Journal Issue

Russian Federation: Selected Issues

International Monetary Fund. European Dept.
Published Date:
August 2015
  • ShareShare
Show Summary Details

Exchange Rate Pass-Through to Inflation. is Russia Different?1

The size of the exchange rate pass-through to inflation is relevant for timely and appropriate policy responses. Estimating the exchange rate pass-through in Russia is difficult because nominal exchange rate fluctuations have been relatively small in the recent past. Large import shares in the CPI basket, and a new exchange rate regime pose challenges to identify underlying inflation, and hence estimate the size of the exchange rate pass-through. This paper estimates exchange rate pass-through to consumer prices for emerging markets using local projection techniques. There is significant evidence of non-linearities and asymmetries. Monetary policy should take into account the time varying, and state dependent nature of the exchange rate pass-through to inflation.

A. Introduction

1. The size of the exchange rate pass-through is relevant for policy makers. Goldberg and Knetter (1997) define exchange rate pass-through (ERPT) as the percentage change in local currency import prices resulting from a 1 percent change in the exchange rate between the exporting and importing countries. Estimating the size of exchange rate pass-through and understanding the main drivers of domestic prices in advanced economies have been amongst the most debated topics in academic and policy circles in recent decades. Less is known for emerging markets (EMs). Having a good understanding and correctly measuring the transmission mechanism from exchange rate fluctuations to domestic prices is crucial for policymakers to implement appropriate, timely and effective policy responses.

2. It is very hard to estimate the exchange rate pass-through to inflation, after a long period under a relatively rigid exchange rate regime. The main challenge is that nominal exchange rate time series do not exhibit enough volatility. A large variety of methods can be used, but it is not possible to find meaningful estimates, if exchange rate fluctuations are small. For emerging open economies, the challenges are even bigger due to several elements. Large exogenous fluctuations (i.e., terms of trade shocks) and structural breaks pose an additional challenge to identify underlying inflation. Large shares of food and imports in consumption and investment goods, add an extra layer of complication. Hence, accurate estimations of the exchange rate pass-through to consumer prices in Russia are challenging.

3. Pass-through estimates for EMs can be a useful benchmark for Russia. Major EM currencies depreciated significantly during 2014 due to a variety of shocks and policy responses (Figure 1). The aim of this paper is to document and estimate ERPT to domestic prices in EMs focusing on three specific dimensions: non-linearities, asymmetries and importance of exchange rate regimes. Using local projection techniques, we estimate state dependent impulse responses for the exchange rate pass-through to consumer prices in a panel of 28 emerging countries. We find that the percentage change in consumer prices due to a one percent change in the nominal effective exchange rate is 22 percent after 12 months. We also find evidence of non-linearities and asymmetries.

Figure 1.Exchange Rate Depreciation in Emerging Markets

4. Russia changed its nominal anchor amid significant external shocks. The central bank of Russia (CBR) changed Russia’s nominal anchor in November 2014 as part of the formal move to an inflation targeting regime. As a result, the FX rule to conduct regular interventions was formally eliminated and CBR’s predetermined FX intervention policy came to a halt. As geopolitical tensions rose and oil prices tank, pressures on the ruble intensified during November and December 2014. Consequently, under the new floating exchange rate regime, and as a result of these 2 big exogenous shocks, the nominal exchange rate depreciated substantially in the course of 2 months.

5. ERPT in Russia is time varying. Contrary to expectations, the EPRT following the ruble’s recent sharp depreciation turned out to be big, and fast (Figure 2).2 The large import component in both tradable and non-tradable goods appears as an important element explaining the fast surge in inflation. Both consumer and producers prices rose sharply during (and after) the Ruble depreciation. Identifying, and measuring the ERPT in Russia after the implementation of the new nominal anchor is difficult. Hence focusing on a sample of EMs comparable to Russia, can shed light to better understand the time varying nature of the ERPT.

Figure 2.Exchange Rate Pass-through to Inflation in Russia

B. Stylized Facts

6. Stylized facts are useful to pin down the main differences between EMs and AEs. Based on unconditional relationships, it is possible to identify 4 stylized facts related to inflation levels, volatility, and the relationship between inflation and exchange rate movements, as well as the relationship between inflation volatility and exchange rate volatility. Figure 3 plots the average inflation, depreciation/appreciation, and the corresponding volatilities for each country, over the sample period 1980-2014.

Figure 3.Inflation, Depreciation and Respective Volatilities

7. Four stylized facts that distinguish EMs from AEs:

  • Inflation appears to be higher and more volatile in EMs, compared to AEs.

  • Higher inflation seems to be associated with higher depreciation rates in EMs, compared to AEs.

  • Depreciation rates appear to be positively associated with depreciation volatility, whereas for AEs there is no apparent relationship.

  • Depreciation volatility and inflation volatility appear to be positively associated in EMs, which does not appear to be the case for AEs.

8. ERPT tends to be higher in EMs than in AEs. Taylor (2000), similarly, links the degree of ERPT to the exchange rate regime and more specifically to the presence of inflation targeting. After the adoption of an IT framework, countries tend to experience a lower the pass-through. The rationale is that IT succeeds in keeping inflation low and this, through expectations of persistent low inflation, pushed firms to keep their prices broadly constant in order to be able to remain competitive. In other words, a low inflation environment causes a reduction in firms’ pricing power that in turn leads to a decline in ERPT. Ca’ Zorzi et al. (2005) argue that the private sector in EMs has fewer hedging instruments available. In a not fully competitive market, this could imply that exchange rates fluctuations are transmitted more into prices. Recent empirical studies document a decrease in ERPT in EMs due to the adoption of IT.

9. EMs tend to be more exposed (and sometimes are more vulnerable) to terms of trade shocks. Compared to AEs, commodities account for a large share in production and exports of many EMs. Terms of trade shocks are sometimes translated to abrupt and relatively large fluctuations in exchange rates. The share of commodity prices (i.e., food and fuel) in EMs consumer price index (CPI) baskets tends to be larger than in AEs. Commodity prices fluctuations can be big and volatile, hence introducing sudden and, often times, large movements in exchange rates, domestic import prices and domestic inflation.

10. Asymmetries, non-linearities, and diffrences across exchange rate regimes exist. The size of the pass-through, and of course the sign, can be different depending on whether the exchange rate is appreciating or depreciating. The size of the pass-though may also be different depending on the size of the depreciation/appreciation. The specific choice of the nominal anchor can also affect the size of the pass-through. Under flexible exchange rate regimes, where the monetary authority targets some monetary aggregate, the size of the pass-thourgh may be different than under inflation targeting. Last, but not least, credibility in the conduct of monetary policy can also affect the size and speed of the pass-through.

Figure 4.Assymetries, Non Linearities and Inflation Targeting

C. Results. Size and State Dependent Nature of the ERPT in EMs

11. The exchange rate pass-through in EMs is about 22 percent after 12 months. Estimating the baseline model for the full sample of EMs (28 countries) indicates that exchange rate pass-through on consumer prices is around 22 percent after 12 months after the initial shock and it reaches 25 percent after two years.3

Table 1.ERPT Coefficients — Linear Model

12. EMs are usually exposed to terms of trade shocks, which in turn affect their exchange rate. The size, frequency and nature of these shocks varies across time and across countries, but the key issue for (appropriate) policy responses is to have a good understanding of non-linearities. It is crucial to have a sense of whether the effect of an exchange rate shock on domestic prices can be different (in size) depending on the magnitude of the shock.4 In line with the stylized facts presented above, the pass-through from exchange rate fluctuations to domestic prices appears to be higher, the larger the change in the exchange rate (Figures 5 and 6).

Figure 5.Exchange Rate Pass-through during 10 Percent Depreciation Episodes

Figure 6.Exchange Rate Pass-through during 20 Percent Depreciation Episodes

Table 2.Exchange Rate Pass-through Coefficient in the Non-linear Model
Horizon - monthsERPT – 10% episodesERPT – 20% episodes

13. The response of inflation to large depreciations is faster and larger than during normal times. After 1 month from the initial shock the ERPT is almost 20 percent and it reaches 40 percent after 6 months. Depreciations of at least 20 percent or more have an even faster effect on prices.5 ERPT for depreciations’ episodes of at least 10 percent or more, is equal to 40 after 6 months and 60 percent after 12 months, while in normal times it is 11 percent. For depreciation episodes of at least 20 percent or more, the ERPT is equal to 43 percent after 6 months and 50 percent after 12 months. So, for larger depreciations, the initial effect is larger but after few months the exchange rate pass-through stabilizes around 40 percent.

14. There is significant evidence of asymmetries in the first 8 months after the initial shock: Appreciation episodes generate a positive reaction in inflation that is not statistically significant. Depreciation episodes are characterized by about 40 percent pass-through after 12 months compared to less than 10 percent for appreciation episodes.

Figure 7.ERPT during Appreciation versus Depreciation

Table 3.Coefficients Appreciation versus Depreciation

15. The EPRT for inflation targeters is considerably lower than the one for non-inflation targeters. It is well documented that inflation targeting regimes tend to be associated with lower inflation volatility and lower pass-through from depreciation to inflation, compared to non-inflation targeting regimes (i.e., predetermined exchange rate regimes). Non inflation targeters display more than 20 percent pass-through after 12 months.6

Figure 8.ERPT during Inflation Targeting versus Other Regimes

D. Conclusion and Policy Implications

16. Pass-through from depreciation to inflation in emerging markets is time varying. Asymmetries non-linearities, and the specific exchange rate regime (i.e., fixed or floating) do affect the size of the pass-through. Emerging Markets display about 22 percent pass-through from depreciation to inflation after 12 months, and about 40 percent after 12 months when: (i) taking into account asymmetries between depreciation and appreciation, (ii) distinguishing between inflation targeters and non-inflation targeters, and (iii) depreciations are large (more than 10, and/or 20 percent yoy). These results confirm that pass-through from exchange rate to domestic prices is higher in emerging markets than in advanced economies.

17. Assymetries, non-linearities and a new nominal anchor call for prudent policies. The main policy implication to be drawn from the evidence presented in this paper is that policy makers should exert caution in conducting monetary policy after (and during) large depreciation episodes, especially after changes in the economy’s nominal anchor. The intuition is simple: nominal variables can react differently (in terms of size, speed, transmission mechanisms, required adjustments, etc) under alternative exchange rate regimes, and depending on the size of the shocks. Appropriate policy reactions should take this into consideration and avoid assuming time invariant, and/or state independent parameters to characterize the response of nominal variables after large depreciation episodes. A natural implication of our findings is that monetary authorities should be cautious when assessing the transmission (and its speed) of exchange rate shocks to inflation. Premature policy responses, without sufficient evidence and understanding of non-linear and state contingent dynamics, could prove detrimental for price stability and pose a challenge to anchoring inflation expectations.

Appendix I. Empirical Strategy

1. Jordà (2005) local projection method (LPs) is flexible and easy to implement. It enables to estimate the dynamic response of inflation to exchange rate movements allowing to capture non-linearities and asymmetries. Asymmetries are defined as the difference between appreciation and depreciation episodes and nonlinearities as: (i) depreciation above or below certain thresholds, and (ii) countries operating under IT vs. others without an explicit IT framework. LPs are a flexible semi-parametric technique to estimate impulse responses which directly estimate a sequence of linear projections of the future value of the dependent variable on the current information set (Killian and Kim 2009).

2. LPs technique is useful for tracing the dynamic response of variables to a shock. LPs methods (as opposed to a standard vector autoregressive model—VAR) do not involve any non-linear transformation of the estimated slope coefficients to obtain impulse responses, and dynamic multipliers depend only on the quality of the local approximation (Jordà et al. 2013). Compared to VARs, LPs regressions are more robust to lags misspecification. If a VAR is a poor representation of the data generating process (DGP), impulse response functions (IRFs) are biased.

3. LPs methods have also some limitations. First, as the forecasting horizon increase, observations from the end of the sample are lost. Second, the IRFs obtained from LPs methods might show significant oscillations at longer horizons. At long horizons, LPs produce substantial oscillations that are not present in other methods. Due to the limitations of LPs methods in small samples, highlighted by Killian and Kim (2009) we use monthly data, to guarantee the longest possible sample.

4. The baseline specification for the linear model is:

LPs generate new estimates for each forecast horizon h, regressing the dependent variable at t+h on the available information set at time t. IRFs are the estimated slope coefficients of the projections. Where δcpii,t is the yoy percent change in the CPI of country i at time t, δNEERit is the yoy percent change in the nominal effective exchange rate, crisist is a dummy equal to one from 2009 to 2012 to proxy for the financial crisis, Δcppii,t* is a proxy for foreign prices. εi,t+h is an error term capturing all other sources of variation in inflation between t and t+h. The coefficient βh traces the response of inflation at time t+h to a depreciation/appreciation occurred in time t.

5. Country by country estimations turn out to be unstable and not robust. For Russia, this is especially true given that prior to 2015 the exchange rate regime, and the nominal anchor of the economy were different. Hence, panel estimations may shed some light to understand the relationship between exchange rate fluctuations and domestic prices across crountries. The intuition is simple: a panel delivers a larger number of observations and, in principle, more variability.

6. The effect on domestic prices can be larger during episodes of “large” depreciations. In order to test whether large (relative to “normal times”) depreciations have more than proportional effects on domestic prices an episode of “high” depreciation can be defined according to two alternative criteria. Depreciation is large when:

  • The monthly year-on-year (yoy) percent change is larger than 10 percent, or

  • The monthly yoy percent change is larger than 20 percent.

A dummy variable depisode is defined as follows:

Where Ψ is a threshold defined according to 1 or 2 above.

Introducing the interaction term between the dummy variable “depisode” and the rate of depreciation, we allow the coefficient βh to be different across periods of “high” depreciation versus “normal times”.

7. Depreciation and appreciation episodes have asymmetric effects on inflation. A depreciation can cause inflation to increase, but an appreciation episode need not reduce inflation by the same proportion. In order to explicitly account for this asymmetry between appreciation and depreciation the following specification can be used:

Where ddepr is defined as follows:

Hence the exchange rate pass-through to domestic inflation during depreciation period is:

8. EPRT can be different under different exchange rate regimes. A common, and well documented result in the literature on EPRT, is that inflation targeting, succeeding in keeping inflation low, causes a reduction in firms’ pricing power that in turns lead to a decrease in the ERPT. The following specification test this hypothesis:

Where IT is defined as follows:

Hence the exchange rate pass-through to inflation in targeting regimes is:


Prepared by Francesca Caselli and Agustin Roitman.

Before the sharp depreciation of the ruble at the end of 2014, most ERPT estimates were around 0.1 to 0.15. These estimates probably relied on historical data (and likely fail to acknowledge the time varying nature of the ERPT).

For details on the methodology, estimation, and alternative specifications see Appendix 1.

In other words, whether a say, 5 percent depreciation would have the same pass-through as a say, 20 percent depreciation on domestic prices.

See Appendix 1 for details.

See Appendix 1 for details.

Other Resources Citing This Publication