Monetary Decentralization Case of Armenia
bdesilets
0 views
39 slides
Sep 29, 2025
Slide 1 of 39
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
About This Presentation
The article “Monetary Decentralization: The Case of Armenia” (2006) analyzes Armenia’s transition from the Soviet ruble to its own currency, the dram, within the framework of optimal currency area theory. It examines Armenia’s monetary management, trade patterns, monetary policy preferences,...
Slide Content
MONETARY DECENTRALIZATION
THE CASE OF ARMENIA
February 2006
ABSTRACT
I analyze the efficiency of the monetary decentralization of the Soviet Union for the case of Armenia
using the following criteria: monetary management, trade, monetary policy preferences (prices, GDP,
wages, interest rates), financial flows, factor mobility and banking. I determine that the issuance of
Armenia’s new currency, the dram, was efficient and resulted in substantial gains for the Armenian
economy. I also look to the future to analyze the possibility of Armenia continuing as an independent
currency zone and the efficiency of it entering into monetary union with Russia and the CIS, Europe, the
US and Iran.
JEL Classification
E42: Monetary Systems; Standards; Regimes; Government and the Monetary System
E50: Monetary Policy, Central Banking and the Supply of Money and Credit
Keywords: Armenia, Soviet Union, Transition Economies, Monetary Economics, Dollarization, EU
Enlargement
Word Count: 9,332
MONETARY DECENTRALIZATION
THE CASE OF ARMENIA
I. INTRODUCTION
No one branch of economic theory offers complete answers for whether certain countries should issue
their own currencies, rely on currencies issued by other countries or shed their own currencies for
supranational currencies. The most relevant theories come from the field of optimal currency areas. The
foundation articles by Mundell (1961), McKinnon (1963) and Kenen (1969) identify some basic
parameters for analyzing the benefits of joining a currency union versus maintaining a separate currency.
Some authors have built on these three works in an attempt to establish a formal model, such as Bayoumi
(1994). However, there is far more interesting reading on the empirical side of the literature.
Eichengreen (1992), Krugman (1993), Alesina and Grilli (1992) and Sachs and Sala-i-Martin (1992) all
entered the debate over European monetary union that heightened in the early 1990s. From the same
period is Bertola (1993) who focuses on economic more than monetary integration but offers some
THE CASE OF ARMENIA
February 2006
ABSTRACT
I analyze the efficiency of the monetary decentralization of the Soviet Union for the case of Armenia
using the following criteria: monetary management, trade, monetary policy preferences (prices, GDP,
wages, interest rates), financial flows, factor mobility and banking. I determine that the issuance of
Armenia’s new currency, the dram, was efficient and resulted in substantial gains for the Armenian
economy. I also look to the future to analyze the possibility of Armenia continuing as an independent
currency zone and the efficiency of it entering into monetary union with Russia and the CIS, Europe, the
US and Iran.
JEL Classification
E42: Monetary Systems; Standards; Regimes; Government and the Monetary System
E50: Monetary Policy, Central Banking and the Supply of Money and Credit
Keywords: Armenia, Soviet Union, Transition Economies, Monetary Economics, Dollarization, EU
Enlargement
Word Count: 9,332
MONETARY DECENTRALIZATION
THE CASE OF ARMENIA
I. INTRODUCTION
No one branch of economic theory offers complete answers for whether certain countries should issue
their own currencies, rely on currencies issued by other countries or shed their own currencies for
supranational currencies. The most relevant theories come from the field of optimal currency areas. The
foundation articles by Mundell (1961), McKinnon (1963) and Kenen (1969) identify some basic
parameters for analyzing the benefits of joining a currency union versus maintaining a separate currency.
Some authors have built on these three works in an attempt to establish a formal model, such as Bayoumi
(1994). However, there is far more interesting reading on the empirical side of the literature.
Eichengreen (1992), Krugman (1993), Alesina and Grilli (1992) and Sachs and Sala-i-Martin (1992) all
entered the debate over European monetary union that heightened in the early 1990s. From the same
period is Bertola (1993) who focuses on economic more than monetary integration but offers some
Monetary Decentralization: The Case of Armenia 2
interesting perspectives. Carlino and DeFina (1999), among others, have analyzed differences among
economic regions in the US and their reactions to federal monetary policy.
From the developing world have come articles focusing on the costs and benefits of dollarization and the
effects of industrial country policies on output and real effective exchange rates in the developing world.
The most interesting of these articles include Schmitt-Grohe and Uribe (2000), Calvo, Leiderman and
Reinhart (1993) and Del Negro and Obiols-Homs (1999).
Literature on the topic of monetary decentralization is especially scant. Nordhaus, Peck and Richardson
(1991) discuss the economic consequences of the breakup of the Soviet Union and touch on the issue of
monetary decentralization but only in the wider framework of economic reform. They reference the
breakup of the Austro-Hungarian Empire as the most relevant example for the breakup of the Soviet
Union since, unlike other European empires and similar to the Soviet Union, the Austro-Hungarian
Empire was a land empire of adjacent states that shared a single currency.
Other relevant examples of monetary decentralization are those of African monetary unions. The East
African Currency Board (EACB), Communauté Financière Africaine (CFA) and Rand Monetary Area
(RMA) all began as rule-based arrangements with fixed exchange rates, full convertibility, limits on
central bank lending to governments and minimum levels of foreign reserves. All of them subsequently
underwent rule changes that lowered the credibility of their monetary policies. Countries that did not
adhere to limits on lending to government and minimum levels of foreign reserves suffered from high
inflation and crowding out of private investment
1
.
The countries of the EACB (Kenya, Tanzania and Uganda) all experienced increased inflation and
increased lending to government after their monetary union fell apart in 1966. During the monetary
1
Guillaume and Stasavage 2000.
interesting perspectives. Carlino and DeFina (1999), among others, have analyzed differences among
economic regions in the US and their reactions to federal monetary policy.
From the developing world have come articles focusing on the costs and benefits of dollarization and the
effects of industrial country policies on output and real effective exchange rates in the developing world.
The most interesting of these articles include Schmitt-Grohe and Uribe (2000), Calvo, Leiderman and
Reinhart (1993) and Del Negro and Obiols-Homs (1999).
Literature on the topic of monetary decentralization is especially scant. Nordhaus, Peck and Richardson
(1991) discuss the economic consequences of the breakup of the Soviet Union and touch on the issue of
monetary decentralization but only in the wider framework of economic reform. They reference the
breakup of the Austro-Hungarian Empire as the most relevant example for the breakup of the Soviet
Union since, unlike other European empires and similar to the Soviet Union, the Austro-Hungarian
Empire was a land empire of adjacent states that shared a single currency.
Other relevant examples of monetary decentralization are those of African monetary unions. The East
African Currency Board (EACB), Communauté Financière Africaine (CFA) and Rand Monetary Area
(RMA) all began as rule-based arrangements with fixed exchange rates, full convertibility, limits on
central bank lending to governments and minimum levels of foreign reserves. All of them subsequently
underwent rule changes that lowered the credibility of their monetary policies. Countries that did not
adhere to limits on lending to government and minimum levels of foreign reserves suffered from high
inflation and crowding out of private investment
1
.
The countries of the EACB (Kenya, Tanzania and Uganda) all experienced increased inflation and
increased lending to government after their monetary union fell apart in 1966. During the monetary
1
Guillaume and Stasavage 2000.
Monetary Decentralization: The Case of Armenia 3
union, Kenya served as a disciplinarian for the other two countries, vetoing any increases in borrowing
limits and keeping its own borrowing below 50 percent of the allowable ceiling most of the time. After
the monetary union, Kenya performed best of the three countries. In the period following the breakup
(1967-1973), Kenya’s budget deficit actually decreased whereas Tanzania’s nearly doubled and Uganda’s
rose nearly 20 percent. Kenya’s average annual GDP growth rate nearly doubled, whereas Tanzania’s
dropped from 10.1 percent to -1.5 percent. Uganda’s growth rate increased from 3.9 to 11.5 percent,
which is attributed to improved terms of trade
2
.
From all of these authors in all of these fields of economic theory come a few basic parameters for
analyzing the choices of monetary regime. They are summarized below.
Monetary management. For developing countries that have adopted another country’s currency or have
pegged their currencies to another currency, the overriding issue has been monetary management.
Usually, mismanagement is accompanied by a public sector that is too large and a government with little
or no fiscal discipline. The result is the issuance of far more currency than is needed for the economy,
leading to inflation and even hyperinflation. If there is no trust in the central bank, which usually is
correlated with low or no central bank independence, the country may eventually adopt another country’s
currency or fix its exchange rate to another country’s currency and limit issuance of its own currency to a
one-to-one ratio with reserves of the peg currency in what nowadays is referred to as a currency board
arrangement.
In the case of monetary decentralization, the issue of monetary management works in the other direction.
Countries generally issue their own currencies because of poor monetary management by the center.
Often, this is accompanied by or directly follows war, as in the case of the former Yugoslavia
3
, financial
2
Guillaume and Stasavage 2000.
3
See Desilets 2002.
union, Kenya served as a disciplinarian for the other two countries, vetoing any increases in borrowing
limits and keeping its own borrowing below 50 percent of the allowable ceiling most of the time. After
the monetary union, Kenya performed best of the three countries. In the period following the breakup
(1967-1973), Kenya’s budget deficit actually decreased whereas Tanzania’s nearly doubled and Uganda’s
rose nearly 20 percent. Kenya’s average annual GDP growth rate nearly doubled, whereas Tanzania’s
dropped from 10.1 percent to -1.5 percent. Uganda’s growth rate increased from 3.9 to 11.5 percent,
which is attributed to improved terms of trade
2
.
From all of these authors in all of these fields of economic theory come a few basic parameters for
analyzing the choices of monetary regime. They are summarized below.
Monetary management. For developing countries that have adopted another country’s currency or have
pegged their currencies to another currency, the overriding issue has been monetary management.
Usually, mismanagement is accompanied by a public sector that is too large and a government with little
or no fiscal discipline. The result is the issuance of far more currency than is needed for the economy,
leading to inflation and even hyperinflation. If there is no trust in the central bank, which usually is
correlated with low or no central bank independence, the country may eventually adopt another country’s
currency or fix its exchange rate to another country’s currency and limit issuance of its own currency to a
one-to-one ratio with reserves of the peg currency in what nowadays is referred to as a currency board
arrangement.
In the case of monetary decentralization, the issue of monetary management works in the other direction.
Countries generally issue their own currencies because of poor monetary management by the center.
Often, this is accompanied by or directly follows war, as in the case of the former Yugoslavia
3
, financial
2
Guillaume and Stasavage 2000.
3
See Desilets 2002.
Monetary Decentralization: The Case of Armenia 4
crises, as in the case of the provinces of Argentina
4
, or a combination of political and financial crises, as
in the case of the former Soviet Union.
Trade. Trade figures into the determination of optimal currency arrangements in at least two ways. First,
the size of the nontradable sector is viewed as an important parameter for a separate currency. The larger
the nontradable sector, the more justified is a separate currency. Second, two regions with a large amount
of trade ought to consider sharing a currency or fixing their exchange rates, to eliminate exchange rate
risks and facilitate that trade.
An important aspect of trade with respect to monetary regimes is its endogeneity. For example, trade
within the new country of Bosnia-Herzegovina increased and trade with its neighbors decreased after the
introduction of its new currency, the convertible marka, and the accompanying customs regime.
Monetary policy preferences. Different countries and regions have different preferences for the conduct
of monetary policy based on their own economic conditions. Those conditions include variations in
prices, wages and GDPs. Countries or regions whose prices, wages and GDPs are highly correlated will
demand the same monetary policies. Tradeoffs such as that between inflation and employment also could
be included among monetary policy preferences.
As with trade, monetary policy preferences are derived from factors that are somewhat endogenous to the
choice of monetary regimes. For example, pre-euro European countries were more diversified than US
regions precisely because they had separate currencies, so diversified to hedge against industry-specific
shocks. US regions, meanwhile, were free to specialize without concern for industry-specific shocks
because the overall US economy was diversified and featured a common currency and fiscal system
5
.
4
See Standard and Poor’s 2002.
5
Krugman 1993.
crises, as in the case of the provinces of Argentina
4
, or a combination of political and financial crises, as
in the case of the former Soviet Union.
Trade. Trade figures into the determination of optimal currency arrangements in at least two ways. First,
the size of the nontradable sector is viewed as an important parameter for a separate currency. The larger
the nontradable sector, the more justified is a separate currency. Second, two regions with a large amount
of trade ought to consider sharing a currency or fixing their exchange rates, to eliminate exchange rate
risks and facilitate that trade.
An important aspect of trade with respect to monetary regimes is its endogeneity. For example, trade
within the new country of Bosnia-Herzegovina increased and trade with its neighbors decreased after the
introduction of its new currency, the convertible marka, and the accompanying customs regime.
Monetary policy preferences. Different countries and regions have different preferences for the conduct
of monetary policy based on their own economic conditions. Those conditions include variations in
prices, wages and GDPs. Countries or regions whose prices, wages and GDPs are highly correlated will
demand the same monetary policies. Tradeoffs such as that between inflation and employment also could
be included among monetary policy preferences.
As with trade, monetary policy preferences are derived from factors that are somewhat endogenous to the
choice of monetary regimes. For example, pre-euro European countries were more diversified than US
regions precisely because they had separate currencies, so diversified to hedge against industry-specific
shocks. US regions, meanwhile, were free to specialize without concern for industry-specific shocks
because the overall US economy was diversified and featured a common currency and fiscal system
5
.
4
See Standard and Poor’s 2002.
5
Krugman 1993.
Monetary Decentralization: The Case of Armenia 5
Financial flows. One way to offset the disparate effects that monetary policy has on different regions
within a currency zone is through a shared fiscal system. Unemployment insurance and other social
safety nets will transfer income from regions that benefit from monetary policy to those that do not.
Sachs and Sala-i-Martin (1992) find that between increased transfers and reduced tax payments, the US
fiscal system offsets approximately 40 percent of regional shocks, including those caused by monetary
policy, such as changes in interest rates. Other financial flows, including foreign direct investment and
private remittances also help to transfer income among regions.
Banking. Another way to offset the disparate effects of monetary policies and to respond to demand
shocks is through the private banking system, which generally has been ignored by authors of currency
area and monetary regime literature. The main role of the banking system is to channel savings into
investment. Insofar as the savers and investors are divided into different regions, a banking system that
operates in both can serve the same function as a fiscal system that operates in both, allocating funds to
regions that have experienced demand shocks so they can invest in new technologies and develop new
goods and services that are in higher demand. Banks benefit from monetary unions because if their loans
and deposits are in the same currency, they do not have to worry about exchange rate risk.
Geographic factor mobility. The higher the geographic mobility of factors of production, the less
justified is a separate currency. This is because factors such as labor will migrate from regions whose
industries are suffering from decreased demand to regions that are experiencing increased demand. If
there are obstacles to mobility, such as language barriers or citizenship requirements, then separate
currencies are more justified. If the region suffering from decreased demand has a separate currency, it
can allow its currency to depreciate, thereby inviting external investment to re-orient its economy toward
industries experiencing increased demand.
Financial flows. One way to offset the disparate effects that monetary policy has on different regions
within a currency zone is through a shared fiscal system. Unemployment insurance and other social
safety nets will transfer income from regions that benefit from monetary policy to those that do not.
Sachs and Sala-i-Martin (1992) find that between increased transfers and reduced tax payments, the US
fiscal system offsets approximately 40 percent of regional shocks, including those caused by monetary
policy, such as changes in interest rates. Other financial flows, including foreign direct investment and
private remittances also help to transfer income among regions.
Banking. Another way to offset the disparate effects of monetary policies and to respond to demand
shocks is through the private banking system, which generally has been ignored by authors of currency
area and monetary regime literature. The main role of the banking system is to channel savings into
investment. Insofar as the savers and investors are divided into different regions, a banking system that
operates in both can serve the same function as a fiscal system that operates in both, allocating funds to
regions that have experienced demand shocks so they can invest in new technologies and develop new
goods and services that are in higher demand. Banks benefit from monetary unions because if their loans
and deposits are in the same currency, they do not have to worry about exchange rate risk.
Geographic factor mobility. The higher the geographic mobility of factors of production, the less
justified is a separate currency. This is because factors such as labor will migrate from regions whose
industries are suffering from decreased demand to regions that are experiencing increased demand. If
there are obstacles to mobility, such as language barriers or citizenship requirements, then separate
currencies are more justified. If the region suffering from decreased demand has a separate currency, it
can allow its currency to depreciate, thereby inviting external investment to re-orient its economy toward
industries experiencing increased demand.
Monetary Decentralization: The Case of Armenia 6
Industrial factor mobility. The more easily factors of production can shift among industries within the
same region, the less justified is a currency union. This is because if a region with high industrial factor
mobility experiences decreased demand for its products, it simply can reorient its factors of production to
another industry that is experiencing increased demand, alleviating the need for exchange rate
adjustments to help increase investment and competitiveness.
Economies of scale/Vehicle currencies. The existence of separate currencies increases transaction costs.
This leads to the demand for a common international trading currency, or vehicle currency, a function
now performed by the euro and the US dollar, among other regional currencies. Within a modern
financial system, there also exists demand for widely accepted and traded units of capital
6
. On the
international level, this demand currently is met by US Treasury bills. Within smaller economies, this
demand is met by national government bonds, mortgage bonds and other financial instruments.
This thesis attempts to analyze each of these parameters for the case of the Republic of Armenia.
Armenia is an interesting case study because it is a very small country with fewer than 4 million people
and an annual GDP of less than US$4 billion so offers a test for the efficiency of a small currency area.
The findings of studying Armenia then can be compared to studies of large currency areas such as the
euro zone. Armenia has a floating exchange rate, which means its currency truly is separate and distinct
from other currencies. Finally, Armenia’s currency is new so, to the extent that comparable data can be
found from before and after the introduction of the new currency, the effects of the new currency can be
measured.
Measurements are not available for all of the parameters of monetary regime efficiency. Indeed, one of
the main challenges in economic research on transition and developing economies is the lack of formal
and reliable data. However, some data is available for most of these parameters, if only after some
6
Ingram 1959.
Industrial factor mobility. The more easily factors of production can shift among industries within the
same region, the less justified is a currency union. This is because if a region with high industrial factor
mobility experiences decreased demand for its products, it simply can reorient its factors of production to
another industry that is experiencing increased demand, alleviating the need for exchange rate
adjustments to help increase investment and competitiveness.
Economies of scale/Vehicle currencies. The existence of separate currencies increases transaction costs.
This leads to the demand for a common international trading currency, or vehicle currency, a function
now performed by the euro and the US dollar, among other regional currencies. Within a modern
financial system, there also exists demand for widely accepted and traded units of capital
6
. On the
international level, this demand currently is met by US Treasury bills. Within smaller economies, this
demand is met by national government bonds, mortgage bonds and other financial instruments.
This thesis attempts to analyze each of these parameters for the case of the Republic of Armenia.
Armenia is an interesting case study because it is a very small country with fewer than 4 million people
and an annual GDP of less than US$4 billion so offers a test for the efficiency of a small currency area.
The findings of studying Armenia then can be compared to studies of large currency areas such as the
euro zone. Armenia has a floating exchange rate, which means its currency truly is separate and distinct
from other currencies. Finally, Armenia’s currency is new so, to the extent that comparable data can be
found from before and after the introduction of the new currency, the effects of the new currency can be
measured.
Measurements are not available for all of the parameters of monetary regime efficiency. Indeed, one of
the main challenges in economic research on transition and developing economies is the lack of formal
and reliable data. However, some data is available for most of these parameters, if only after some
6
Ingram 1959.
Monetary Decentralization: The Case of Armenia 7
patching among different sources. The most difficult measurements are those that attempt to show the
changes in the Armenian economy from before and after the breakup of the Soviet Union.
II. MONETARY MANAGEMENT
For a developing country, the decision to join a currency union or to peg its currency to another currency
usually is the result of poor monetary management and inflation. In the transition countries of the former
Soviet Union (FSU), the opposite situation existed. Countries decided to issue their own currencies
because of poor monetary management by the central Soviet government. In January 1992, Russia
liberalized prices which caused immediate and steep inflation. The price level increased 15-fold or 1,000
percent from 1991-1992 then another 30 percent in 1992-1993. A cash shortage ensued, exacerbated by
the fact that only Russia could issue rubles. This shortage was a major motivation for the former Soviet
republics to seek alternative monetary arrangements
7
.
Not everyone agreed that the individual republics should have issued their own currencies. The IMF
initially advised against it and attempted to establish a shared currency among separate central banks
8
. As
Nordhuas, Peck and Richardson (1991) noted, “…although adopting one’s own national currency is no
less seductive than having a national airline, it is a perilous course
9
.” For Armenia, the airlines venture
was perilous as eventually it had to remit a large equity stake in its national airline to Russia in exchange
for a debt write-off. History will show if Armenia’s currency will suffer a similar fate. Belarus, for one,
is on course for monetary reunification with Russia
10
.
In 1992, there already existed nineteen de facto currencies. The ruble accounts in the banking systems of
each republic were not convertible among each other and had different exchange rates. In Russia, the
7
Knobl, Sutt and Zavoico 2002.
8
Sachs 2005.
9
Nordhaus, Peck and Richardson 1991.
10
UNECE 2004.
patching among different sources. The most difficult measurements are those that attempt to show the
changes in the Armenian economy from before and after the breakup of the Soviet Union.
II. MONETARY MANAGEMENT
For a developing country, the decision to join a currency union or to peg its currency to another currency
usually is the result of poor monetary management and inflation. In the transition countries of the former
Soviet Union (FSU), the opposite situation existed. Countries decided to issue their own currencies
because of poor monetary management by the central Soviet government. In January 1992, Russia
liberalized prices which caused immediate and steep inflation. The price level increased 15-fold or 1,000
percent from 1991-1992 then another 30 percent in 1992-1993. A cash shortage ensued, exacerbated by
the fact that only Russia could issue rubles. This shortage was a major motivation for the former Soviet
republics to seek alternative monetary arrangements
7
.
Not everyone agreed that the individual republics should have issued their own currencies. The IMF
initially advised against it and attempted to establish a shared currency among separate central banks
8
. As
Nordhuas, Peck and Richardson (1991) noted, “…although adopting one’s own national currency is no
less seductive than having a national airline, it is a perilous course
9
.” For Armenia, the airlines venture
was perilous as eventually it had to remit a large equity stake in its national airline to Russia in exchange
for a debt write-off. History will show if Armenia’s currency will suffer a similar fate. Belarus, for one,
is on course for monetary reunification with Russia
10
.
In 1992, there already existed nineteen de facto currencies. The ruble accounts in the banking systems of
each republic were not convertible among each other and had different exchange rates. In Russia, the
7
Knobl, Sutt and Zavoico 2002.
8
Sachs 2005.
9
Nordhaus, Peck and Richardson 1991.
10
UNECE 2004.
Monetary Decentralization: The Case of Armenia 8
ruble accounts were convertible to cash rubles only at a discount of 40 percent. In December 1991,
Ukraine became the first to issue a formal alternative to the ruble, called the karbovanets. At first, it
traded at par with the ruble even when the ruble accounts did not
11
. The Baltic countries soon followed
suit. In May 1992, Latvia introduced its own ruble that circulated alongside the Russian ruble until it left
the ruble zone in July. Estonia left the ruble zone when it introduced its own currency in June. Lithuania
followed suit in July. Armenian introduced its dram on November 22, 1993.
In all republics, foreign currency circulation increased, substituting for a formal inter-republican
payments system as identified by Fischer as a necessary facility for transition
12
. The situation was similar
to that which existed in the former Yugoslavia after the new republics issued their own currencies but the
Deutsch Mark circulated among all of the republics
13
. In establishing new currencies, the FSU republics
faced the need for foreign exchange, of which they had none or very little. Officials at the Bank of
Estonia suggested pledging the country’s state-owned forests in lieu of international currency
14
.
The first stage of Armenia’s transition (1991-1995) included monetary financing of the budget deficit and
rapid devaluation of the dram, both of which led to hyperinflation in 1993-1994
15
. Claims on general
government increased from 63 million drams in 1992 to more than 33 billion drams in 1996
16
. Prices rose
nearly 500 percent in 1991-1992, another 450 percent in 1992-1993 and nearly doubled in 1993-1994
before tapering off and returning to normal levels in 1996-1997
17
. The Central Bank of Armenia (CBA)
chose a stabilization policy with the exchange rate as the anchor given the lack of confidence in the new
11
Fischer, Nordhaus and Summers 1992.
12
Fischer, Nordhaus and Summers 1992.
13
Desilets 2002.
14
Adalbert, Sutt and Zavoico 2002.
15
Grigorian, Khachatryan and Sargsyan 2004.
16
IMF International Financial Statistics (IFS).
17
United Nations Economic Commission for Europe.
ruble accounts were convertible to cash rubles only at a discount of 40 percent. In December 1991,
Ukraine became the first to issue a formal alternative to the ruble, called the karbovanets. At first, it
traded at par with the ruble even when the ruble accounts did not
11
. The Baltic countries soon followed
suit. In May 1992, Latvia introduced its own ruble that circulated alongside the Russian ruble until it left
the ruble zone in July. Estonia left the ruble zone when it introduced its own currency in June. Lithuania
followed suit in July. Armenian introduced its dram on November 22, 1993.
In all republics, foreign currency circulation increased, substituting for a formal inter-republican
payments system as identified by Fischer as a necessary facility for transition
12
. The situation was similar
to that which existed in the former Yugoslavia after the new republics issued their own currencies but the
Deutsch Mark circulated among all of the republics
13
. In establishing new currencies, the FSU republics
faced the need for foreign exchange, of which they had none or very little. Officials at the Bank of
Estonia suggested pledging the country’s state-owned forests in lieu of international currency
14
.
The first stage of Armenia’s transition (1991-1995) included monetary financing of the budget deficit and
rapid devaluation of the dram, both of which led to hyperinflation in 1993-1994
15
. Claims on general
government increased from 63 million drams in 1992 to more than 33 billion drams in 1996
16
. Prices rose
nearly 500 percent in 1991-1992, another 450 percent in 1992-1993 and nearly doubled in 1993-1994
before tapering off and returning to normal levels in 1996-1997
17
. The Central Bank of Armenia (CBA)
chose a stabilization policy with the exchange rate as the anchor given the lack of confidence in the new
11
Fischer, Nordhaus and Summers 1992.
12
Fischer, Nordhaus and Summers 1992.
13
Desilets 2002.
14
Adalbert, Sutt and Zavoico 2002.
15
Grigorian, Khachatryan and Sargsyan 2004.
16
IMF International Financial Statistics (IFS).
17
United Nations Economic Commission for Europe.
Monetary Decentralization: The Case of Armenia 9
currency and lack of money market instruments
18
. In the second stage of transition (1995-2002), the CBA
defined price stability as its main objective, reducing its concern for the exchange rate
19
.
In 1998, the CBA stopped direct financing of the budget and began securitizing government debt
20
. From
the beginning of 1999, the CBA began securitizing external financing of the government by linking
issuance of treasury bills to the disbursement of external financing
21
. The CBA sterilizes some of the
influx of external financing by issuing bonds. Flows from the exterior are insensitive to the Armenian
interest rate
22
. The CBA has a limited ability to sterilize based on its small portfolio of domestic money
market instruments. However, it does sterilize to some extent. My analysis shows that with a one-quarter
lag, current transfers explain more than 63 percent of the variation in claims on general government
(TBILL) between the end of 1999 and the beginning of 2005:
(1) (0.63)
2
TBILL = 33 + (0.39)T +
Basically, the central government has been sterilizing public and private transfers into the economy.
III. TRADE
McKinnon (1963) stresses the importance of the nontradable sector and notes that a separate currency is
more efficient for an area that has a large nontradable sector, with the exchange rate fixed to a
representative basket of nontradable goods and services. As the size of the tradable sector increases, so
do the arguments in favor of a fixed exchange rate, or currency union.
18
Grigorian, Khachatryan and Sargsyan 2004.
19
Grigorian, Khachatryan and Sargsyan 2004.
20
Grigorian, Khachatryan and Sargsyan 2004.
21
Grigorian, Khachatryan and Sargsyan 2004.
22
Grigorian, Khachatryan and Sargsyan 2004.
currency and lack of money market instruments
18
. In the second stage of transition (1995-2002), the CBA
defined price stability as its main objective, reducing its concern for the exchange rate
19
.
In 1998, the CBA stopped direct financing of the budget and began securitizing government debt
20
. From
the beginning of 1999, the CBA began securitizing external financing of the government by linking
issuance of treasury bills to the disbursement of external financing
21
. The CBA sterilizes some of the
influx of external financing by issuing bonds. Flows from the exterior are insensitive to the Armenian
interest rate
22
. The CBA has a limited ability to sterilize based on its small portfolio of domestic money
market instruments. However, it does sterilize to some extent. My analysis shows that with a one-quarter
lag, current transfers explain more than 63 percent of the variation in claims on general government
(TBILL) between the end of 1999 and the beginning of 2005:
(1) (0.63)
2
TBILL = 33 + (0.39)T +
Basically, the central government has been sterilizing public and private transfers into the economy.
III. TRADE
McKinnon (1963) stresses the importance of the nontradable sector and notes that a separate currency is
more efficient for an area that has a large nontradable sector, with the exchange rate fixed to a
representative basket of nontradable goods and services. As the size of the tradable sector increases, so
do the arguments in favor of a fixed exchange rate, or currency union.
18
Grigorian, Khachatryan and Sargsyan 2004.
19
Grigorian, Khachatryan and Sargsyan 2004.
20
Grigorian, Khachatryan and Sargsyan 2004.
21
Grigorian, Khachatryan and Sargsyan 2004.
22
Grigorian, Khachatryan and Sargsyan 2004.
Monetary Decentralization: The Case of Armenia 10
For Armenia, trade as a percentage of GDP peaked in 1994 at 112 percent. However, this is due less to an
increase in trade and more to a decrease in GDP. As GDP has increased, trade as a percentage of GDP
generally has decreased, except for 2001-2003 in which both trade as a percentage of GDP and GDP were
on the increase.
So, it could be said that during 1995-2001 the increasing relative importance of Armenia’s non-tradable
sector justified a separate currency. However, since the percentage of trade as a percentage of GDP never
dipped below 70 percent and since 2001 it has risen to 88 percent, this justification for a separate currency
is weak. For comparison, in the US, more than half of states’ labor forces and outputs are oriented toward
internal markets
23
. In addition, Armenia has a history as a very open economy in which trade has
represented a high percentage of GDP.
In Soviet times, Armenia’s trade as a percentage of GDP is reported to have been more than 100 percent.
Avanesyan and Freinkman (2003) estimate that in 1987 trade as a percentage of GDP was 103 percent,
which they say was common for the smaller republics of the FSU. In general, even though the Soviet
Union was isolated from the rest of the world, it was highly integrated and very open internally. In 1988,
inter-republican trade represented 21 percent of Soviet GDP whereas in the European Community, trade
among members and with the rest of the world represented only 14 percent of GDP
24
. The Soviet
republics also were very specialized, making them highly dependent on inter-republican trade
25
. In this
way, the Soviet Union was similar to the US where Krugman (1993) observes that regions are free to
specialize given the stability of common fiscal and monetary systems
26
.
23
Krugman 1993.
24
Nordhaus, Peck and Richardson 1991.
25
Fischer, Nordhaus and Summers 1992.
26
Casella (1993) reminds us that trade-induced specialization is the main source of gains from trade at any level.
For Armenia, trade as a percentage of GDP peaked in 1994 at 112 percent. However, this is due less to an
increase in trade and more to a decrease in GDP. As GDP has increased, trade as a percentage of GDP
generally has decreased, except for 2001-2003 in which both trade as a percentage of GDP and GDP were
on the increase.
So, it could be said that during 1995-2001 the increasing relative importance of Armenia’s non-tradable
sector justified a separate currency. However, since the percentage of trade as a percentage of GDP never
dipped below 70 percent and since 2001 it has risen to 88 percent, this justification for a separate currency
is weak. For comparison, in the US, more than half of states’ labor forces and outputs are oriented toward
internal markets
23
. In addition, Armenia has a history as a very open economy in which trade has
represented a high percentage of GDP.
In Soviet times, Armenia’s trade as a percentage of GDP is reported to have been more than 100 percent.
Avanesyan and Freinkman (2003) estimate that in 1987 trade as a percentage of GDP was 103 percent,
which they say was common for the smaller republics of the FSU. In general, even though the Soviet
Union was isolated from the rest of the world, it was highly integrated and very open internally. In 1988,
inter-republican trade represented 21 percent of Soviet GDP whereas in the European Community, trade
among members and with the rest of the world represented only 14 percent of GDP
24
. The Soviet
republics also were very specialized, making them highly dependent on inter-republican trade
25
. In this
way, the Soviet Union was similar to the US where Krugman (1993) observes that regions are free to
specialize given the stability of common fiscal and monetary systems
26
.
23
Krugman 1993.
24
Nordhaus, Peck and Richardson 1991.
25
Fischer, Nordhaus and Summers 1992.
26
Casella (1993) reminds us that trade-induced specialization is the main source of gains from trade at any level.
Monetary Decentralization: The Case of Armenia 11
Soviet Armenia’s economy was characterized by large plants whose suppliers and customers were located
in other parts of the Soviet Union. This arrangement ensured interdependence during the Soviet Union as
well as a painful transition experience, since the plants were not linked to the rest of Armenia’s
economy
27
. The breakup of the Soviet Union was devastating for Armenia’s trade. From 1990-1993,
trade decreased from more than US$4 billion to almost nothing. In real terms, trade never really
recovered since prices increased so much in 1991-1994 and this was a one-time adjustment to achieve
price liberalization.
In 1988-1994, Russia was Armenia’s main trading partner, followed by Belarus and Ukraine. After the
breakup, the smaller republics had a great incentive to open to the world economy because of the
increased tariffs on imports from the other Soviet republics and as a method for tempering domestic
monopolies and price increases
28
. This led to the development of trade relationships with a very different
set of countries than those from the Soviet period. Russia and other FSU members remained significant
trading partners, but they gradually decreased in importance.
In 1998, Armenia’s trade with the EU surpassed its trade with Russia and in 2000 it surpassed its trade
with the entire FSU. In 2001-2002, trade with the Middle East surpassed that with Russia and the FSU.
The EU and the Middle East were important partners from the beginning. Belgium, the US and Iran rose
to prominence as individual countries.
For my analysis, what is most important is Armenia’s trade with existing or likely currency unions, so I
focus on the CIS, EU, US and Iran. Trade with these currency areas ranges from 5-24 percent depending
on the period. In the latest period, 2000-2003, trade with the EU surpassed the CIS and now amounts to
27
Avanesyan and Freinkman 2003.
28
Nordhaus, Peck and Richardson 1991.
Soviet Armenia’s economy was characterized by large plants whose suppliers and customers were located
in other parts of the Soviet Union. This arrangement ensured interdependence during the Soviet Union as
well as a painful transition experience, since the plants were not linked to the rest of Armenia’s
economy
27
. The breakup of the Soviet Union was devastating for Armenia’s trade. From 1990-1993,
trade decreased from more than US$4 billion to almost nothing. In real terms, trade never really
recovered since prices increased so much in 1991-1994 and this was a one-time adjustment to achieve
price liberalization.
In 1988-1994, Russia was Armenia’s main trading partner, followed by Belarus and Ukraine. After the
breakup, the smaller republics had a great incentive to open to the world economy because of the
increased tariffs on imports from the other Soviet republics and as a method for tempering domestic
monopolies and price increases
28
. This led to the development of trade relationships with a very different
set of countries than those from the Soviet period. Russia and other FSU members remained significant
trading partners, but they gradually decreased in importance.
In 1998, Armenia’s trade with the EU surpassed its trade with Russia and in 2000 it surpassed its trade
with the entire FSU. In 2001-2002, trade with the Middle East surpassed that with Russia and the FSU.
The EU and the Middle East were important partners from the beginning. Belgium, the US and Iran rose
to prominence as individual countries.
For my analysis, what is most important is Armenia’s trade with existing or likely currency unions, so I
focus on the CIS, EU, US and Iran. Trade with these currency areas ranges from 5-24 percent depending
on the period. In the latest period, 2000-2003, trade with the EU surpassed the CIS and now amounts to
27
Avanesyan and Freinkman 2003.
28
Nordhaus, Peck and Richardson 1991.
Monetary Decentralization: The Case of Armenia 12
15 percent of GDP. Based on this parameter, if Armenia were to join an existing or likely currency area,
it would choose to join the euro zone.
As Table 1 shows, Armenia’s trade displays interesting relationships with its money supply and banking
system deposits. Trade with the CIS and Iran is negatively correlated with Armenia’s money supply.
This indicates that trade with these countries occurs in currencies other than the dram, possibly in the
currencies of these countries or possibly in US dollars which circulate widely in the FSU
29
. Trade with
the CIS and Iran is even more negatively correlated with the foreign currency penetration indicators.
Even though this trade might use US dollars, it probably is not transacted through the banking system, so
the data would not reflect it.
Trade with the EU is highly correlated with both the money supply and foreign currency penetration,
indicating that most of this trade is transacted through the banking system. Armenia’s trade with the EU
is more highly correlated with foreign currency penetration than with the money supply, indicating that
most of its trade is conducted through foreign currency bank accounts.
Trade with the CIS, Iran and EU explains more than 90 percent of the variation in this measure of foreign
currency penetration. Trade with the CIS and Iran, of course, have negative coefficients and trade with
the EU has a positive coefficient.
Trade with the US is very highly correlated with Armenia’s money supply (85 percent with M1 and 84
percent with M2). Trade with the CIS, EU and US explains more than 77 percent of the variation in M2,
however the t-statistics for the CIS and EU are low. Regressing M2 on trade with just the US yields an R
2
of more than 65 percent and a very high t-statistic:
29
Feige (2003) estimates that in 2001, 29 percent of the US$580 billion of US currency that circulates outside of the
US circulated in the transition countries of Eastern Europe and the FSU.
15 percent of GDP. Based on this parameter, if Armenia were to join an existing or likely currency area,
it would choose to join the euro zone.
As Table 1 shows, Armenia’s trade displays interesting relationships with its money supply and banking
system deposits. Trade with the CIS and Iran is negatively correlated with Armenia’s money supply.
This indicates that trade with these countries occurs in currencies other than the dram, possibly in the
currencies of these countries or possibly in US dollars which circulate widely in the FSU
29
. Trade with
the CIS and Iran is even more negatively correlated with the foreign currency penetration indicators.
Even though this trade might use US dollars, it probably is not transacted through the banking system, so
the data would not reflect it.
Trade with the EU is highly correlated with both the money supply and foreign currency penetration,
indicating that most of this trade is transacted through the banking system. Armenia’s trade with the EU
is more highly correlated with foreign currency penetration than with the money supply, indicating that
most of its trade is conducted through foreign currency bank accounts.
Trade with the CIS, Iran and EU explains more than 90 percent of the variation in this measure of foreign
currency penetration. Trade with the CIS and Iran, of course, have negative coefficients and trade with
the EU has a positive coefficient.
Trade with the US is very highly correlated with Armenia’s money supply (85 percent with M1 and 84
percent with M2). Trade with the CIS, EU and US explains more than 77 percent of the variation in M2,
however the t-statistics for the CIS and EU are low. Regressing M2 on trade with just the US yields an R
2
of more than 65 percent and a very high t-statistic:
29
Feige (2003) estimates that in 2001, 29 percent of the US$580 billion of US currency that circulates outside of the
US circulated in the transition countries of Eastern Europe and the FSU.
Monetary Decentralization: The Case of Armenia 13
(2) (0.65)
2
M2 = -256 + (3.0)(TRADEUS) +
This indicates that the CBA is fairly disciplined in issuing currency only when it is backed by US dollars.
That assumption is further supported by the fact that the dram/US$ exchange rate explains nearly 80
percent of the variation in M2:
(3) (0.79)
2
M2 = -370 + (0.97)EADM/$
***{INSERT TABLE 1 HERE}***
IV. MONETARY POLICY PREFERENCES
Central banks set monetary policy based on the economic activity within the regions they oversee. An
increase in GDP generally signals the need for more currency to be issued, to facilitate economic growth
and to prevent liquidity crises. Inflation of prices generally signals the need for increased interest rates, to
rein in frivolous lending by the banking system and preserve the value of money. For these reasons,
regions that experience similar movements in GDP, prices and wages will demand similar monetary
policies.
PRICES
As Table 2 shows, the only price level that is more than 70 percent correlated with Armenia’s is
Turkmenistan’s. However, this was only in the initial transition period, 1992-1995. Turkmenistan also
(2) (0.65)
2
M2 = -256 + (3.0)(TRADEUS) +
This indicates that the CBA is fairly disciplined in issuing currency only when it is backed by US dollars.
That assumption is further supported by the fact that the dram/US$ exchange rate explains nearly 80
percent of the variation in M2:
(3) (0.79)
2
M2 = -370 + (0.97)EADM/$
***{INSERT TABLE 1 HERE}***
IV. MONETARY POLICY PREFERENCES
Central banks set monetary policy based on the economic activity within the regions they oversee. An
increase in GDP generally signals the need for more currency to be issued, to facilitate economic growth
and to prevent liquidity crises. Inflation of prices generally signals the need for increased interest rates, to
rein in frivolous lending by the banking system and preserve the value of money. For these reasons,
regions that experience similar movements in GDP, prices and wages will demand similar monetary
policies.
PRICES
As Table 2 shows, the only price level that is more than 70 percent correlated with Armenia’s is
Turkmenistan’s. However, this was only in the initial transition period, 1992-1995. Turkmenistan also
Monetary Decentralization: The Case of Armenia 14
was one of Armenia’s top trading partners in the early transition period, but given its fiscal and political
situation, it is unlikely that the two would forge a currency union.
Armenia’s Caucasus neighbors, Azerbaijan and Georgia consistently rank among the most correlated in
terms of price levels. Armenia’s relations with Azerbaijan are poor and a trade blockade has existed
between the two countries since the transition began. Armenia has good relations with Georgia and
depends on it for access to its Black Sea ports. A Caucasus monetary union might prove efficient if
Armenia’s relations with Azerbaijan were to improve.
Another interesting aspect of price correlations is that Russia and many countries of the FSU and Eastern
Europe consistently rank among the most correlated. While one of the main reasons for Armenia’s
decision to issue its own currency was poor monetary management by the central Soviet government in
Moscow, there still exists economic evidence such as these correlations that a monetary union among
some FSU and Eastern European countries might be efficient.
The main US price level as well as the price levels for various regions of the US consistently rank among
the least correlated with Armenia. Many members of the Eurozone, most notably Germany, also rank
among the least correlated. These figures bode poorly for any plans for Armenia to dollarize or euroize.
***{INSERT TABLE 2 HERE}***
GDP
For the period 1995-2005, Armenia’s GDP is highly correlated with many countries of the FSU and
Eastern Europe. In general, the correlation rates are higher for GDP than for prices. Whereas the most
was one of Armenia’s top trading partners in the early transition period, but given its fiscal and political
situation, it is unlikely that the two would forge a currency union.
Armenia’s Caucasus neighbors, Azerbaijan and Georgia consistently rank among the most correlated in
terms of price levels. Armenia’s relations with Azerbaijan are poor and a trade blockade has existed
between the two countries since the transition began. Armenia has good relations with Georgia and
depends on it for access to its Black Sea ports. A Caucasus monetary union might prove efficient if
Armenia’s relations with Azerbaijan were to improve.
Another interesting aspect of price correlations is that Russia and many countries of the FSU and Eastern
Europe consistently rank among the most correlated. While one of the main reasons for Armenia’s
decision to issue its own currency was poor monetary management by the central Soviet government in
Moscow, there still exists economic evidence such as these correlations that a monetary union among
some FSU and Eastern European countries might be efficient.
The main US price level as well as the price levels for various regions of the US consistently rank among
the least correlated with Armenia. Many members of the Eurozone, most notably Germany, also rank
among the least correlated. These figures bode poorly for any plans for Armenia to dollarize or euroize.
***{INSERT TABLE 2 HERE}***
GDP
For the period 1995-2005, Armenia’s GDP is highly correlated with many countries of the FSU and
Eastern Europe. In general, the correlation rates are higher for GDP than for prices. Whereas the most
Monetary Decentralization: The Case of Armenia 15
correlated price levels fall in the 40-50 percent range, the most correlated GDPs are correlated in the 70-
90 percent rage. The US is much more correlated in terms of GDP rates than with prices, at 68 percent.
Especially when viewed in light of the price level correlations, it seems that were Armenia to join a
currency union, it would fare best in one with Kazakstan, Kyrgyzstan, Ukraine, some of Eastern Europe
and even the Baltics (basically, the FSU and Eastern Europe). Comparable GDP data was not available
for some important countries, including Azerbaijan, Georgia, Russia, Germany and France.
A MODEL WITH PRICES AND GDP
In Desilets (2003), I apply the model of Alesina and Grilli (1992) to analyze the efficiency of a currency
union for the CIS. In the model, the CIS central bank minimizes its losses, LCIS, which can be caused by
inflation, CIS, and the deviation of output from an expected level, x
*
CIS:
(4) LCIS = ½ E[
2
CIS + b(xCIS – x
*
CIS)
2
].
Output is determined by the Phillips curve relation:
(5) xCIS = (CIS –
e
CIS) + ,
Later, a separate loss function is introduced for the individual countries within the CIS:
(6) L
i
= ½ E[
2
CIS + (yi – y
*
i)
2
],
where
(7) yi = (CIS –
e
CIS) + I,
correlated price levels fall in the 40-50 percent range, the most correlated GDPs are correlated in the 70-
90 percent rage. The US is much more correlated in terms of GDP rates than with prices, at 68 percent.
Especially when viewed in light of the price level correlations, it seems that were Armenia to join a
currency union, it would fare best in one with Kazakstan, Kyrgyzstan, Ukraine, some of Eastern Europe
and even the Baltics (basically, the FSU and Eastern Europe). Comparable GDP data was not available
for some important countries, including Azerbaijan, Georgia, Russia, Germany and France.
A MODEL WITH PRICES AND GDP
In Desilets (2003), I apply the model of Alesina and Grilli (1992) to analyze the efficiency of a currency
union for the CIS. In the model, the CIS central bank minimizes its losses, LCIS, which can be caused by
inflation, CIS, and the deviation of output from an expected level, x
*
CIS:
(4) LCIS = ½ E[
2
CIS + b(xCIS – x
*
CIS)
2
].
Output is determined by the Phillips curve relation:
(5) xCIS = (CIS –
e
CIS) + ,
Later, a separate loss function is introduced for the individual countries within the CIS:
(6) L
i
= ½ E[
2
CIS + (yi – y
*
i)
2
],
where
(7) yi = (CIS –
e
CIS) + I,
Monetary Decentralization: The Case of Armenia 16
The beta terms in the loss functions represent the tradeoff between inflation and output. If the CIS central
bank is more conservative than the Armenian central bank, Armenia gains from lower inflation by
remaining in a monetary union.
For 1990-2001, inflation in Armenia averaged 819 percent. For the CIS as a whole, it averaged 345
percent and for Russia it averaged 274 percent. However, in 1996-2001, Armenian inflation cooled to an
annual average of 7 percent, whereas in Russia it remained high at 36 and for the CIS at 37 percent
30
. In
this respect, Armenia gained from its own monetary management which proved better that the CIS’s and
Russia’s.
The error terms in the output functions of the above model represent output shocks to the economies in
question. The higher correlated are these shocks, the more justified is a monetary union. The data shows
that Russia, Kazakstan, Belarus and Ukraine are most highly correlated so would benefit most from a
CIS-wide monetary policy, whereas Tajikistan, Turkmenistan and the countries of the Caucasus region
would benefit least because of the large economic distance between them and the CIS average.
Regression analysis shows that the percentage of value added to GDP by the agriculture sector and a
dummy variable for the Caucasus region explain more than 80 percent of the relative deviation in output
among CIS countries
31
:
(8) (0.66)[(i/CIS) – 1] = (0.012)AG + (0.64)CAUC +
Interestingly, growth in Armenia during 1989-2001 did not suffer as much as growth for Russia or for the
CIS as a whole. The average growth rate for the period was -0.9 for Armenia, -2.6 for Russia and -2.9 for
30
UNECE 2002.
31
Desilets 2003.
The beta terms in the loss functions represent the tradeoff between inflation and output. If the CIS central
bank is more conservative than the Armenian central bank, Armenia gains from lower inflation by
remaining in a monetary union.
For 1990-2001, inflation in Armenia averaged 819 percent. For the CIS as a whole, it averaged 345
percent and for Russia it averaged 274 percent. However, in 1996-2001, Armenian inflation cooled to an
annual average of 7 percent, whereas in Russia it remained high at 36 and for the CIS at 37 percent
30
. In
this respect, Armenia gained from its own monetary management which proved better that the CIS’s and
Russia’s.
The error terms in the output functions of the above model represent output shocks to the economies in
question. The higher correlated are these shocks, the more justified is a monetary union. The data shows
that Russia, Kazakstan, Belarus and Ukraine are most highly correlated so would benefit most from a
CIS-wide monetary policy, whereas Tajikistan, Turkmenistan and the countries of the Caucasus region
would benefit least because of the large economic distance between them and the CIS average.
Regression analysis shows that the percentage of value added to GDP by the agriculture sector and a
dummy variable for the Caucasus region explain more than 80 percent of the relative deviation in output
among CIS countries
31
:
(8) (0.66)[(i/CIS) – 1] = (0.012)AG + (0.64)CAUC +
Interestingly, growth in Armenia during 1989-2001 did not suffer as much as growth for Russia or for the
CIS as a whole. The average growth rate for the period was -0.9 for Armenia, -2.6 for Russia and -2.9 for
30
UNECE 2002.
31
Desilets 2003.
Monetary Decentralization: The Case of Armenia 17
the CIS
32
. This is all the more impressive since Armenia’s initial drop in output was among the highest
for the FSU; its turnaround was much stronger and quicker than those of the other countries. So, Armenia
gained from both lower inflation and higher output by assuming management of its own currency.
WAGES
Many of the same Eastern European countries that are most correlated with Armenia in terms of prices
also are highly correlated with Armenia in terms of wages. Whereas the most correlated price levels are
correlated at only 40-50 percent, wages are correlated in the range of 80-100 percent. This lends more
weight to the argument that if Armenia were to enter into a currency union, it should do so with countries
of Eastern Europe and the FSU. Specifically, wages in Kazakstan and Kyrgyzstan both are correlated at
higher than 95 percent. Two of the Baltic states, Latvia and Lithuania, also rank in the 90s (Estonia was
not included in the data set). As with price correlations, Poland and Croatia both appear high in the list.
Comparable wage data was not readily available for some important countries such as Russia, Germany
and the US. Again, it seems clear that were Armenia to enter a currency union, it would be most efficient
to do so with the FSU and Eastern Europe.
INTEREST RATES
Research on transition and developing countries in general indicates that changes in domestic interest
rates have a low impact on aggregate demand. The credit channel usually is weak given weak financial
markets. The exchange rate usually has a much higher impact on price fluctuations than in developed
countries
33
. For example, Calvo, Leiderman and Reinhart (1993) show that for the period 1988-1992, half
of the variance in real exchange rates for 10 Latin American countries can be explained by variations in
U. S. interest rates. My own regressions show that with a two-quarter lag, the US interest rate explains
32
UNECE 2002.
33
Grigorian, Khachatryan and Sargsyan 2004.
the CIS
32
. This is all the more impressive since Armenia’s initial drop in output was among the highest
for the FSU; its turnaround was much stronger and quicker than those of the other countries. So, Armenia
gained from both lower inflation and higher output by assuming management of its own currency.
WAGES
Many of the same Eastern European countries that are most correlated with Armenia in terms of prices
also are highly correlated with Armenia in terms of wages. Whereas the most correlated price levels are
correlated at only 40-50 percent, wages are correlated in the range of 80-100 percent. This lends more
weight to the argument that if Armenia were to enter into a currency union, it should do so with countries
of Eastern Europe and the FSU. Specifically, wages in Kazakstan and Kyrgyzstan both are correlated at
higher than 95 percent. Two of the Baltic states, Latvia and Lithuania, also rank in the 90s (Estonia was
not included in the data set). As with price correlations, Poland and Croatia both appear high in the list.
Comparable wage data was not readily available for some important countries such as Russia, Germany
and the US. Again, it seems clear that were Armenia to enter a currency union, it would be most efficient
to do so with the FSU and Eastern Europe.
INTEREST RATES
Research on transition and developing countries in general indicates that changes in domestic interest
rates have a low impact on aggregate demand. The credit channel usually is weak given weak financial
markets. The exchange rate usually has a much higher impact on price fluctuations than in developed
countries
33
. For example, Calvo, Leiderman and Reinhart (1993) show that for the period 1988-1992, half
of the variance in real exchange rates for 10 Latin American countries can be explained by variations in
U. S. interest rates. My own regressions show that with a two-quarter lag, the US interest rate explains
32
UNECE 2002.
33
Grigorian, Khachatryan and Sargsyan 2004.
Monetary Decentralization: The Case of Armenia 18
more than 60 percent of the variation in Armenia’s real effective exchange rate from 1995-2004. The
percentage rises to more than 67 percent for the period 2000-2004:
(9) (0.67)
2
(EADM/$)/(P$/PADM) = 73.8 + (4.4)iUS +
Del Negro and Obiols-Homs (1999) show that for the period 1970-1997, most shocks to Mexico’s output
and prices originated from shocks to US industrial production, interest rates and consumer and
commodity prices
34
. My own regression shows that variations in US GDP and prices explain 64 percent
of the variation in Armenian GDP. US interest rates were not significant in explaining Armenian GDP:
(10) (0.64)
2
Y-ARM = (49,966,792)YUS + (-1,287,736,574)PUS +
Poghosyan (2003) finds that the impact of domestic interest rates on domestic money demand is
insignificant, which indicates that the dram is not generally used for saving. He also finds a limited
response in the demand for money with respect to prices (20 percent) and explains this by the high
sensitivity of prices to the dram/US$ exchange rate. He concludes that a price increase indicates a dram
depreciation, causing a flight from the dram and further foreign currency penetration of the economy.
This indicates the population’s hypersensitivity to inflation. I did not find evidence of this relationship in
analyzing the price level and the percentage of deposits held in foreign currency. Variations in the T-bill
interest rate explain approximately 60 percent of the variation in the difference between dram deposits
(DD) and foreign currency deposits (FCD) at private banks:
(11) (0.59)
2
(FCD – DD) = 104,616 + (-1,803)iADM +
34
Schmitt-Grohe and Uribe 2000.
more than 60 percent of the variation in Armenia’s real effective exchange rate from 1995-2004. The
percentage rises to more than 67 percent for the period 2000-2004:
(9) (0.67)
2
(EADM/$)/(P$/PADM) = 73.8 + (4.4)iUS +
Del Negro and Obiols-Homs (1999) show that for the period 1970-1997, most shocks to Mexico’s output
and prices originated from shocks to US industrial production, interest rates and consumer and
commodity prices
34
. My own regression shows that variations in US GDP and prices explain 64 percent
of the variation in Armenian GDP. US interest rates were not significant in explaining Armenian GDP:
(10) (0.64)
2
Y-ARM = (49,966,792)YUS + (-1,287,736,574)PUS +
Poghosyan (2003) finds that the impact of domestic interest rates on domestic money demand is
insignificant, which indicates that the dram is not generally used for saving. He also finds a limited
response in the demand for money with respect to prices (20 percent) and explains this by the high
sensitivity of prices to the dram/US$ exchange rate. He concludes that a price increase indicates a dram
depreciation, causing a flight from the dram and further foreign currency penetration of the economy.
This indicates the population’s hypersensitivity to inflation. I did not find evidence of this relationship in
analyzing the price level and the percentage of deposits held in foreign currency. Variations in the T-bill
interest rate explain approximately 60 percent of the variation in the difference between dram deposits
(DD) and foreign currency deposits (FCD) at private banks:
(11) (0.59)
2
(FCD – DD) = 104,616 + (-1,803)iADM +
34
Schmitt-Grohe and Uribe 2000.
Monetary Decentralization: The Case of Armenia 19
V. FINANCIAL FLOWS
Sachs and Sala-i-Martin (1992) observe that if a monetary union is aligned with a fiscal union, the fiscal
system can be used to counter asymmetric disturbances, including those caused by monetary policy.
They find that aligned fiscal and monetary systems facilitate inter-regional insurance in the sense that all
regions can finance the deficits of regions facing disturbances. Casella (1993) observes that in Italy,
transfers from the north to the south are seen by many as a counter-migratory force intended to avoid
social tensions.
For Armenia, the disintegration of the Soviet Union included the breakdown of a system of various types
of inter-republican income transfers. According to Avanesyan and Freinkman (2003), Armenia’s balance
of transfers within the Soviet system was -6.6 percent of GDP due to the major tax payments sent to
Moscow by its profitable enterprises. However, Armenia and other smaller Soviet republics probably ran
deficits in other services, such as education, health and research and development that were provided free
by institutions in Moscow or other regional centers
35
. In addition, their enterprises received a variety of
subsidies that in market-based accounting would be considered foreign direct investment.
Armenia’s current account deficit was greater than 16 percent in 1987. It was financed mostly from the
Soviet Union’s central investment budget
36
. In other words, Armenia paid more into the Moscow budget
than it received in tax transfers, but at the same time ran a current account deficit since it was importing
more than it exported. The current account deficit was financed by capital inflows in the form of
investment.
Since independence, Armenia has continued to rely on transfers from abroad. From 2000-2005, Direct
Investment, Capital Account Credits, Exceptional Financing and Current Transfers (all IMF definitions)
35
Avanesyan and Freinkman 2003.
36
Avanesyan and Freinkman 2003.
V. FINANCIAL FLOWS
Sachs and Sala-i-Martin (1992) observe that if a monetary union is aligned with a fiscal union, the fiscal
system can be used to counter asymmetric disturbances, including those caused by monetary policy.
They find that aligned fiscal and monetary systems facilitate inter-regional insurance in the sense that all
regions can finance the deficits of regions facing disturbances. Casella (1993) observes that in Italy,
transfers from the north to the south are seen by many as a counter-migratory force intended to avoid
social tensions.
For Armenia, the disintegration of the Soviet Union included the breakdown of a system of various types
of inter-republican income transfers. According to Avanesyan and Freinkman (2003), Armenia’s balance
of transfers within the Soviet system was -6.6 percent of GDP due to the major tax payments sent to
Moscow by its profitable enterprises. However, Armenia and other smaller Soviet republics probably ran
deficits in other services, such as education, health and research and development that were provided free
by institutions in Moscow or other regional centers
35
. In addition, their enterprises received a variety of
subsidies that in market-based accounting would be considered foreign direct investment.
Armenia’s current account deficit was greater than 16 percent in 1987. It was financed mostly from the
Soviet Union’s central investment budget
36
. In other words, Armenia paid more into the Moscow budget
than it received in tax transfers, but at the same time ran a current account deficit since it was importing
more than it exported. The current account deficit was financed by capital inflows in the form of
investment.
Since independence, Armenia has continued to rely on transfers from abroad. From 2000-2005, Direct
Investment, Capital Account Credits, Exceptional Financing and Current Transfers (all IMF definitions)
35
Avanesyan and Freinkman 2003.
36
Avanesyan and Freinkman 2003.
Monetary Decentralization: The Case of Armenia 20
amounted to 15-20 percent of GDP. Current Transfers are by far the largest in the group. Alone they
amounted to 8-11 percent of GDP or between 200 and 400 million dollars.
In addition to transfers from other governments, Armenia also benefits from private remittances. Since
independence, between 500,000 and 1 million Armenians have left to work in other countries
permanently. In addition, each year thousands of Armenians leave to work in other countries temporarily.
In 2003, that number was approximately 180,000. There are more than 8 million Armenian diaspora
living throughout the world
37
. Many of these Armenians abroad send money home in the form of private
remittances. From 1998-2003, these remittances increased from US$133 million to US$289 million.
According to Banaian and Roberts (2005), the largest group to send remittances is the Armenian Diaspora
that left the country before independence. They account for approximately 60 percent of the total amount
of remittances sent. Most remittances come from Russia (68 percent in 2002), followed by the
US/Canada (17 percent), Europe (9 percent) and other CIS countries (4 percent).
VI. FACTOR MOBILITY
The importance of factor mobility in determining the efficiency of a currency area always has been
recognized in economic literature. Mundell (1961) discusses the issue and concludes that:
If the world can be divided into regions within each of which there is factor mobility and between
which there is factor immobility, then each of these regions should have a separate currency
which fluctuates relative to all other currencies.
Mundell was not the first to address the issue of factor mobility. In his own article, he cites the debate
over a single currency for Western Europe and notes that Meade (1957) warns against a common
37
Banaian and Roberts 2005.
amounted to 15-20 percent of GDP. Current Transfers are by far the largest in the group. Alone they
amounted to 8-11 percent of GDP or between 200 and 400 million dollars.
In addition to transfers from other governments, Armenia also benefits from private remittances. Since
independence, between 500,000 and 1 million Armenians have left to work in other countries
permanently. In addition, each year thousands of Armenians leave to work in other countries temporarily.
In 2003, that number was approximately 180,000. There are more than 8 million Armenian diaspora
living throughout the world
37
. Many of these Armenians abroad send money home in the form of private
remittances. From 1998-2003, these remittances increased from US$133 million to US$289 million.
According to Banaian and Roberts (2005), the largest group to send remittances is the Armenian Diaspora
that left the country before independence. They account for approximately 60 percent of the total amount
of remittances sent. Most remittances come from Russia (68 percent in 2002), followed by the
US/Canada (17 percent), Europe (9 percent) and other CIS countries (4 percent).
VI. FACTOR MOBILITY
The importance of factor mobility in determining the efficiency of a currency area always has been
recognized in economic literature. Mundell (1961) discusses the issue and concludes that:
If the world can be divided into regions within each of which there is factor mobility and between
which there is factor immobility, then each of these regions should have a separate currency
which fluctuates relative to all other currencies.
Mundell was not the first to address the issue of factor mobility. In his own article, he cites the debate
over a single currency for Western Europe and notes that Meade (1957) warns against a common
37
Banaian and Roberts 2005.
Monetary Decentralization: The Case of Armenia 21
currency because of the lack of labor mobility in the region whereas Scitovsky (1958) favors a common
currency because he believes it will induce capital mobility.
In the same article, Mundell also notes that:
In the real world, of course, currencies are mainly an expression of national sovereignity, so that
actual currency reorganization would be feasible only if it were accompanied by profound
political changes. The concept of an optimum currency area therefore has direct practical
applicability only in areas where political organization is in a state of flux, such as in ex-colonial
areas and in Western Europe.
One could argue that political organization is in a constant state of flux. In the last half of the Twentieth
Century, we see obvious political reorganization in the former colonies of Africa, in the former
Yugoslavia and in the FSU and Eastern Europe among other places. So, Armenia presents itself as an
area for which the concept of an optimum currency area has direct practical application. Therefore, it is
important to examine the mobility of factors in Armenia to determine the practicality of Armenia’s own
currency and the use of other currencies within its territory.
One version of Bertola’s economic growth model (1993) predicts the disappearance of peripheral regions
after economic integration, due to the flight of capital and labor. Bertola solves this problem by
introducing the concept of interdependent technologies, in which the level of technology in each region
depends not only on its own investment, but also on investment in other regions. If local currencies
increase the level of local investment, they might help to sustain peripheral regions during and after
economic integration, although a model must be developed to assess the extent of these effects in the face
of decreased foreign investment due to costs associated with exchange rate uncertainty.
currency because of the lack of labor mobility in the region whereas Scitovsky (1958) favors a common
currency because he believes it will induce capital mobility.
In the same article, Mundell also notes that:
In the real world, of course, currencies are mainly an expression of national sovereignity, so that
actual currency reorganization would be feasible only if it were accompanied by profound
political changes. The concept of an optimum currency area therefore has direct practical
applicability only in areas where political organization is in a state of flux, such as in ex-colonial
areas and in Western Europe.
One could argue that political organization is in a constant state of flux. In the last half of the Twentieth
Century, we see obvious political reorganization in the former colonies of Africa, in the former
Yugoslavia and in the FSU and Eastern Europe among other places. So, Armenia presents itself as an
area for which the concept of an optimum currency area has direct practical application. Therefore, it is
important to examine the mobility of factors in Armenia to determine the practicality of Armenia’s own
currency and the use of other currencies within its territory.
One version of Bertola’s economic growth model (1993) predicts the disappearance of peripheral regions
after economic integration, due to the flight of capital and labor. Bertola solves this problem by
introducing the concept of interdependent technologies, in which the level of technology in each region
depends not only on its own investment, but also on investment in other regions. If local currencies
increase the level of local investment, they might help to sustain peripheral regions during and after
economic integration, although a model must be developed to assess the extent of these effects in the face
of decreased foreign investment due to costs associated with exchange rate uncertainty.
Monetary Decentralization: The Case of Armenia 22
Krugman’s (1993) discussion of the case of the US Commonwealth of Massachusetts as a possible
destiny for some European states is very relevant. He finds that US regions that experience negative
demand shocks do not generally respond with decreases in the price level sufficient to attract new export
industries. Instead, regions simply shed labor and capital and never regain their previous output levels.
Krugman provides the example of California--where employment rose 86 percent relative to New York
during the period from 1960 to 1988 though the relative per capita income between the two states
remained unchanged. He adds that traditional manufacturing states have seen their employment shares
steadily decline since World War II.
The history of trade barriers in Europe leads Krugman to hypothesize that European economies are less
specialized than US regions. Krugman finds that migration played a very small role in European
countries, but that real wages indeed have adjusted over time. Changes in exchange rates have helped
these adjustments. If Massachusetts had its own currency, real wages could have adjusted more easily to
accommodate increased investment.
Bertola and Krugman’s arguments are most relevant for Armenia. While I was not able to obtain data on
Soviet investment in Armenia, surely it was much greater than current investment. While this may not be
obvious in Armenia’s capital Yerevan, where Italians have built a five-star hotel, the IFC has helped
finance the renovation by Marriott of another grand hotel and the French have taken over the cognac
factory, it is strikingly obvious in the smaller cities and throughout the countryside, where factories are
closed and more than half of the population is unemployed. This has occurred even though, unlike
Massachusetts, Armenia did introduce its own currency. Armenia has shed both labor and capital since
the transition began.
According to even the most conservative estimates (IFS), Armenia has lost nearly 10 percent of its
population since 1985 and has lost more population than most other FSU countries. The percentage of
Krugman’s (1993) discussion of the case of the US Commonwealth of Massachusetts as a possible
destiny for some European states is very relevant. He finds that US regions that experience negative
demand shocks do not generally respond with decreases in the price level sufficient to attract new export
industries. Instead, regions simply shed labor and capital and never regain their previous output levels.
Krugman provides the example of California--where employment rose 86 percent relative to New York
during the period from 1960 to 1988 though the relative per capita income between the two states
remained unchanged. He adds that traditional manufacturing states have seen their employment shares
steadily decline since World War II.
The history of trade barriers in Europe leads Krugman to hypothesize that European economies are less
specialized than US regions. Krugman finds that migration played a very small role in European
countries, but that real wages indeed have adjusted over time. Changes in exchange rates have helped
these adjustments. If Massachusetts had its own currency, real wages could have adjusted more easily to
accommodate increased investment.
Bertola and Krugman’s arguments are most relevant for Armenia. While I was not able to obtain data on
Soviet investment in Armenia, surely it was much greater than current investment. While this may not be
obvious in Armenia’s capital Yerevan, where Italians have built a five-star hotel, the IFC has helped
finance the renovation by Marriott of another grand hotel and the French have taken over the cognac
factory, it is strikingly obvious in the smaller cities and throughout the countryside, where factories are
closed and more than half of the population is unemployed. This has occurred even though, unlike
Massachusetts, Armenia did introduce its own currency. Armenia has shed both labor and capital since
the transition began.
According to even the most conservative estimates (IFS), Armenia has lost nearly 10 percent of its
population since 1985 and has lost more population than most other FSU countries. The percentage of
Monetary Decentralization: The Case of Armenia 23
the population employed in Armenia has decreased steadily throughout the transition, from 45 percent in
1994 to 37 percent in 2003
38
. Not only is Armenia losing its population, but it appears to be losing its
most productive population. This is not surprising if it is supposed that higher levels of education
increase labor mobility. Apart from education, those most interested in working will travel to work if
there are no jobs in their home country.
Even though Armenia has lost at least 10 percent (some estimate closer to 20 percent) of its population
and perhaps as much of its labor force, that does not mean that these workers do not contribute to the
Armenian economy. It has a large Diaspora population and many more recent emigrants. A significant
number of Armenians abroad remit some portion of their salaries to families in Armenia. This group
includes seasonal workers, emigrant workers and Diaspora. The theme of remittances is more fully
discussed in Section V.
As I have show, the labor force of Soviet Armenia proved to be highly mobile. Beginning in the mid-
1990s, Armenia began shedding its labor force. Mundell might examine the destination of the mobile
labor force and determine that Armenia should adopt the currency of the destination country. The
problem is that Armenia’s mobile labor force does not choose one currency zone as its destination, but at
least three: Europe, Russia and the US. In terms of both Diaspora population and remittances, Armenia
proves most integrated with Russia and the FSU.
McKinnon (1963) includes a special note on factor mobility. He recognizes the importance of geographic
factor mobility as described by Mundell, but also stresses the importance of industrial factor mobility. He
38
I use IFS data for population. According to it, Armenia had 3.545 million people in 1990, 3.372 people in 1993
and 3.026 people in 2004. Others estimate that the decrease in population is even greater. Banaian and Roberts
(2005) estimate that approximately 1 million people have left Armenia in 1992-2003. Surely some of the
discrepancy is accounted for by diaspora Armenians, many of whom still consider themselves citizens of Armenians
and remit portions of their salaries home.
the population employed in Armenia has decreased steadily throughout the transition, from 45 percent in
1994 to 37 percent in 2003
38
. Not only is Armenia losing its population, but it appears to be losing its
most productive population. This is not surprising if it is supposed that higher levels of education
increase labor mobility. Apart from education, those most interested in working will travel to work if
there are no jobs in their home country.
Even though Armenia has lost at least 10 percent (some estimate closer to 20 percent) of its population
and perhaps as much of its labor force, that does not mean that these workers do not contribute to the
Armenian economy. It has a large Diaspora population and many more recent emigrants. A significant
number of Armenians abroad remit some portion of their salaries to families in Armenia. This group
includes seasonal workers, emigrant workers and Diaspora. The theme of remittances is more fully
discussed in Section V.
As I have show, the labor force of Soviet Armenia proved to be highly mobile. Beginning in the mid-
1990s, Armenia began shedding its labor force. Mundell might examine the destination of the mobile
labor force and determine that Armenia should adopt the currency of the destination country. The
problem is that Armenia’s mobile labor force does not choose one currency zone as its destination, but at
least three: Europe, Russia and the US. In terms of both Diaspora population and remittances, Armenia
proves most integrated with Russia and the FSU.
McKinnon (1963) includes a special note on factor mobility. He recognizes the importance of geographic
factor mobility as described by Mundell, but also stresses the importance of industrial factor mobility. He
38
I use IFS data for population. According to it, Armenia had 3.545 million people in 1990, 3.372 people in 1993
and 3.026 people in 2004. Others estimate that the decrease in population is even greater. Banaian and Roberts
(2005) estimate that approximately 1 million people have left Armenia in 1992-2003. Surely some of the
discrepancy is accounted for by diaspora Armenians, many of whom still consider themselves citizens of Armenians
and remit portions of their salaries home.
Monetary Decentralization: The Case of Armenia 24
explains that factors do not need to migrate between geographic areas if they can migrate between
industries in the same area.
In considering the reorganization of the labor force by industry, I observe that Agriculture has absorbed
labor from Industry and from the Public Sector. Based on that observation, I would assume that the new
Armenian economy does not offer desirable jobs for the most educated and most motivated workers. For
this reason, these workers leave the country to seek employment elsewhere, either because it is more
gratifying or because it offers better wages. This implies that some industries have in effect relocated or
been replaced by other industries altogether and that, in general, industrial factor mobility in Armenia is
very low, thereby justifying its accession to another currency area.
VII. BANKING
In general, economic literature on currency areas ignores the private banking sector. Including a
discussion of the private banking sector severely complicates the discussion over currency areas, but also
highlights some very important issues. For example, as I discussed in Section IV, a country would want
to join a currency area if its prices, wages and GDP were correlated with the pre-existing currency area.
From a banking perspective, the opposite would be true. Banks try to balance their portfolios to hedge
against losses. For this reason, they seek diverse portfolios of assets whose returns are negatively
correlated
39
.
Banks are free to purchase assets from other currency areas but they then accept a degree of foreign
exchange risk. If they can purchase assets whose returns are negatively correlated but which are from the
same currency area, they eliminate foreign exchange risk and improve the overall return of their assets.
Many banks, including public institutions such as the World Bank, do just this by lending in dollars or
another international currency even to clients outside of the dollar currency area. This reduces foreign
39
See Saunders 1997 for a general discussion of international bank management.
explains that factors do not need to migrate between geographic areas if they can migrate between
industries in the same area.
In considering the reorganization of the labor force by industry, I observe that Agriculture has absorbed
labor from Industry and from the Public Sector. Based on that observation, I would assume that the new
Armenian economy does not offer desirable jobs for the most educated and most motivated workers. For
this reason, these workers leave the country to seek employment elsewhere, either because it is more
gratifying or because it offers better wages. This implies that some industries have in effect relocated or
been replaced by other industries altogether and that, in general, industrial factor mobility in Armenia is
very low, thereby justifying its accession to another currency area.
VII. BANKING
In general, economic literature on currency areas ignores the private banking sector. Including a
discussion of the private banking sector severely complicates the discussion over currency areas, but also
highlights some very important issues. For example, as I discussed in Section IV, a country would want
to join a currency area if its prices, wages and GDP were correlated with the pre-existing currency area.
From a banking perspective, the opposite would be true. Banks try to balance their portfolios to hedge
against losses. For this reason, they seek diverse portfolios of assets whose returns are negatively
correlated
39
.
Banks are free to purchase assets from other currency areas but they then accept a degree of foreign
exchange risk. If they can purchase assets whose returns are negatively correlated but which are from the
same currency area, they eliminate foreign exchange risk and improve the overall return of their assets.
Many banks, including public institutions such as the World Bank, do just this by lending in dollars or
another international currency even to clients outside of the dollar currency area. This reduces foreign
39
See Saunders 1997 for a general discussion of international bank management.
Monetary Decentralization: The Case of Armenia 25
exchange risk for the bank, but it also may increase the default risk of the loan by requiring the borrower
to repay the loan in dollars even though his or her revenues are in another currency. The management of
the local currency is in the hands of the local government, not in the hands of the borrower, so the
borrower in effect takes on the responsibility of monetary management by agreeing to repay his or her
loan in dollars. Indeed, lending in dollars is cited as one of the main causes of the financial crises that
plagued developing and transition countries during the past 10 years
40
.
Theoretically, developing countries offer higher returns on investment because they are growing at a
faster rate than developed countries. In reality, however, most developing countries experience more
inflation than developed countries so agreeing to repay a loan in dollars when your revenues are in a local
currency is very risky since the exchange rate may deteriorate during the loan term.
BANKING IN ARMENIA
In the late 1980’s, there were 52 branches of the State Bank of the Soviet Union in Armenia. In
December 1991, this bank was converted to the National Bank of Armenia and in March 1993, it was
converted to the Central Bank of Armenia
41
. The first cooperative bank, Haik, was established in
Armenia in October 1988. This was the beginning of the development of a wide network of banks
throughout Armenia
42
. The banking crisis in 1994-1995 forced 37 banks to cease their activities
43
. In
2002, the CBA placed 8 out of the country’s 30 banks under temporary administration
44
. Between 2000
and 2004, the number of banks decreased from 31 to 20. Some of this decrease was due to a gradual
increase in capital requirements, but some also to management issues
45
.
40
Bernanke 2005.
41
CBA 1999, 2002.
42
CBA 1999.
43
Ohanian, Safdari and Scannell 2003.
44
Grigorian 2003.
45
Beddies et al 2005.
exchange risk for the bank, but it also may increase the default risk of the loan by requiring the borrower
to repay the loan in dollars even though his or her revenues are in another currency. The management of
the local currency is in the hands of the local government, not in the hands of the borrower, so the
borrower in effect takes on the responsibility of monetary management by agreeing to repay his or her
loan in dollars. Indeed, lending in dollars is cited as one of the main causes of the financial crises that
plagued developing and transition countries during the past 10 years
40
.
Theoretically, developing countries offer higher returns on investment because they are growing at a
faster rate than developed countries. In reality, however, most developing countries experience more
inflation than developed countries so agreeing to repay a loan in dollars when your revenues are in a local
currency is very risky since the exchange rate may deteriorate during the loan term.
BANKING IN ARMENIA
In the late 1980’s, there were 52 branches of the State Bank of the Soviet Union in Armenia. In
December 1991, this bank was converted to the National Bank of Armenia and in March 1993, it was
converted to the Central Bank of Armenia
41
. The first cooperative bank, Haik, was established in
Armenia in October 1988. This was the beginning of the development of a wide network of banks
throughout Armenia
42
. The banking crisis in 1994-1995 forced 37 banks to cease their activities
43
. In
2002, the CBA placed 8 out of the country’s 30 banks under temporary administration
44
. Between 2000
and 2004, the number of banks decreased from 31 to 20. Some of this decrease was due to a gradual
increase in capital requirements, but some also to management issues
45
.
40
Bernanke 2005.
41
CBA 1999, 2002.
42
CBA 1999.
43
Ohanian, Safdari and Scannell 2003.
44
Grigorian 2003.
45
Beddies et al 2005.
Monetary Decentralization: The Case of Armenia 26
The inter-bank credit market is virtually nonexistent and banks deal mostly with the CBA for their credit
needs. Grants and low-interest loans from donors provide opportunities for corruption. Grigorian (2003)
suggests the possibility of permitting full foreign ownership of the banking system with some
requirements for domestic investments
46
. In terms of capital, foreign ownership already accounts for 53
percent of the banking system. Only one international bank, HSBC of London, operates in Armenia. Its
assets represent 18 percent of the total banking system and its deposits a quarter of total deposits
47
.
Deposits increased steadily from 1994-2003, which reflected growing confidence in the system.
However, as Graph 1 shows, a notable feature about the deposits is that the vast majority of them are held
in foreign currency. Foreign currency deposits accounted for approximately 70-80 percent of total
deposits during most of the 1994-2003 period. Most of these foreign currency deposits are in US dollars,
but some are in other currencies, such as the euro, the Russian ruble and other currencies.
***{INSERT GRAPH 1 HERE}***
Armenia receives a large amount of external flows. During 2000-2005, Direct Investment, Capital
Account Credits, Exceptional Financing and Current Transfers amounted to 15-20 percent of GDP.
Current Transfers are by far the largest in the group. Alone they amounted to 8-11 percent of GDP or
between 200 and 400 million US dollars.
This results in foreign currency penetration
48
. For a general measure of the extent of foreign currency
penetration, I divide Quasi-money by M2. Quasi-money is defined as Time, Savings and Foreign
Currency Deposits of resident sectors other than government
49
. This is a valid proxy since resident dram
46
Grigorian 2003.
47
Beddies et al 2005.
48
Grigorian, Khachatryan and Sargsyan 2004.
49
IFS.
The inter-bank credit market is virtually nonexistent and banks deal mostly with the CBA for their credit
needs. Grants and low-interest loans from donors provide opportunities for corruption. Grigorian (2003)
suggests the possibility of permitting full foreign ownership of the banking system with some
requirements for domestic investments
46
. In terms of capital, foreign ownership already accounts for 53
percent of the banking system. Only one international bank, HSBC of London, operates in Armenia. Its
assets represent 18 percent of the total banking system and its deposits a quarter of total deposits
47
.
Deposits increased steadily from 1994-2003, which reflected growing confidence in the system.
However, as Graph 1 shows, a notable feature about the deposits is that the vast majority of them are held
in foreign currency. Foreign currency deposits accounted for approximately 70-80 percent of total
deposits during most of the 1994-2003 period. Most of these foreign currency deposits are in US dollars,
but some are in other currencies, such as the euro, the Russian ruble and other currencies.
***{INSERT GRAPH 1 HERE}***
Armenia receives a large amount of external flows. During 2000-2005, Direct Investment, Capital
Account Credits, Exceptional Financing and Current Transfers amounted to 15-20 percent of GDP.
Current Transfers are by far the largest in the group. Alone they amounted to 8-11 percent of GDP or
between 200 and 400 million US dollars.
This results in foreign currency penetration
48
. For a general measure of the extent of foreign currency
penetration, I divide Quasi-money by M2. Quasi-money is defined as Time, Savings and Foreign
Currency Deposits of resident sectors other than government
49
. This is a valid proxy since resident dram
46
Grigorian 2003.
47
Beddies et al 2005.
48
Grigorian, Khachatryan and Sargsyan 2004.
49
IFS.
Monetary Decentralization: The Case of Armenia 27
time deposits represent only 7 percent of total resident deposits. Using this measure, I observe that
foreign currency penetration of the economy ranged from 25 to 60 percent in 1994-1999 and since then
has hovered around 50 percent.
Another measure of foreign currency penetration is the percentage of total deposits (TD) that are foreign
currency deposits (FCD). This measure is even higher, hovering around 70 percent:
(12) FCD/TD = 0.71
Feige (2003) attempts to calculate the amount of dollars in circulation in Armenia and develops two
alternative measures of foreign currency penetration. First, he determines the percentage of currency in
circulation represented by US dollars. He adds drams in circulation (DC) to dollars in circulation (US$C)
and calculates the percentage of the total represented by dollars in circulation. For Armenia, that
percentage is 62 for the year 2001:
(13) US$C/(DC + US$C) = 0.62
Feige also calculates what he calls a comprehensive dollarization index. Instead of simply calculating the
percentage of foreign currency deposits (FCD) in M2, he calculates the percentage of foreign currency
deposits plus US dollars in circulation in M2 plus US dollars in circulation. That percentage for Armenia
is 68.1 for the year 2001:
(14) US$C + FCD/(M2 + US$C) = 0.68
This is much higher than what Feige calls the traditional dollarization index:
time deposits represent only 7 percent of total resident deposits. Using this measure, I observe that
foreign currency penetration of the economy ranged from 25 to 60 percent in 1994-1999 and since then
has hovered around 50 percent.
Another measure of foreign currency penetration is the percentage of total deposits (TD) that are foreign
currency deposits (FCD). This measure is even higher, hovering around 70 percent:
(12) FCD/TD = 0.71
Feige (2003) attempts to calculate the amount of dollars in circulation in Armenia and develops two
alternative measures of foreign currency penetration. First, he determines the percentage of currency in
circulation represented by US dollars. He adds drams in circulation (DC) to dollars in circulation (US$C)
and calculates the percentage of the total represented by dollars in circulation. For Armenia, that
percentage is 62 for the year 2001:
(13) US$C/(DC + US$C) = 0.62
Feige also calculates what he calls a comprehensive dollarization index. Instead of simply calculating the
percentage of foreign currency deposits (FCD) in M2, he calculates the percentage of foreign currency
deposits plus US dollars in circulation in M2 plus US dollars in circulation. That percentage for Armenia
is 68.1 for the year 2001:
(14) US$C + FCD/(M2 + US$C) = 0.68
This is much higher than what Feige calls the traditional dollarization index:
Monetary Decentralization: The Case of Armenia 28
(15) FCD/M2 = 0.47
In any case, my proxy of Quasi-money divided by M2 serves as a good estimate of foreign currency
penetration in the Armenian economy.
For 2000-2005, Current Transfers (T) explain more than 63 percent of the variation in the amount of
foreign currency deposits at private banks, implying that most of those transfers end up in foreign
currency bank accounts:
(16) (0.63)
2
FCD = 47,104 + (810)T +
The transfers that are not deposited in foreign currency bank accounts are exchanged for local currency so
end up in the central bank. The amount of total deposits (TD) and the amount of foreign currency
deposits at private banks together explain more than 92 percent of the variation in foreign assets at CBA:
(17) (0.93)
2
FA = 160,730 + (6.9)TD - (0.79)FCD +
Demand deposits carry a positive coefficient of 6.9 and foreign currency deposits one of -0.79. This
supports the assumption that external flows are exchanged for domestic currency and deposited in private
banks. The flows that are not exchanged are deposited in foreign currency accounts in private banks.
For 2003-2004, the US dollar represents the vast majority of CBA reserves at 65 percent. The euro is
next at 21 percent, followed by the British pound at 10 percent. Russian rubles account for a very small
percentage of CBA reserves. In 2004, CBA net assets of non-convertible currency averaged only 12
(15) FCD/M2 = 0.47
In any case, my proxy of Quasi-money divided by M2 serves as a good estimate of foreign currency
penetration in the Armenian economy.
For 2000-2005, Current Transfers (T) explain more than 63 percent of the variation in the amount of
foreign currency deposits at private banks, implying that most of those transfers end up in foreign
currency bank accounts:
(16) (0.63)
2
FCD = 47,104 + (810)T +
The transfers that are not deposited in foreign currency bank accounts are exchanged for local currency so
end up in the central bank. The amount of total deposits (TD) and the amount of foreign currency
deposits at private banks together explain more than 92 percent of the variation in foreign assets at CBA:
(17) (0.93)
2
FA = 160,730 + (6.9)TD - (0.79)FCD +
Demand deposits carry a positive coefficient of 6.9 and foreign currency deposits one of -0.79. This
supports the assumption that external flows are exchanged for domestic currency and deposited in private
banks. The flows that are not exchanged are deposited in foreign currency accounts in private banks.
For 2003-2004, the US dollar represents the vast majority of CBA reserves at 65 percent. The euro is
next at 21 percent, followed by the British pound at 10 percent. Russian rubles account for a very small
percentage of CBA reserves. In 2004, CBA net assets of non-convertible currency averaged only 12
Monetary Decentralization: The Case of Armenia 29
million drams while net assets of convertible currency averaged 146 billion drams
50
. CBA holds most of
its reserves in the form of foreign currency deposits, but it also holds securities, mostly government
Treasury bills, and correspondent accounts with foreign banks
51
.
Armenian banks not only hold deposits in foreign currencies but also lend in foreign currencies. While
the percentage of lending in foreign currencies appears to be decreasing, it remained above 70 percent in
2003, exposing borrowers, especially producers of nontradables, to high risks
52
. Most of these foreign
currency loans are in US dollars, but some also are in euros, Russian rubles and other currencies.
Armenia has one of the highest foreign currency penetration rates of the FSU and Eastern Europe, at 49
percent. Only its neighbor Azerbaijan ranks higher in this list. Armenia also is one of the least monetized
economies of the FSU and Eastern Europe at only 11.86 percent
53
.
Armenia has extended relatively little credit to its private sector. Credit to the private sector amounts to
only 8.5 percent of GDP. In this respect, it leads only Azerbaijan, Belarus and Kyrygzstan. The average
for the CIS is 9.7 percent and for the FSU is 17.7 percent
54
. The problems with the banking sector
discussed above contribute to this situation. Since the crises in Mexico in 1994, East Asia in 1997-1998,
Russia in 1998, in Brazil in 1999 and Argentina in 2002, investors and bankers are wary of lending to the
developing world. Some of the structural issues that led to these crises were overvalued exchange rates
and too much short-term debt denominated in foreign currencies
55
. Another major issue is the lack of
institutional capacity to identify and manage profitable projects. This is the case in Armenia, which
suffered its own mini crisis in 2002 when the government overtook several banks. In general, since 1998
50
CBA July 2005.
51
CBA Annual Report 2004.
52
Grigorian 2003 and IMF 2002.
53
CBA Annual Report 2004.
54
CBA Annual Report 2004.
55
Bernanke 2005.
million drams while net assets of convertible currency averaged 146 billion drams
50
. CBA holds most of
its reserves in the form of foreign currency deposits, but it also holds securities, mostly government
Treasury bills, and correspondent accounts with foreign banks
51
.
Armenian banks not only hold deposits in foreign currencies but also lend in foreign currencies. While
the percentage of lending in foreign currencies appears to be decreasing, it remained above 70 percent in
2003, exposing borrowers, especially producers of nontradables, to high risks
52
. Most of these foreign
currency loans are in US dollars, but some also are in euros, Russian rubles and other currencies.
Armenia has one of the highest foreign currency penetration rates of the FSU and Eastern Europe, at 49
percent. Only its neighbor Azerbaijan ranks higher in this list. Armenia also is one of the least monetized
economies of the FSU and Eastern Europe at only 11.86 percent
53
.
Armenia has extended relatively little credit to its private sector. Credit to the private sector amounts to
only 8.5 percent of GDP. In this respect, it leads only Azerbaijan, Belarus and Kyrygzstan. The average
for the CIS is 9.7 percent and for the FSU is 17.7 percent
54
. The problems with the banking sector
discussed above contribute to this situation. Since the crises in Mexico in 1994, East Asia in 1997-1998,
Russia in 1998, in Brazil in 1999 and Argentina in 2002, investors and bankers are wary of lending to the
developing world. Some of the structural issues that led to these crises were overvalued exchange rates
and too much short-term debt denominated in foreign currencies
55
. Another major issue is the lack of
institutional capacity to identify and manage profitable projects. This is the case in Armenia, which
suffered its own mini crisis in 2002 when the government overtook several banks. In general, since 1998
50
CBA July 2005.
51
CBA Annual Report 2004.
52
Grigorian 2003 and IMF 2002.
53
CBA Annual Report 2004.
54
CBA Annual Report 2004.
55
Bernanke 2005.
Monetary Decentralization: The Case of Armenia 30
the loan to deposit ratio for the Armenian banking system has been falling, even while deposits have been
rising.
The anxiety surrounding lending to the developing world leads bankers and governments to demand
investments that are perceived as less risky. Savers in the developing world then end up financing the US
current account deficit
56
. This confirms the assertion of McKinnon (1963). He notes that the liquidity
value of money decreases with the size of its area of use and with bad monetary management, giving
incentives to savers to demand more of another currency. Because of this, he concludes that small
countries with weak currencies will finance the balance of payments deficits of large countries with more
liquid currencies, even when investment returns are higher in the small countries.
Armenia does have its own “tradable capital units” in the form of Treasury bills. The main role of these
bills in Armenia is to provide the banking system with a liquid, risk-free asset. The high interest rates
paid on the bills effectively subsidize the nation’s banks and help them cover costs and even losses in
other activities. Interest rates on the bills decreased from a high of more than 70 percent in 1997 to 4
percent in 2004. In 1999, private bank holdings of treasury bills surpassed CBA holdings. Reducing the
number of bills in circulation or lowering their interest rate would have a severe impact on the nation’s
banks, though it could result in an increase in efficiency
57
.
There is a market for T-bills outside the banking system as well. In 2004, the number of T-bills held by
non-bank investors amounted to almost 6.8 billion drams or nearly 14 percent of all outstanding T-bills.
There is only a small market for T-bills outside Armenia. In 2004, the amount of T-bills owned by non-
residents reached only 0.18 percent or 90 million drams
58
.
56
Bernanke 2005.
57
Grigorian 2003.
58
CBA Annual Report 2004
the loan to deposit ratio for the Armenian banking system has been falling, even while deposits have been
rising.
The anxiety surrounding lending to the developing world leads bankers and governments to demand
investments that are perceived as less risky. Savers in the developing world then end up financing the US
current account deficit
56
. This confirms the assertion of McKinnon (1963). He notes that the liquidity
value of money decreases with the size of its area of use and with bad monetary management, giving
incentives to savers to demand more of another currency. Because of this, he concludes that small
countries with weak currencies will finance the balance of payments deficits of large countries with more
liquid currencies, even when investment returns are higher in the small countries.
Armenia does have its own “tradable capital units” in the form of Treasury bills. The main role of these
bills in Armenia is to provide the banking system with a liquid, risk-free asset. The high interest rates
paid on the bills effectively subsidize the nation’s banks and help them cover costs and even losses in
other activities. Interest rates on the bills decreased from a high of more than 70 percent in 1997 to 4
percent in 2004. In 1999, private bank holdings of treasury bills surpassed CBA holdings. Reducing the
number of bills in circulation or lowering their interest rate would have a severe impact on the nation’s
banks, though it could result in an increase in efficiency
57
.
There is a market for T-bills outside the banking system as well. In 2004, the number of T-bills held by
non-bank investors amounted to almost 6.8 billion drams or nearly 14 percent of all outstanding T-bills.
There is only a small market for T-bills outside Armenia. In 2004, the amount of T-bills owned by non-
residents reached only 0.18 percent or 90 million drams
58
.
56
Bernanke 2005.
57
Grigorian 2003.
58
CBA Annual Report 2004
Monetary Decentralization: The Case of Armenia 31
Kenen (1969) notes the need for a liquid currency with a stable value to serve as the standard for
allocating capital among regions. He suggests that the currency of the largest region or the region with
the best monetary management would be converted to the inter-regional standard. Ingram (1959)
analyzes the balance of payments between North Carolina and other States in the US. He finds that the
availability of a large stock of tradable capital, i.e. federal government bonds, solves transfer problems
and other balance of payments pressures and could be more important than a single currency in
facilitating balance of payments adjustments. He notes that the establishment of the Federal Reserve
System coincided with the increased use of general government bonds by commercial banks. Ingram
concludes that the development of an integrated capital market, mostly in the form of a large stock of
tradable capital units, eliminated regional payment pressures previously experienced in the US. The
existence of nontradable capital units separates economic systems and produces balance of payment
pressures.
In line with Kenen and Ingram’s observations, Armenia as well as most of the rest of the world has relied
on the euro and US dollar as standards for allocating capital and solving transfer problems. However, as
the financial crises in developing and transition countries during the past ten years have shown, reliance
on foreign currencies has its drawbacks and even helps to cause financial crises. This phenomenon
supports the arguments in favor of Armenia’s maintenance of its own free-floating currency. On the other
hand, the hyperinflation of the early transition years justifies foreign currency deposits and free exchange
which discipline the CBA into keeping inflation low. Now that inflation in Armenia is in check, use of
the dram may increase, but as a small and open economy, it should expect foreign currencies to continue
to circulate in its economy and it should facilitate this circulation as it now does by permitting foreign
currency bank accounts and foreign currency lending.
Kenen (1969) notes the need for a liquid currency with a stable value to serve as the standard for
allocating capital among regions. He suggests that the currency of the largest region or the region with
the best monetary management would be converted to the inter-regional standard. Ingram (1959)
analyzes the balance of payments between North Carolina and other States in the US. He finds that the
availability of a large stock of tradable capital, i.e. federal government bonds, solves transfer problems
and other balance of payments pressures and could be more important than a single currency in
facilitating balance of payments adjustments. He notes that the establishment of the Federal Reserve
System coincided with the increased use of general government bonds by commercial banks. Ingram
concludes that the development of an integrated capital market, mostly in the form of a large stock of
tradable capital units, eliminated regional payment pressures previously experienced in the US. The
existence of nontradable capital units separates economic systems and produces balance of payment
pressures.
In line with Kenen and Ingram’s observations, Armenia as well as most of the rest of the world has relied
on the euro and US dollar as standards for allocating capital and solving transfer problems. However, as
the financial crises in developing and transition countries during the past ten years have shown, reliance
on foreign currencies has its drawbacks and even helps to cause financial crises. This phenomenon
supports the arguments in favor of Armenia’s maintenance of its own free-floating currency. On the other
hand, the hyperinflation of the early transition years justifies foreign currency deposits and free exchange
which discipline the CBA into keeping inflation low. Now that inflation in Armenia is in check, use of
the dram may increase, but as a small and open economy, it should expect foreign currencies to continue
to circulate in its economy and it should facilitate this circulation as it now does by permitting foreign
currency bank accounts and foreign currency lending.
Monetary Decentralization: The Case of Armenia 32
VIII. CONCLUSIONS
My analysis of output and inflation shows that the decision by the newly independent Republic of
Armenia to issue its own currency was a wise and efficient decision. After the initial years of transition,
Armenia’s inflation rate slowed to a low 7 percent, 30 percent lower than Russia’s and the CIS’s as a
whole. Armenia also suffered from a smaller reduction in output than Russia and the CIS as a whole.
However, the issuance of the separate currency did not generate the investment that some models might
predict and a deeper analysis of foregone investment caused by Armenia’s exit from its fiscal and
monetary union with the FSU is required.
Armenia’s current de facto multi-currency system is efficient. Foreign currency deposits and lending
result in higher levels of economic activity than under a system of forced use of the domestic currency.
This is because savers face a lower risk in holding euro or US dollar bank accounts and lending in these
currencies. The free exchange regime also imposes a limit on domestic inflation, forcing the CBA to
curtail inflation. As I have shown, the demand for the dram and for dram deposits are highly sensitive to
the domestic inflation rate.
As confidence in the dram increases, so too will dram deposits and overall use of the dram. Dram
deposits have been on the rise. The main obstacle to confidence in the currency now is a poor financial
system based on unprofitable lending.
How long the dram will last and whether Armenia will join another currency area is somewhat of a
political question but also somewhat of a financial question. If Armenia were to be presented with large
amounts of investment, such as EU Structural Funds, it may be worth trading off its own currency for
such investment.
VIII. CONCLUSIONS
My analysis of output and inflation shows that the decision by the newly independent Republic of
Armenia to issue its own currency was a wise and efficient decision. After the initial years of transition,
Armenia’s inflation rate slowed to a low 7 percent, 30 percent lower than Russia’s and the CIS’s as a
whole. Armenia also suffered from a smaller reduction in output than Russia and the CIS as a whole.
However, the issuance of the separate currency did not generate the investment that some models might
predict and a deeper analysis of foregone investment caused by Armenia’s exit from its fiscal and
monetary union with the FSU is required.
Armenia’s current de facto multi-currency system is efficient. Foreign currency deposits and lending
result in higher levels of economic activity than under a system of forced use of the domestic currency.
This is because savers face a lower risk in holding euro or US dollar bank accounts and lending in these
currencies. The free exchange regime also imposes a limit on domestic inflation, forcing the CBA to
curtail inflation. As I have shown, the demand for the dram and for dram deposits are highly sensitive to
the domestic inflation rate.
As confidence in the dram increases, so too will dram deposits and overall use of the dram. Dram
deposits have been on the rise. The main obstacle to confidence in the currency now is a poor financial
system based on unprofitable lending.
How long the dram will last and whether Armenia will join another currency area is somewhat of a
political question but also somewhat of a financial question. If Armenia were to be presented with large
amounts of investment, such as EU Structural Funds, it may be worth trading off its own currency for
such investment.
Monetary Decentralization: The Case of Armenia 33
Many economic variables point to the efficiency of a currency area among some Eastern European
countries. However, political institutions are not in place to develop such a possibility. The possibility of
a common currency for the CIS is more feasible. Given that Armenia’s financial and labor markets
continue to be highly integrated with those of Russia, this might make good economic sense in the future.
Russia’s monetary management is not attractive for potential accession countries. However, the large
amount of Armenian debt held by Russia, investments by Russia in Armenia and large amounts of trade
between Armenia, Russia and the rest of the CIS indicate that these economies always will be highly
interconnected. In addition, many of the variables that affect monetary policy preferences point to the
efficiency of a currency union with countries of the FSU.
Many economic variables point to the efficiency of a currency area among some Eastern European
countries. However, political institutions are not in place to develop such a possibility. The possibility of
a common currency for the CIS is more feasible. Given that Armenia’s financial and labor markets
continue to be highly integrated with those of Russia, this might make good economic sense in the future.
Russia’s monetary management is not attractive for potential accession countries. However, the large
amount of Armenian debt held by Russia, investments by Russia in Armenia and large amounts of trade
between Armenia, Russia and the rest of the CIS indicate that these economies always will be highly
interconnected. In addition, many of the variables that affect monetary policy preferences point to the
efficiency of a currency union with countries of the FSU.
Monetary Decentralization: The Case of Armenia 34
TABLES AND GRAPHS
Table 1: Trade and Money Correlations
Trade w/CIS Trade w/Iran
Trade
w/EU2015 Trade w/US
M1 - end of year -51% -28% 48% 85%
M2 - end of year -61% -39% 56% 84%
For. dep. as % of total dep. -84% -55% 54% 32%
Quasi-money/M1 - end of year -87% -66% 61% 36%
Quasi-money/M2 - end of year -88% -69% 63% 39%
Source: Author’s calculations based on IFS
Table 2: Price Correlations
1992-1995 1996-1999 2000-2004
Armenia 1.00 Armenia 1.00 Armenia 1.00
Turkmenistan 0.93 Poland 0.64 Kyrgyzstan 0.53
Uzbekistan 0.55 Hungary 0.62 Latvia 0.52
Greece 0.48 Poland 0.58 Kazakhstan 0.52
Georgia 0.48 Estonia 0.53 Croatia 0.51
Kazakhstan 0.40 Bosnia Herzegovina 0.51 Georgia 0.46
Azerbaijan 0.40 Lithuania 0.49 Poland 0.45
Hungary 0.37 Latvia 0.45 Russian Federation 0.45
Belarus 0.35 Croatia 0.45 Azerbaijan 0.45
Poland 0.34 Georgia 0.43 Ukraine 0.45
Source: Author’s calculations based on Eurostat, UNECE and BLS
TABLES AND GRAPHS
Table 1: Trade and Money Correlations
Trade w/CIS Trade w/Iran
Trade
w/EU2015 Trade w/US
M1 - end of year -51% -28% 48% 85%
M2 - end of year -61% -39% 56% 84%
For. dep. as % of total dep. -84% -55% 54% 32%
Quasi-money/M1 - end of year -87% -66% 61% 36%
Quasi-money/M2 - end of year -88% -69% 63% 39%
Source: Author’s calculations based on IFS
Table 2: Price Correlations
1992-1995 1996-1999 2000-2004
Armenia 1.00 Armenia 1.00 Armenia 1.00
Turkmenistan 0.93 Poland 0.64 Kyrgyzstan 0.53
Uzbekistan 0.55 Hungary 0.62 Latvia 0.52
Greece 0.48 Poland 0.58 Kazakhstan 0.52
Georgia 0.48 Estonia 0.53 Croatia 0.51
Kazakhstan 0.40 Bosnia Herzegovina 0.51 Georgia 0.46
Azerbaijan 0.40 Lithuania 0.49 Poland 0.45
Hungary 0.37 Latvia 0.45 Russian Federation 0.45
Belarus 0.35 Croatia 0.45 Azerbaijan 0.45
Poland 0.34 Georgia 0.43 Ukraine 0.45
Source: Author’s calculations based on Eurostat, UNECE and BLS
Monetary Decentralization: The Case of Armenia 35
Graph 1: DEPOSITS, 1994-2003
0.0
20.0
40.0
60.0
80.0
100.0
120.0
140.0
160.0
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Year
Billions of Drams
Drams
Total Deposits
-Foreign currency
-Domestic currency
Source: IMF.
Graph 1: DEPOSITS, 1994-2003
0.0
20.0
40.0
60.0
80.0
100.0
120.0
140.0
160.0
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Year
Billions of Drams
Drams
Total Deposits
-Foreign currency
-Domestic currency
Source: IMF.
Monetary Decentralization: The Case of Armenia 36
REFERENCES
Alesina, Alberto and Vittorio Grilli. (1992). "The European Central Bank: Reshaping Monetary Politics
in Europe" in Establishing a Central Bank: Issues in Europe and Lessons from the US. Cambridge:
Cambridge University Press.
Avanesyan, Vahram and Lev Freinkman. (2003). Impact of External Shocks on the Armenian Economy at
the Outset of Transition. Prepared for the Armenia International Policy Research Group Conference,
“Armenia: Recent Economic Trends and Growth Prospects” January 2003. Washington, D. C.: World
Bank.
Banaian, King and Bryan Roberts. (2005). Remittances in Armenia: Size, Impacts and Measures to
Enhance Their Contribution to Development. Working Paper 05/01, January 2005. Washington, DC:
Armenia International Policy Research Group.
Bayoumi, Tamim. (1994). A Formal Model of Optimum Currency Areas. London: Centre for Economic
Policy Research.
Beddies, Christian, Enrique Gelbard, Garbis Iradian, Jimmy McHugh and Laure Redifer. (2005). Growth
and Poverty Reduction in Armenia: Achievements and Challenges. Washington, DC: IMF.
Beddies, Christian and Oleh Havrylyshyn. (2003). “Dollarisation in the Former Soviet Union: From
Hysteria to Hysteresis” in Comparative Economic Studies. Vol. 45.
Belkindas, Misha and Olga Ivanaov (eds.). (1995). “Foreign Trade Statistics in the USSR and Successor
States”. Studies of Economies in Transformation. No. 18. Washinton, D. C.: World Bank.
Bernanke, Ben. (2005). Remarks by Governor Ben S. Bernanke on the Global Saving Glut and the US
Currency Account Deficit. Speech delivered at the Sandridge Lecture, Virginia Association of Economics,
Richmond, Virginia. March 10, 2005. Washington, DC: Board of Governors of the Federal Reserve
System.
Bertola, Giuseppe. (1993). “Models of Economic Integration and Localized Growth” in Adjustment and
Growth in the European Monetary Union by Francesco Giavazzi and Francisco Torres (eds.). Cambridge:
Cambridge University Press.
Bureau of Economic Analysis. (2002). Regional Accounts Data. Washington, D. C.: BEA.
Calvo, Guillermo, Leonardo Leiderman and Carmen Reinhart. (1993). “Capital Inflows and Real
Exchange Rate Appreciation in Latin America: The Role of External Factors”. International Monetary
Fund Staff Papers. No. 40. March 1993, pp. 108-151.
Carlino, Gerald and Robert DeFina. (1999). The Differential Regional Effects of Monetary Policy:
Evidence from the U. S. States. Journal of Regional Science, Vol. 39, No. 2, pp. 339-358. Malden (MA):
Blackwell Publishers.
Casella, Alessandra. (1993). Discussion of Krugman’s “Lessons of Massachusetts for EMU” in
Adjustment and Growth in the European Monetary Union by Francesco Giavazzi and Francisco Torres
(eds.). Cambridge: Cambridge University Press.
Central Bank of Armenia (CBA). (July 2005). Monetary Overview. Yerevan: CBA.
REFERENCES
Alesina, Alberto and Vittorio Grilli. (1992). "The European Central Bank: Reshaping Monetary Politics
in Europe" in Establishing a Central Bank: Issues in Europe and Lessons from the US. Cambridge:
Cambridge University Press.
Avanesyan, Vahram and Lev Freinkman. (2003). Impact of External Shocks on the Armenian Economy at
the Outset of Transition. Prepared for the Armenia International Policy Research Group Conference,
“Armenia: Recent Economic Trends and Growth Prospects” January 2003. Washington, D. C.: World
Bank.
Banaian, King and Bryan Roberts. (2005). Remittances in Armenia: Size, Impacts and Measures to
Enhance Their Contribution to Development. Working Paper 05/01, January 2005. Washington, DC:
Armenia International Policy Research Group.
Bayoumi, Tamim. (1994). A Formal Model of Optimum Currency Areas. London: Centre for Economic
Policy Research.
Beddies, Christian, Enrique Gelbard, Garbis Iradian, Jimmy McHugh and Laure Redifer. (2005). Growth
and Poverty Reduction in Armenia: Achievements and Challenges. Washington, DC: IMF.
Beddies, Christian and Oleh Havrylyshyn. (2003). “Dollarisation in the Former Soviet Union: From
Hysteria to Hysteresis” in Comparative Economic Studies. Vol. 45.
Belkindas, Misha and Olga Ivanaov (eds.). (1995). “Foreign Trade Statistics in the USSR and Successor
States”. Studies of Economies in Transformation. No. 18. Washinton, D. C.: World Bank.
Bernanke, Ben. (2005). Remarks by Governor Ben S. Bernanke on the Global Saving Glut and the US
Currency Account Deficit. Speech delivered at the Sandridge Lecture, Virginia Association of Economics,
Richmond, Virginia. March 10, 2005. Washington, DC: Board of Governors of the Federal Reserve
System.
Bertola, Giuseppe. (1993). “Models of Economic Integration and Localized Growth” in Adjustment and
Growth in the European Monetary Union by Francesco Giavazzi and Francisco Torres (eds.). Cambridge:
Cambridge University Press.
Bureau of Economic Analysis. (2002). Regional Accounts Data. Washington, D. C.: BEA.
Calvo, Guillermo, Leonardo Leiderman and Carmen Reinhart. (1993). “Capital Inflows and Real
Exchange Rate Appreciation in Latin America: The Role of External Factors”. International Monetary
Fund Staff Papers. No. 40. March 1993, pp. 108-151.
Carlino, Gerald and Robert DeFina. (1999). The Differential Regional Effects of Monetary Policy:
Evidence from the U. S. States. Journal of Regional Science, Vol. 39, No. 2, pp. 339-358. Malden (MA):
Blackwell Publishers.
Casella, Alessandra. (1993). Discussion of Krugman’s “Lessons of Massachusetts for EMU” in
Adjustment and Growth in the European Monetary Union by Francesco Giavazzi and Francisco Torres
(eds.). Cambridge: Cambridge University Press.
Central Bank of Armenia (CBA). (July 2005). Monetary Overview. Yerevan: CBA.
Monetary Decentralization: The Case of Armenia 37
Central Bank of Armenia (CBA). (2004). Annual Report. Yerevan: CBA.
Central Bank of Armenia (CBA). (2004). Monetary Policy Program of the Republic of Armenia. Yerevan:
CBA.
Central Bank of Armenia (CBA). (2004). Statistical Bulletin of the Central Bank of Armenia. Yerevan:
CBA.
Central Bank of Armenia (CBA). (1999). Banking System of Armenia: Development, Regulation and
Supervision. Yerevan: CBA.
Del Negro, M. and F. Obiols-Homs. (1999). “Has Monetary Policy Been So Bad That it is Better to Get
Rid of It? The Case of Mexico” Unpublished Paper, September 1999. ITAM: Mexico City.
Desilets, Brien. (2003). Local Currencies: Theory and Evidence. Barcelona: Universidad de Barcelona
[unpublished academic paper].
Desilets, Brien. (2002). Bosnia and Herzegovina: Fiscal and Monetary Development and the Role of the
IMF. Barcelona: Universidad de Barcelona [unpublished academic paper].
Eichengreen, Barry. (1992). “Should the Maastricht Treaty Be Saved?”. Princeton Studies in
International Finance. No. 74. December 1992. Princeton: International Finance Section, Department of
Economics, Princeton University.
Feige, Edgar. (2003). “Dynamics of Currency Substitution, Asset Substitution and De facto Dollarisation
and Euroisation in Transition Countries” in Comparative Economic Studies. Vol. 45.
Fischer, Stanley, William Nordhaus and Lawrence Summers. (1992). “Stabilization and Economic
Reform in Russia” in Brookings Papers on Economic Activity. Vol. 1992, No. 1, pp. 77-126.
Grigorian, David. (2003). Banking Sector in Armenia: What Would it Take to Turn a Basket Case into a
Beauty Case? Working Paper 03/07, January 2003. Washington, D. C.: Armenia International Policy
Research Group.
Grigorian, David, Armine Khachatryan and Grigor Sargsyan. (2004). Exchange Rate, Money and Wages:
What is Driving Prices in Armenia? Working Paper 04/08, January 2004. Washington, D. C.: Armenia
International Policy Research Group.
Guillaume, Dominique and David Stasavage. (2000). “Improving Policy Credibility: Is There a Case for
African Monetary Unions?” in World Development. Vol. 28. No. 8, pp. 1391-1407.
Ingram, James. November (1959). “State and Regional Payments Mechanisms” in The Quarterly Journal
of Economics. Vol. 73, issue 4 pp. 619-632. Boston: MIT Press.
IMF. (2002). Article IV Consultation. Washington, D. C.: IMF.
Kenen, Peter B. (1969). "The Theory of Optimum Currency Areas: An Eclectic View," in Monetary
Problems of the International Economy. By Robert Mundell and A. K. Swaboda (eds.). Chicago:
University of Chicago Press.
Central Bank of Armenia (CBA). (2004). Annual Report. Yerevan: CBA.
Central Bank of Armenia (CBA). (2004). Monetary Policy Program of the Republic of Armenia. Yerevan:
CBA.
Central Bank of Armenia (CBA). (2004). Statistical Bulletin of the Central Bank of Armenia. Yerevan:
CBA.
Central Bank of Armenia (CBA). (1999). Banking System of Armenia: Development, Regulation and
Supervision. Yerevan: CBA.
Del Negro, M. and F. Obiols-Homs. (1999). “Has Monetary Policy Been So Bad That it is Better to Get
Rid of It? The Case of Mexico” Unpublished Paper, September 1999. ITAM: Mexico City.
Desilets, Brien. (2003). Local Currencies: Theory and Evidence. Barcelona: Universidad de Barcelona
[unpublished academic paper].
Desilets, Brien. (2002). Bosnia and Herzegovina: Fiscal and Monetary Development and the Role of the
IMF. Barcelona: Universidad de Barcelona [unpublished academic paper].
Eichengreen, Barry. (1992). “Should the Maastricht Treaty Be Saved?”. Princeton Studies in
International Finance. No. 74. December 1992. Princeton: International Finance Section, Department of
Economics, Princeton University.
Feige, Edgar. (2003). “Dynamics of Currency Substitution, Asset Substitution and De facto Dollarisation
and Euroisation in Transition Countries” in Comparative Economic Studies. Vol. 45.
Fischer, Stanley, William Nordhaus and Lawrence Summers. (1992). “Stabilization and Economic
Reform in Russia” in Brookings Papers on Economic Activity. Vol. 1992, No. 1, pp. 77-126.
Grigorian, David. (2003). Banking Sector in Armenia: What Would it Take to Turn a Basket Case into a
Beauty Case? Working Paper 03/07, January 2003. Washington, D. C.: Armenia International Policy
Research Group.
Grigorian, David, Armine Khachatryan and Grigor Sargsyan. (2004). Exchange Rate, Money and Wages:
What is Driving Prices in Armenia? Working Paper 04/08, January 2004. Washington, D. C.: Armenia
International Policy Research Group.
Guillaume, Dominique and David Stasavage. (2000). “Improving Policy Credibility: Is There a Case for
African Monetary Unions?” in World Development. Vol. 28. No. 8, pp. 1391-1407.
Ingram, James. November (1959). “State and Regional Payments Mechanisms” in The Quarterly Journal
of Economics. Vol. 73, issue 4 pp. 619-632. Boston: MIT Press.
IMF. (2002). Article IV Consultation. Washington, D. C.: IMF.
Kenen, Peter B. (1969). "The Theory of Optimum Currency Areas: An Eclectic View," in Monetary
Problems of the International Economy. By Robert Mundell and A. K. Swaboda (eds.). Chicago:
University of Chicago Press.
Monetary Decentralization: The Case of Armenia 38
Knobl, Adalbert, Andres Sutt and Basil Zavoico. (2002). The Estonian Currency Board: Its Introduction
and Role in the Early Success of Estonia’s Transition to a Market Economy. IMF Working Paper 02/96,
May 2002. Washington, D. C.: International Monetary Fund.
Krugman, Paul. (1993). “Lessons of Massachusetts for EMU” in Adjustment and Growth in the European
Monetary Union by Francesco Giavazzi and Francisco Torres (eds.). Cambridge: Cambridge University
Press.
McKinnon, Ronald I. "Optimum Currency Areas," American Economic Review. (1963). Vol. 53, pp. 717-
25.
Meade, J. E. (1957). “The Balance of Payments Problems of a Free Trade Area” in Economic Journal.
September 1957. No. 67. pp. 379-396.
Mundell, Robert. (1961). "A Theory of Optimum Currency Areas," American Economic Review. Vol. 51,
pp. 657-65.
Nordhaus, William, Merton Peck and Thomas Richardson. (1991). “Do Borders Matter? Soviet
Economic Reform After the Coup” in Brookings Papers on Economic Activity. Vol. 1991, No. 2, pp. 321-
340.
Ohanian, Rubina, Cyrus Safdari and Nancy Scannell. (2003). A Statistical Analysis of Banks in Armenia.
Working Paper 03/08, January 2003. Washington, D. C.: Armenia International Policy Research Group.
Poghosyan, Tigran. (2003). Demand for Money in Armenia: Partial Adjustment Approach. Prepared for
the Armenia: Recent Economic Trends and Growth Prospects Conference. Washington, D. C.: Armenia
International Policy Research Group.
Sachs, Jeffrey and Xavier Sala-I-Martin. (1992). "Fiscal Federalism and Optimum Currency Areas:
Evidence for Europe from the United States" in Establishing a Central Bank: Issues in Europe and
Lessons from the US. Cambridge: Cambridge University Press.
Sachs, Jeffrey. (2005). The End of Poverty. New York: Penguin Group.
Saunders, Anthony. (1997). Financial Institutions Management: A Modern Perspective. Second Edition.
Chicago: Richard D. Irwin.
Schmitt-Grohé, Stephanie and Martín Uribe. (May 2000). Stabilization Policy and the Costs of Foreign
currency penetration. Social Science Research Network Electronic Library [Unpublished, available
online].
Scitovsky, Tibor. (1958). Economic Theory and Western European Integration. Stanford.
Standard and Poor´s. (28 February 2002). Acuerdos Provinciales en Argentina: ¿Por qué son tan difíciles
de alcanzar?
United Nations Economic Commission for Europe (UNECE). (2004). Economic Survey of Europe. No. 1.
Geneva: UNECE.
United Nations Economic Commission for Europe (UNECE). (2002). Economic Survey of Europe. No. 2.
Geneva: UNECE.
Knobl, Adalbert, Andres Sutt and Basil Zavoico. (2002). The Estonian Currency Board: Its Introduction
and Role in the Early Success of Estonia’s Transition to a Market Economy. IMF Working Paper 02/96,
May 2002. Washington, D. C.: International Monetary Fund.
Krugman, Paul. (1993). “Lessons of Massachusetts for EMU” in Adjustment and Growth in the European
Monetary Union by Francesco Giavazzi and Francisco Torres (eds.). Cambridge: Cambridge University
Press.
McKinnon, Ronald I. "Optimum Currency Areas," American Economic Review. (1963). Vol. 53, pp. 717-
25.
Meade, J. E. (1957). “The Balance of Payments Problems of a Free Trade Area” in Economic Journal.
September 1957. No. 67. pp. 379-396.
Mundell, Robert. (1961). "A Theory of Optimum Currency Areas," American Economic Review. Vol. 51,
pp. 657-65.
Nordhaus, William, Merton Peck and Thomas Richardson. (1991). “Do Borders Matter? Soviet
Economic Reform After the Coup” in Brookings Papers on Economic Activity. Vol. 1991, No. 2, pp. 321-
340.
Ohanian, Rubina, Cyrus Safdari and Nancy Scannell. (2003). A Statistical Analysis of Banks in Armenia.
Working Paper 03/08, January 2003. Washington, D. C.: Armenia International Policy Research Group.
Poghosyan, Tigran. (2003). Demand for Money in Armenia: Partial Adjustment Approach. Prepared for
the Armenia: Recent Economic Trends and Growth Prospects Conference. Washington, D. C.: Armenia
International Policy Research Group.
Sachs, Jeffrey and Xavier Sala-I-Martin. (1992). "Fiscal Federalism and Optimum Currency Areas:
Evidence for Europe from the United States" in Establishing a Central Bank: Issues in Europe and
Lessons from the US. Cambridge: Cambridge University Press.
Sachs, Jeffrey. (2005). The End of Poverty. New York: Penguin Group.
Saunders, Anthony. (1997). Financial Institutions Management: A Modern Perspective. Second Edition.
Chicago: Richard D. Irwin.
Schmitt-Grohé, Stephanie and Martín Uribe. (May 2000). Stabilization Policy and the Costs of Foreign
currency penetration. Social Science Research Network Electronic Library [Unpublished, available
online].
Scitovsky, Tibor. (1958). Economic Theory and Western European Integration. Stanford.
Standard and Poor´s. (28 February 2002). Acuerdos Provinciales en Argentina: ¿Por qué son tan difíciles
de alcanzar?
United Nations Economic Commission for Europe (UNECE). (2004). Economic Survey of Europe. No. 1.
Geneva: UNECE.
United Nations Economic Commission for Europe (UNECE). (2002). Economic Survey of Europe. No. 2.
Geneva: UNECE.
Monetary Decentralization: The Case of Armenia 39
World Bank. (1992 – 1996). Statistical Handbook: States of the former USSR. Washington, D. C.:
World Bank.
World Bank. (1992 – 1996). Statistical Handbook: States of the former USSR. Washington, D. C.:
World Bank.
Tags
Download
Download Slideshow Get the original presentation fileQuick Actions
Statistics
- Views 0
- Slides 39
- Age 63 days
Related Slideshows
1
DTI BPI Pivot Small Business - BUSINESS START UP PLAN
MeljunCortes
29 views
1
CATHOLIC EDUCATIONAL Corporate Responsibilities
MeljunCortes
30 views
11
Karin Schaupp – Evocation; lançamento: 2000
alfeuRIO
29 views
10
Pillars of Biblical Oneness in the Book of Acts
JanParon
26 views
31
7-10. STP + Branding and Product & Services Strategies.pptx
itsyash298
28 views
44
Business Legislation PPT - UNIT 1 jimllpkggg
slogeshk98
31 views