index swap (built by accumulating the daily interbank interest rate). The other party commits to pay a fixed interest rate in foreign currency and the foreign exchange rate variation. At maturity, the settlement is made by netting positions paid in local currency.

SC BCRPs help to control the foreign exchange spot rate, because they allow banks to hedge their positions from their activities in the forward markets without trading US dollars in the foreign exchange spot market. At the same time, they do not affect the money market, because they have no effect on monetary aggregates. As the settlement is on a netting basis, there is no exchange of notional amounts, either at the beginning or at the maturity of the contract. SC BCRPs are placed under an auction mechanism (the foreign fixed interest rate) that is carried out by the central bank.

Paraguay

The Central Bank of Paraguay intervenes in the foreign exchange market to smooth seasonal fluctuations and speculative movements under the authority of Article 50 of the Organic Law of the Central Bank of Paraguay (489/95), Article 3 of which aims to preserve and safeguard the stability of prices and to promote the efficiency and stability of the financial system. The central bank is constantly involved in the foreign exchange market by either buying or selling US dollars. In recent years, such involvement resulted in a substantial accumulation of international reserves.

The central bank uses two mechanisms to intervene in the foreign exchange market: (1) preannounced sales of the US dollars it receives from the ministry of finance to exchange into guaraníes to support its public expenditures; and (2) discretionary interventions, without previous announcement, to address any abrupt market movements. Under the first mechanism, the central bank announces monthly the frequency and size of the following month’s sales. However, the amount is made at the central bank’s discretion and should not exceed the current year’s US dollar proceeds bought from the ministry of finance.

Conclusions

This chapter aimed mainly to explore and present a taxonomy of the different dimensions of foreign exchange intervention implementation. Some of the differences could be attributed to country-specific characteristics, others to the goal or strategy of a central bank to achieve its highest efficacy, which can change over time.

Central banks can differ in the transparency of foreign exchange interventions, for example, in policy transparency or operational transparency. In Latin America, policy transparency is observed more often than not. Operational transparency is used less frequently, to mitigate potential idiosyncratic arbitrage opportunities that could undermine intervention effectiveness.

Regardless of the degree of transparency of foreign exchange intervention, another dimension in this taxonomy is whether the central bank operates under a rules-based or discretionary framework. If under a rules-based framework, foreign exchange intervention oftentimes establishes clear thresholds to trigger the intervention in the market. It can even specify the type and volume of the intervention. At the other extreme, sometimes we have observed Latin American central banks to intervene in a discretionary manner, with no precommitment or announcement. The use of rules and discretion vary by country and over time. Moreover, most countries in the region have used discretion at some point in time and a rules-based framework at others. The costs and benefits of each of these frameworks depend on time-varying needs, the type of shocks to which countries are exposed, the nature of the monetary policy framework in place at that time, and the development of a country’s financial markets, among other things.

For example, in the aftermath of the deepest financial crisis in recent history, emerging market economies experienced large shifts in foreign exchange market conditions, and many central banks adjusted their market operations to the evolving market and policy backdrop. In some cases, concerns for financial stability led to more frequent use of discretionary and less transparent foreign exchange intervention than before. Large and rapidly shifting capital flows and widening currency mismatches seem to have added support to policies aimed at containing exchange rate volatility and providing the private sector with insurance against exchange rate risks.

Going into a more operational territory, we then delve into the details of the type of instrument and modality used for the foreign exchange intervention. Latin America has extensively used a wide variety of intervention instruments, including forwards, swaps, repos, NDFs, and options, as well as US dollar–linked debt. More than one of these instruments has been used by the region’s central banks because of the structure and growing sophistication of their financial markets and the multiplicity of objectives.

Furthermore, to avoid affecting the monetary policy stance when intervening, as given by the market interest rate, central banks have been driven to use instruments that reduce the pressure on the spot foreign exchange rate, while reducing distortions from foreign exchange forward market transactions on interest rates. The choice of instrument has also affected the objective of the intervention, which spans from financial stability (the most frequent one), to building buffers, and in some episodes of large depreciation pressures, price stability (to mitigate the effect and potential nonlinearity of the exchange rate pass-through to domestic prices).

Thus, the wide arrangement of instruments, modalities, and frameworks for foreign exchange intervention in Latin America reflect the permanently shifting needs, structural country characteristics (such as financial deepness and global integration), and short-run objectives of countries in the region. It is not surprising that the taxonomy of foreign exchange interventions is extensive. Using this taxonomy as an encompassing framework, Chapters 7 through 13 detail interventions in Brazil, Chile, Colombia, Costa Rica, Mexico, Peru, and Uruguay, respectively.

References

    ArcherDavid. 2005. “Foreign Exchange Market Intervention: Methods and Tactics.” In Foreign Exchange Market Intervention in Emerging Markets: Motives Techniques and Implications edited by Monetary and Economic Department Bank for International Settlements. BIS Papers No. 24. Basel: Bank for International Settlements4055.

    • Search Google Scholar
    • Export Citation

    BarroRobertJ. andDavid B.Gordon. 1983. “Rules, Discretion, and Reputation in a Model of Monetary Policy.” Journal of Monetary Economics12: 10121.

    • Search Google Scholar
    • Export Citation

    CalvoGuillermoA.1978. “On the Time Consistency of Optimal Policy in a Monetary Economy.” Econometrica46 (6): 141128.

    Canales-KriljenkoJorgeIvan. 2003. “Foreign Exchange Intervention in Developing and Transition Economies: Result of a Survey.” IMF Working Paper 03/95International Monetary FundWashington, DC.

    • Search Google Scholar
    • Export Citation

    Canales-KriljenkoCemKaracadag andRobertoPereira Guimarães. 2003. “Official Intervention in the Foreign Exchange Market: Elements of Best Practice.” IMF Working Paper 03/152International Monetary FundWashington DC.

    • Search Google Scholar
    • Export Citation

    DominguezKathrynM. andJeffreyFrankel. 1993. “Does Foreign Exchange Intervention Work?Institute for International EconomicsWashington, DC.

    • Search Google Scholar
    • Export Citation

    DrazenA. 2003. “Political Economy in Macroeconomics.” Princeton University PressPrinceton, NJ.

    EnochCharles. 1998. “Transparency in Central Bank Operations in the Foreign Exchange Market.” IMF Policy Discussion Papers 98/2International Monetary FundWashington, DC.

    • Search Google Scholar
    • Export Citation

    HungJuan. 1997. “Intervention Strategies and Exchange Rate Volatility: A Noise Trading Perspective.” Journal of International Money and Finance16 (5): 77993.

    • Search Google Scholar
    • Export Citation

    KydlandFinnE. andEdward C.Prescott. 1977. “Rules Rather than Discretion: The Inconsistency of Optimal Plans.” Journal of Political Economy85 (3): 47392.

    • Search Google Scholar
    • Export Citation

    Montoro Carlos andMarcoOrtiz. 2016. “Foreign Exchange Intervention and Monetary Policy Design: A Market Microstructure Analysis.” Working Paper Series DT.N. 2016–008Central Bank of Peru.

    • Search Google Scholar
    • Export Citation

    MussaMichael. 1981. “The Role of Official Intervention.” Occasional Paper No. 6 Group of ThirtyNew York.

    NeelyChristopherJ.2005. “An Analysis of Recent Studies of the Effects of Foreign Exchange Intervention.” Federal Reserve Bank of St. Louis Working Paper No. 2005–030Federal Reserve Bank of St. Louis, MO.

    • Search Google Scholar
    • Export Citation

    RogoffKenneth. 1985. “The Optimal Degree of Commitment to an Intermediate Monetary Target.” Quarterly Journal of Economics100 (4): 116989.

    • Search Google Scholar
    • Export Citation

    SarnoLucio andMark P.Taylor. 2001. “Official Intervention in the Foreign Exchange Market: Is It Effective, and, If So, How Does It Work?Journal of Economic Literature34 (September): 83968.

    • Search Google Scholar
    • Export Citation

    VitalePaolo. 1999. “Sterilized Central Bank Intervention in the Foreign Exchange Market.” Journal of International Economics49 (2): 24567.

    • Search Google Scholar
    • Export Citation
1Bank for International Settlements Working Paper 144/2003: “Transparency versus Constructive Ambiguity in Foreign Exchange Interventions” refers to the IMF Code of Good Practices’ case for enhancing central banks’ intervention transparency. However, it also highlights that the code states that “ . . . there are circumstances where full transparency may not be beneficial to achieve policy goals, . . . the Code recognizes that there may be justifications for limiting certain disclosure practices in situations where increased transparency could endanger the effectiveness of policies, or be potentially harmful to market stability.”
2In Brazil, it is illegal to denominate contracts in foreign currency. Tus, contracts need to be settled in domestic currency even if indexed to US dollars.

    Other Resources Citing This Publication