Exchange Rate and Foreign Exchange Interventions in Guatemala
Author:
International Monetary Fund. European Dept.
Search for other papers by International Monetary Fund. European Dept. in
Current site
Google Scholar
Close

1. In 2005, the Bank of Guatemala (Banguat) adopted an IT framework along with a rule based mechanism for interventions in the Foreign Exchange (FX) market. Following a period of relative disfavor in the late 1980s and early 1990s, many central banks in Emerging Market Economies (EMEs) started to introduce Inflation Targeting (IT) frameworks with de jure floating but de facto rigid exchange rates. At the outset, fixed, predetermined, or quasi-fixed nominal exchange rates were viewed as an effective device for guiding a disinflation program as it was said it could rapidly generate a convergence toward the country for which the nominal exchange rate is anchored.2 It was also implied that such arrangement was transitory.3 Nevertheless, Foreign Exchange Intervention (FXI) has remained a widely used policy instrument. Of those central banks, many continue to use FXI to limit excessive volatility or for reserve adequacy purposes, without targeting any specific level of the exchange rate.4 Guatemala is no exception. Banguat’s monetary framework leverages two instruments: the policy rate and FXI. However, it remains an interest rate-based monetary policy regime (a regime where the interest rate is the main lever that the central bank uses to influence aggregate demand and inflation) as opposed to an exchange rate-based monetary policy regime (a regime where the main lever used to stabilize inflation is the exchange rate).

Abstract

1. In 2005, the Bank of Guatemala (Banguat) adopted an IT framework along with a rule based mechanism for interventions in the Foreign Exchange (FX) market. Following a period of relative disfavor in the late 1980s and early 1990s, many central banks in Emerging Market Economies (EMEs) started to introduce Inflation Targeting (IT) frameworks with de jure floating but de facto rigid exchange rates. At the outset, fixed, predetermined, or quasi-fixed nominal exchange rates were viewed as an effective device for guiding a disinflation program as it was said it could rapidly generate a convergence toward the country for which the nominal exchange rate is anchored.2 It was also implied that such arrangement was transitory.3 Nevertheless, Foreign Exchange Intervention (FXI) has remained a widely used policy instrument. Of those central banks, many continue to use FXI to limit excessive volatility or for reserve adequacy purposes, without targeting any specific level of the exchange rate.4 Guatemala is no exception. Banguat’s monetary framework leverages two instruments: the policy rate and FXI. However, it remains an interest rate-based monetary policy regime (a regime where the interest rate is the main lever that the central bank uses to influence aggregate demand and inflation) as opposed to an exchange rate-based monetary policy regime (a regime where the main lever used to stabilize inflation is the exchange rate).

Exchange Rate and Foreign Exchange Interventions in Guatemala1

1. In 2005, the Bank of Guatemala (Banguat) adopted an IT framework along with a rule based mechanism for interventions in the Foreign Exchange (FX) market. Following a period of relative disfavor in the late 1980s and early 1990s, many central banks in Emerging Market Economies (EMEs) started to introduce Inflation Targeting (IT) frameworks with de jure floating but de facto rigid exchange rates. At the outset, fixed, predetermined, or quasi-fixed nominal exchange rates were viewed as an effective device for guiding a disinflation program as it was said it could rapidly generate a convergence toward the country for which the nominal exchange rate is anchored.2 It was also implied that such arrangement was transitory.3 Nevertheless, Foreign Exchange Intervention (FXI) has remained a widely used policy instrument. Of those central banks, many continue to use FXI to limit excessive volatility or for reserve adequacy purposes, without targeting any specific level of the exchange rate.4 Guatemala is no exception. Banguat’s monetary framework leverages two instruments: the policy rate and FXI. However, it remains an interest rate-based monetary policy regime (a regime where the interest rate is the main lever that the central bank uses to influence aggregate demand and inflation) as opposed to an exchange rate-based monetary policy regime (a regime where the main lever used to stabilize inflation is the exchange rate).

uA003fig01

Nominal Exchange Rate, Nov 1989 – Dec 2022

(GTQ per US$)

Citation: IMF Staff Country Reports 2023, 173; 10.5089/9798400243158.002.A003

2. Since then, Guatemala’s nominal exchange rate has continued to be stable with low volatility. The exchange rate has been solidly anchored to around 7.75 GTQ per U.S. dollars since the launch of the IT regime in January 2005. The monthly volatilty, expressed as the monthly percentage change, was contained within a two percent band, reflecting both the natural equilibrium of supply and demand of U.S. dollars as well as daily FXI to mitigate exchange rate volatility. The FXI rule alone would have allowed a maximum monthly fluctuation of 5.4 percent (under the 2005 fluctuation band limit set at 0.5 percent) and 9.7 percent (under the current fluctuation band limit set at 0.9 percent). This indicates that the stability of the exchange rate is also driven by the supply and demand equilibrium. Not surprisingly, the yearly volatility, expressed as the annual percentage change, displays higher peaks which were systematically corrected in about two and half years. Most notably, the largest depreciation observed was between January 2009 and March 2011 (up to 8.7 percent depreciation relative to the mean of 7.7) during the great financial crisis and was gradually reversed, partially aided by FXI to slow the volatility pressure. The largest appreciation observed was between June 2016 and May 2018 (up to 5.5 percent relative to the mean of 7.7), corresponding to the sudden large increase in remittances inflows, and was reversed rapidly, mainly due to sizable FXI. Since January 2005, the average deviation from the mean is about 0.3 percent, thus highlighting the strong stabilty of the exchange rate over the long run.

uA003fig02
Source: Banguat; IMF staff calculations.

A. A Review of Guatemala’s Foreign Exchange Developments to Date

3. From its creation in 1924 until the mid-1980s, Guatemala’s official currency (the quetzal) was fixed at parity with the U.S. dollars. Despite the end of the Bretton Woods system in 1971, which prompted many countries to start adopting more exchange rate flexibility, Guatemala chose to continue with a fixed exchange rate regime. However, imbalances started to mount following the oil price shock of the 1970s, the global economic recession of the early 1980s, and political and fiscal instability in the country. International reserves declined rapidly, forcing Banguat to restrict transactions in U.S. dollars while selling foreign currency to some importers and financial institutions at a subsidized value. This led Banguat to accrue operating losses, which eroded its capital base. Eventually, in 1986, the quetzal started to devaluate as the pressure was no longer sustainable.

4. During the 1989-2005 period, discretionary FXI was used to stabilize the exchange rate. In 1989, at a time of very high inflation, the transition to a flexible exchange rate regime was advanced through a series of structural reforms to both address the pressure on international reserves and reduce inflation, including by promoting greater autonomy to Banguat to conduct its policies. During the period of 1989-1995, the exchange rate flexibility was restricted by fluctuation bands and FXI. The interventions, which were not sterilized in the money market until 1996, were based on expert judgment on the deviations of the nominal exchange rate vis-a-vis its historical moving average. In 1996, the Electronic Currency Negotiation System (SINEDI)5 was established. SINEDI enabled greater transparency in FX operations, reduced the price uncertainty in the FX market, and increased the volume of transactions significantly. Banguat carried out all FXI transactions via SINEDI. In May 2001, the establishment of the Law of Free Currency Negotiation6 was the beginning of a series of structural reforms that paved the way for the adoption of the IT regime along with a flexible exchange rate regime. Nevertheless, the exchange rate became de facto anchored to the U.S. dollar during that transition period.

Exchange Rate Intervention Rules under the Inflation Targeting Framework

5. In January 2005, the formal regulation of FXI was introduced through the participation rule to moderate volatility without affecting the trend7, but its criteria anchored de facto the exchange rate. The rule-based intervention policy aimed at stabilizing excessive exchange rate volatility, independent of monetary policy. The rule aimed at moderating appreciation pressure from larger remittances inflows that had been occurring since 2004 and to boost exports and economic activity that had had low growth since the early 2000s. The rule established three thresholds defined by predetermined exchange rates. The intervention criteria varied at each threshold, such that interventions were more frequent as the exchange rate continued to appreciate. In December 2006, the participation rule was modified to provide greater exchange rate flexibility while moderating volatility. A fluctuation band was defined (between Q7.60 and Q8.05) with different asymmetric deviation margins that resisted appreciation pressure more forcefully than depreciation pressure, whether the exchange rate is outside or inside the fluctuation band.8 In the end, the rule limited the fluctuations of the exchange rate to a very narrow range. Moreover, the limited permissible fluctuation margin along with the asymmetric properties of the rule suggested a greater intervention of Banguat during appreciation pressure, thus potentially complicating the implementation of the newly established IT framework.

6. At the same time, Banguat introduced certificates of deposits in U.S. to absorb excess liquidity in U.S. dollars in the financial sector when deemed necessary.9 The instrument allowed reducing the amount of U.S. dollars in circulation (through auctions) with a view to smoothing the behavior of the exchange rate without affecting its trend.10 20 However, contrary to the participation rule, this mechanism only withdraws liquidity temporarily as the certificates will eventually mature. This instrument was used in 2017-2018 in response to substantial remittance inflows and lower imports (due to lower oil prices since mid-2014).11

7. Since June 2008, Banguat’s authorities have amended the participation rule to enable greater flexibility, along with clearer and more explicit objectives. In its June 2008 decision, the Monetary Board (Junta Monetaria) eliminated the fluctuation band and the asymmetry of the FXI triggers and proposed clearer and more explicit objectives. These changes marked the beginning of the gradual transition towards more flexibility. In nearly all subsequent years, the criteria of the participation rule evolved to either expand the volatility trigger (from •0.50 percent to •0.90 percent as of January 2023), the maximum amount for interventions (from US$ 8 million per auction to US$ 20 million per auction as of January 2023), or the number of daily auctions (from 3 auctions to 5 auctions as of January 2023). The Monetary Board also explicitly reiterated the possibility to use certificates of deposits in U.S. dollars to withdraw U.S. dollars from the financial system when deemed necessary as well as language to promote a forward market in FX.12

8. In 2013, the Monetary Board introduced the possibility to intervene in the FX market to counteract unusual volatility.13 This new clause enables Banguat to react to exceptional events that impact the spot market by either buying or selling FX, beyond the participation rule. As of December 2022, it was only used once, on March 24, 2020, for a total amount of US$ 111 million (sale).

9. In December 2014, a new FXI instrument (the reserve accumulation rule) was added to Banguat’s toolkit.14 Its explicit objective is to accumulate international reserves while considering the evolution of the monetary and exchange markets so that the fundamental objective of the central bank is not jeopardized, nor distortions are introduced. Since the great financial crisis in 2009, the net international reserves-to-import ratio was below 4 percent and, in mid-2013, the growth rate of net international reserves started to slow down (0.8 percent y-o-y in December 2014 compared to 9.1 percent on average between 2009 and 2013). This new FXI provided the tool for Banguat to increase its level of reserves for precautionary reasons, if deemed necessary.

10. As of January 2023, Banguat relies on three instruments to conduct its exchange rate policy:

  • The Participation rule. An intervention is triggered when the weighted average exchange rate of the sell (buy) transactions15 is less (more) than the five-day moving average reference exchange rate minus (plus) 0.9 percent. If triggered, the central bank currently offers up to a maximum of five daily auctions of US$20 million each in the interbank market.

  • The Reserve Accumulation rule. Currently, the yearly maximum amount to purchase U.S. dollar under the rule is set at US$ 1.5 billion through three increments of US$ 500 million (each increment must be approved by the Monetary Board).

  • The issuance of certificates of deposits in U.S. dollars to reduce the amount of U.S. dollars in circulation.

Evolution of the Participation Rule

article image

The Exchange Rate and Interventions

11. Between 1996 and 2001, the main objective of Banguat’s FXI was to mitigate depreciation pressure, while between 2001 and 2007, the interventions were to reduce appreciation pressure. The depreciation pressure was particularly strong between 1998 and 2000 as remittances inflows decreased and the exchange rate depreciated from Q6.05 to Q7.75 per U.S. dollars (a 28 percent change). Banguat intervened frequently by selling U.S. dollars (up to 73 percent of total volume in the interbank market). After the establishment of the Law of Free Currency Negotiation and a new surge of remittances inflows in 2004, FXI were used to moderate appreciation pressure as well as excessive volatility.

uA003fig03

Monthly Foreign Exchange Interventions, 2004-07

(US$ million)

Citation: IMF Staff Country Reports 2023, 173; 10.5089/9798400243158.002.A003

Source: Banguat.

12. Source: Banguat.Between 2008 and 2015, FXI were frequently used in clusters in response to pressures. The cluster corresponding to the great financial crisis of 2008-2009 is an example of a short-term trend that reversed in early 2010 when appreciation pressures started to materialize again, thus prompting another cluster of interventions in the first quarter of 2011 (purchase of U.S. dollars). Another cluster was in late 2014-early 2015 when appreciation pressures triggered further interventions. Overall, empirical evidence shows that the central bank tends to react more forcefully when leaning against an appreciation, consistent with the threshold effect of the FXI rule.16

uA003fig04

Monthly Foreign Exchange Interventions, 2008-15

(US$ million)

Citation: IMF Staff Country Reports 2023, 173; 10.5089/9798400243158.002.A003

Source: Banguat.

13. Between 2015 and 2022, sizeable interventions coincided with unprecedented increase in remittance inflows. Between 2015 and 2017, a strong appreciation pressure emerged at the same time that import declined driven by lower oil prices, exports increased, and remittance inflows were slightly higher. On February 27, 2017, Banguat reacted by initiating the issuance of certificates of deposits in U.S. dollars to mop up excess FX liquidity (relative to the historical trend).17 As oil prices started to augment in late 2017, imports increased, but remittance inflows continued to rise. The new equilibrium was considered transitory as oil prices continued their upward trajectory and remittances growth started to show some signs of moderation. Eventually in 2018, Banguat activated its reserve accumulation rule which was deemed more efficient at withdrawing excess FX liquidity. It was considered a more structural operation (compared to the participation rule that deals with daily volatility). After a few months of appreciation, the exchange rate eventually returned to its historical level in mid-2018 and the activation of the reserve accumulation rule stopped in 2019. Then, like many other countries, the COVID-19 shock hit Guatemala in March 2020 with a strong, but temporary, impact on FX liquidity. Banguat immediately addressed the situation by injecting US$ 111 million to the interbank FX market (as part of the unusual volatility clause introduced by Banguat in 2013) in addition to the participation rule and the establishment of a repo facility in U.S. dollars to provide liquidity. When the economy started to rebound in late 2020, remittance inflows accelerated (35 percent y-o-y in 2021), putting greater appreciation pressure on the exchange rate. The reserve accumulation rule was re-activated in 202022 and overpowered the participation rule in 2021-22. In late 2022, pressure temporarily eased due to some one-off events related to the U.S. monetary policy tightening (repayments of banks’ credit lines).

uA003fig05

Remittances, 2014-22

(US$ million)

Citation: IMF Staff Country Reports 2023, 173; 10.5089/9798400243158.002.A003

Source: Banguat; IMF staff calculations
uA003fig06

Monthly Foreign Exchange Interventions, 2016-22

(US$ million)

Citation: IMF Staff Country Reports 2023, 173; 10.5089/9798400243158.002.A003

Source: Banguat.

14. Over the past 10 years, the exchange rate volatility was largely contained within the participation rule bands. The reference exchange rate ended up breaching the band only 13 times (equivalent to 99.5 percent of the time). Most of the breaches were minor and the COVID-19 shock in March 2020 explains the largest breaches.18 The activation of the reserve accumulation rule in 2018 also played a significant role as exchange rate volatility narrowed considerably since its application, while at the same time, the participation rule band increased (along with its ceiling for interventions).

Monthly Foreign Exchange Interventions, 2019-22

(US$ million)

article image
Source: Banguat; IMF staff calculations.

Foreign Exchange Interventions, Net Purchase, 2015-22

article image
Source: Banguat; IMF staff calculations.
uA003fig07

Actual Deviation from the Previous 5-day Moving Average and Volatility Band, 2012-22

(percent)

Citation: IMF Staff Country Reports 2023, 173; 10.5089/9798400243158.002.A003

Source: Banguat; IMF staff calculations.

15. The FX interbank market remains shallow and very segmented. Four banks accounted for more than two third of the FX sales (excluding Banguat) and four other banks accounted for nearly two third of the FX purchases (excluding Banguat) in the FX interbank market between 2019 and 2022. The volume of transactions is about US$ 440 million per month on average. Banguat’s purchases of FX account for nearly one third of the volume (against less than 5 percent when a seller).

uA003fig08

Volume and Share of Foreign Exchange Interbank Market Operations

Citation: IMF Staff Country Reports 2023, 173; 10.5089/9798400243158.002.A003

Source: Banguat; IMF staff calculations.

B. Measuring Exchange Rate Volatility in Guatemala

16. There are different methods to measure volatility. Banguat uses the percentage deviation from the previous 5-day moving average. Other typical and simple definitions of exchange rate volatility, most followed in the markets, are the day-to-day or intra-day changes. Absence of timely intra-day information affecting the exchange rate, the day-to-day change indicator is then usually preferred over the other and is often defined as the log return of the exchange rate, rt= log(et/et-1). The following figure shows the distribution of the log return of the exchange rate for

uA003fig09

Exchange Rate Volatility Distribution, 2012-2022

Citation: IMF Staff Country Reports 2023, 173; 10.5089/9798400243158.002.A003

Source: Banguat; IMF staff calculations.

17. Guatemala between 2012 and 2022. The shape of the distribution is a relatively thin bell curve with platykurtic properties (low frequency of outliers), indicating that the returns are clustered around the mean, which illustrates the low exchange rate volatility during the period.

18. There are other measures of volatility.

  • A kernel density estimation (KDE).19 The figure below shows the probability density function of daily volatility for selected countries for the period 2012-2022 as well as the probability density function for Guatemala for both day-to-day volatility and the previous 5-day moving average deviation. Both Colombia and Mexico relied on higher fixed intervention threshold during the period (between ± 2 and ± 7 % and between ± 1 and ± 2 % respectively) and display leptokurtic distributions (fat tails). On the other end, Guatemala’s distribution is leptokurtic (thin tails), reflecting Banguat’s low exchange rate volatility.

uA003fig10

Probability Density Function for Selected Countries, 2012-22

Citation: IMF Staff Country Reports 2023, 173; 10.5089/9798400243158.002.A003

Source: Country authorities; IMF staff calculations.
  • A volatility distribution can be derived from the historic distribution conditional on a set of exogenous variables. History-based triggers are not dynamic, not market-based, and could have strong limitations, especially when volatility is low (see Lafarguette and Veyrune (2021)).20 This conditional value-at-risk would depend on a set of variables corresponding to selected “determinants” of exchange rate (such as the VIX, oil prices, bid-ask spreads). The conditional predictive density of exchange rate can then be defined with a vector of explanatory variables. This methodology allows the intervention region to evolve every day as a function of market conditions. Using the determinants proposed for Guatemala, we find that the explanatory power of those variables is very limited.21

  • A Monte-Carlo value-at-risk method can be used to simulate projected exchange rate returns over thousands of possible iterations. However, the probability distribution for the chosen risk factors can be difficult to model beyond the use of normal distributions.

1

Prepared by Alexandre Nguyen-Duong. Includes extensive discussions with Maria A. Oliva.

2

This arrangement was behind the stabilization efforts in Argentina, Chile, and Mexico among others.

3

Many authors argued that in countries with an inflationary problem, after a short initial period with a pegged exchange rate, a more flexible regime should be adopted (Dornbusch 1997, Bruno 1995, Sachs, Tornell and Velasco 1995). In other words, the exchange rate should be a nominal anchor only in the initial phase of inflation stabilization, given the concept of the so-called "impossible trinity" (Obstfeld, Shambaugh and Taylor, 2005 and Aizenman, 2011).

4

The impact of FXI on the exchange rate is notoriously difficult to measure with econometric techniques, which implies that the effectiveness of FXI could vary greatly overtime and across countries.

5

SINEDI is a software, administered by the Bolsa Nacional de Valores (BNV), which records the amounts and exchange rates of foreign exchange transactions carried out daily between Guatemala's financial institutions.

6

Ley de Libre Negociacion de Divisas. In 2001 -02, three financial bills were approved by Congress to legalize the free exchange of foreign currencies, amend the constitutional law of Banguat, and improve the law governing banks and financial institutions. These new laws helped consolidate the liberalization of the capital account and paved the way for the establishment of the inflation-targeting framework.

7

The adoption of the rules based FXI was also accompanied with the commitment to gradually provide greater flexibility based on the compliance with the following principles: (i) be consistent with a monetary scheme of explicit inflation targets, (ii) based on explicit, transparent, and understandable rules for the markets, (iii) eliminate the discretion of Banguat's participations, and (iv) minimize volatility without affecting the trend.

8

Interventions to purchase U.S. dollars is triggered under a 0.5 percent deviation margin criterion (or 0.1 percent is the exchange rate is lower than 7.6). Interventions to sell U.S. dollars is triggered under a 1 percent deviation margin (or 0.5 percent if the exchange rate is greater than 8.05).

9

Recepcion de depositos a plazo en dolares de los estados unidos de america (JM-99-2004, September 2004).

10

Banguat would collect U.S. dollars at a high interest rate and place those U.S. dollars in the international market at a lower rate. In the end, such scheme could be proven counterproductive as it creates incentives for economic agents to borrow U.S. dollars in the international market at the low interest and have them renumerated (risk-free) at Banguat at a higher interest rate, such exacerbating the increase in U.S. dollars inflows.

11

The use of the instrument was eventually abandoned in favor of the reserve accumulation rule which was proven more efficient at mopping up FX liquidity.

12

Banks and financial institutions were asked to systematically communicate to Banguat, daily, financial characteristics of forward contracts related to FX. The information would be then published in the Banguat's website (JM-161-2008).

13

JM-1121-2013.

14

JM-133-2014

15

Any transactions greater than US$20,000 between a bank and bank or between a bank and a private entity triggers the rule for the day.

16

See Juan Catalan-Herrera (2016).

17

Certificates were issued with various maturities (91, 182, and 273 days) for a total of US$ 439 million (44 operations in 2017 and 32 operations in 2018). This instrument was stopped in 2018 as demand for such operations declined overtime and replace with the reserve accumulation rule.

18

Large depreciation pressure on March 24-26 which rebounded (mostly due to the moving average property of the formula) outside the band on March 30-April 1. On March 25, 2020, Banguat directly intervene to sell US$ 111 million.

19

A KDE estimates a function defined as the sum of kernel functions on every data point (in this case, the log return of the exchange rate or the percentage deviation from the previous 5-day moving average). The process can be used to estimate the probability dennsity function of historical exchange rate volatility.19 The analysis uses a gaussian kernel function fx=1nhi=1neu2/22πxx1h with the Silverman's rule of thumb bandwidth h=s43n1/5 to estimate the probability density function.

20

See IMF Working Paper No. 2021/032. The exchange rate at risk model defines the percentile at a given threshold of the conditional distribution of the exchange rate returns (i.e., based on a value-at-risk methodology)

21

The best fitting shows an R square of 7 percent (compared to 28 percent in the example proposed in the paper for Mexico).

  • Collapse
  • Expand
Guatemala: Selected Issues
Author:
International Monetary Fund. European Dept.