Annual Conference | May 17-18, 2008 Exchange Rate Dynamics in Armenia: How predictable are they? Vahé Heboyan PhD Candidate, International Economics &

Annual Conference | May 17-18, 2008

Exchange Rate Dynamics in Armenia:

How predictable are they?

Vahé HeboyanPhD Candidate, International Economics &

Development

AIPRG Conference | May 17-18, 2008 | Washington DC ¡ Vahé Heboyan Slide 2

Outline

Motivation Theoretical background Methodology Data and series Preliminary results Future steps


Importance of Exchange Rates

The most important relative price in international finance

Getting the right exchange rate is a key objective for international investors, MNCs, and scientists.

Poor exchange rate policies were blamed for economic crisis in the developing world

Exchange rate directly affects economic performance International competitiveness Inflation Output and FDI Currency crisis


70.000

80.000

90.000

100.000

110.000

120.000

130.000

140.000

150.000

160.000

170.000

1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

0.0

20.0

40.0

60.0

80.0

100.0

120.0

140.0

160.0

NEER REER $/AMD €/AMD RR/AMD

Exchange Rates in Armenia


Motivation

Government Rapid growth in

remittances from abroad

Economic growth Weakening US$ Price increase in

world markets

Non-Government Those opposing Gov.

policies authorities engineered

appreciation to pocket hard currency and benefit gov-connected importers

Price increases for most goods Monopolized imports?

(Smbat Nasibyan, Chairman, Converse Bank)

Exporters have articulated for more intervention

Increasing controversies over the factors behind appreciation


Central Bank’s Position

Monetary policy priority: Inflation management 3-4±1.5 % annually

Appreciation has helped to prevent complete pass-through of increasing world oil prices to the domestic market

Appreciation creates unique opportunities for local businesses to invest in advanced foreign technologies as they become cheaper Increased productivity Increased competitiveness


PhD Dissertation

Objective: survey, adapt, and extend empirical models to the benefit of understanding and practically analyzing exchange rate dynamics and behavior in Armenia Understand the behavior and determinants of

exchange rate in Armenia Understand the factors behind rapid appreciation of

Armenia’s currency Examine the dynamics of the real exchange rate and

its variation from its long-run equilibrium path Examine adequacy of theoretical models in ER

determination and policy decision making in Armenia Analyze the exchange rate pass-through into prices

and impact on macro-economic performance

Exchange Rate Dynamics and Macroeconomic Impact in Transition Countries: An Empirical Investigation of Armenia’s Exchange Rate


Paper Objectives Find the ‘best’ modeling approach that

helps to explain exchange rate dynamics and determinants in Armenia

Scientifically explain ER dynamics and appreciation Market vs. non-market forces Stop different speculative theories

Examine applicability in monetary policy decision making process How stable is the model?

Series Time span

Out-of-sample prediction ?


Exchange Rate Determination Random Walk: ERs evolve according to

a random walk, thus cannot be predicted

UIP: expected change in ER determined by interest differentials

PPP: constant equilibrium ER Balassa-Samuelson: PPP fails as a result

of international productivity differentials Mundell-Fleming: a standard for

understanding the role of monetary and fiscal policies in an open economy

Monetary Approach: Flexible-price monetary model Dornbusch’s “overshooting” model

Portfolio Balance Approach Other


Overall Consensus There is no “best” model Long run: based on economic fundamentals Short run: expectations and capital

movements

Choice depends on data frequencies, time span, degree of country’s development, and even the particular pair of bilateral exchange rates (Rosenberg, 2003).

Selection criteria Assumptions reflecting reality in the country(s)

of interest Explanatory power


Concept of Misalignment and ERER

Edwards (1989, 1994) Sustained deviations of the actual

real exchange rate from its long-run equilibrium level.

What is the Equilibrium RER? Long-run equilibrium is achieved at

the point when stock-point equilibrium is achieved for all agents of the economy (Driver and Westaway,2004)


Empirical Approaches to Equilibrium ERs

Source: Driver and Westaway 2004

UIP PPP Balassa-Samuelson

Monetary Models

CHEERs ITMEERs BEERs FEERs DEERs APEERs PEERs NATREX SVARs DSGE

Name Uncovered Interest Parity

Purchasing Power Parity

Balassa-Samuelson

Monetary and Portfolio balance models

Capital Enhanced Equilibrium Exchange Rates

Intermediate Term Model Based Equilibrium Exchange Rates

Behavioural Equilibrium Exchange Rates

Fundamental Equilibrium Exchange Rates

Desired Equilibrium Exchange Rates

Atheoretical Permanent Equilibrium Exchange Rates

Permanent Equilibrium Exchange Rates

Natural Real Exchange Rates

Structural Vector Auto Regression

Dynamic Stochastic General Equilibrium models

Theoretical Assumption

The expected change in the exchange rate determined by interest differentials

Constant Equilibrium Exchange Rate

PPP for tradable goods. Productivity differentials between traded and non-traded goods

PPP in long run (or short run) plus demand for money

PPP plus nominal UIP without risk premia

Nominal UIP including a risk premia plus expected future movements in real exchange rates determined by fundamentals

Real UIP with a risk premia and/ or expected future movements in real exchange rates determined by fundamentals

Real exchange rate compatible with both internal and external balance. Flow not full stock equilibrium

As with FEERs, but the definition of external balance based on optimal policy

None As BEERs As with FEERs, but with the assumption of portfolio balance (so domestic real interest rate is equal to the world rate)

Real exchange rate affected by supply and demand (but not nominal) shocks in the long run

Models designed to explore movements in real and/ or nominal exchange rates in response to shocks

Relevant Time Horizon

Short run Long run Long run Short run Short run (forecast)

Short run (forecast)

Short run (also forecast)

Medium run Medium run Medium run/ Long run

Medium run / Long run

Long run Short (and short) run

Short and long run

Statistical Assumptions

Stationarity (of change)

Stationary Nonstationary Nonstationary Stationary, with emphasis on speed of convergence

None Nonstationary Nonstationary Nonstationary Nonstationary (extract permanent component)

Nonstationary (extract permanent component)

Nonstationary As with theoretical

As with theoretical

Dependent Variable

Expected change in the real or nominal

Real or nominal

Real Nominal Nominal Future change in the nominal

Real Real effective

Real effective

Real Real Real Change in real

Change relative to long-run steady state

Estimation Method

Direct Test for stationarity

Direct Direct Direct Direct Direct Underlying balance

Underlying balance

Direct Direct Direct Direct Simulation


Behavioral Equilibrium Exchange Rate

Popularized by Edwards (1989, 1994) MacDonald (1997)

Assessing whether movements in the RER represent a misalignment or movements are results of changes in economic fundamentals.

Edwards’ (1989, 1994) model is tailored to developing countries with an aim to understand the relationship b/w RER and economic fundamentals.

Has been used extensively by IMF and others.


Basic Model Structure Real exchange rate

e* - ERER, in turn a function of fundamentals Zt - an index of macroeconomic policies (i.e.

the rate of growth of domestic credit) Zt*- sustainable level of macroeconomic

policies (i.e. the rate of increase of demand for domestic money)

St - nominal exchange rate PMPR - spread in the parallel market for

foreign exchange θ,λ, φ and ψ are positive parameters that

capture the most important dynamic aspects of the adjustment process.

* *1

1 1

log log log

log log

t t t t t

t t t t

e e e Z Z

S S PMPR PMPR


Basic Model Structure Equilibrium Real Exchange Rate (ERER)

FUND - set of fundamental variables that are assumed to have a determining effect on the ERER Terms of trade for goods (TOT). Is defined as the price ratio

of the country’s exports over imports. Government spending (GOV). Usually defined as the

government consumption of the non-tradable goods. Market openness (OPEN). A proxy for trade controls/

restrictions. Technological progress/productivity (TECHPRO). Allows

capturing the famous Balassa-Samuelson effect. Investment (INV). Debt service (DS). Is defined as a share of exports. Net foreign assets (NFA). Is a proxy for the country’s net

external position and is defined as a share of GDP. Aid flows (AID) as a share of exports. Controls over capital flows (CAPCTRL). Similar to the

market openness liberalization will have impact on the ERER.

*0log t i tt

e FUND


Basic Model Structure Equation to be estimated using conventional

methods.

POLICY - single notation of macroeconomic policy variables

NOMDEV - stands for nominal devaluation γ’s are combinations of β’s and θ.

1log log 1 logt i it t i it t te FUND e POLICY NOMDEV


Methodology

Issues of stationarity in time series

Co-integration & error-correction Chudik and Mongardini (2007)

Determination of variables’ order of integration becomes uncertain due to poor performance of unit root tests for small samples


Methodology Gregory and Hansen (1996)

“the standard tests for cointegration are not appropriate, since they presume that the cointegrating vector is time-invariant under the alternative hypothesis (p. 100)”

if there exists a cointegration, the standard ADF test may not reject the null, thus wrongly concluding that there is no long-run relationship.

Gregory, Nason, and Watt (1996) the power of standard ADF test decreases

sharply when a structural break is present. technological progress, economic crises, changes in

the people’s preferences, policy or regime shifts, and institutional developments,

very typical to developing and transition countries.


Methodology Traditional econometric cointegration

approaches (e.g. Johansen’s) require the series to be integrated to the same order => introducing a further degree of uncertainty

into the analysis of level relationships especially in a transition country setting.

Pesaran, Shin, and Smith (2001) – Bounds Testing testing for the existence of level relationship

between variables irrespective of whether the underlying variables are stationary, integrated to the order of one, or a mixture of the two.

this approach has been successful and superior to the traditional Johansen cointegration test in a small sample.


ARDL

Autoregressive Distributed Lag Modeling by Pesaran and Shin (1999) estimation is independent of the

order of integration in the variables => provides statistically better

results (Mongardini, 1998). General structure ,

1 1 1

y xp pn

t i t i ji j t ii j i

y y x c


Error correction model

Bounds test (Pesaran, Shin, and Smith, 2001) F-stat for existence of the level

relationship Hypothesis

Joint hypothesis

'0:0: ,00, xyxyyxyxyy HandH

1 1

1 , , 1 ,1 1 1 0

y xp pn n

t yy t yx j j t i t i ji j t ij i j i

y y x y x c

'0:0: ,11, xyxyyxyxyy HandH

xyxyy

xyxyy

HHH

HHH,

,

111

000


Estimation

Econometric Tamplate developed by Chudik (2006) Estimates using ARDL User friendly, Windows based interface Modified to fit our purpose

Data selection Small sample size requires variable

selection Potential 9 explanatory variables Up to 4 commonly used in the

empirical literature


Data and Series

1996:Q1 – 2007:Q4Variable

Description Log

REER Real equilibrium exchange rate. Is a multilateral CPI based real effective exchange rate. It is defined in terms of Armenian Dram per unit of foreign currency, so that an increase (decrease) in REER represents depreciation (appreciation).

Yes

ToT Terms of trade for Armenia is defined as the ratio of export price index over import price index.

Yes

GOV Government consumption as a share of GDP relative to that of foreign trading partners.

Yes

OPEN

Openness to trade: exports plus imports as a share of GDP. Yes

NFA Net foreign assets as a share of GDP. No

INV Direct investments in rep. economy as a share of GDP relative to that of the trading partners.

Yes

EXCRE

Excess supply of domestic credit. Measured as the rate of growth of domestic credit minus the lagged rate of growth of real GDP.

No

PROD

Measure of productivity. Proxied by per capita real GDP relative to that of the trading partners.

Yes

DS Debt service as a share of exports. No

MONDEV

Nominal devaluation. Nominal effective exchange rate is used.

No


Preliminary Results

126 models in various specifications

Selection criteria existence of a long-run cointegration

using the bounds testing; the number of statistically significant

variables at 5 percent significance level; and

the number of correct signs of the parameters as predicted by the economic theory.


Two best models – Model A

4.1

4.2

4.3

4.4

4.5

4.6

4.7

4.8

4.9

5.0

19961 19971 19981 19991 20001 20011 20021 20031 20041 20051 20061 20071

ERER REER ERER_smoothed

3.8 4.5 2.3 5.00.48 0.80 0.05 0.63ERER GOV NFA INV EXCRE


Two best models – Model B

NOMDEVEXCRENFAGOVERER 6.45.83.75.3

03.088.061.128.0

4.0

4.2

4.4

4.6

4.8

5.0

5.2

5.4

19961 19971 19981 19991 20001 20011 20021 20031 20041 20051 20061 20071

ERER REER ERER_Smoothed


Misalignment

Model A

Model B

-40%

-30%

-20%

-10%

0%

10%

20%

30%

40%

50%

19961 19971 19981 19991 20001 20011 20021 20031 20041 20051 20061 20071

Misalignment Misalignment_smoothed

-80%

-60%

-40%

-20%

0%

20%

40%

60%

19961 19971 19981 19991 20001 20011 20021 20031 20041 20051 20061 20071

Misalignment Misalignment_smoothed


Conclusions

Preliminary results confirm RER misalignment

However, very sensitive to set of fundamentals

Both models have statistically strong and yield similar misalignment trends, however they differ in the magnitude.


Future research is needed

Further analysis are necessary to study the sensitivity of the results to the choice of the variable set.

Given theoretical ambiguity of the signs of majority of variables, additional work is required to make the right inference about the results.


Additional studies Short run volatility as a function of

expectations (incl. political dummy) and monetary aggregates

Pass through into important price of important products will be examined critical policy implication in order to

determine the appropriate monetary policy response

can influence the future direction of the economic, social, as well as political developments in the country.

IMF (2007): “Look for ways to reduce monopolistic practices in the import business, with a view to increasing the pass-through of exchange rate changes to domestic prices”

Thank You

Comments and suggestions are gladly welcome !!!

Annual Conference | May 17-18, 2008 Exchange Rate Dynamics in Armenia: How predictable are they? Vahé Heboyan PhD Candidate, International Economics &

Documents

armenia slide

appreciation slide

determinants of exchange

aiprg conference

right exchange rate

exchange rate passthrough

real exchange rate

developing world exchange