An economy's real exchange rate plays an important role in resource allocation as it determines the relative price of tradable and nontradable goods. Exchange rate misalignment occurs when the real exchange rate becomes disassociated from its equilibrium value, which may arise due to an economic shock with a significantly destabilizing effect, frictions in the foreign exchange market that prohibit efficient price discovery, or a policy intervention such as supporting a currency peg. The misalignment results in misallocation of resources between the tradable and nontradable sectors, and tends to create an unsustainable balance of payments position, possibly leading to a currency crisis.
A voluminous literature with extensive empirical evidence has shown that prolonged exchange rate misalignment, in particular currency overvaluation, has a negative effect on growth, export performance and the general outward orientation of the economy. In addition, it has been a triggering factor for balance of payments and currency crises in developing countries, which necessitated substantial economic adjustment in the aftermath of these crises.
Against the backdrop of the currency crisis that Egypt experienced during 2016, and given the aforementioned implications of exchange rate misalignment, this paper aims to estimate Egypt's equilibrium real exchange rate and exchange rate misalignment over the period 2001Q3–2017Q3. To the best of our knowledge, this is the first attempt to estimate the intensity and duration of exchange rate misalignment during the currency crisis of 2016, and it marks the first contribution of this paper.
The crisis led to a sizable exchange rate devaluation when the Central Bank of Egypt (CBE) announced the adoption of a floating exchange rate regime as part of a wide-raging economic reform program in agreement with the International Monetary Fund (IMF) to rein in the fiscal deficit and revive economic competitiveness. Our attempt to estimate the exchange rate misalignment is a bid to provide deeper understanding of how the crisis emerged and unraveled, and also to offer lessons for the future conduct of monetary policy in Egypt.
In addition to estimating the historical exchange rate misalignment, we also show how the current misalignment is likely to evolve over the medium term (specifically over the period 2017Q4–2020Q4), which marks the second contribution of this paper. Offering an outlook for the future path of the equilibrium real exchange rate constitutes a departure from earlier studies on the Egyptian economy in which the future outlook was disregarded. We chart out the course of the equilibrium real exchange rate using the latest IMF projections of the economic fundamentals for Egypt and its main trade partners.
Estimating the equilibrium real exchange rate and exchange rate misalignment has been the subject of a large literature adeptly summarized in Edwards (1994), Hinkle and Montiel (1999), and Williamson (1994). There are generally two strands in this literature: the so-called behavioral approach based on Edwards (1989b), and the fundamentals approach of Williamson (1985), where their primary difference pertains to the role ascribed to current account sustainability in the analysis. A third but less widespread method, known as the “external sustainability” approach, is outlined in International Monetary Fund (2006) and it links the equilibrium real exchange rate to a sustainable net foreign assets position.
Both the fundamentals approach of Williamson (1985) and the external sustainability approach discussed in IMF (2006) require an assumption about the level of some exogenous variable that is consistent with the simultaneous attainment of internal and external balance.1 This exogenous variable is the level of capital inflows in the case of Williamson (1985), and the economy's net foreign assets in the external sustainability approach. This has been a point of criticism of both methods since the resulting estimate of exchange rate misalignment is highly sensitive with regard to this assumption.
The approach adopted in this paper is based on the behavioral approach of Edwards (1989b), and its extensions in Elbadawi (1994), Baffes, Elbadawi, and O'Connell (1999), Edwards and Savastano (2000), and Edwards (1994) among others. The appeal of this method relates to its strong theoretical foundations, which we discuss in detail in the body of the paper, in addition to its suitability in a developing economy context as emphasized by Clark and MacDonald (1998). For the objectives of this paper, the Edwards (1989b) model offers three distinct advantages. First, it explicitly highlights the interaction between exchange rate misalignment and changing economic fundamentals, which deepens our understanding of the confluence of factors behind Egypt's recent currency crisis. Second, it circumvents the need to make ad hoc assumptions about some exogenous variable (e.g. the level of capital inflows, or net foreign assets) that is consistent with internal and external balance. Third, and quite importantly for our objectives, it allows for constructing projections of the equilibrium real exchange rate and exchange rate misalignment based on forecasts of the fundamentals.
Our empirical results show that right before the floatation of the Egyptian pound on 3 November 2016, the real exchange rate was overvalued by 24% in 2016Q3. The floatation of the currency led to a significant depreciation in which the currency lost about 32% of its value in real terms, which led to a reversal from overvaluation to undervaluation of 14.7% in 2016Q4. The undervaluation increased further to 22.3% in 2017Q1 before retreating to 18.5% by 2017Q3. This was caused by exchange rate overshooting which is in line with the predictions of the Dornbusch (1976) model. It is also consistent with the exchange rate behavior during the preceding currency crisis of 2002–2003, in which successive devaluations were effected to correct for exchange rate misalignment at the time.
According to our projections for 2017Q4–2020Q4, the current exchange rate undervaluation is expected to disappear rapidly. Since Egypt's inflation is projected to remain high (relative to trade partners), the real effective exchange rate will witness a gradual increase, reaching the equilibrium level by 2019Q1, at which time the undervaluation would cease to exist. Starting 2019Q2, the misalignment will switch sign to indicate currency overvaluation. Our projections show that if the nominal exchange rate stabilizes at its level in 2017Q3 (i.e. 17.73 pounds per US dollar), the currency will be overvalued by 13.1% in 2020Q4. If, on the other hand, the nominal exchange rate tracks its projected equilibrium path, we expect to see some nominal depreciation. Specifically, the exchange rate for the US dollar is expected to reach 20.05 pounds per US dollar by 2020Q4.
The projections are of course subject to a margin of uncertainty, which is further amplified by a potential structural break given the adoption of a new exchange rate regime. Sensitivity analysis undertaken by combining forecasts from different vintages of estimation around 2016Q4 (the time of the regime change) shows that the projected nominal exchange rate may be in the range of 18.21–18.83 pounds per US dollar by the end of 2020 if less weight is placed on the 2017Q3 data in estimation. The wide range in the projections is reflective of the fact that it is an inherently difficult task to accurately measure exchange rate misalignment around times of a regime change.
The rest of the paper is organized as follows: Section 2 provides the theoretical background and discusses the set of economic fundamentals considered for the estimation of the equilibrium real exchange rate. Section 3 provides a brief overview of exchange rate developments in Egypt during the period of study, while Section 4 outlines the econometric methodology. Section 5 presents the empirical results for the in-sample analysis, and Section 6 provides the medium-term projections. Section 7 concludes the paper and highlights policy implications for the future conduct of monetary policy in Egypt.
2 Theoretical Framework
In this section, we present the theoretical framework starting with a definition of the nominal and real effective exchange rates, followed by a discussion of the concept of the equilibrium real exchange rate, and how to estimate it. The nominal effective exchange rate (NEER) is a weighted average of bilateral nominal exchange rates for the local currency against the currencies of the main trade partners. For the
One limitation of the NEER is that it does not control for the price level differentials between the local economy and the economies of the trade partners, which is an important dimension of how economic competitiveness evolves over time. The real effective exchange rate (REER), which addresses this shortcoming, is given by
where the price level at time
In theoretical macroeconomic models, the REER is often defined as the relative price of nontradables to tradables. Let
Nurkse (1945) was among the first economists to discuss the concept of the equilibrium real exchange rate (ERER), which he defined as the real exchange rate value that simultaneously achieves internal and external balance. Edwards (1989b) utilized this concept to outline a coherent methodology to estimate the ERER using observed economic fundamentals. The model of Edwards (1989b), hereinafter referred to as the Edwards model, along with its extensions in Elbadawi (1994), Baffes, Elbadawi, and O'Connell (1999), and Edwards and Savastano (2000), have been a workhorse for a large body of applied research on exchange rate misalignment.
In this paper, we utilize the Edwards model to estimate Egypt's ERER and use it to obtain a measure of the historical exchange rate misalignment. The single equation that arises from the Edwards model is of the form:
Based on the existing literature, with particular emphasis on the literature focusing on developing countries, we consider six determinants of the ERER which constitute the variables in
The productivity differential is one crucial factor that may cause systematic deviations from the purchasing power parity-implied exchange rate over the short and medium terms. The acknowledgement of its importance as a fundamental determinant of the ERER is due to the seminal works of Balassa (1964) and Samuelson (1964). Both studies highlight the shortcomings of the purchasing power parity theory of the exchange rate, and argue that the productivity growth differential between the tradables and nontradables sectors can change relative prices in the economy, thereby inducing movements in the real exchange rate.
The Balassa-Samuelson effect is illustrated in a two-sector, two-good economy, denoting the tradables and nontradables sectors. Market forces are assumed to prevail in the two sectors, and the law of one price is assumed to hold for the tradables sector. It is also assumed that wages in the tradables sector are determined by the productivity level in that sector, with wages equalizing across the two sectors. In this highly stylized setting, if wages rise in the tradables sector due to a productivity increase, the wage level will also increase in the nontradables sector. Since there was no matching productivity increase in the latter, this results in an increase in the price of nontradables, which in turn leads to an appreciation of the real exchange rate.
The more recent literature has shown that deviations from the Balassa-Samuelson assumptions make the impact of the productivity differential on the real exchange rate rather ambiguous. This occurs since the real appreciation due to the Balassa-Samuelson is often countered, and sometimes dominated by factors with a depreciating effect on the real exchange rate due to the law of one price failing to hold in practice. These factors include home bias in the consumption of tradables (Benigno and Thoenissen 2003; Lee and Tang 2003), imperfect substitutability and product variety (MacDonald and Ricci 2002), a large nontradable component in the tradable good (MacDonald and Ricci 2001), and the impact of price liberalization during economic transitions (Froot and Rogoff 1995); see Égert, Halpern, and MacDonald 2006a for a detailed discussion. While the majority of studies have documented that an increase in the productivity differential tends to have a positive impact on the real exchange rate, Égert, Lommatzsch, and Lahrèche-Révil (2006b) showed that the evidence is mixed.
The Balassa-Samuelson effect identifies supply-side factors as determinants of the ERER. As argued by Froot and Rogoff (1995), and as demonstrated in Ostry (1994) and Lane and Milesi-Ferretti (2004), demand side factors such as an increase in government consumption or investment may also influence relative prices, which in turn affects the ERER. An increase in either variable may induce a real appreciation if this increase tends to be more heavily associated with the demand for nontradables, which increases their relative price.
A country's terms of trade, defined as the unit price of exports relative to the unit price of imports, also influences the ERER, but its impact is undetermined a priori due to the presence of income and substitution effects given a change in the terms of trade (see Edwards 1989b for a discussion). On the one hand, a decline in the terms of trade triggers a negative income effect with the falling purchasing power shrinking the demand for nontradables, which lowers their relative price. On the other hand, a substitution effect shifts consumption from tradables to nontradables thereby increasing the price of the latter. Empirical evidence for developing economies suggests that the income effect is generally dominant.
Given the connection between the real exchange rate and external balance, the economy's stock of net foreign assets is another important determinant of the ERER. A decline in net foreign assets indicates that an adjustment to the trade balance is needed to ensure sustainable servicing of foreign liabilities and this adjustment is typically induced by a nominal depreciation (see Obstfeld and Rogoff 1995b). Lane and Milesi-Ferretti (2004) show that a deterioration in a country's net foreign assets position tends to be associated with real depreciation, and they provide supporting cross-country evidence. On the other hand, Égert, Lommatzsch, and Lahrèche-Révil (2006b) argue that the impact of net foreign assets on the real exchange rate may go in the opposite direction during periods of economic transition. They showed that exchange rate appreciation may co-exist with a deterioration in net foreign assets in transition economies with initially low levels of indebtedness, and given an improved outlook on future growth, which generally indicates higher ability to repay accumulated debt in the long run.
Finally, trade openness also affects the ERER over the medium term. Often accompanied by trade liberalization, trade openness lowers the cost of imports, which shifts demand from nontradables to tradables. This in turn leads to a decline in the price of nontradables causing a real depreciation. Obstfeld and Rogoff (1995a) and Hau (2002) also show that closed economies are subject to greater real exchange rate volatility given that they have higher overall price rigidity due to an oversized nontradables sector.
Changes in the economic fundamentals induce movements in the ERER, but these changes are expected to occur in a gradual fashion, since cyclical variation and transitory noise in these variables are ignored when constructing the ERER estimate. Generally, an increase in the ERER is an indication that the economic fundamentals are changing in a way that supports a real appreciation of the currency. This could be due to a surge in government spending, an increase in investment that unevenly affects the demand for nontradables, or a build-up of net foreign assets. On the other hand, an increase in trade openness is expected to lead to a lower ERER (i.e. a depreciated real exchange rate) to be consistent with the demand shift from nontradables to tradables. The observed real exchange rate may of course deviate from the ERER, however it is severe and prolonged misalignment that imparts drastic effects on the economy.
The effects of exchange rate misalignment have been well documented in the literature. One notable impact is the negative impact of exchange rate overvaluation on economic growth as demonstrated in Cottani, Cavallo, and Khan (1990), Dollar (1992), Fischer (1993), Ghura and Grennes (1993), Easterly, Loayza, and Montiel (1997), Razin and Collins (1999), and Rodrik (2008). Razin and Collins (1999) further document evidence of nonlinearity in the impact of exchange rate misalignment on growth: only high levels of overvaluation hurt growth, while moderate to high undervaluation seems conducive to higher growth. Dollar (1992) and Sachs and Warner (1995) also establish a negative connection between exchange rate overvaluation and the overall level of economic openness. Rodrik (2008) argues that developing economies with undervalued currencies tend to have a large tradables sector, relative to the nontradables sector, which has a positive impact on economic growth over the long run.
In addition to its long term impact on growth and economic openness, exchange rate misalignment has been a culprit in many currency crises. In a seminal paper, Krugman (1979) outlines the anatomy of how, under a fixed exchange rate regime, a balance of payments crisis manifested in falling international reserves may lead to a full-fledged currency crisis where a speculative attack on the domestic currency results in a significant devaluation. Empirically, Frankel and Rose (1996) and Kaminsky and Reinhart (1999) find currency overvaluation to be a significant predictor of currency crashes.
While the causes for exchange rate misalignment have varied historically across countries, there is strong evidence that adopting inflexible exchange rate regimes, particularly a fixed exchange rate, is a significant contributing factor (see, for example, Coudert and Couharde 2009 and Holtemöller and Mallick 2013). Under a fixed exchange rate regime, the real exchange rate tends to witness appreciation, particularly if inflation in the home economy significantly exceeds that in the economies of the trade partners.
Previous studies of exchange rate misalignment which focused exclusively on Egypt include Mongardini (1998), Mohieldin and Kouchouk (2003), Hosni (2015), and Hosni and Rofael (2015). Despite studying different time periods, a main finding in these studies is the presence of significant exchange rate misalignment around times of economic turmoil, specifically during the late 1980s and late 1990s. Both periods ended with a significant nominal depreciation in the Egyptian pound, and with the exchange rate overshooting its equilibrium value. Mohieldin and Kouchouk (2003) argue that the misalignment was a result of an unsustainable mix of fiscal and monetary policies that was incompatible with the fixed exchange rate regime adopted historically in Egypt. These studies corroborate the significance of the above-mentioned six economic fundamentals, with varying degrees of significance given different sample periods.
Also, Drine and Rault (2003) and Brixiova, Égert, and Essid (2014) considered Egypt in cross-country studies of exchange rate misalignment. Drine and Rault (2003) showed that government consumption and the level of trade openness are significant determinants of the ERER, while Brixiova, Égert, and Essid (2014) found net foreign assets, trade openness and the terms of trade to be the most important economic fundamentals with regard to determining the ERER.
3 Recent Exchange Rate Developments in Egypt
In this section, we document Egypt's exchange rate developments during the period of study, which is 2001Q3–2017Q3. Figure 1 includes the historical nominal exchange rate for the US dollar (top left panel), Egypt's trade-weighted inflation differential with its trade partners (top right panel), the NEER (bottom left panel) and the REER (bottom right panel).3 During the period 2001Q3–2004Q4, the nominal exchange rate against the US dollar depreciated by more than 50% from 4.04 pounds per dollar to 6.22 pounds per dollar. This was due to a series of successive devaluations to counter pressure on the exchange rate due to overvaluation at the time (see Mohieldin and Kouchouk 2003; Hosni and Rofael 2015). Afterwards, the pound strengthened to 5.35 pounds per dollar in 2008Q3 before gradually losing ground to the US dollar, witnessing a sharp depreciation after it was floated in 2016Q4. Since the floatation, the nominal exchange rate reached 18.03 pounds per dollar in 2017Q2 before appreciating slightly in 2017Q3 recording 17.73 pounds per dollar.
The NEER shows a similar trend signaling that the Egyptian pound witnessed a general depreciating trend against the currencies of all trade partners. Recall that the NEER and REER series are constructed such that an increase means appreciation. With regard to the REER, the period 2001Q3–2003Q4 shows real depreciation in the currency and an improvement in external competitiveness due to successive devaluations. However, starting 2004, the REER started to appreciate due to the rising rate of inflation in Egypt relative to its trade partners (see the top right panel), and also partly due to a nominal appreciation in the Egyptian pound during 2005Q1–2008Q3.4
As the inflation differential between Egypt and its trade partners declined with the economic downturn in Egypt starting 2011, coupled with a gradual nominal depreciation of the pound, the REER went down slightly before witnessing a significant increase since the beginning of 2014. During 2014Q1–2015Q4, the currencies of most of Egypt's trade partners witnessed a strong retreat against the US dollar.5 Since the Egyptian pound witnessed only a mild depreciation against the US dollar (around 13% during this period), this resulted in a significant rise in the REER. In addition, Egypt's inflation recorded an average of 9.8% compared to a weighted-average rate of inflation at about 1.7% for Egypt's main trade partners.
As the REER showed significant appreciation during this period, and as it became strongly detached from its equilibrium level as we show later in the empirical analysis, a parallel currency market started to emerge, which gradually acquired an increasing share of foreign exchange transactions. The parallel market exchange rate premium increased considerably during the summer months of 2016 amid a drying up of foreign exchange liquidity (see the top left panel of Figure 2).6 By the end of the summer of 2016, the majority of foreign exchange transactions in Egypt were conducted at the parallel market rate.
Against the background of a general deterioration in external accounts starting FY 2010/2011 (see the top right panel of Figure 2), and the exchange rate becoming significantly overvalued since the beginning of 2015, the attempt to defend the currency peg led to a decline in international reserves. Reserves went down from a peak level of USD 35.2 billion in FY 2009/2010 to USD 17.6 billion in FY 2015/2016 (see the bottom left panel of Figure 2). A further reduction in net international reserves could have occurred had it not been for a surge in capital and financial inflows (see the bottom right panel of Figure 2), which were primarily in the form of government-to-government grants and short-term loans. On 11 August 2016, the Egyptian authorities signed a staff-level agreement with the IMF to obtain a three-year Extended Fund Facility (EFF) for USD 12 billion. The agreement stipulated a series of economic reforms, which paved the way for the floatation of the pound on 3 November 2016.
Right before the floatation, the parallel market exchange rate was quite volatile from day to day due to the presence of intense speculation. After the floatation, the official market rates became increasingly aligned with quoted rates in the parallel market leading to a gradual disappearance of the latter. The floatation of the currency, and the concurrent defensive hike in interest rates attracted significant financial flows to Egypt in the form of carry trade (see the bottom right panel of Figure 2), which led to a significant improvement in the balance of payments in 2016/2017 despite the continued current account deficit (see the top right panel of Figure 2).
The adoption of a floating exchange rate was one of the fundamental pillars of the EFF agreement with the IMF. It marks a significant departure from the policy of pegging the exchange rate to the US dollar as an anchor currency, which lasted for most of Egypt's recent history. The EFF agreement also included other reforms such as the removal of subsidies on energy products and publicly-provided utilities (e.g. electricity, water and transportation) through gradual price increases. The enactment of these measures coupled with the pass-through effect due to the floatation led to an unprecedented surge in the annual rate of inflation, which exceeded 30% during the early months of 2017.
4 Econometric Methodology
The Edwards model stipulates that a significant level of exchange rate misalignment is unsustainable, and should lead to a correction to bring the actual exchange rate closer to its equilibrium level. Given this premise, cointegration is a natural modeling framework to estimate the ERER. As discussed initially in Engle and Granger (1987), and later on in Johansen (1988, 1991), one must first test for the nonstationary of the variables employed in the model. If found nonstationary, then following the Engle and Granger (1987) method, a long run regression is estimated for the levels of the variables. Based on (4), which is the long run regression, one can interpret
Given the aforementioned economic fundamentals, the long run regression is given by:
Once the cointegration vector
with the exchange rate misalignment at time
Johansen (1988, 1991) developed an alternative method which is capable of finding more than one cointegration relationship, if existing.
If the variables in
While the Johansen method is deemed superior to that of Engle and Granger (1987) in terms of the power properties of the cointegration test, it is often problematic in application when it indicates the presence of more than one cointegration relationship without a clear theoretical interpretation for each one. In addition, the Johansen method involves fitting a VAR model as a first step, and therefore suffers from the curse of dimensionality problem when dealing with a large number of variables and a relatively short time series.8
5 Empirical Results
5.1 Data Sources
The data sample used in the construction of the REER and in estimating the model parameters is for the period 2001Q3–2017Q3. The trade partners used in the analysis are the EU (29.7%), US (7.8%), China (6.9%), Turkey (3.8%), Russia (3.0%) and the UK (2.5%), with the trade shares over the period 2006–2015 reported in brackets. The trade shares are obtained using data from Trade Map (www.trademap.org). The bilateral exchange rates and consumer price indices used in the construction of the REER are obtained from the IMF International Financial Statistics and OECD online databases.
The GDP components and net foreign assets for Egypt are obtained from the Ministry of Planning and the Central Bank of Egypt. For the terms of trade, we constructed a commodity terms of trade index using global price indices available from the IMF World Economic Outlook (WEO) database. These include agricultural raw materials, cereals, fuel, meat, metals, sugar and vegetable oils. The corresponding shares for Egypt's exports and imports with regard to these commodity groups are obtained from Trade Map. For per capita GDP, which is used to construct the productivity differential, data for the trade partners are obtained from the IMF WEO database and Eurostat. The corresponding time series are plotted in Figure 4.9
5.2 Unit Root Tests
In Table 1, we report the results of the ADF test for the presence of a unit root, for both the levels and first differences of the variables employed in the study. These are: the real effective exchange rate (REER), the productivity differential (PD), the share of investment in GDP (INV), the share of government consumption in GDP (GOV), terms of trade (TOT), net foreign assets as a percentage of GDP (NFA), and trade openness (OPEN).
Augmented Dickey-Fuller (ADF) unit root test results.
Notes: All variables are in natural logarithm except for NFA. For the ADF test, the test specification assumed an intercept for the all the variables, and a trend was also included for PD and TOT due to the presence of an upward trend in the levels during the sample period. For lag selection in the ADF test, the Schwarz information criterion was used.
*** and ** mark statistical significance at the 1% and 5% levels of significance, respectively.
The results indicate that all of the variables are I(1), and are stationary in first differences. The null hypothesis of a unit root in first differences is uniformly rejected for all the variables at the 1% level of significance, except for OPEN where the hypothesis is rejected at the 5% level of significance.
5.3 Estimation Results
In Table 2, we report the parameter estimates for the long run regression in (5). All of the estimated coefficients have the expected signs and are statistically significant at the 1% level of significance. The only exception is the coefficient on TOT, which is not statistically significant. Generally, the model has a good fit with an R2 of 0.88. The normality of the residuals is not rejected at the 1% level of significance using the Jarque-Bera and Anderson and Darling (1952) tests, with the latter providing more power to detect misspecification in the tails of the distribution. The Ramsey RESET test was also carried out, and it indicates the appropriateness of the linear specification.
Determinants of the equilibrium real exchange rate: long run estimates.
|Coefficient estimates||Standard error||Standardized coefficients|
Notes: All variables are in natural logarithm except for NFA. For an estimated coefficient
*** marks statistical significance at the 1% level of significance.
The signs of the coefficients are all consistent with economic theory, and their magnitude is comparable to the estimates found in the empirical literature for other developing economies. The positive signs of the coefficients on INV and GOV indicate that an increase in investment and government consumption tends to be concentrated in the nontradables sector, which leads to a real appreciation. The TOT has a positive yet statistically insignificant coefficient. Despite its insignificance, we still include it in the model since it is found to be significant when using sub-samples that do not include the observations post the currency floatation. This is discussed later in Section 6.2. The positive and statistically significant coefficient for NFA reflects a strong connection between the real exchange rate and Egypt's net indebtedness, with no evidence of intertemporal dynamics by which improved future growth prospects could have ameliorated the exchange rate depreciation during the crises of 2002–2003 and 2016. This is partly attributed to the high level of public debt in Egypt during the sample period; see Égert, Lommatzsch, and Lahrèche-Révil (2006b) for a discussion of the contrasting experience of the Baltic states (Estonia, Latvia and Lithuania) at the inception of their economic transition, particularly their low level of indebtedness.
The last column reports the standardized coefficients to gauge the relative importance of each explanatory variable by correcting for different levels of variability and the effect of units of measurement. The estimates show that OPEN seems to be the most influential among all the economic fundamentals with a standardized coefficient of −0.829, while PD and NFA come next with standardized coefficients equal to 0.504 and 0.318, respectively. Analysis of variance also shows that OPEN provides the largest contribution to explaining the variation in the REER, which is 44.3%, followed by PD at 19.8%. The decomposition of the total variation shows that GOV and NFA come in at third and fourth places with a contribution of 14.3% and 11.0%, respectively.
With regard to parameter stability, Figure 3 shows recursive parameter estimates, which are estimated by increasing the sample size by one observation at a time. The results show that the estimated parameters are relatively stable once the sample size is large enough, with the exception of the coefficient on GOV, which shows an upward trend. However, there is evidence of a structural break in the parameter values starting 2016Q4, the quarter in which the fixed exchange rate regime was abandoned. This change is more pronounced for the coefficients on PD and OPEN, which happen to be the two most influential determinants of the ERER. There is also a notable change in the value of the coefficient on GOV. Visual inspection of Figure 3 indicates that this structural break is not transitory in nature, therefore, we conduct sensitivity analysis in Section 6.2 to check the robustness of our projections to this structural break.
Following Engle and Granger (1987), we conduct the ADF unit root test for
Error correction model estimates.
|Coefficient estimates||Standard error||Standardized coefficients|
Notes: All variables are in natural logarithm except for NFA. For an estimated coefficient
*** marks statistical significance at the 1% level of significance.
As for the coefficient of the ECM term, it has the correct sign and is statistically significant at the 1% level of significance. The coefficient magnitude indicates a relatively fast speed of adjustment to past disequilibria, and this is primarily due to the high pass-through from the exchange rate to domestic inflation, which leads to fast appreciation in the REER after a significant devaluation. This behavior had been evident during the crisis of 2002–2003 and, more recently, during the crisis of 2016.
We also implemented the Johansen (1988, 1991) method by estimating a VAR model, which incorporated all the variables including the REER.10 The analysis rendered the following results: the trace statistic indicated the presence of 4 cointegration relationships, while the maximum eigenvalue statistic suggested the presence of 3 cointegration relationships. This is also encountered in Hosni (2015), and the author chose to eliminate some variables from the model to address the problem. In addition to suggesting an inconceivably large number of cointegration relationships, the results of the Johansen method also reveal that the coefficients of the cointegrating vectors have uninterpretable signs and implausible magnitudes. Finally, the test for multivariate normality of the residuals is also rejected at conventional levels of significance returning a p-value of 0.003.11
5.4 Constructing the Equilibrium Real Exchange Rate
As discussed in Section 4, we use the permanent components of the economic fundamentals to construct our estimate of the ERER. Figure 4 displays the time series for the economic fundamentals along with their long run trends.
The permanent components of the economic fundamentals are used along with the parameter estimates of the long run regression in Table 2 to construct the ERER series. Figure 5 displays the REER (gray line), the ERER (red smooth line) and the exchange rate misalignment as the area graph displayed around the horizontal axis.
Figure 5 shows that the ERER has varied over time along with changes in the economic fundamentals. In general, the REER has tracked the ERER well, however there were periods during which the two variables became strongly disassociated. During 2001Q3–2002Q4, the exchange rate was significantly overvalued. A strong nominal devaluation in 2003Q1 led to overshooting, which reversed the misalignment such that the exchange rate became undervalued. By 2003Q2 the exchange rate was undervalued by 11.6% rising to 12.6% in 2003Q4 before the misalignment started to retreat. From 2006Q1 onwards, the REER along with the ERER were appreciating, which indicates that the improving fundamentals of the Egyptian economy helped sustain real currency appreciation. The improvement in the fundamentals during this period is attributed to strong productivity growth relative to trade partners, an improvement in the terms of trade, and a notable increase in net foreign assets (see Figure 4).
After some mild undervaluation during 2013Q1–2014Q2, the situation drastically changed to strong overvaluation starting 2015Q1 due to the dual effect of REER appreciation and a decline in the ERER. The rise in the REER was due to the US dollar strengthening against the currencies of Egypt's other trade partners, and the increase in Egypt's inflation differential, particularly since 2016Q2 as it soared to double-digit figures (see Figure 1). On the other hand, the decline in the ERER was driven by the deterioration in the economic fundamentals, mainly the productivity differential, investment, terms of trade and net foreign assets (see Figure 4). Despite the increase in government consumption and the general decline in trade openness during this period, their combined effect was not sufficient to counter the depreciating impact of the other fundamentals.
Although the correct policy response at the time was to let the Egyptian pound to depreciate to match the decline in the ERER and thereby avoid currency overvaluation, the monetary authority opted instead for defending the currency peg, which proved to be an untenable stance. In 2016Q3, right before the floatation, the Egyptian pound was overvalued by about 24%. With the floatation of the currency, exchange rate overshooting occurred resulting in the currency being undervalued in real terms by about 14.7% in 2016Q4, and increasing further to reach 22.3% undervaluation by 2017Q1, before showing a gradual decline. The decline in the misalignment more recently is due to the fast appreciation of the REER because of the surge in domestic inflation. Towards the end of the sample period in 2017Q3, the exchange rate was undervalued by about 18.5%.
5.5 Exchange Rate Evolution under A Flexible Exchange Rate Regime: A Counterfactual Scenario
In the previous subsection, we argued that the policy response to the exchange rate overvaluation during the period 2015Q1–2016Q3 was suboptimal. In this subsection, we present a counterfactual scenario to answer the following question: If the exchange rate was determined without intervention since the beginning of 2015, what would have been the path of the nominal exchange rate for the US dollar? Using the estimated ERER, we back out the equilibrium nominal exchange rate for the US dollar that would have prevailed over the period 2015–2016 if it were left to be determined by market forces. We focus on the period 2015–2016 since this is the period in which a parallel market for foreign exchange emerged and had become the predominant foreign exchange market before the floatation of the currency.
In Table 4, we show the evolution of the equilibrium nominal exchange rate for the US dollar, along with the official and parallel market rates. The figures clearly show that the overshooting of the exchange rate after the floatation could have been avoided had the exchange rate been determined more flexibly to reflect the changing economic fundamentals during this turbulent period. Specifically, the nominal exchange rate at the time should have been around 12.24 pounds per US dollar. Unsurprisingly, the parallel market rate was tracking the rise in the equilibrium rate during this period, hence it was more reflective of the pound's fair value. However, the parallel market rate started to increase notably, exceeding its equilibrium value in 2016Q2. We conjecture that this occurred due to the emergence of huge speculative demand for foreign exchange amid a dearth of liquidity. It was also coupled with a growing state of anxiousness about an inevitable strong devaluation at some undetermined date. The delay in deciding to devalue or float the Egyptian pound was certainly a compounding factor.
Evolution of the nominal exchange rate for the US dollar (EGP per USD) under the counterfactual scenario.
|Official rate||Equilibrium rate||Parallel market rate|
The value for 2017Q1 is the average of the daily rates quoted in the parallel market from 1 January 2017 until 6 February 2017, the date after which the parallel market ceased to exist.
Between 2016Q2 and 2016Q4, the black market rate increased from 10.8 pounds per US dollar to 17.16 pounds per US dollar, which is an increase of 58.9%. This rendered any traded foreign currency quite a valuable asset with abnormal returns, which served to boost speculative demand for these currencies. Most likely, this has also hampered efficient price discovery post the floatation as the parallel market persisted for a while to test the monetary authority's resolve in committing to the new regime. During this period, the parallel market rates recorded new highs with the exchange rate reaching 20 pounds per US dollar at one point, however by early February 2017 the parallel market ceased to exist.
6 Medium-Term Projections
In this section, we provide projections for the ERER until the end of 2020, along with the likely path for the evolution of the current misalignment, i.e. current undervaluation of the Egyptian pound. This is undertaken by constructing projections for the six economic fundamentals to obtain projections for the ERER. In addition, we also construct projections for the REER to show the likely path for exchange rate misalignment over the short to medium term. The source for Egypt's projections is the Egypt IMF Country Report No. 17/290 (September 2017). The projections for the trade partners are obtained from the IMF WEO database (October 2017 edition) and Eurostat. Figure 6 shows the projected fundamentals (dotted gray line) along with their long run values obtained using the Hodrick-Prescott filter.
6.1 Equilibrium Real Exchange Rate Projections for 2017Q4–2020Q4
As shown in Figure 7, the projections indicate that the REER will be subject to gradual appreciation since Egypt's inflation is expected to remain significantly above the rate of inflation in the trade partners until the end of the projections period in 2020Q4. This path is displayed by the solid gray line. Such evolution is based on the assumption that the nominal exchange rate for the US dollar remains at its 2017Q3 level, which is 17.73 pounds per US dollar, and given the IMF inflation projections for Egypt and the trade partners. Over the same period, the ERER is showing a decline, indicating that the equilibrium exchange rate will be more depreciated relative to its value in 2017Q3.
The decline in the ERER, which places it on a depreciating path, is due to the lower share of government consumption in GDP (GOV) given fiscal consolidation plans, the projected increase in trade openness (OPEN) and an expected deterioration in the terms of trade (TOT) (see Figure 6). The impact of these variables dominates that of the increase in investment (INV) and net foreign assets (NFA), which have an appreciating effect on the ERER.
Given the combined impact of an appreciating REER and a declining ERER, the exchange rate misalignment will disappear by 2019Q1. At that point in time, the equilibrium nominal exchange rate is expected to be 17.70 pounds per US dollar. As the REER keeps appreciating, the currency is projected to be overvalued by around 13.1% by the end of 2020Q4. This occurs since the nominal exchange rate is assumed to remain at 17.73 pounds per US dollar, while the equilibrium value will be around 20.05 pounds per dollar by the end of 2020.
The assumption that the nominal exchange rate remains fixed at 17.73 pounds per US dollar is not innocuous. It implies that the exchange rate will not be allowed to respond to changing fundamentals, which should not be the case if the currency is truly floating. We now make an alternative assumption that the exchange rate will be depreciating in nominal terms by 1% every quarter starting from 2017Q4 until 2020Q4. This path is given by the dotted gray line in Figure 7. It shows that the REER appreciation path is now milder since two opposing forces will be at play: the high inflation differential will be pushing the REER to rise, while the nominal depreciation will be pushing it down. It is evident that the effect of the inflation differential will dominate. Under this 1%-per-quarter nominal depreciation path, the exchange rate undervaluation will dissipate more slowly, and the nominal exchange rate will be approximately at its equilibrium level of 20.05 pounds per US dollar by 2020Q4.
It is worth emphasizing that the exchange rate values quoted in this analysis are subject to a nontrivial margin of uncertainty for two reasons. First, the exercise of estimating the ERER around times of significant economic adjustment, as in Egypt's case, is a challenging task. We already showed evidence of a structural break in the value of the long run parameters, which necessitates some sensitivity analysis to ascertain the robustness of the results. Second, the projected ERER is based on the projections of many variables for Egypt and its trade partners, therefore, it is somewhat sensitive to changes in the paths of these variables. Naturally, the margin of uncertainty is higher for the longer horizon.
6.2 Robustness Check: A Forecast Combination Approach
As indicated earlier, there is evidence of a structural break in the parameters of the long run regression. This presents a challenge since it occurs at the end of the sample period, and this is particularly relevant since we use the estimated parameters to construct future projections of the ERER. One way to ameliorate the impact of this break is by utilizing various point estimates of the parameters to construct alternative forecast paths under each set of parameter values. This is based on the idea of forecast combinations which has been shown to provide robust results when confronted with structural breaks (see, for example, Pesaran and Timmermann 2007; Tian and Anderson 2014).
To this end, we utilize different sets of parameter estimates before and after the regime change in 2016Q4. Specifically, we consider the period 2016Q3–2017Q3 for a total of five quarters. We then obtain parameter estimates from different samples that end in each respective quarter, and use it to construct projections for the period 2017Q4–2020Q4. The implied ERER path under each set of estimates is displayed in Figure 8, with the vintage reported next to each trajectory.
One can obtain projections of the ERER by averaging the five trajectories reported in Figure 8. As can be gleaned from the figure, the combined forecast will have a more appreciated level relative to the one based on the 2017Q3 vintage, noting that the latter was the basis for the analysis in Section 6.1. We combine the trajectories using both unweighted and weighted averaging schemes, the difference being that the latter gives relatively more weight to the recent vintages.12 According to the combined forecast path, the estimates show that the fair value for the Egyptian pound will be in the range of 18.21–18.83 pounds per US dollar in 2020Q4, where the lower and upper limits of this range correspond to the unweighted and weighted schemes, respectively.
This exercise shows the sensitivity of the projections to how the regime change of 2016Q4 is addressed in the analysis, which, in turn, invokes the question of whether recent data should be given less weight through averaging the estimates from different vintages. The answer depends on one's view as to whether the break is a genuine one or if it is transitory in nature. While this is a matter of speculation since there is not enough observations post the break to undertake more extensive analysis, we conjecture that recent data should be given full weight when it comes to near-future projections of the ERER given that the new parameter values have rather persisted since the 2017Q1 vintage. The analysis in Section 6.1, in which the projected nominal exchange rate in 2020Q4 is 20.05 pounds per US dollar, is based on the most recent vintage (i.e. 2017Q3), which gives full weight to the most recent data.
7 Concluding Remarks
Although the floatation of the Egyptian pound led to overshooting, with the exchange rate becoming undervalued by about 22.3% right after the floatation, this undervaluation is only temporary. The strong devaluation in the currency sparked an unprecedented inflationary wave in Egypt, which is exerting upward pressure on the REER, signifying that the competitiveness gains from the devaluation are dissipating fast. Our estimates show that if the exchange rate is left to be determined by market forces, then the nominal exchange rate for the US dollar will have to depreciate from its 2017Q3 level of 17.73 pounds per US dollar. This paper has shown some likely paths for the real and nominal exchange rates under different sets of assumptions, which are indicative of the underlying trends. The main conclusion from the analysis is that the change in Egypt's economic fundamentals in the last few years warrants a more depreciated, and hence more competitive, exchange rate to maintain internal and external balance.
The excessive overshooting of the exchange rate in the aftermath of the floatation is due in no small part to the delay in devaluing the Egyptian pound during the course of 2016, or even earlier during 2015. Given legitimate concerns at the time about the inflationary impact of the devaluation, postponing this move has inadvertently fueled speculative demand for foreign currencies. This has in turn contributed to exchange rate overshooting, which created an inflationary wave that surpassed expectations by a substantial margin.
Going forward, the recent crisis offers important lessons and also poses questions, which the policymakers must ponder if Egypt is to successfully devise a new framework for monetary policy. Since the emerging monetary policy framework is an inflation-targeting regime, a natural question to consider is whether the institutional prerequisites for inflation targeting are in place. Masson, Savastano, and Sharma (1997) and Mishkin (2004) emphasize that the scope for the conduct of independent monetary policy in developing countries is often hindered by the excessive reliance on the monetization of the budget deficit. This form of fiscal dominance makes it difficult, if not impossible, for the monetary authority to control inflation, and it tends to undermine the credibility of the monetary policy framework. An additional constraint that often exists is having shallow capital markets, which reduces the effectiveness of the monetary transmission mechanism from policy instruments to targets such as output and inflation.
The success of inflation targeting in Egypt not only necessitates the de facto independence of the monetary authority from fiscal pressure, but also requires continuous communication with the public to adequately anchor inflation expectations. The operationalization of inflation targeting goes well beyond the introduction of numerical inflation targets, as its success critically hinges upon the adoption of an information-intensive approach to monetary policy conduct, including enhanced transparency, in addition to holding the monetary authority accountable if it fails to meet the announced inflation targets.
Convincing the public that the CBE's overriding objective is price stability is not an easy task, especially if there is skepticism about the absence of fiscal dominance. In contrast to an exchange rate peg, inflation targeting allows the monetary authority to smooth output fluctuations over the business cycle using its policy instruments. However, the commitment to a low and stable rate of inflation must be the overriding objective if conflict arises. Instating this notion in the public's mind takes effort and time, and it needs to be a central component of the monetary authority's drive to build its credibility with regard to preserving price stability.
This in turn hinges upon economic agents gradually shifting their focus from a long-established nominal anchor, which is the nominal exchange rate, to a new anchor which is the announced inflation target. Initially, this requires a true abandonment of the exchange rate peg. Currently, the observed exchange rate volatility appears artificially low, and this will not serve the CBE over the medium term to re-anchor expectations around the announced inflation target. It is natural to worry about the impact of exchange rate volatility on inflation, however, one must also acknowledge that the high pass-through from the exchange rate to inflation is largely due to the exchange rate having been a firmly-established nominal anchor for far too long.
Needless to say, bringing inflation down to single digits is needed before inflation targeting can be meaningfully adopted. Most inflation-targeting economies have targets in the range of 2–3% annual inflation. Some countries (e.g. Brazil, India, Russia and Vietnam) chose targets in the higher range of 4–5%. Achieving similar inflation rates in the near future is a challenge given the commitment of the Egyptian government to continue on a course of price liberalization for energy products and public utilities as government subsidies are removed. However, the adoption of a carefully-designed communication strategy could help a great deal in managing the course of monetary policy during this transition.
On the operational side, there are two important issues to consider. The first relates to the effectiveness of the policy tools at the disposal of the monetary authority to affect output and inflation outcomes. The second issue pertains to the availability of timely and accurate data, which enables policymakers to make the right decisions in real time. With regard to the second issue, there is no doubt that both the accuracy and timeliness of economic data could be improved, especially when it comes to the real sector. As for the monetary policy transmission mechanism, we believe that it is not very well understood in the case of Egypt given that the majority of the adult population is unbanked, and also given the existence of a large informal sector. This issue warrants further study.
The above points are not meant to discourage the CBE's move to inflation targeting. Adopting inflation targeting is a real opportunity to institute a successful and sustainable monetary policy regime in Egypt. Rather, they are intended to highlight some pertinent issues that are critical for the success of inflation targeting in Egypt. Needless to say, having a sustainable monetary policy framework in Egypt will not, on its own, solve all of Egypt's economic woes. The root causes of many of Egypt's economic problems relate to structural weaknesses. In particular, the alarmingly low rate of national savings, which indicates an insufficient rate of capital formation, places a binding constraint on the rate of growth of potential output. Complementary reforms are needed to encourage private saving and investment, with incentives to expand the economy's potential to export goods and services with high domestic value added. This is when the economy could really reap the fruits of an improved macroeconomic policy framework.
I would like to thank Hoda Elabbadi for excellent research assistance. I would also like to thank the Editor, an anonymous referee, Karim Abadir, Abla Abdelatif, Nadine Abdel Raouf, Sherifa Abdelrazek, Mohamed Abu Basha, Ahmed Eissa, Mohamed Farid, Hany Genena and Hoda Selim for helpful comments and suggestions.
Anderson, T. W., and D. A. Darling. 1952. “Asymptotic theory of certain goodness-of-fit criteria based on stochastic processes.” Annals of Mathematical Statistics 23: 193–212.
Baffes, J., I. A. Elbadawi, and S. A. O'Connell. 1999. “Single equation estimation of the equilibrium real exchange rate.” In Exchange Rate Misalignment: Concepts and Measurement for Developing Countries, edited by L.E. Hinkle and P.J. Montiel, Oxford: Oxford University Press.
Balassa, B. 1964. “The purchasing power parity doctrine: A reappraisal.” Journal of Political Economy 72: 584–596.
Benigno, G., and C. Thoenissen. 2003. “Equilibrium exchange rates and capital and supply side performance.” Economic Journal 113: 103–124.
Brixiova, Z., B. Égert, and T. H. A. Essid. 2014. “The real exchange rate and external competitiveness in Egypt, Morocco and Tunisia.” Review of Middle East Economics and Finance 10: 1–27.
Clark, P. B., and R. MacDonald. (1998). Exchange rates and economic fundamentals: A methodological comparison of BEERs and FEERs. IMF Working Paper 98/67.
Cottani, J. A., D. F. Cavallo, and M. S. Khan. 1990. “Real exchange rate behavior and economic performance in LDCs.” Economic Development and Cultural Change 39: 61–76.
Coudert, V., and C. Couharde. 2009. “Currency misalignments and exchange rate regimes in emerging and developing countries.” Review of International Economics 17: 121–136.
Dollar, D. 1992. “Outward-oriented developing economies really do grow more rapidly: Evidence from 95 LDCs, 1976–1985.” Economic Development and Cultural Change 40: 523–544.
Dornbusch, R. 1976. “Expectations and exchange rate dynamics.” Journal of Political Economy 84: 1161–1176.
Drine, I., and C. Rault. (2003). On the long-run determinants of real exchange rates for developing countries: Evidence from Africa, Latin America and Asia. William Davidson Institute working paper No. 571, University of Michigan Business School.
Easterly, W., N. Loayza, and P. Montiel. 1997. “Has Latin America's post-reform growth been disappointing?.” Journal of International Economics 43: 287–311.
Edwards, S. 1989a. “Exchange rate misalignment in developing countries.” The World Bank Research Observer 4: 3–21.
Edwards, S. 1989b. Real Exchange Rates, Devaluation and Adjustment. Cambridge, MA: MIT Press.
Edwards, S. 1994. “Real and Monetary Determinants of Real Exchange Behavior: Theory and Evidence from Developing Countries.” In Estimating Equilibrium Exchange Rates, Ed. J. Williamson, Washington, DC: Institute for International Economics.
Edwards, S., and M. Savastano. 2000. “Exchange rates in emerging economies: What do we know? What do we need to know?.” In Economic Policy Reform: The Second Stage, Ed. A. Krueger, Chicago: University of Chicago Press.
Égert, B., L. Halpern, and R. MacDonald. 2006a. “Equilibrium exchange rates in transition economies: Taking stock of the issues.” Journal of Economic Surveys 20: 257–324.
Égert, B., K. Lommatzsch, and A. Lahrèche-Révil. 2006b. “Real exchange rates in small open OECD and transition economies: Comparing apples with oranges?.” Journal of Banking and Finance 30: 3393–3406.
Elbadawi, I. 1994. “Estimating long-run equilibrium real exchange rates.” In Estimating Equilibrium Exchange Rates, Ed. J. Williamson, Washington, DC: Institute for International Economics.
Engle, R. F., and C. W. J. Granger. 1987. “Co-integration and error correction: Representation, estimation, and testing.” Econometrica 55: 251–276.
Fischer, S. 1993. “The role of macroeconomic factors in growth.” Journal of Monetary Economics 32: 485–511.
Frankel, J. A., and A. K. Rose. 1996. “Currency crashes in emerging markets: An empirical treatment.” Journal of International Economics 41: 351–366.
Froot, K. A., and K. Rogoff. 1995. “Perspectives on PPP and long-run real exchange rates.” In Handbook of International Economics, Eds. G.M. Grossman and K. Rogoff, Vol. 3. Amsterdam: Elsevier Science Publishers.
Ghura, D., and T. J. Grennes. 1993. “The real exchange rate and macroeconomic performance in sub-Saharan Africa.” Journal of Development Economics 42: 155–174.
Hau, H. 2002. “Real exchange rate volatility and economic openness: Theory and evidence.” Journal of Money, Credit, and Banking 34: 611–629.
Hinkle, L. E., and P. J. Montiel. 1999. Exchange Rate Misalignment: Concepts and Measurement for Developing Countries. Oxford: Oxford University Press.
Holtemöller, O., and S. Mallick. 2013. “Exchange rate regime, real misalignment and currency crises.” Economic Modelling 34: 5–14.
Hosni, R. 2015. “An assessment of the real exchange rate misalignments in Egypt: An application of the behavioral equilibrium approach.” Global Journal of Human-Social Science 15: 37–51.
Hosni, R., and D. Rofael. 2015. “Real exchange rate assessment in Egypt: Equilibrium and misalignments.” Journal of Economics and International Finance 7: 80–97.
International Monetary Fund (2006). Methodology for Consultative Group on Exchange Rate (CGER) Assessments. IMF Research Department Report, Washington, DC.
Johansen, S. 1988. “Statistical analysis of cointegration vectors.” Journal of Economic Dynamics and Control 12: 231–254.
Johansen, S. 1991. “Estimation and hypothesis testing of cointegration vectors in Gaussian vector autoregressive models.” Econometrica 59: 1551–1580.
Kaminsky, G., and C. M. Reinhart. 1999. “The twin crises: The causes of banking and balance of payments problems.” American Economic Review 89: 473–500.
Krugman, P. R. 1979. “A model of balance-of-payments crises.” Journal of Money, Credit, and Banking 11: 311–325.
Lane, P. R., and G. M. Milesi-Ferretti. 2004. “The transfer problem revisited: Net foreign assets and real exchange rates.” Review of Economics and Statistics 86: 841–857.
Lee, J., and M. K. Tang. (2003). Does productivity growth lead to appreciation of the real exchange rate?. International Monetary Fund Working Paper 03/154.
MacDonald, R., and L. Ricci. (2001). PPP and the Balassa–Samuelson effect: The role of the distribution sector. International Monetary Fund Working Paper 01/38.
MacDonald, R., and L. Ricci (2002). Purchasing power parity and new trade theory. International Monetary Fund Working Paper 02/32.
Masson, P. R., M. A. Savastano, and S. Sharma. (1997). The scope for inflation targeting in developing countries. International Monetary Fund Working Paper 97/130.
Mishkin, F. S. (2004). Can inflation targeting work in emerging market countries? NBER Working Paper No. 10646.
Mohieldin, M., and A. Kouchouk. (2003). On exchange rate policy: The case of Egypt 1970–2001. Economic Research Forum Working Paper No. 0312.
Mongardini, J. (1998). Estimating Egypt's equilibrium real exchange rate. International Monetary Fund Working Paper 98/5.
Nurkse, R. 1945. Conditions on international monetary equilibrium. Essays in International Finance 4 (Spring). Princeton, NJ: Princeton University Press.
Obstfeld, M., and K. Rogoff. 1995a. “Exchange rate dynamics redux.” Journal of Political Economy 103: 624–660.
Obstfeld, M., and K. Rogoff. 1995b. “The mirage of fixed exchange rates.” Journal of Economic Perspectives 9: 73–96.
Ostry, J. D. 1994. “Government purchases and relative prices in a two-country world.” Economic Record 70: 149–161.
Pesaran, M. H., and A. Timmermann. 2007. “Selection of estimation window in the presence of breaks.” Journal of Econometrics 137: 134–161.
Razin, O., and S. M. Collins. 1999. “Real exchange rate misalignments and growth.” In The Economics of Globalization: Policy Perspectives from Public Economics, Eds. A. Razin and E. Sadka, Cambridge: Cambridge University Press.
Rodrik, D. 2008. “The real exchange rate and economic growth.” Brookings Papers on Economic Activity 2: 365–412.
Sachs, J., and A. Warner. 1995. “Economic reform and the process of global integration.” Brooking Papers on Economic Activity 1: 1–95.
Samuelson, R. 1964. “Theoretical notes on trade problems.” Review of Economics and Statistics 46: 145–154.
Tian, J., and H. M. Anderson. 2014. “Forecasting under structural break uncertainty.” International Journal of Forecasting 30: 161–175.
Williamson, J. 1985. The exchange rate system. Policy Analyses in International Economics. Vol. 5. Washington, DC: Institute for International Economics.
Williamson, J., Ed. 1994. Estimating Equilibrium Exchange Rates. Washington, DC: Institute for International Economics.
Internal balance refers to equilibrium in the labor market and the market for nontradable goods, while external balance is attainted when the economy is intertemporally solvent, that is when the current and projected current account deficits (surpluses) match the economy's net foreign assets (liabilities).
The long run values of the fundamentals refer to their permanent components when cyclical variation is ignored; see Edwards (1989b), Elbadawi (1994), and Baffes, Elbadawi, and O'Connell (1999). In the empirical literature, they are typically obtained using some filtering method, e.g. the Hodrick-Prescott filter.
Details about the trade partners used in the construction of the NEER and the REER are given in Section 5.1.
As we show later on, this real appreciation was in fact supported by improving economic fundamentals, and therefore did not cause exchange rate misalignment.
The cumulative depreciation against the US dollar during 2014Q1–2015Q4 amounted to 25.1% for the euro, 4.4% for the Chinese yuan, 31.1% for the Turkish lira, 88.5% for the Russian ruble and 8.9% for the Sterling pound.
In 2016Q1, the official and parallel market rates were 8.02 and 9.06 pound per US dollar, respectively. By 2016Q3, the official and parallel market rates recorded 8.86 and 12.39 pound per US dollar.
It is worth noting that Égert, Halpern, and MacDonald (2006a) argue that the estimates of currency misalignment obtained from time series models, as opposed to panel methods, should be interpreted as measures of short-term misalignment where the deviations from equilibrium tend to be short-lived. In other words, time series models may fail to identify long-lasting currency over- or undervaluation.
In the empirical analysis, we employ the two methods and show that the results of the Johansen method are difficult to interpret since they suggest more than one cointegration relationship, with coefficient signs and magnitudes that cannot be reconciled with economic theory or corresponding findings from the literature. We discuss this in detail in Section 5.3.
For variables available only at the annual frequency, cubic spline interpolation has been used to obtain the quarterly observations. Seasonal adjustment for the variable INV has been applied to remove its strong seasonal pattern using the X-13 ARIMA-SEATS algorithm, the details of which can be found in the U.S. Census Bureau manual available at https://www.census.gov/ts/x13as/docX13AS.pdf.
The Schwarz criterion suggested the inclusion of 4 lags in the VAR model.
The detailed results are not reported in the interest of brevity, but are available from the author upon request.
For the weighted average, we use exponentially declining weights given by