Chapter

Chapter 4. Implications of Global Financial Integration for Monetary Policy in Latin America

Author(s):
Luis I. Jacome H., Yan Carriere-Swallow, Hamid Faruqee, and Krishna Srinivasan
Published Date:
October 2016
Share
  • ShareShare
Show Summary Details
Author(s)
Yan Carrière-Swallow and Bertrand Gruss 

Financial asset markets around the world have grown increasingly interconnected, and many economies in Latin America have been no exception. This chapter assesses two aspects of monetary policy in the region for which financial integration is expected to have important implications: monetary autonomy from global financial conditions, and exchange rate pass-through. It also explores how improving the credibility of policy frameworks and establishing a strong nominal anchor has helped the region to deal with the trade-offs along these two aspects that financial integration has likely accentuated.

The first aspect that we will consider is the degree to which monetary policy in Latin America has been constrained by monetary policy decisions taken abroad. In discussions surrounding the normalization of U.S. monetary policy after the global financial crisis, a range of observers have questioned whether Latin American central banks would be able to maintain an accommodative stance—and thus allow the exchange rate to adjust in response to a larger interest rate differential—or if they would see the need to follow U.S. monetary policy despite weak domestic demand.

The second aspect is a reexamination of the relationship between exchange rate movements and inflation in the region. In a context of increased financial integration and volatile exchange rates, the ability of central banks to deliver stable inflation depends crucially on a limited degree of exchange rate pass-through. Indeed, a concern during the adoption of inflation-targeting regimes was precisely that exchange rate pass-through had historically been high in the region.

As monetary policies in many advanced economies have resorted to unconventional measures since the global financial crisis, and emerging market exchange rates have undergone large fluctuations in response to a series of global shocks, these topics have attracted a good deal of attention in the recent literature and policy debate.

Financial Integration and Its Implications for Monetary Policy

Latin American economies have gradually removed restrictions on the movement of international capital over the past four decades. Panel 1 of Figure 4.1 shows how de jure measures of capital account policies have evolved in selected Latin American countries since 1970—the index ranges from zero to one, where a larger number represents greater openness. While restrictions have gradually become less pervasive over this period, a particularly marked movement toward greater openness has taken place since the early 1990s.

Figure 4.1.Measures of Financial Integration in Latin America since 1970

Sources: Chinn and Ito (2008); Lane and Milesi-Ferretti (2007); and authors’ calculations.

Note: The Chinn and Ito (2006, 2008) index is based on information regarding restrictions in the International Monetary Fund’s Annual Report on Exchange Arrangements and Exchange Restrictions. It is normalized between zero and one, and a higher value indicates more openness to cross-border capital transactions.

In a region long characterized by low savings rates, opening the capital account offered the promise of allowing more plentiful foreign savings to finance productive investment, thus spurring development. As foreign financial intermediaries entered domestic markets, increased competition would improve efficiency among local incumbents. And finally, the ability of governments and firms to borrow abroad would allow them to smooth the impact of idiosyncratic shocks.

The removal of restrictions on the movement of international capital has indeed been accompanied by a gradual increase in capital flows to and from the region. Panel 2 of Figure 4.1 reports a de facto measure of financial integration, which corresponds to the sum of external assets and external liabilities over gross domestic product, based on data compiled by Lane and Milesi-Ferretti (2007).

Increased financial integration with global markets was also expected to carry certain well-recognized costs. In terms of risks, integrated economies face increased exposure to shocks from abroad such as sudden capital reversals, which can impose large costs and even trigger financial crises in extreme cases (see a discussion in Obstfeld 1998).

The move to greater openness had important implications for monetary policymaking. As international arbitrage of domestic assets became ever more prevalent under more interconnected financial markets, policymakers were confronted with a choice under the open-economy trilemma: either control the exchange rate but give up monetary policy, or maintain the autonomy to set policy interest rates in line with domestic objectives for inflation and activity while letting the exchange rate fluctuate.

During the financial liberalization agendas implemented in parts of Latin America during the 1970s, authorities had opted to keep exchange rates fixed. The transition to greater openness during the 1990s was characterized by a more eclectic and gradual approach that, in most cases, involved an increasing degree of exchange rate flexibility. This was in part due to memories of the accumulation of external imbalances that culminated in the debt crisis of the early 1980s. But it also responded to recognition of the importance of retaining monetary policy autonomy in the face of large external shocks, especially as countercyclical fiscal policy had often proved difficult to implement.

Panel 1 of Figure 4.2 shows that the number of countries with pegged exchange rates had declined by the 1990s. In some cases, the exchange rate arrangement evolved into a managed float or soft peg, rather than a fully flexible exchange rate, and the number of pegs in Latin America has actually increased somewhat recently—including due to the adoption of the U.S. dollar as legal tender in some countries. But the de facto degree of exchange rate flexibility in the region generally increased over the past few decades, as shown in panel 2 of Figure 4.2—where a lower reading of the exchange rate stability index denotes more flexibility.

Figure 4.2.Transitioning toward More Exchange Rate Flexibility

Sources: Aizenman, Chinn, and Ito (2008); Klein and Shambaugh (2008); and authors’ calculations.

Note: The Klein and Shambaugh (2008) classification of pegs and soft pegs closely follows the method employed by Obstfeld and Rogoff (1995). The Aizenman, Chinn, and Ito (2008) index is based on the annual standard deviation of each country’s monthly exchange rate with respect to its base country, and has been normalized to take values between zero and one. A larger value denotes less exchange rate flexibility. EME = emerging market economies. LA5 = Brazil, Chile, Colombia, Mexico, and Peru.

While there has been no recipe, gradual transitions toward greater financial openness have allowed regulation and supervision regimes to mature and local financial markets to deepen alongside the increasing participation of nonresidents. In many cases, countries have opted to maintain some degree of discretion in the use of capital flow measures—a strategy that the International Monetary Fund has come to broadly endorse in light of international experience (IMF 2012).

In many countries, there has been an effort to extend the availability and use of financial instruments to hedge the currency mismatches that can make corporate and bank balance sheets vulnerable in a context of greater exchange rate volatility. As policy frameworks have allowed greater exchange rate flexibility, this generates an endogenous incentive for private actors to pay the price of insuring against exchange rate risk. In turn, this has reduced pressure on independent central banks to intervene in foreign exchange markets.

Regulatory and prudential reforms also helped address financial vulnerabilities in a context of more volatile currencies. One clear example is the de-dollarization agenda implemented in several Latin American countries. The set of policies included the introduction of prudential measures to create incentives to internalize the risks of dollarization.1 These measures included active management of reserve requirement differentials and setting higher provision requirements for foreign currency loans and tighter limits on the banks’ net open position. An explicit strategy for the development of a capital market in local currency through the issuance of long-term public bonds in domestic currency also played a key role, as it facilitated bank funding and pricing of long-term loans in domestic currency. In a context of macroeconomic stability—with a notable reduction of inflation levels—and sustained exchange rate appreciation, financial dollarization ratios in Bolivia, Paraguay, Peru, and Uruguay decreased by almost 30 percentage points on average (García-Escribano and Sosa 2011). Despite this progress, however, dollarization levels are still relatively high in some countries.

In order to successfully transition to a new monetary regime geared toward domestic price stability under flexible exchange rates, establishing a strong nominal anchor was critical. Many countries in the region adopted inflation targeting in the late 1990s and early 2000s as a means of establishing such an anchor in the form of a simple, easily communicated policy framework. However, adopting a formal target for inflation with some degree of legislative oversight meant that scrutiny would befall the central bank in the event that the target went unmet. In a region where currency adjustments had been historically associated with periods of hyperinflation, containing high exchange rate pass-through presented a challenge for meeting inflation objectives.

How have these countries fared in their pursuit of a domestic nominal anchor? As we have shown, the change toward openness was followed by larger capital flows and higher exchange rate volatility. Panel 1 of Figure 4.3 shows the evolution of inflation in selected Latin American countries since 1990, where bars correspond to average inflation in 1990–95, 1996–2001, 2002–07, 2008–13, and 2014–16. In those countries that have adopted an inflation target, average inflation has decreased markedly. But the figure also makes clear that in other countries, reducing inflation to levels consistently below 10 percent—a threshold that some empirical work has identified as imposing high costs from misallocation in terms of output and employment (Reis 2013)—remains a challenge.

Figure 4.3.Reduction in the Level and Volatility of Inflation since 1990

Source: IMF, Information Notice System.

Note: For clarity, values in panel 1 and 2 have been capped at 15. Argentina’s CPI data after December 2006 correspond to private analysts’ estimates.

Since price instability is the ultimate source of welfare costs that central banks seek to reduce, the variability of inflation matters alongside the average rate of inflation. Panel 2 of Figure 4.3 shows the standard deviation of inflation during the same time periods. Importantly, the reduction of the level of inflation has also been accompanied by a marked decrease in the volatility of inflation. Finally, as was noted in Chapter 3, these achievements have been accompanied by an increased anchoring of inflation expectations, with average forecasts falling in line with targets, and the degree of disagreement among forecasters being strongly reduced. That is to say, private agents increasingly tend to agree on the future trajectory of prices, and their projections broadly coincide with central banks’ stated objectives.

Whether this improved monetary performance can be attributed to the adoption of inflation targeting is open to debate, as international evidence on the impact of implementing such a monetary framework is somewhat mixed.2 We now turn to the analysis of two aspects of monetary policymaking in Latin America that can shed light on the link between their evolving policy frameworks and observed monetary performance.

The first aspect is related to the ability of central banks to tailor monetary policy to domestic objectives in a context of greater financial integration. The impossible trinity or policy trilemma in open economies (Mundell 1963; Obstfeld and Taylor 1998) states that if the monetary authority of a financially open economy fixes the exchange rate, they will be unable to tailor monetary policy according to domestic objectives such as ensuring output and price stability. The corollary is that autonomous monetary policy is possible as long as the exchange rate is allowed to float. However, many studies have found that the pass-through of international to domestic interest rates remains significant even for countries with flexible exchange rates (see, for instance, Edwards 2015). Some have even questioned the validity of the trilemma based on the large co-movement of interest rates and other asset prices across integrated economies (Rey 2015). Have Latin American central banks been able to pursue autonomous monetary policy after opening up their capital accounts?

The second aspect is related to the ability to deliver price stability in a context of greater exchange rate variability. Even if the central bank retains autonomy to set policy interest rates in line with domestic objectives, higher exchange rate variability may result in excessive inflation volatility if the consumer prices are highly sensitive to the exchange rate. Under such a scenario, it may not be possible to effectively offset such price variability with monetary policy given its lengthy transmission mechanism, resulting in difficulties to anchor inflation expectations. The extent of exchange rate pass-through had indeed been historically large in the region, so a natural question is how it was possible to deliver low and stable inflation with more exchange rate volatility. Have the monetary reforms been accompanied by a reduction in pass-through rates in the region?

Spillovers and Monetary Policy Autonomy in Latin America

Since the early 2000s, there has been substantial co-movement of interest rates across a large sample of advanced and emerging market economies.3 We analyze this aspect of the global financial cycle by estimating a global factor of short- and long-term interest rates, real output growth, and Consumer Price Index (CPI) inflation, which corresponds to the principal component of the time series of each variable across a panel of 60 countries.4

Short-term interest rates exhibit a positive correlation with the global component in most countries. Within Latin America, the largest degree of co-movement with the global component is observed among those economies that have become the most integrated with global financial markets (Brazil, Chile, Colombia, Mexico, and Peru, denoted LA5). In these countries, the average correlation of short-term interest rates with the global factor has been slightly above 0.7, which is comparable to that of financially integrated economies in Asia. The degree of co-movement with global short-term rates is smaller for other Latin American countries, such as Costa Rica, Honduras, and Paraguay. The patterns are similar in the case of long-term interest rates.

But does a high correlation of global and domestic interest rates indicate that open economies with flexible exchange rates lack the autonomy to set an independent monetary policy? As we will argue, not necessarily.

As shown in Figure 4.4, the degree of co-movement of interest rates with respect to the corresponding global factor varies over time, and fluctuations tend to mimic the variations in synchronization of business cycles across countries. For instance, the degree of synchronicity of short-term interest rates in the LA5 economies with the global factor reached particularly high levels following the common shock from the global financial crisis and has since moderated somewhat as real cycles have diverged during the postcrisis period. This observation highlights the importance of accounting carefully for the co-movement in business cycles if we want to infer the extent of monetary policy autonomy from estimates of interest rate interdependence—an exercise to which we turn later in this section.

Figure 4.4.Rolling Correlation across Selected Latin American Economies and the Global Common Factor

Source: Authors’ calculations.

Note: Average correlation across selected macroeconomic variables in Brazil, Chile, Colombia, Mexico, and Peru (LA5) against the global component for the corresponding variable; four-year moving average.

We first use a set of country-specific vector autoregressions (VARs) and monthly data since the early 2000s to quantify the linkages between interest rates in the United States and those in a sample of 43 advanced and emerging market economies.5 All models include changes in short- or long-term interest rates as domestic variables—depending on the question—and changes in the U.S. federal funds rate or in the 10-year Treasury bond yields as external or exogenous variables.6 Following the results in Chen, Mancini-Griffoli, and Sahay (2014), we also include the Chicago Board Options Exchange Volatility Index (VIX) as an external variable to account for changes in global risk sentiment.

We start by exploring the reaction of domestic short-term rates to changes in the U.S. federal funds rate.7 Short-term rates react quite differently across our sample of countries in response to a movement in the federal funds rate (circles in panel 1 of Figure 4.4).8 The average response to a 100-basis-point hike in the federal funds rate for a broad sample of emerging markets outside Latin America is less than 10 basis points, while the average response for advanced economies is about 30 basis points. Within Latin America, the response of Mexican and Peruvian short-term rates to the same shock is as high as 95 basis points and 80 basis points, respectively. Short-term rates in Argentina, Bolivia, Chile, Costa Rica, and Uruguay react only by 20 to 40 basis points. In Colombia, in turn, their response is close to zero, and it is negative—though not statistically significant—in the case of Brazil.

There is nothing necessarily surprising or inherently undesirable about domestic financial conditions being synchronized with those of international financial markets. For instance, countries with strong trade and financial linkages with the United States—such as Canada and Mexico—will tend to have an economic cycle that is highly synchronized with the U.S. cycle. In such cases, changes in domestic financial conditions may be broadly aligned with U.S. financial conditions, without compromising price and output stabilization objectives.

The situation would be different, however, if domestic financial conditions are driven by foreign conditions that are out of sync with the domestic business cycle. In this case, monetary policy may deviate from the inward-looking policy objectives of local central banks, with consequent costs in terms of output and price stability.

To narrow our analysis to spillovers from U.S. interest rates that would signal constrained monetary autonomy, we follow the two-stage procedure presented in Caceres, Carrière-Swallow, and Gruss (2016), which partials out the systematic policy response to changes in domestic macroeconomic conditions.9 The intuition behind the procedure is as follows. We first impose the null hypothesis that domestic monetary policy is inward-looking, focusing exclusively on stabilizing domestic output and prices. But at times, monetary policy decisions will have deviated from what the central bank’s policy rule would predict on the basis of local developments alone. We then estimate whether these deviations can be accounted for by movements in U.S. monetary policy. If so, we conclude that monetary autonomy is constrained to some degree.

More specifically, in the first stage we estimate a Taylor-type rule for the dynamic relationship between domestic interest rates and 12-month-ahead forecasts of inflation and output growth, as reported by Consensus Economics. These market forecasts are meant to capture changes in the economic outlook due to both idiosyncratic and global factors.

Ideally we would employ the central banks’ internal forecasts used to inform the policy decision, but these are publicly available for only a small number of countries. Using market forecasts instead is subject to two potential limitations. First, there is a timing problem because they are not collected on the day of monetary policy decisions, such that information sets will differ from those available to policymakers. This could potentially bias the assessment of monetary autonomy.10 As a robustness check, we verify that using alternative forecast vintages does not significantly alter the results. The second concern is that market forecasts may incorporate expected policy responses. If the central bank is fully credible, the argument goes, market forecasts might not react to a shock that affects domestic conditions, as agents anticipate that the central bank will do whatever is necessary to neutralize its effects. However, it is generally believed that monetary policy affects economic outcomes with a significant delay, such that movements in 12-month-ahead market forecasts should be highly correlated with movements in the central bank’s internal forecasts.

The residuals or unexplained components from the first-stage estimations can be interpreted as deviations from the historical policy reaction function that characterizes the central bank’s efforts to stabilize the domestic cycle. Of course, these unexplained interest rate movements likely reflect other central bank objectives beyond preserving price stability, including financial stability concerns related to exchange rate fluctuations, and could well be welfare-enhancing.11 Nonetheless, they entail changes in domestic monetary conditions beyond what can be attributed to the central bank’s usual response to inflation and output developments.

In the second stage, we seek to quantify to what extent these unexplained interest rate movements can be attributed to changes in U.S. interest rates. To do so, we estimate a second set of country-specific VAR models including the residuals from the first-stage estimation, the federal funds rate, and the VIX. As before, the coefficient matrices are restricted to ensure that global variables are not affected by domestic variables. We expect the impulse response of the Taylor-rule residuals to shocks from the federal funds rate to be nonzero where monetary autonomy is constrained, so we denote these responses autonomy-impairing spillovers.

The autonomy-impairing spillovers from U.S. to domestic short-term interest rates (depicted with bars in panel 1 of Figure 4.5) are generally smaller than the overall responses reported earlier (20 basis points lower on average). This suggests that an important portion of the co-movement in interest rates is simply a reflection of synchronized business cycles, and thus cannot be taken to mean that the central bank lacks monetary autonomy.

Figure 4.5.Spillovers to Short- and Long-Term Interest Rates

Source: Authors’ calculations.

Note: Responses in basis points following a 100 basis point increase in the corresponding U.S. rate. ADV = advanced economies; EME = emerging market economies; LA5 = Brazil, Chile, Colombia, Mexico, and Peru.

However, the estimated autonomy-impairing spillovers are statistically significant in 8 of the 46 advanced and emerging market economies in the sample: Canada, Hong Kong Special Administrative Region, Israel, Mexico, Peru, Saudi Arabia, Singapore, and Taiwan Province of China.12 The average spillover among these economies is quite large—about 40 basis points—but there is substantial heterogeneity across countries. Interestingly, these economies include some with fully flexible exchange rates and well-established central bank credibility, such as Canada and Israel.

In Latin America, spillovers associated with limited autonomy are significant and large in the cases of Mexico and Peru—about 70 basis points and 50 basis points, respectively—but smaller and not statistically significant in the other countries. This is likely related to the tight financial linkages with the United States in the former and the high degree of dollarization in the latter. In a context in which the U.S. and Latin American business cycles are out of sync, our results would suggest that a co-movement with U.S. rates would be more likely in Mexico and Peru than elsewhere.

In sum, we find that a large portion of the response of short-term interest rates to movements in U.S. rates can be attributed to the synchronicity of business cycles across countries. However, we also find that movements in U.S. rates generate significant spillovers to domestic short-term rates in several countries—including a few in Latin America—above and beyond what can be explained by standard business-cycle co-movement. Overall, there seems to be an important amount of heterogeneity in terms of the ability to set short-term rates according to domestic objectives, which motivates an analysis of the determinants of monetary autonomy in the next section.

Many important economic decisions taken by firms, households and financial institutions depend on longer-term interest rates. Moreover, the U.S. Federal Reserve has conducted monetary policy by influencing the longer end of the yield curve through quantitative easing and forward guidance since the policy rate hit the zero lower bound during the global financial crisis. So before turning to the determinants of monetary autonomy, we explore the response of domestic interest rates at the longer-end of the yield curve to changes in U.S. interest rates. To this end, we estimate country-specific VAR models that include the long-term interest rate as the domestic variable of interest, and U.S. interest rates (the federal funds rate or the 10-year U.S. Treasury bond yields, depending on the question) and the VIX as external variables.

We find that changes in the federal funds rate do not have an important impact on domestic long-term interest rates elsewhere (dashes in panel 1 of Figure 4.5). A notable exception in Latin America is Mexico, where long-term rates are estimated to rise by about 40 basis points following a 100 basis point hike in the U.S. policy rate.

However, movements in 10-year U.S. bond yields typically have a greater impact on corresponding domestic rates, and the responses are more similar across countries (dots in panel 2 of Figure 4.5). After a 100 basis point increase in U.S. bond yields, long-term rates in emerging market and advanced economies increase by an average of 35 and 55 basis points, respectively. The average response of long-term rates in the LA5 economies is even larger, at about 90 basis points. Brazil stands out with a response of about 130 basis points, followed by Colombia at 120 basis points. The response in the other LA5 countries lies between 55 and 75 basis points.

For completeness, we also report estimates of spillovers to long-term rates obtained by implementing the two-stage procedure described above—shown with bars in panel 2 of Figure 4.5—but find that they are essentially the same as the overall response from simple VAR models. An interpretation of this result is that the short-term domestic macroeconomic outlook has little bearing on long-term sovereign interest rates, which are in turn heavily impacted by movements in corresponding U.S. rates.

In an additional exercise (not reported here due to space considerations), we reestimate the country-specific VAR models including long-term interest rates in the domestic block, but substituting the U.S. bond yields in the external block with its decomposition into the expected path of short-term interest rates and the term premium. We find that the term premium is a much more important driver of Latin American long-term rates than the expected path for short-term rates, suggesting that conventional U.S. Federal Reserve policy has not been a major driver of international long-term rates. While this might suggest a secondary role for U.S. monetary policy in driving the global financial cycle in this asset class, it should be noted that unconventional monetary policy measures adopted since the global financial crisis appear to have influenced the term premium by reducing uncertainty surrounding the future rate path.

Determinants of Monetary Autonomy

In the previous section we reported that the response of domestic short-term rates to changes in the U.S. federal funds rate varies significantly across countries even after controlling for domestic economic conditions, suggesting that monetary autonomy is limited in some cases but not in others. What explains this heterogeneity in monetary policy autonomy? According to the trilemma theory, the main determinants of autonomy are the degree of exchange rate flexibility and capital account openness. More recently, Rey (2015) has argued that autonomy can only be achieved by restricting the capital account, challenging the dimensions of the trilemma’s central trade-off. Our results seem to lay somewhere in between. Some countries with flexible exchange rates are constrained by significant spillovers from global financial conditions. Meanwhile, many other countries in Latin America do seem to enjoy full autonomy to set monetary policy according to domestic objectives, regardless of the global financial cycle. We should note that this does not imply that these countries are isolated from financial conditions abroad, since these likely have important impacts on their domestic macroeconomic variables. Rather, these global developments do not cause domestic financial conditions to deviate from the objectives of stabilizing output and inflation and are thus indistinguishable from a fully inward-looking monetary policy.

The large variability in estimates of spillovers associated with limited autonomy across countries suggests that there may also be other factors affecting the trade-offs underlying monetary autonomy. This section reports the results from a panel VAR estimation approach that exploits the differences in spillovers across countries and explores how these are affected by policy frameworks and other characteristics. Following the setup used in Towbin and Weber (2013), an interacted-panel VAR (IPVAR) model is used to exploit heterogeneity in fundamentals across countries and also over time. The model includes the same variables that are in the country-specific VAR models—that is, the first-stage Taylor-rule residuals, the U.S. federal funds rate, and the VIX—but estimates the dynamic relationships in a panel setting.

A regular panel VAR would force the coefficient matrices to be the same for all countries. In the IPVAR model, instead, the model coefficients are functions of country-specific fundamentals that can, in addition, vary over time. Once the model is estimated, we can then obtain different impulse response functions that are conditional on particular values of each country-specific fundamental.

We start by verifying the basic predictions of the trilemma by conditioning the impulse-response functions on the exchange rate regime. Panel 1 of Figure 4.6 compares the estimated autonomy-impairing spillovers under alternative exchange rate regimes (as defined in Reinhart and Rogoff, 2004, and Ilzetzki, Reinhart, and Rogoff, 2009), conditional on an open capital account. We find that maintaining a fully flexible exchange rate regime sharply reduces the degree of spillovers to domestic short-term rates. For a country with high financial openness, the spillover effect declines from about 40 basis points under a fixed exchange rate to about 14 basis points under a floating exchange rate and to only four basis points under a fully flexible regime—and the response under a fully flexible exchange rate regime is actually indistinguishable from zero at a 10 percent confidence level.13

Figure 4.6.Determinants of Monetary Autonomy

Sources: Authors’ calculations.

Note: The impulse responses are estimated with an interacted-panel vector autoregression. The exchange rate regime classification used for the conditional impulse responses in panel 1 follows Ilzetzki, Reinhart, and Rogoff (2009). The impulse responses reported in panel 2 are conditional on different values of an index capturing the degree of anchoring of inflation expectations that is inversely related to the disagreement among Consensus Economics inflation forecasts. “Low” (High) corresponds to the third (seventh) decile of the distribution of this index in the sample, while “Max” corresponds to its maximum value.

Other results (not reported here due to space considerations) also confirm that constraining capital mobility reduces the extent of autonomy-impairing spillovers: when we condition on a floating exchange rate and the degree of capital account openness is set to match the value of the index corresponding to the first decile in our sample, the spillover drops from 14 basis points to practically zero.

Finally, we explore whether the strength of policy frameworks affects the extent of autonomy-impairing spillovers. To this end, we extend the model to include a third fundamental and construct a country-specific index for the degree of anchoring of inflation expectations and monetary policy credibility.14 Panel 2 of Figure 4.6 shows that, for a given policy choice along the capital account openness and exchange rate flexibility dimensions, a smaller degree of disagreement among professional forecasters of inflation reduces the extent of spillovers that are associated with limited monetary autonomy. The spillover response drops by about 10 basis points when the index of policy credibility moves from a value corresponding to the third decile of its distribution to its maximum level in our sample. We interpret a smaller degree of forecast disagreement as suggesting that inflation expectations are better anchored, probably as a result of more predictable and credible monetary policy.

This result is important for understanding the benefits of inflation targeting. While alternative policy frameworks could be implemented under a policy of flexible exchange rates, the benefits in terms of monetary autonomy are greatest when the framework delivers better-anchored inflation expectations. By helping to align private agents’ expectations about the future path of inflation, inflation-targeting regimes may have helped to deliver greater monetary autonomy from global financial conditions.

So, have Latin American countries been able to implement autonomous monetary policy despite greater financial integration? Edwards (2015) suggests that policy autonomy in the region has been quite limited recently, with monetary policy in the region’s inflation-targeting countries following the Federal Reserve to a considerable extent. Our results are more mixed: while Mexico and Peru do exhibit spillovers from U.S. monetary policy, Brazil, Colombia, and Chile do not. They also suggest that anchoring inflation expectations in a context of more flexible exchange rates has helped to increase the extent of monetary autonomy in the region.

Exchange Rate Pass-Through

A key question for Latin America at the time of transitioning to more flexible exchange rate regimes was the implications it would carry in terms of inflation volatility.15 The sensitivity of domestic prices to changes in the exchange rate had historically been high—indeed, pegging the currency to the U.S. dollar had been extensively used in the region to stabilize prices after periods of high inflation. Delivering price stability under flexible exchange rates in a context of high exchange rate pass-through to consumer prices would naturally pose an important challenge to the region’s incipient inflation-targeting regimes.

As we have shown, the region’s inflation-targeting regimes have generally been able to deliver lower and more stable inflation. This took place in spite of these regimes being characterized—by design—by increased volatility of the exchange rate (Figure 4.2). This is an indication that more stable inflation does not seem to be due to a more stable external environment over this period.

We begin by exploring whether the rate of exchange rate pass-through has changed over time in Latin America. Our empirical estimation of exchange rate pass-through to consumer prices is based on a standard specification (Campa and Goldberg 2005; and Gopinath 2015). The cumulative response is estimated for each country using Jordà’s (2005) local projection method (LPM):

where pi, t denotes the natural logarithm of the domestic price level in country i at period t; NEER is the natural logarithm of the import-weighted nominal effective exchange rate;16 Δ is a first difference operator; h denotes the horizon of the cumulative response; and εi,t is a random disturbance. The vector X includes a set of control variables (and their lags) that may affect both the exchange rate and inflation, to reduce concerns about omitted variable bias: international oil and food prices in U.S. dollars; the cost of production in countries from which country i imports (proxied by the import-weighted producer price index of trading partners, as in Gopinath 2015);17 and local demand conditions (proxied by the Hodrick-Prescott cyclical component of industrial production).

Figure 4.7 reports our exchange rate pass-through estimates β0h across regional country groupings for estimations carried out in three overlapping time periods: 1995–2007, 1999–2011, and 2003–15. Pass-through estimates have declined substantially from the first to the last estimation period, with particularly marked declines among emerging market economies. In Latin America, declines in exchange rate pass-through have been broad-based, but pass-through has reached particularly low levels among the region’s large inflation targeters. The pass-through in those countries has become comparable to estimates among advanced economies.

Figure 4.7.Estimated Exchange Rate Pass-Through over Time

Source: Authors’ calculations.

Note: Exchange rate pass-through coefficients refer to the estimated increase in the price level two years after a 1 percent increase (depreciation) of the nominal effective exchange rate. LA5 = Brazil, Chile, Colombia, Mexico, and Peru.

What may explain the downward trend in exchange rate pass-through in emerging markets and, in particular, in Latin America? We next explore whether this phenomenon can be related to the transformation of monetary policy frameworks and its implications for the inflationary environment.

The exchange rate pass-through to consumer prices can be considered as a two-stage process. In the first stage, the prices at the border and in local currency of imported goods adjust to variations in the exchange rate. In the second stage, this gets reflected in consumer prices. The overall pass-through to consumer prices will thus depend on the sensitivity of import prices in local currency to exchange rate movements, and on the degree of import content in domestic consumption.18 In consequence, the product of the exchange rate pass-through to import prices and the import content of final household consumption provides a rough benchmark for the expected first-round effects of currency depreciations on consumer prices.19 As a first approximation and in line with the results in Gopinath (2015), we can assume complete pass-through to import prices, in which case the benchmark would simply correspond to the import content of consumption.20

But the overall pass-through to consumer prices might differ from this benchmark related to the adjustment of tradable prices if, for instance, rigidities in the labor or product markets, or poorly anchored inflation expectations, lead to second-round effects of currency movements. In an unstable monetary environment, in particular, the impact of currency depreciation on inflation can be amplified by changes in inflation expectations that, in turn, affect price- and wage-setting decisions, feeding back into actual inflation.

Figure 4.7 also plots the import content of domestic consumption alongside our estimates of exchange rate pass-through. A few observations are noteworthy. First, the import content has increased gradually over time in all countries, reflecting a process of increased trade integration. Second, the gap between exchange rate pass-through estimates and the import content benchmark has declined in all regions, including Latin America. Among the LA5 economies, exchange rate pass-through estimates in the most recent period are actually smaller than the import share of consumption—a feature that had previously been true only among advanced economies. We interpret this as evidence that second-round effects have become less pervasive.

The fact that second-round effects seem to have eased over time points to improvements in monetary frameworks as one of the factors behind the decline in pass-through to consumer prices. Starting by effects at the border, and given that prices are generally assumed to be sticky in the currency they are invoiced, the literature has documented how exchange rate pass-through depends on whether the price of imported goods is set in foreign or domestic currency. As such, the larger the fraction of imports that are invoiced in foreign currency, the higher the pass-through of the exchange rate to import prices in local currency. Indeed, Gopinath (2015) finds a strong correlation between the rate of pass-through to import prices and the share of foreign currency invoicing, both at short and long horizons, for a sample of 18 advanced and 6 emerging market economies.

There is scarce data on the currency of invoicing in Latin America, but what is available points to widespread usage of the U.S. dollar. For instance, Gopinath (2015) reports that the average fraction of imports invoiced in dollars over the last few years in Argentina and Brazil has been about 88 and 84 percent, respectively. In the case of Colombia and Peru, this share is as high as 99 and 94 percent. The share in other Latin American countries is likely to also be very high, which would suggest that the pass-through to import prices is large. Indeed, Gopinath (2015) estimates nearly complete import price pass-through for Argentina, Brazil, and Mexico.

The currency of invoicing is precisely one channel through which improvements in monetary frameworks may lead to lower pass-through to import prices and, ultimately, to consumer prices. Devereux, Engel, and Storgaard (2004) argue that the currency of invoicing is an endogenous decision, with agents choosing to price their goods in the currency that most reliably holds its value. By this argument, delivering price stability would deliver an endogenous fall in the pass-through of the exchange rate to import prices.

But a change in the currency of invoicing is only one of the reasons that may explain a reduction in pass-through to consumer prices, and arguably not the most important in Latin America so far. There are other factors that are particularly relevant for the extent of second-round effects. For instance, the explicit indexation of contracts for goods and services can trigger adjustments in the price of nontradables as well. Where nominal anchors are weak and market participants doubt the central bank’s commitment to price stability, indexation practices may become more widespread. More generally, a perception that an observed change in the price of tradables may actually reflect underlying inflationary pressures, rather than a transitory relative-price adjustment following an external shock, is more likely to lead price-setters to raise their prices in turn if nominal anchors are not well established.

Along these lines, Taylor (2000) argues that lower and more stable inflation is a factor behind the decline in exchange rate pass-through. In particular, he argues that an environment of lower inflation is associated with lower expected persistence of cost and price changes. Using a microeconomic model of price setting, he shows that more stable inflation leads to a reduction in the extent to which firms pass through cost increases—including those that are due to exchange rate movements—to the price of their products. A straightforward implication of this argument is that variations in pass-through should not be taken as exogenous to the inflationary environment.

One way to explore the validity of this hypothesis is to see how the extent of second-round effects is related to the inflationary environment. Figure 4.8 shows the correlation of our estimated second-round effects—that is, the difference between pass-through estimates and their benchmark—at the country level with average inflation during the estimation sample, the volatility of inflation, and the metric of the degree of inflation expectation anchoring mentioned in the previous section.

Figure 4.8.The Inflationary Environment and Second-Round Effects from Depreciations

Source: Authors’ calculations.

Note: The vertical axis shows the estimated second-round effects, constructed as the difference between the exchange rate pass-through estimate and the benchmark based on the import content of consumption for each individual country over 2000–15. The variables in the horizontal axis denote averages over the sample used to estimate the exchange rate pass-through.

Indeed, the extent of second-round effects seems to depend on the inflationary environment. In particular, the higher and the more volatile inflation is, the larger the gap between the exchange rate pass-through and its benchmark. Also, second-round effects are larger where inflation expectations are less anchored. In a more formal analysis of these relationships, Carrière-Swallow and others (forthcoming) address possible reverse causality concerns. The results support the view that the degree of anchoring of inflation expectations is a key determinant of the extent of second-round effects from depreciations on consumer prices.

How were better-anchored inflation expectations achieved in Latin America? Many central banks have adopted an inflation-targeting regime precisely in order to anchor inflation expectations. We explore the relationship between the level of pass-through and monetary regimes by comparing the pass-through estimate for two groups of emerging market economies, with and without inflation-targeting regimes. The pass-through estimates, as well as the average degree of anchoring of expectations over each estimation sample, are reported in Figure 4.9. Inflation expectations are better anchored among countries with inflation-targeting regimes, and the extent of exchange rate pass-through is smaller in these economies. The comparison of results across samples also reveals that, as the degree of anchoring of inflation expectations improved over time, the extent of exchange rate pass-through decreased in both groups.

Figure 4.9.Inflation Targeting, Anchored Inflation Expectations, and Exchange Rate Pass-Through

Source: Authors’ calculations.

Note: Exchange rate pass-through coefficients refer to the estimated increase in the consumer price level two years after a 1 percent increase (depreciation) of the nominal effective exchange rate. Expectation anchoring is inversely related to the disagreement among Consensus Economics inflation forecasts; a higher value is associated with better-anchored inflation expectations. IT denotes countries with inflation-targeting regimes in place. The sample is restricted to 31 emerging markets.

Of course, causal relationships cannot be inferred from these results. But the correlation of pass-through estimates with the inflation environment—and in particular the degree of anchoring of inflation expectations—and with the monetary regime suggests that credible monetary policy, supported by an institutional framework that allows central banks to fulfill their mandate independently of fiscal and other considerations, effectively lowers the exchange rate pass-through to consumer prices. In doing so, it attenuates trade-offs and makes monetary policy easier to implement.

As Figure 4.10 makes clear, exchange rate pass-through continues to vary a great deal across Latin American countries. In many countries in the region, pass-through estimates remain substantially above the import-content benchmark, suggesting that second-round effects remain relevant. Some of these economies have been reluctant to let their exchange rates float for a variety of reasons, including the difficulty they might have in delivering price stability under a more volatile exchange rate. Our results suggest that the recent experience of other emerging market economies offers hope in this regard. As frameworks oriented toward price stability become well established and inflation expectations are anchored, exchange rate pass-through appears to decline substantially. In line with Taylor’s (2000) argument, it seems that there may be virtuous endogeneity between the monetary policy regime and the ease with which it can reach its price stability objectives.

Figure 4.10.Exchange Rate Pass-Through Estimates in Latin America, by Country

(Percent)

Source: Authors’ calculations.

Note: Exchange rate pass-through coefficients refer to the estimated increase in the consumer price level two years after a 1 percent increase (depreciation) of the nominal effective exchange rate. Pass-through estimates for individual countries are obtained from country-specific regressions, while average regional pass-through correspond to panel model estimates. Solid bars denote statistically significant responses at the 10 percent confidence level. ADV = advanced economies; EM = emerging market economies; LA = Latin America; LA5 = Brazil, Chile, Colombia, Mexico, and Peru.

Conclusions

Over the past few decades, Latin America has embarked on a process of removing restrictions on the movement of international capital, leading to stronger integration with global financial markets. This brought important implications for monetary policymaking in the region and required an overhaul of macroeconomic frameworks. Most notably, many countries in the region moved from fixed exchange rate regimes to more flexible arrangements—a necessary condition to maintain monetary policy autonomy in a context of unrestricted capital mobility. The transition to greater financial integration and more flexible exchange rate regimes required, in turn, the establishment of a strong nominal anchor to deliver price stability. Many Latin American countries adopted inflation-targeting frameworks in the late 1990s and early 2000s as a means to establish such an anchor.

Despite claims to the contrary and widespread doubts prior to its implementation, inflation-targeting regimes with flexible exchange rates seem to have delivered the goods, in the sense that they have facilitated a reduction of inflation and inflation volatility and have been associated with increasingly well-anchored inflation expectations. Two considerations underpinned doubts about whether such a monetary framework would be able to succeed in delivering price stability in the region. First, many argued that the predictions of the classical monetary trilemma did not necessarily hold. That is, small and open economies could lack monetary autonomy even if they let their exchange rates fluctuate freely. Second, even if the central bank is able to set policy rates autonomously, monetary policy may fail to deliver price stability and anchor inflation expectations if the sensitivity of domestic prices to the exchange rate is large.

This chapter has presented evidence suggesting that the region’s inflation targeters seem to have overcome these concerns through a virtuous cycle of increased credibility, lower exchange rate pass-through, and stronger autonomy. By better anchoring inflation expectations, the new frameworks facilitated a larger degree of monetary autonomy and a decline in the extent of exchange rate pass-through. This, in turn, enhanced the ability of monetary policy to reach its price stability objectives.

This study has found that the large degree of correlation of interest rates across countries does not necessarily reflect a lack of monetary autonomy, but is largely a consequence of co-movement in business cycles. After controlling for domestic economic conditions, there is much weaker evidence that interest rates in small and open economies—and in Latin America in particular—respond to global financial variables such as U.S. policy rates. But we also find evidence suggesting that certain Latin American economies do not enjoy full autonomy to set monetary conditions according to domestic price and output stability objectives, and are led to follow the financial conditions prevailing abroad.

Importantly, the analysis of cross-country heterogeneity reveals that the degree of monetary autonomy is related to the economic policy framework that is in place. Our results support the predictions of the classical trilemma: more exchange rate flexibility allows for greater monetary autonomy, even when the capital account is unrestricted. But they also emphasize the importance of predictable and credible monetary policy. A stronger nominal anchor—captured by better-anchored inflation expectations—is associated with a higher degree of monetary autonomy.

These findings do not mean that financial spillovers are irrelevant for the region—we do find, for instance, particularly large spillovers to domestic long-term interest rates from fluctuations in the U.S. term premium. But they suggest that, with careful design and implementation of their policy frameworks, central banks in Latin America can enjoy the substantial monetary autonomy they need to achieve their domestic objectives even in the face of shifting foreign financial conditions.

The ability to deliver price stability in Latin America has also been made possible by a substantial reduction in the extent of exchange rate pass-through to consumer prices. Our findings suggest that second-round effects from currency depreciations have reached unperceivable levels in many countries. This, in turn, seems to be largely due to improvements in the inflation environment and, in particular, to the anchoring of inflation expectations that resulted from monetary reforms and the implementation of inflation-targeting frameworks in the region.

References

    AizenmanJ.M. D.Chinn and H.Ito.2008. “Assessing the Emerging Global Financial Architecture: Measuring the Trilemma’s Configurations over Time.NBER Working Paper 14533National Bureau of Economic ResearchCambridge, Massachusetts.

    ArmasA.A.Ize and E.Levy eds. 2006. Financial Dollarization—The Policy Agenda.New York: Palgrave Macmillan for the International Monetary Fund.

    BritoR. D. and B.Bystedt.2010. “Inflation Targeting in Emerging Economies: Panel Evidence.Journal of Development Economics91(2): 198210.

    BursteinA.B.Eichenbaum and S.Rebelo.2005. “Large Devaluations and the Real Exchange Rate.Journal of Political Economy113(4): 74284.

    CaceresC.Y.Carrière-Swallow and B.Gruss.2016. “Global Financial Integration and Monetary Policy Autonomy.IMF Working Paper 16/108International Monetary FundWashington, DC.

    CampaJ. M. and L. S.Goldberg.2005. “Exchange Rate Pass-Through into Import Prices.Review of Economics and Statistics87(4): 67990.

    Carrière-SwallowY.B.GrussN.Magud and F.Valencia. Forthcoming. “Monetary Policy Credibility and Exchange Rate Pass-Through.IMF Working PaperInternational Monetary FundWashington, DC.

    CarstensA. and L. I.Jácome.2005. “Latin American Central Bank Reform: Progress and Challenges.IMF Working Paper 05/114International Monetary FundWashington, DC.

    CéspedesL. F.R.Chang and A.Velasco.2014. “Is Inflation Targeting Still on Target? The Recent Experience of Latin America.International Finance17(2): 185207.

    ChenJ.T.Mancini-Griffoli and R.Sahay.2014. “Spillovers from United States Monetary Policy on Emerging Markets: Different This Time?IMF Working Paper 14/240International Monetary FundWashington, DC.

    ChinnM. D. and H.Ito.2006. “What Matters for Financial Development? Capital Controls, Institutions, and Interactions.Journal of Development Economics81(1): 16392.

    ChinnM. D. and H.Ito.2008. “A New Measure of Financial Openness.Journal of Comparative Policy Analysis10(3): 30922.

    ClaroS. and L.Opazo.2014. “Monetary Policy Independence in Chile.BIS Working Paper 78Bank for International SettlementsBasel.

    De GregorioJ.2014. “Some Challenges for Financial Policies in Emerging Markets.Paper presented at the Conference on Financial Sector Development: Policies to Promote and Strengthen Local Capital MarketsOctober 2–3Graduate Institute, Geneva.

    DevereuxM. B.C.Engel and P. E.Storgaard.2004. “Endogenous Exchange Rate Pass-Through When Nominal Prices Are Set in Advance.Journal of International Economics63: 26391.

    DovernJ.U.Fritsche and J.Slacalek.2012. “Disagreement among Forecasters in G7 Countries.Review of Economics and Statistics94(4): 108196.

    EdwardsS.2015. “Monetary Policy Independence under Flexible Exchange Rates: An Illusion?World Economy38: 77387.

    García-EscribanoM. and S.Sosa.2011. “What Is Driving Financial De-dollarization in Latin America?IMF Working Paper 11/10International Monetary FundWashington, DC.

    GonçalvesC. E. and J. M.Salles.2008. “Inflation Targeting in Emerging Economies: What Do the Data Say?Journal of Development Economics85: 31218.

    GopinathG.2015. “The International Price System.NBER Working Paper 21646National Bureau of Economic ResearchCambridge, Massachusetts.

    IlzetzkiE.C. M.Reinhart and K.Rogoff.2009. “Exchange Rate Arrangements Entering the 21st Century: Which Anchor Will Hold?University of Maryland, College Park, and Harvard UniversityCambridge, Massachusetts. Unpublished.

    International Monetary Fund (IMF). 2012. “The Liberalization and Management of Capital Flows: An Institutional View.IMF Policy PaperWashington, DC.

    International Monetary Fund (IMF). 2015. “To Hike or Not To Hike: Is That an Option For Latin America? Assessing Monetary Policy Autonomy.” Chapter 3 in Regional Economic Outlook: Western Hemisphere.Washington, DCOctober.

    International Monetary Fund (IMF). 2016. “Exchange Rate Pass-Through in Latin America.” Chapter 4 in Regional Economic Outlook: Western Hemisphere.Washington, DCApril.

    JordàÒ.2005. “Estimation and Inference of Impulse Responses by Local Projections.American Economic Review95(1): 16182.

    KleinM. W. and J. C.Shambaugh.2008. “The Dynamics of Exchange Rate Regimes: Fixes, Floats, and Flips.Journal of International Economics75(1): 7092.

    LaneP. and G. M.Milesi-Ferretti.2007. “The External Wealth of Nationals Mark II: Revised and Extended Estimates of Foreign Assets and Liabilities.Journal of International Economics73: 22350.

    MundellR.1963. “Capital Mobility and Stabilization Policy under Fixed and Flexible Exchange Rates.Canadian Journal of Economics and Political Science29(4): 47585.

    ObstfeldM.1998. “The Global Capital Market: Benefactor or Menace?Journal of Economic Perspectives12(4): 930.

    ObstfeldM.2015. “Trilemmas and Tradeoffs: Living with Financial Globalization.” In Global Liquidity Spillovers to Emerging Markets and Policy Responses edited by C.RaddatzD.Saravia and J.Ventura1378. Santiago: Central Bank of Chile.

    ObstfeldM. and K.Rogoff.1995. “The Mirage of Fixed Exchange Rates.Journal of Economic Perspectives9(4): 7396.

    ObstfeldM. and A. M.Taylor.1998. “The Great Depression as a Watershed: International Capital Mobility over the Long Run.” In The Defining Moment: The Great Depression and the American Economy in the Twentieth Century edited by M. D.BordoC.Golden and E. M.White.Cambridge, Massachusetts: National Bureau of Economic Research.

    ReinhartC. and K.Rogoff.2004. “The Modern History of Exchange Rate Arrangements: A Reinterpretation.Quarterly Journal of Economics119(1): 148.

    ReisR.2013. “Central Bank Design.Journal of Economic Perspectives27(4): 1744.

    ReyH.2015. “Dilemma Not Trilemma: The Global Financial Cycle and Monetary Policy Independence.NBER Working Paper 21162National Bureau of Economic ResearchCambridge, Massachusetts.

    TaylorJ. B.2000. “Low Inflation, Pass-Through, and the Pricing Power of Firms.European Economic Review44(7): 1389408.

    TowbinP. and S.Weber.2013. “Limits of Floating Exchange Rates: The Role of Foreign Currency Debt and Import Structure.Journal of Development Economics101: 17994.

See Armas, Ize, and Levy (2006) for a discussion of policies to deal with financial dollarization, and Chapter 9 for an in-depth discussion of the Peruvian case.

For instance, Gonçalves and Salles (2008) find that developing countries that adopted inflation targeting benefited from a larger fall in inflation and inflation volatility than those that did not. However, Brito and Bystedt (2010) find that this reduction in inflation came at the cost of lower average output growth. So while it may be true that inflation targeting allowed central banks to fight inflation more aggressively, it may not have reduced the costs associated with disinflation in terms of economic activity. See Céspedes, Chang, and Velasco (2014) for a review of the recent experience with inflation targeting in a subset of Latin American countries.

The analysis in this section was originally reported in IMF (2015).

This principal component corresponds to the linear combination of the underlying series that best summarizes the total variance in the data.

The quantitative exercises in this chapter are focused on the effects of changes in U.S. interest rates, as these are a key driver of global financial conditions (see, for instance, Rey, 2015).

All model specifications share the assumption that domestic variables do not affect global variables, as is thought to be the case for Latin America’s small and open economies.

We use interest rates on short-term government bonds (with a maturity of about three months). Even if this is not the monetary policy instrument, the short-term market interest rate should be closely linked to changes in the monetary policy stance.

Throughout the chapter we focus on the estimated cumulative impulse response functions after 12 months to allow transmission to be fully realized.

The justification for the use of this procedure over other empirical methods for assessing monetary autonomy is described in detail in Caceres, Carrière-Swallow, and Gruss (2016).

To ensure that forecasts are predetermined with respect to policy decisions, we use lagged market forecasts, even though this may come at the cost of reducing their information content further.

Consider the case of a central bank that decides to increase interest rates in the face of a shock that would otherwise lead to exchange rate depreciation. Our procedure identifies the part of the rate increase that can be explained by its concern for the second-round effects on inflation, as captured by its historical behavior. The remainder is considered unexplained, even though it could correspond to an explicit intent to contain vulnerabilities from balance sheet mismatches in order to preserve financial stability.

It should be noted that these estimates reflect historical average effects, and thus cannot fully capture improvements in policy frameworks that have been implemented since 2000.

This finding is in line with the panel analysis of Obstfeld (2015) for a similar broad sample of countries, and with the narrative approach in Claro and Opazo (2014) and De Gregorio (2014) for the case of Chile.

The degree of anchoring of inflation expectations is constructed as the inverse of a four-year moving average of the standard deviation of inflation forecasts reported by Consensus Economics at a 12-month fixed horizon. This metric is thought to provide a proxy for monetary policy credibility, since the more predictable a central bank’s reaction function is, the less likely are forecasters to disagree about the future path of inflation. Disagreement has also been found to be closely related to de jure measures of central bank independence (Dovern, Fritsche, and Slacalek 2012).

The analysis in this section was originally reported in IMF (2016). The transition to flexible exchange rates in the region was not always an orderly and gradual process, and in many cases came amid a traumatic event characterized by a sharp depreciation, aggravating concerns about pass-through. See Carstens and Jácome (2005) and Chapter 2 in this volume for discussions of central bank reforms in the region.

Similarly to Gopinath (2015), we use a nominal effective exchange rate metric with weights given by imports rather than by total trade, and allow these weights to vary each year.

World inflation or trade-weighted consumer prices are common alternatives to control for changes in exporting countries’ production costs. The drawback of those alternatives is the preponderance of nontraded goods and services in consumer price indices. Trade-weighted export prices are also problematic, as they may already reflect pricing decisions by exporters.

The import content of domestic consumption includes the share of imported goods in the consumption basket, but also the share of intermediate inputs in the production of domestic goods and services that are consumed domestically.

Similarly to Burstein, Eichenbaum, and Rebelo (2005) and Gopinath (2015), we measure the total import content of households’ final demand using input-output tables. See Carrière-Swallow and others (forthcoming) for more details.

Carrière-Swallow and others (forthcoming) report an alternative benchmark constructed by estimating the exchange rate pass-through to import prices in local currency.

    Other Resources Citing This Publication