Transcript (Financial) Markets in Transition Economies

DOLLARIZATION AND UNDEVELOPED CAPITAL
(FINANCIAL) MARKETS IN TRANSITION
ECONOMIES
Ruslan Piontkivsky,
International Centre for Policy Studies, Kyiv, Ukraine;
and National University of Kiev-Mohyla Academy, Kyiv, Ukraine
GDN Global Development Conference, December 9-12, 2001
The author is thankful for the support
to the Economic Education and Research Consortium – Russia
Definitions
 Dollarization: a situation in which residents of a country
hold a share of their assets in the form of foreign currency
and foreign currency denominated assets

replacing medium of exchange, unit of account, and store of value
functions of domestic money
 Currency substitution (CS) (direct currency substitution):
foreign currency is used for payments

replacing the medium of exchange function of domestic money
 Asset substitution (AS) (indirect currency substitution):
foreign currency is mainly used as a store of value
Objectives of the Study
Why it is important:
 Budget deficit financing becomes more inflationary
 Probability of a banking crisis increases due to a mismatch of assets and
liabilities as well as a lower ability to pay of foreign currency borrowers
 The structure of dollarization—whether it represents more CS or AS—is
crucial in the choice of exchange rate regime or monetary aggregate as
an operating goal of monetary policy
Objectives of the study:
 Determine dollarization factors in transition economies
 Test whether financial market developments are determinants of
dollarization in transition economies ;
 Draw conclusions regarding policies aimed at minimizing adverse
effects of dollarization
Hypotheses
 The level of dollarization is determined by return and risk
characteristics of assets denominated in national and foreign
currencies;
 The level of dollarization depends on the extent of financial
markets development;
Where This Project Fits
Existing Literature
 Early CS models (Calvo and Rodriguez, 1977; Leviatan, 1981)—assume
only two assets: domestic and foreign currency
 Asset portfolio balance models—explicitly assume the existence of
bonds denominated in each currency



Sequential portfolio balance model (Miles, 1978)
Dynamic optimization model (Bufman and Leiderman, 1993)
“Unrestricted” portfolio balance model (Branson and Henderson, 1985;
Cuddington, 1983; Thomas, 1985)
 Balance sheet models (Ize and Levy-Yeyati, 1998)—consider
dollarization from two sides of the financial intermediary; the absence of
foreign currency in circulation among the assets
Theory
 Thomas (1985)
R  R
S 2  SS *
 f  b 
 2
*2
2
*
A(V )( S  S  2SS ) (S  SS * )  (S *2  SS * )
 Consumers maximize their expected utility by choosing real
consumption level c and asset portfolio structure (f —foreign currency,
b—foreign bonds)
 Optimal choice (dollarization ratio) depends on the difference of real
returns on foreign and domestic bonds (R-R), domestic and foreign
prices volatility (S and S, prices’ instantaneous standard deviations), and
attitude to risk (V(A)—Arrow-Pratt measure of relative risk aversion )
Citations
Savastano (1996):
 “The relative importance of foreign currency as an inflation
hedge will be inversely related to the economy’s level of
financial development.”
Cuddington (1989):
 “Extending the Thomas paper to an environment where
there are goods and capital markets imperfections would be
one way of yielding an appropriate empirical specification
on which to base tests of the importance of CS in LDCs.”
Model Specification
 We assume that foreign prices are constant
DR 
 RR
RR

1

1

DR

A(V )( S 2 )
A(V )( S 2 )
 log( 1  DR ) it  log( A(V )) i   1 log( RR ) it   2 log( S 2 ) it  uit
 “Net” relative return on domestic bonds can appear to be
negative, we change net return into gross return
log( 1  DR ) it   i   1 log( 1  RR ) it   2 log( S 2 ) it  u it
 For small DR and RR, one can approximate
(  DR ) it   i   1 RR it   2 log( S 2 ) it  u it
Financial Markets Development
The extent of financial markets development affects the
dollarization level through the following channels:
 The change in domestic assets portfolio risk characteristics. Let d
consist of a set of assets n (mainly, bank deposits). As n increases—
emergence of enterprises’ stocks and bonds as well as government
bonds—opportunities for domestic portfolio diversification increase;
 Banking system development. If banks offer new instruments allowing
its clients to hedge off inflation and/or devaluation (i.e., indexed
deposits), then inflation volatility becomes less damaging, therefore,
the dollarization decreases;
 Due to borrowing constraints, CS and dollarization are not entirely
independent. So, the dollarization might depend on the foreign trade
turnover, as agents are not able to offset trade flows with financial
instruments
Data 1
 quarterly data for 6 transition economies—Ukraine, Russia, Czech
Republic, Slovak Republic, Poland, and Romania
 Most of the data from the International Financial Statistics (IFS)
 1991–1993 until Q4’2000
dollarization ratio:
 ratio of the sum of foreign currency deposits in the country to the money
supply (to total volume of deposits (incl. those in domestic currency) and
domestic currency in circulation)
b
DRM 
mbd
Data 2
Real relative return of domestic bonds RR  i  (i *  e e )
 difference between weighted domestic currency deposit rate and LIBOR
on three-months’ deposits in the U.S. dollar, taking into account
expected devaluation (changes in average quarterly exchange rate of the
domestic currency to the U.S. dollar)
Inflation volatility indicator S 2
 for quarterly data, we calculate it as variance of monthly inflation levels
(taking monthly CPI data)
Preliminary Estimation Results
 Empirical estimations do not support static exchange rate expectations.
For Russia, Ukraine, Poland, and Slovak Republic, the best results were
achieved for the specification, where rational exchange rate expectations
are modeled, while for Romania and Czech Republic adaptive
expectations fit better;
 For all the countries in consideration, except for Slovak and Czech
Republic, inflation volatility and relative returns are significant and
correctly signed at least at 5% (for Ukraine), while for Russia, Romania,
and Poland p-value is less than 1%;
 Poland has the best results in the sample;
 Incorporation of the foreign cash in circulation into the dollarization
ratio, used for estimation, leads to different results for Ukraine and
Russia. In case of Russia, the statistical properties of the regression
improve, while for Ukraine they deteriorate.
Policy Implications
 Relative return on assets and inflation volatility are determinants
of dollarization in transition economies
 Results of the Thomas model imply an alternative explanation of
the hysteresis. According to the model, the increased relative
return on domestic assets does not lead to the reduced
dollarization if it is accompanied with an increase in inflation
volatility.
 Estimation results offer a potential to fight the dollarization
resorting to such instruments as high domestic interest rates,
predictable exchange rate, and stability (not necessarily decline)
of inflation rate

Inflation targeting and sending signals to market about exchange rate
dynamics seems to be a preferred instrument
Financial Market Development
and dollarization (D)
 T. Beck, A. Demirguc-Kunt, R. Levine “A New Database of
Financial Development and Structure”, June 1999
 Transition economies (8): Russia, Ukraine, Poland,
Romania, Czech Republic, Latvia, Lithuania, Estonia
 Developing countries (5): Turkey, Egypt, Argentina, Bolivia,
Chile
 Averages over the 1990s, when available
D and Deposit Money Banks’
Assets
0,5
Lat
Bol
0,45
Tur Lit
Arg
Rus
0,4
Dollarization Ratio
0,35
Ukr
0,3
Rom
Est
0,25
Egypt
Pol
0,2
0,15
0,1
Cze
Chi
0,05
0
0
0,1
0,2
0,3
0,4
0,5
0,6
0,7
Deposit Money Banks' Assets, share in GDP
0,8
0,9
D and Private Credit
0,5
Lat
0,45
Lit
Dollarization Ratio
0,4
Rus
Ukr
0,35
0,3
Bol
Tur
Arg
Rom
Est
0,25
Egypt
Pol
0,2
0,15
Cze
0,1
Chi
0,05
0
0
0,1
0,2
0,3
0,4
0,5
Private credit of Deposit Money Banks and Other Financial
Institutions, share of GDP
0,6
D and Stock Market Value
Traded
0,5
Lat
0,45
Lit
Dollarization Ratio
0,4
Tur
Rus
0,35
Arg
Rom
0,3
0,25
Pol
0,2
0,15
0,1
Chi
Cze
0,05
0
0
0,02
0,04
0,06
0,08
0,1
Stockmarket total value traded to GDP
0,12
0,14
D and Foreign Trade Turnover
Dollarization Ratio
0,5
0,45
Bol
0,4
Tur
Lat
Lit
Rus
Arg
0,35
Rom
0,3
Ukr
Est
0,25
Egypt
Pol
0,2
0,15
0,1
Chi
Cze
0,05
0
0
20
40
60
80
100
120
Foreign Trade Turnover to GDP
140
160