Exchange Rate Pass Through In Republic Of Moldova Finance Essay
Efficient monetary policy is one of the essential conditions for a transition economy to achieve economic stability and sustainable growth. According to the monetary policy, two main objectives the central bank of republic of Moldova pursues are: domestic (price stability) and external (exchange rate stability) perspectives. In the economic theory there is a clearly defined pattern between exchange rates and price levels through the law of one price and purchasing power parity theory.
The notion of “pass-through effect” in broader sense can be defined as “the process how home prices change in response to changes of exchange rates” (Menon 1995, McCarthy 2000, Hufner and Schoder 2002), or as a partial elasticity of prices with respect to the exchange rate.
The problem of examination the exchange rate pass-through (henceforth ERPT) attracts special attention due to several reasons. Firstly, the volatility of exchange rates has increased dramatically in recent years. Since the Breton Woods system of fixed exchange rate collapsed in the early 1970s and most countries allowed exchange rates to be changed daily, the volatility of exchange rates has become one of the main problems for monetary policy regulations. Secondly, high globalization process, trade integration and, therefore, trade flows, foreign direct investments led to close interaction between countries, cross-country capital flows, which exacerbate exchange rate volatility.
The problem of exchange rate volatility is particularly important for small open economies, with high amount of cross-border trade operations and in countries where prices are simultaneously affected by both internal and external factors. Republic of Moldova is a good example of such a case.
This research work aims to estimate the exchange rate pass-through on domestic prices by category of good produced and consumed in the republic of Moldova, i.e long-run relationship between the exchange rate, the price level and controlling variables such as money supply, world oil prices, GDP Gap, etc.
For these purposes VAR model, based on Dornbusch (1987) Monetary model for a small open economy with floating exchange rate, Granger causality test, cumulative impulse responses of inflation to exchange rate shock on domestic inflation, as well as variance decomposition techniques will be used.
Although there is plenty of literature available, which studies this topic for group of countries and for individual countries as well, for republic of Moldova there have not yet been studies on this topic. Most of previous research have concentrated on the pass-through of a country’s exchange rate fluctuations to its import prices (for example, Goldberg and Knetter (1997)), some of the investigations concentrated on the effect of ERPT from exchange rate and import prices to domestic inflation. There are also a number of studies on the pass-through to domestic producer and consumer prices; some examples include Parsley and Popper (1998), the ERPT estimation in European Union (Fouquin et al. 2001), emerging markets, such as Russia (Dobrynskaya 2005) and Belarus (Tsesliuk 2002), Ukraine (Bandura 2010).
The focus is to estimate ERPT on domestic price level represented by consumer price indices (CPI), producer price indices (PPI) monthly time series.
Thus, the data will include Nominal Effective Exchange Rate, taken from IMF Financial Statistics, the data on CPI, PPI, which are taken from National Bureau of Statistics of rep. of Moldova. Moreover, CPI data are divided by groups of goods and services and PPI data are divided by Market, Economic Activities.
Why is it important to study the relationship between price level and exchange rate? It is a well-known fact that the overall country’s competitiveness and macroeconomic stability depends on the effectiveness of exchange rate policy. The relationship between inflation and economic activity is very strong. Inflation affects all aspects of the economy, from consumer spending, business investment and employment rates to government programs, tax policies, and interest rates.
The pass-through effect is known to be most evident in countries with a larger import share of domestic demand (McCarthy 2000) and higher PTE means greater exposure of an open economy to the external shocks on the world prices. Moldova represents the country where export accounts for less than 45 % of import. (Table 1, Graph 2, Appendix). In addition, Frankel, Parsley and Wei (2005), Dobrynska (2005) mentioned that smaller countries, and less developed countries generally tend to have a higher pass-through of exchange rates on prices. Some investigations also demonstrate that the exchange rate pass-through would be higher for countries with more volatile monetary policy (Devereux and Engel 2001, Bacchetta and van Wincoop 2001, Taylor 2000).
The other important fact, which should be mentioned, is the ambition of CB of republic of Moldova to adopt full-fledged inflation-targeting regime, which by-turn requires competent understanding of the behavior of import and export prices and a lot of effort from CB authorities to study the experience of already existing countries with IT  in regulating the ERPT to provide efficient monetary policy.
The effect of ERPT in Moldova is suspected to be higher and the speed is faster than in countries with lower inflation rates (most of which are industrial), (for example Choudri and Hakura 2001; Ross 1998; Kuijs 2001). However, it may be diminishing during the last years because of CB policy of inflation targeting. Recent research suggests that the pass-through of exchange rate changes into domestic inflation has declined in many countries since the 1990s due to increased attention on inflation stabilization from many central banks’ authorities.
This work will be valuable for monetary policy authority, for National Bureau for price regulations, as well as for business companies which are engaged in export/import operations and even for foreign trade-partners, and investors. For the monetary authorities, it is crucial to know the exact amount of PTE (pass-through effect) to correctly build macroeconomic models, to design various policies aiming to make some sectors of the economy more flexible and competitive, predict domestic inflation of the import share of consumer goods and introduce adequate optimal targeting monetary policies. Taylor (2000) suggested that the effect of pass-through is systematically related to monetary policy. For export/ import companies it’s important to know how sudden change in exchange rate will be reflected in prices, and therefore in loses or profits of a company.
The responsiveness of prices to movements in the nominal exchange rate has important implications for the transmission of shocks. Exchange rates always reflect fundamental macroeconomic situation of a country. Currency system in a country should be so robust to be able to prevent economic crises and distortions.
The work is structured as follows. In the next section I present comprehensive theoretical and empirical literature review on exchange rate pass-through. Theoretical background and methodology of estimation is provided in section three. The description of data can be found in section four. The main results and conclusions are presented in section four and five, respectively.
Modeling and estimating the relationship between the exchange rates and prices, whether expressed in consumer/producer price indexes, import or export prices have been given considerable attention since early 70-ties after Breton-Woods system of exchange rates collapsed. However, the theory and prerequisites for this study are dating back even to Cassel (1918), the first economist who popularized the law of one price and PPP theory of exchange rates and developed it in the form, we observe it now. Moreover, earlier studies of PPP, according to Officer (1982), are dating back at least to scholars at the University of Salamanca in the fifteenth and sixteenth centuries.
There are two main reasons to explain the responsiveness of prices to exchange rate changes. The first one arises from the interest to explain empirically the law of one price and PPP theory. The evidence suggests that the long-run relative PPP holds remarkably well with negligible deviations, even though there can be substantial short-run deviations from PPP theory (Goldfajn and Werlang 2000; Gagnon 2004; Burnstein, Eichenbaum, and Rebelo 2007).
The second reason is to estimate the effect of exchange rate fluctuations on current account, balance of payments. Many economists, public authorities are worried about the relationship between exchange rate volatility and its influence on the trade balance through trade prices and volume. The theory of the relationship between exchange rate and trade balance is based on Marshall-Lerner condition; Lausen-Metzler effect and ‘J’ curve.
The theory of exchange rates and prices starts from the basic Monetary models: IS-LM model for closed economy, Mundell-Flemming model for small open economy.
The existing literature studying the effect of exchange rates on prices started with Haberler (1949) as mentioned in Krugman (1986), and was restated by Dornbush (1985, 1987). In the later research work the author explains the adjustment of relative prices to exchange rate movements from the industrial organization approach. All the models studied predict that once the exchange rate appreciates, this would lead to the declining of import prices.
Later on the industry approach to ERPT was illustrated by Feenstra (1989), who showed the evidence of a symmetric response of import prices to changes in the bilateral exchange rate and an import tariff; Aizenman (1985, 1986) and Giovannini (1985) investigate price-setting behavior in the context of exchange rate movements, concentrating on short-term issues of transactions cost and uncertainty rather than on the large, persistent movements in the real exchange rates.
Major works on cross-country analysis of ERPT can be divided due to countries’ classification: works on developed, developing or emerging countries, CIS countries, etc.
Developed countries. Among the works on developed countries are Anderton (2003), Campa and Goldberg (2004), Campa et al. (2005), Gagnon and Ihrig (2004) and McCarthy (2000). McCarthy (2000), using the multi-country panel regression study suggests insignificant impact of the volatility of exchange rate on prices for some developed countries, and the positive (significant) effect of the exchange rate changes on inflation for other developed countries such as Japan and France, where inflation rate was relatively higher. Campa et al. (2005) has performed an empirical analysis of transmission rates from exchange rate movements to import prices of the countries in EMU. The results obtained showed high rate of transmission, although incomplete, in the short-run, and different across industries and countries.
Gagnon and Ihrig (2004) also comes to the conclusion that countries in which either the level or the standard deviation of inflation declined substantially tend to have large declines in estimated rates of pass-through.
Hahn (2003), using quarterly data analyzes the ERPT for the Euro Area. The author adds 3-month interest rate as a control variable to the model, estimated by McCarthy (2000), interpreting by the fact that interest rate quite well approximates inflation pattern in the country with some lag of time.
Developing countries. There are a few studies that investigate the exchange rate pass-through in developing countries, such as Loungani and Swagel (2001), Garcia and Restrepo (2001), Goldfajn and Werlang (2000), Frankel, Parsley and Wei (2005) and Barhoumi (2006). Loungani and Swagel (2001), using a panel of 53 developing countries: African countries – 16, Asian – 11, South American – 19, and Mediterranean – 7, in most countries found positive relationship between the exchange rate depreciation and the inflation rate . The author also uses dummy variables to control for the impact of different exchange rate regimes on the exchange rate pass-through in the countries. The case of developing countries with the floating exchange rate suggests that the impact of exchange rate depreciation on the price changes is positive and statistically significant.
Emerging Countries. Ca’ Zorzi et al. (2007) estimated effect of ERPT in 12 emerging markets in Asia, Latin America, and Central and Eastern Europe. The authors’ results turned out to contradict the common expectation that in ‘emerging’ countries ERPT is always higher than in “developed” countries. In fact, for emerging countries with lower inflation rates, (most notably the Asian countries), the ERPT was found to be low and similar to the levels of developed economies. This evidence that exchange rate pass-through is smaller in countries with lower inflation is also proven in Choudri and Hakura (2001), Ross (1998), Kuijs (2001). Research corroborates that pass-through from exchange rate changes to consumer prices has been extremely low over the past two decades for a large group of countries that have pursued stable monetary policies, including Ihrig, Marazzi, and Rothenberg (2006), Bailliu and Fujii (2004), and Sekine (2006).
Korhonen et al. (2005) assess the extent and speed of exchange rate pass-through in CIS countries, using vector autoregressive regressions, impulse functions and variance decompositions techniques and compare it with the benchmark group of emerging market economies. The result of ERPT for CIS countries appeared to be higher than for the other emerging countries. Checking variables for stationarity and coming to the conclusion that there is no cointegrating relationship, the authors estimate vector autoregressive (VAR) models without error correction terms.
Besides works that estimate exchange rate pass-through for groups of countries, a big variety of research s that examine exchange rate pass-through separately by countries exist.
Billmeier et al. (2002) using Error Correcting Model (ECM) for Croatia ERPT came to the conclusion that the retail price index moves positively with the exchange rate, indicating that a 10 percent devaluation results in a 3.3 percent rise of the retail price index.
For Russia Dobrynska et al. (2005) applied two-step procedure of constructing Error Correction Model (ECM) using Johansen cointegration analysis at the first step. The authors found that a 50% of the PTE on consumer prices vanishes almost entirely within one month. Comparison of PTE in Russia and other countries (France, Germany, Italy, Netherlands, Spain) reveals that pass-through elasticity is much stronger in Russia than in European countries, confirming high dependence of Russian economy on foreign markets.
Before estimating results of ERPT for republic of Moldova, it’s valuable to know the economic situation in its neighbor-countries. For Romania, Gueorguiev (2003) and later Cozmanca et al. (2009). estimate the exchange rate pass-through into import prices, producer prices and several different measures of consumer prices indices. The results obtained reveal almost complete pass-through into import prices, which is not contradicting to the theory of PPP and incomplete pass-through into producer and consumer prices.
The ERPT for Ukraine, estimated by Bandura (2010), is equal to 14 % for PPI and 9 % for CPI aggregate indexes, which is lower than existing in literature 30 % for developing countries. This research differs from existing works for Ukraine, because the author examines the ERPT effect separately for different groups of goods and services with the scope to separate effect on imported goods consumption and domestic goods.
Also, it should be mentioned that existing research can be classified by ERPT to what prices are used for evaluation. Most part of existing research is concentrated on the effects of exchange rate changes on import prices Goldberg and Knetter (1997).There are several works, which study PTE on producer and consumer prices (for example Woo (1984), Feinberg (1986, 1989), Parsley and Popper (1998), McCarthy (2000)); several consider ERPT on the export prices (e.g. Klitgaard (1999), Dwyer, Kent and Pease (1993)).
From technical point of view, there exist two methods to estimate ERPT. First one is based on pass-through regression. The other and most popular used is the structural vector autoregression (VAR) methodology. The VAR models were introduced by Christopher Sims (1972, 1980, 1986). This modeling strategy to estimate ERPT was developed for advanced countries by McCarthy (2000). Later on Hahn (2003) analyzes the ERPT for the euro area. The analysis is based on quarterly data covering the time period 1970Q2 to 2002Q.
Besides the work of Gigineishvili (2007), which estimates transmission mechanism through interest rates channel in the republic of Moldova, there is no work which would explore exchange rate pass-through in the republic of Moldova.
Theoretical background and methodology
The theory of ERPT is based on the Quantity Theory in an Open Economy, the so-called ‘four-way equivalence theorem’ with floating Exchange rates, including Purchasing Power parity theory, Interest Rate Parity, expectations theory of exchange rates, the International Fisher effect and the Fisher Open hypothesis.
Let me briefly describe each of these concepts before passing to the Monetary model.
Purchasing Power Parity Theory
Purchasing Power Parity Theory, basically, refers to the law of one price working at macro-level.
The formula which corresponds to the ‘the law of one price’ can be written in the following way:
where - the price of good i in the home country at time t , - home-currency price of foreign exchange; -the foreign price of good i at time t. According to the equation, the absolute version of the LOOP  tells that the same good should have the same price across countries if prices are expressed in terms of the same currency of denomination.
There also exists a weaker relative version, which can be expressed with the following formula:
where -domestic price in the next period, -foreign price in the next period, - exchange rate of the next period.
The relative version argues that the deviation, if any, between the prices of some good in the two countries in one time period also holds in the next period.
The summation of all the traded goods in each country provides the absolute version of the PPP hypothesis :
where the weights in the summation satisfy
Taking variables in logarithms, the absolute PPP condition can be derived as:
It can be easily seen that the real exchange rate can be expressed as
Here the attention should be paid to the ambiguity or duality of PPP formulation: under fixed exchange rates prices should adjust and, on the contrary, under flexible rates, it is the exchange rate which adjusts to the long-run requirements of PPP.
The PPP theory is based on quantity theory of money and several assumptions, such as full-price flexibility, no transportation costs, no tariffs, no fixed investments necessary for arbitrage (i.e. complete spatial arbitrage) and no other impediments to trade flows. Any violation of these conditions can, in principle, cause a violation of this law.
Interest rate parity reflects the relationship between spot and forward exchange rates adjusted with interest rate.
International Fisher effect
International Fisher effect comes from the Fisher effect for closed economy and which states that an expected change in the current exchange rate between any two currencies is approximately equivalent to the difference between the two countries' nominal interest rates for that time.
where i-nominal interest rate for home country, s –exchange rate in home country, -nominal interest rate for foreign country.
Fisher open hypothesis or the hypothesis of perfect asset substitutability
Fisher open hypothesis reflects the idea that difference in the interest rates should be underpinned by the movement in the spot rate of exchange.
This research is based on the Dornbush (1976) model, the first and most famous monetary model of exchange rates.
The model can be described by several equations.
First equation stands for uncovered interest rate parity, where i-nominal interest rate, -real interest rate, -inflation.
Second equation comes from purchasing power parity, so that the exchange rate is expected to adjust partially toward an equilibrium value.
A function of specific form, used by Dornbush (1987), which comes from the function of the demand for money M=m(i, ) , is expressed as:
or in logarithms
Rearranging and substituting second equation (2) into four equation (4), we obtain
This is a key equation, which relates the current exchange rate to the level of prices.
Graphically, Dornbusch model can be represented in the following way: in the graph below the vertical and horizontal axis stand for price level and exchange rate, respectively. The interrelation between prices and exchange rate is expressed by AA-line with equilibrium point P. AS the result of quantity of money reduction the new long-run equilibrium (point R) is formatted.
AA –line shows the combination of price levels and exchange rates that clear the market.
In general, econometric studies reveal that the volatility of relative prices is considerably lower than the volatility of nominal exchange rates and provide rejection of the LOOP for a very broad range of goods. Isard (1977) represents first attempt to estimate the LOOP for a number of traded goods and number of countries. The results received provide significant deviation from the LOOP and are highly correlated with exchange rate movements. Giovanni (1988) along with Bui and Pippinger (1990), Fraser, Taylor and Webster (1991) proved the same outcomes.
Among the recent literature, Engle and Rogers (1996) use CPI data for US and Canadian cities found that distance is one of the distortion factor for the law of one price validity. Also, they showed that ‘border effect’ has the same impact as additional 2500 to 23000 extra miles of distance between the cities considered.
There are two basic approaches for empirical estimation of ERPT. First one is based on ‘pass-through simple regression’. Before the 1980s many economists used linear regression on (de-trended) non-stationary time series data, which was criticized as a dangerous approach that could produce spurious correlation by Engle and Granger (1987).
An alternative approach is to use either VAR procedure or VEC procedure for cointegrated variables. VAR Methodology was introduced by Christopher Sims (1972, 1980, 1986). A vector autoregression is a generalization of the AR (p) model to the multivariate case. McCarthy (2000) first used this method for advanced countries to estimate ERPT. Later Choudhri et al. (2005) provided structural VAR evidence on the degree of exchange rate pass-through for several countries.
After estimating VAR model, additional estimation techniques can be applied. For example, variance decomposition shows for each variable the ratio of the forecast error variance that is attributable to its own shocks and to shocks stemming from the other (upstream) variables, or, in other words, how important variation in one variable is for explaining variation in other variable. Impulse-response function can be used to assess the speed and extent of the pass-through, it shows the estimated response of each variable to an impulse in one of the other variables.
Data and descriptive analysis
The choice of the dataset is guided by the theoretical background, previous literature experience and data availability for the republic of Moldova. The analysis is based on monthly data covering the period between 2000M01 and 2010M12 for CPI and PPI.
The variables used are:
CPI: Consumer price indices, which stand for demand-side inflation.
The CPI index (by type of products) published by National Bureau of Statistics of Republic of Moldova was used. The series was normalized (considering Dec1999=100) and transformed into logarithm. This transformation doesn’t change the path of the variables and, hence, is completely in line with the majority of the empirical works.
The data used represents monthly indices for foodstuff product total and individually by components/ groups of food-stuff products (bread and bakery products, milk and dairy products, fish and fish products, etc), non-foodstuff products total amount and by categories (clothing, footwear, carpets, etc.) and services (communication services, electricity supply, etc.). It would be easier to restrict estimation on the effect of CPI only, however, since there is no monthly data on indices of import/export goods, estimating ERPT effect on disaggregated data of CPI will give possibility to analyze the scale of ERPT on imported, exported and produced at home groups of goods separately. Following Korhonen (2005) I seasonally adjust the data on CPI (X-12 procedure in EViews 6.0) to avoid shocks of seasonality.
PPI: Producer price indices, which stand for supply-side inflation.
The PPI index (disaggregated into major industry groups) published by National Bureau of Statistics of Republic а Moldova is used. The series are normalized (considering Dec 1999=100) and transformed into logarithm. The data used is monthly indices for total product of industry and individually by category of industry (mining and quarrying, manufacturing industry, manufacture of food products and beverages, etc). The data on PPI is also seasonally adjusted.
Exchange rate: I use nominal effective exchange rate (henceforth, NEER). The series of nominal Effective Exchange rate cover periods 2000 M1-2010 M12 for estimating the effect on CPI and PPI. The primary source of data is International Financial Statistics, series code 921..NECZF.
Making choice which control variables to include in the model is judged by the theory, previous experience in estimating ERPT and availability of data. Based on the Dornbusch model from the equation (5), it would be correctly to include such variable as money supply, GDP level, interest rate.
Money supply M2. In the existing literature on ERPT there are several examples of which aggregate of money supply to include in the equation. Choosing between Money Supply M1 and M2 I acted on the premise that republic of Moldova represents a high dollarized economy and income from deposits’ interest represent additional source of income for a big part of the population. The M2 data is taken from IMF International Financial Statistics and is already seasonally adjusted. The data used is monthly and is represented in millions, national currency. M2 money aggregate includes money in circulation, current account balances and demand deposits in local and foreign currencies. Formally, M2 aggregate equals money base times the money multiplier.
GDP: The data on GDP is provided only quarterly. In the literature different tricks can be found how to bypass such data imperfection. There are examples of using different proxies for GDP level: Dobrynska (2005) for Russia ERPT uses real consumption, there are few examples of using industrial indices (Bandura (2010) for Ukraine), nominal average salary. Using different proxies for GDP is explained by high correlation between output level and these variables. For example, in republic of Moldova average wage correlates with GDP with coefficient 0.91.
The second approach is data interpolation, which I used in this work. For this purpose there are two commonly used methods: the quadratic method and the cubic spline method (Brown and Williams (1972)). I used the later, since it gives more smooth interpolated time series.
Also, there are two possibilities, whether to include Output Gap, which represents detrended data or to include real GDP data in the estimation equation. The outcome after inclusion one of them doesn’t differ much in our case.
Oil price: Price of oil is the spot price of crude oil on New-York Mercantile Exchange (NYMEX), provided by Bloomberg.
According to McCarthy (2000) oil price captures supply shocks, identified from the dynamics of oil price inflation.
Interest rate: The inclusion of the interest rate is explained by the interest rate parity from equation (1). Because there are some periodical gaps in dataset for money market rates, I will use deposit rate and lending rate.
Descriptive statistics of explanatory variables is presented in Table 1, descriptive statistics of CPI and PPI, separately, by category is represented in Tables 2 and 3, respectively.
There is clearly observed trend of CPI growth over the analyzed period. Overall, prices on consumer foods and services tripled from the beginning of 2000 till the end of 2010. The most vulnerable to inflation processes categories of products and services are: grapes, fresh fruits, meat and meat products, sugar, nuts, public alimentation, medicaments, communal services. These results should have been expected since the biggest part of all these products except grapes and nuts are mostly imported. Quite surprising appears situation with grapes, fresh fruits and nuts, which are considered to be national product and are in abundant quantity in republic of Moldova and represent the biggest part of export. As can be seen from Figure 1, (Figure 2 for seasonal adjusted data) line for CPI is quite volatile with some significant deviations from the trend in periods 2002-2003, 2007-2010. The similar situation is observed for PPI. Most vulnerable production processes to inflation process are: electricity and heat, gas and water supply, total manufacturing industry (particularly, production, processing and preserving of meat and meat products, manufacture of food products and beverages, manufacture of wine), mining and quarrying. The inputs for electricity and gas supply industries are almost all imported (gas is imported from Russia, electricity and other fuel is partly imported from Ukraine). Due to this reason since 2010 NBM has started to target core inflation2, not base inflation, explaining this fact by inability of monetary policy to affect uncontrollable by NBM inflation processes. Period between 2008 and 2010 is characterized by marked drop in prices for CPI and PPI, exchange rate as well. This period is characterized by political instability in the country, change of political power, World Financial Crisis.
The dynamics of Nominal vs. average official exchange rates ( LEI/USD and LEI/EUR) are displayed in Figure 3.
Figure 1. Time profiles of NEER and national price indices.
Figure 2. Time profiles of NEER and national price indices, seasonally adjusted.
Figure 3. Time profiles of NEER and average official rates MDL/USD and MDL/EUR.
Before starting to estimate relationship between prices and exchage rates empirically, it is better firstly to familiarize with host of factors such as the structure of the economy, its specific characteristics, the extent of financial market development, the history of inflation, etc.
The monetary transmission mechanism in republic of Moldova is mainly represented by three channels: through exchnage rate, interest rate and inflation expectations. The scheme below is visual image of transmission mechanism in republic of Moldova. There are several works which study pass-through effect through interest rate (IDIS Viitorul Report, Gigineishvili (2007)). All of them converge to the inference of weak interest rate pass-through effect. Combining with the fact that in many countries with undeveloped financial markets, import dependence of inputs and consumption, and lack of central bank credibility, foreign exchange channel is considered to be one of the influential factors for prices changes.
Sceme: Transmission mechanism in Republic of Moldova.
Money market operations
Interbank market operations
Monetary policy instruments
Monetary policy influence
Market interest rate
Liquidity and credits
Expectations on exchange rate
Expectations on inflation
Source: Monetary policy annual book. National Bank of republic of Moldova.
So, the empirical analysis started with data refinining. Firstly, all data was checked fot stationarity of the time series with unit root tests. I test whether the assumed time series are I (1). To do that I employ the very standard Augmented Dickey-Fuller test (ADFt). First, I test for the unit roots in the cases when intercept and trend is present in the regression, then when there is the intercept only, and finally without intercept and trend. If I am not able to reject the null hypothesis about the unit root I run the ADF test on the first differences of the original time series.
The results, as expected, reveal that time series of consumer/producer price indices and exchange rates in the republic of Moldova are nonstationary. Therefore, the time series data should be taken in difference. The results of ADF test for differenced time series for regressors and regressands are represented in Tables 5, 6 respectively.
After checking for the order of integration for individual variables, the order of cointegration between the set of variables is checked. The cointegration analysis started from the Stock and Watson (1988), which stated that if examining variables share common stochastic trend then they are cointegrated. Such a relationship between variables can have a side effect on their mutual behavior.
There exist two tests which estimate if variables are integrated. First one was developed by Engle and Granger (1987) and Granger (1988) and the second was developed by Johansen (1988). Both tests reveal no cointegration. The results are presented in Table 7 for CPI by category and in Table 8 for PPI by category.
Since the data checking for unit root test, turned out to be non-stationary and there is no cointegration between variables, so VAR in differences is the most appropriate procedure for estimating ERPT.
For this purposes, the set of two equations is estimated:
- inflation level, EX-exchange rate, i-interest rate, X- control variables (gdp, money supply, etc.)
The set of equation is estimated using VAR procedure. The time lags for each type of indices were determined individually by running ‘varsoc’ command in Stata, which gives the lags that produce best fit and statistically significant coefficient estimates. The results of lags selection are represented in the Table 6 (Appendix).
The results of ERPT estimation is represented in Table 8. The found result contradicts to existing in literature for developing economies. In most works the ERPT for developing countries is found to be higher than for developed countries and account for around or more than 50 percent. (Dobrynska (2005) for Russia, for PPI around 50%, and for CPI more than 50 % of ERPT). The pass-through appears rather weak and quite interesting result is that the relationship between exchange rate and prices is observed to be reverse.
The insignificance of GDP coefficient can be explained by the fact that around 40% of GDP is represented by currency inflows from abroad. However, no reliable data are available for monthly inflow of remittances, so to be able to estimate their effect on transmission mechanism.
Since the effect of GDP is in most cases insignificant, but money supply and interest rates represent the main channels of transmission.
In VARs the direction of causality can be tricky, and can’t really be ascertained by just looking at coefficients. In theory, if the purchasing power parity holds, it can be both prices and exchange rate which adjust. And it can be that prices react first. One needs to look at variance decomposition and Granger causality tests.
A Wald test is commonly used to test for Granger causality. In a VAR model, under the null hypothesis that variable exchange rate does not Granger cause variable inflation, all the coefficients on the lags of variable ‘exchange rate’ will be zero in the equation for variable ‘inflation’ and ,on the contrary, will be zero on the lags of variable ‘inflation’ for the variable ‘exchange rate’. The results from Granger causality (tables 11, 12 for CPI and PPI, respectively) and Variance Decomposition prove the fact of reverse pass-through effect.
The weak reaction of pass-through mechanism can be explained by high market liquidity, which can’t be sterilized with real market interest rates. The high liquidity is observed from foreign currency inflows into moldavian economy from abroad, mainly represented by remittances. In 2009, according to Development Prospects Group, World Bank report, Moldova occupied 4th place in the list of countries with highest percent of remittances in GDP. Lack of monetary market profoundness and its undeveloped infrastructure makes impossible to control liquidity and therefore, inflation in the country. The results reveal imprudent exchange rate policy from the CB of republic of Moldova.
IDIS ‘Viitorul’ Report: ‘Analiza tendenţilor pieţei monetare şi a mecanismelor de transmisie monetară asupra inflaţiei. Oportunitatea aplicării instrumentelor de ţintire a inflaţiei în Republic Moldova.
Bache, Ida Wolden. 2006. Econometrics of Exchange Rate Pass-Through. In Doctoral dissertation in Economics. No6. Norges Bank. DO
Bandura, Pavel. 2010. A Study of Exchange Rate Pass-through in Ukraine. In KSE MA Thesis. Kyiv: Kyiv School of Economics.
Bogdan Cozmanca, and Florentina Manea. 2009. Exchange rate Pass- through into Romanian Price Indices: A VAR Approach. In Advances in Economic and Financial Research-DOFIN Working Paper series, 34.
Ca’ Zorzi, Michele, Hahn Elke, and Marcelo Sánchez. Exchange Rate Pass-Through in Emerging Markets. In Working Paper Series. No 739 / March 2007.
Campa, Jose Manual, and Goldberg, Linda S. 2004, Exchange Rate Pass-Through into Import Prices. In CEPR Discussion Paper. No. 4391.
Campa, Jose Manual, Goldberg, Linda S. and Jose Manual, González-Mínguez. 2005. Exchange Rate Pass-Through to Import Prices in the Euro Area. In Federal Reserve Bank of New York Staff Paper No. 219.
Cassel, Gustav. 1918. Abnormal Deviations in International Exchanges. Economic Journal, Vol. 28, pp. 413-15.
Cassel, Gustav. 1922. Money and Foreign Exchange after 1914 (London: Constable).
Choudhri, Ehsan U., Hamid Faruqee, and Dalia S. Hakura. 2005. Explaining the Exchange Rate Pass-Through in Different Prices. Journal of International Economics, 65, 349–374.
Dobrynskaya Viktoria, and Levando Dmitriy. 2005. A Study of Exchange Rate Pass-Through in Russia. In HSE Series Research in Economics and Finance. Moscow: Higher School of Economics.
Dornbush, Rudiger. 1987. Exchange Rate and Prices. American Economic Review, vol. 77, No. 1, 93-106.
Engel, Charles. 2002b. The Responsiveness of Consumer Prices to Exchange Rates: A Synthesis of Some New Open-Economy Macro Models. In The Manchester School Supplement, 1–15.
Engel, Charles. 2002a. Expenditure Switching and Exchange Rate Policy. NBER Macroeconomics Annual 17: 231–72.
Engle, Robert, and Clive Granger. 1987. Cointegration and Error Correction: Representation, Estimation, and Testing. Econometrica, Vol. 55, pp. 251-76.
Faruqee, H. 2004. Exchange-rate pass-through in the euro area: The Role of Asymmetric Pricing Behaviour. In IMF Working Paper No. 04/14.
Feenstra, Robert.1995. Empirical Studies of Trade Policy. In Grossman and Rogoff (eds.) Handbook of International Economics. Amsterdam, North Holland.
Ganev, Georgy, Krisztina Molnar, Krzysztof Rybinski and Przemyslaw Wozniak. 2002. Transmission Mechanism of Monetary Policy in Central and Eastern Europe. In CASE Report, No.52.
Goldberg, Pinelopi Koujianou and Michael, M. Kneter. 1997. Goods Prices and Exchange Rates. What Have We Learned? Journal of Economic Literature. Vol. 35, No. 3, 1243-1272.
Granger, Clive, and Paul Newbold. 1974. Spurious Regressions in Econometrics. Journal of Econometrics, Vol. 2, pp. 111-20.
Gueorguiev, N. 2003. Exchange rate pass through in Romania. In IMF Working Paper, 130/2003.
Hahn, E. 2003. Pass-Through of External Shocks to Euro Area Inflation, In ECB Working Paper No. 243.
Hooper, Peter, and Mann, Catherine. Exchange Rate Pass-Through in the 1980-s: The Case of U.S. Import of Manufactures. In Brookings papers on Economics Activity.
Johansen, S., (1988) “Statistical Analysis of Cointegration Vectors”, Journal of Economic
Dynamics and Control, Vol. 2 (June-September), pp. 231-54.
Korhonen, Iikka and Paul Wachtel. 2006. A Note on Exchange Rate Pass-Through in CIS Countries. In Research in International Business and Finance, Vol. 20, No. 2, 215-226.
Krugman, Paul. 1986. Pricing to Market when the Exchange Rate Changes. In NBER Working Paper, No. 1926.
McCarthy John. 2000. Pass-Through of Exchange Rates and Import Prices to Domestic Inflation in Some Industrialized Economies. In NBER Working paper.
Michael, B. Devereux, and Charles Engel. 1999. The Optimal Choice of Exchange-Rate Regime: Price-Setting Rules and Internationalized Production. In NBER Working Papers 6992, National Bureau of Economic Research.
Mihaljek, Dubravko and Marc, Klau. 2008. Exchange Rate Pass-Through in Emerging Market Economes: What has changed and Why? In Transmission Mechanisms for Monetary Policy in Emerging Market Economies. In BIS Paper No. 35.
Officer, L.H. 1982. Purchasing Power Parity and Exchange Rates: Theory, Evidence and Relevance. In Greenwich, Connecticut: JAI Press.
Parsley, David, and Helen Popper. 1998. Exchange Rates, Domestic Prices and Central Bank Actions: Recent US Experience. Southern Economic Journal 64(April): 957-972.
Rogoff, Kenneth. 1992. Traded Goods Consumption Smoothing and the Random Walk Behavior of the Real Exchange Rate. Monetary and Economic Studies, Vol. 10, pp. 1-29.
Tsesliuk, Siarhei. 2002. Exchange Rate and Prices in an Economy with Currency Substitution: The Case of Belarus. In KSE EERC MA Thesis. Kyiv: Kyiv School of Economics.
Woo, Wing T. Exchange Rates and the Prices of Nonfood, Nonfuel products.2: 1984. Brookings papers on economic activity. 511-30.,
Brown and Williams. 1972. Interpolation in Economic Time Series. Research Exercise Note. 5/72, CSO.
Need help with your literature review?
Our qualified researchers are here to help. Click on the button below to find out more:
In addition to the example literature review above we also have a range of free study materials to help you with your own dissertation: