Khuôn khổ lạm phát mục tiêu có phù hợp với việt nam hay không

MINISTRY OF EDUCATION AND TRAINING
UNIVERSITY OF ECONOMICS, HO CHI MINH CITY
FULBRIGHT ECONOMICS TEACHING PROGRAM




Nguyen Thi Thanh Huyen


IS INFLATION TARGETING APPROPRIATE FOR
VIETNAM?


MASTER IN PUBLIC POLICY DISSERTATION









Ho Chi Minh City, 2012

MINISTRY OF EDUCATION AND TRAINING
UNIVERSITY OF ECONOMICS, HO CHI MINH CITY
FULBRIGHT ECONOMICS TEACHING PROGRAM



Nguyen Thi Thanh Huyen

IS INFLATION TARGETING APPROPRIATE FOR
VIETNAM?

Public Policy Major
Code: 603114

MASTER IN PUBLIC POLICY DISSERTATION


SUPERVISOR
Dr. JONATHAN R. PINCUS






Ho Chi Minh - 2012

i

CERTIFICATION
I certify that I wrote this thesis myself.
I certify that the study has not been submitted for any other degrees.
I certify that any help received and all sources used have been acknowledged in this thesis
with the best of my knowledge.
The study does not necessarily reflect the views of the Ho Chi Minh City
Economics University or Fulbright Economics Teaching Program.

Author

Nguyen Thi Thanh Huyen

ii



ACKNOWLEDGEMENTS
This master of public policy dissertation could not be completed if I have not
received the help and encouragement from many people. First, I am very grateful to
my supervisor, Dr. Jonathan R. Pincus, who kept an eye on the progress of my work
and was always helpful when I need his advice. His advices, supports, criticisms and
comments helped me to deeply understand the overall knowledge related my
dissertation. I also convey a special thank to Dr. Vu Thanh Tu Anh, the first teacher
who encourange me to begin this dissertation and many other teachers that were
always support me during the time I studied in Fulbright Economic Teaching
Programme.
I gratefully acknowledge the financial support of FETP, without which this
dissertation would not have been done.

I also express my thankfulness to my colleagues in the Monetary Policy
Department, State Bank of Vietnam who are very helpful and enthusiastic colleagues,
for their sharing experiences and data to complete the dissertation. I also would like
to convey special thanks to Mr. Nguyen Xuan Thanh, Mrs. Nguyen Thi Kim Chau,
Mr. Tran Thanh Phong, Mr. Tran Thanh Thai, Mr. Truong Minh Hoa and Ms. Hoang
Ngoc Lan, who assisted me a lot of things in terms of adminstration to stabilize the
study here.
At the same time, I owe a large debt of gratitude to my parents and mother-in-
law, who helped to arrange family works for me to study. Last, I thank my husband
and daughters for their patience and support for me.


iii



CONTENTS
CERTIFICATION i
ACKNOWLEDGEMENTS ii
CONTENTS iii
ABBREVIATIONS v
LIST OF GRAPHS vi
LIST OF TABLES vii
ABSTRACT viii
CHAPTER 1: INTRODUCTION 1
CHAPTER 2: THEORETICAL AND EMPIRICAL OVERVIEW ON I.T 3
2.1. Conceptual framework 3
2.1.1. Supporting ideas for I.T 3
2.1.2. The model of I.T 4
2.1.3. Issues in the implementation of I.T 5

2.2. Global evidence on I.T. adoptation 6
2.2.1 Evidence supporting I.T 6
2.2.2 Evidence against I.T 8
2.3. The prerequisites of I.T 10
2.4. Experiences of inflation targeters 12
2.4.1 Experiences of industrial country inflation targeters 12
2.4.2 Experiences of developing country inflation targeters 14
2.4.2.1 Chile 16
2.4.2.2 South Korea 18
2.4.2.3 Brazil 20
CHAPTER 3: THE MONETARY POLICY FRAMEWORK IN VIETNAM 23
iv



3.1 Independence of SBV 23
3.2 Monetary policy implementation in Vietnam 23
3.2.1 Monetary instruments 24
3.2.2 Monetary policy transmission mechanism 25
3.2.3 The strategy of monetary policy 26
3.2.4 Monetary policy effectiveness 30
CHAPTER 4: TESTING I.T PREREQUISITES FOR VIETNAM 33
4.1 Economic structure 33
4.1.1 A small, open economy with liberalized capital and trade 33
4.1.2 Dollarization and goldization 35
4.1.3 Exchange rate pass-through effect 36
4.2 Health of the financial system 37
4.3 Analytical capability of SBV 39
4.4 Central bank independence 39
CHAPTER 5: CONCLUSION AND RECOMMENDATIONS 43

5.1 Conclusion 43
5.2 REER monetary policy framework is also problematic 44
5.3 Recommendations 45
REFERENCES 47

v



ABBREVIATIONS
ADB Asian Development Bank
CBI Central Bank Indepedence
CPI Consumer Price Index
FCD Foreign Currency Deposits
FDI Foreign Direct Investment
Fed Federel Reserve System
FII Foreign Indirect Investment
GDP Gross Domestic Products
GSO General Statistics Office
I.T Inflation Targeting
IFS International Funds Statistics
IMF International Monetary Fund
MoF Ministry of Finance
NCM New Concensus Macroeconomics
NEER Nominal Effective Exchange Rate
OECD Organization for Economic Coorperation and Development
OLS Ordinary Least Square
RER Real Exchange Rate
REER Real Effective Exchange Rate
SBV State Bank of Vietnam

U.K United Kingdom
USD United State Dollar
VAR Vector Autoregression
VND Vietnam Dong
WB World Bank




vi



LIST OF GRAPHS
Graph 3.1: Velocity of money in Vietnam 29
Graph 3.2: CPI, M2 and GDP growth rate of Vietnam 31
Graph 3.3: REER, NEER and Relative CPI of Vietnam 32
Graph 4.1: Trade deficit and openness of Vietnam economy 34
Graph 4.2: Capital inflow structure in Vietnam 34
Graph 4.3: USD/VND Exchange Rate 35
Graph 4.4 : Dollarization in Vietnam 36
Graph 4.5: Stock market capitalization in some countries 38
Graph 4.6 : Bond market development in Vietnam 38
Graph 4.7: Seigniorage to GDP in Vietnam 40
Graph 4.8: Real deposite rate in Vietnam 41
Graph 4.9: Fiscal balance to GDP in Vietnam 42

vii



LIST OF TABLES
Table 3.1: Exchange rate adjustment milestones of SBV 28
Table 3.2: Total reserves in months of imports in Vietnam 28





viii



ABSTRACT
Monetary policy is one of the most crucial macroeconomic policies in an economy
in general and in Vietnam in particular. Considering the low effectiveness of monetary
policy in Vietnam, many economists think that inflation targeting (I.T) is a best choice for
the future. The dissertation takes a comprehensive look at the I.T framework in theoretical
and empirical terms, the conditions as well as the global experiences in industrial and
developing countries. In theory, the I.T model does not mention other determinants of
inflation such as real shocks, exchange rate movements and fiscal roots. It is surprising that
although I.T is applied nearly everywhere but not all countries are successful and I.T is
often not the key determinant of success. Other factors are also important, such as fiscal
descipline, the economic structure, financial system strength, central bank indepedence and
political support for the inflation target.
The disseration also tests the conditions of Vietnam using some prerequisites for
I.T suggested by the IMF and finds that Vietnam is not suitable for this framework because

the country’s economic characteristics include a weak financial system and a lack of
central bank indepedence. Although the experience in some successful countries like Chile
and South Korea show that those conditions are not wholly met in these countries, at least
Chile and South Korea had good fiscal discipline and central bank independence. In
Vietnam, fiscal dominance exists with continuous fiscal deficits. Thus, to improve the
prevailing monetary policy framework, I.T may not be the best choice for Vietnam. REER
targeting is an alternative but it is also problematic. Vietnam is a highly open economy
with a high pass-through effect and dollarization. The dissertation recommends that at first
Vietnam should improve institutional conditions for monetary policy and strengthen
financial markets, and at the same time restructure the economy and impose fiscal
discipline on the government. Then choosing I.T or REER targeting or another monetary
framework will depend on political considerations and the desired balance between growth
and price stability.

1



CHAPTER 1: INTRODUCTION
Monetary policy is one of the most important macroeconomic policies in an
economy. The prevailing monetary policy framework in Vietnam now is a mix exchange
rate peg and monetary framework. But the effectiveness of monetary policy in Vietnam
is not easy to estimate as the central bank must target several objectives, including low
inflation, rapid growth and exchange rate stability. Besides, inflation has returned to
Vietnam in recent years and makes many economists think of I.T as the best choice but
it related many choices and conditions. These are the dynamics for me to carry out this
dissertation to answer the following questions
: First, is inflation targeting (I.T)
appropriate for Vietnam? Why and why not? And second, what are the implications
for Vietnam’s monetary policy framework? The dissertation uses institutional analysis

of monetary policy, case studies, descriptive statistics and results of other qualitative
researchs to answer research questions. The secondary data used are macroeconomics
fundamentals and finance data in Vietnam in the period 2000-2010.

I.T is applied in many industrial as well as developing countries and the adoption
of I.T based on specific theoretical framework and some key assumptions regarding
institutional prerequisites to using the inflation rate as a nominal anchor. There are very
few research mentioning adoption I.T in Vietnam (Le Anh Tu Packard, 2007; Mai Thu
Hien, 2007) and most of them gave general suggestions only. The objectives of the
dissertation is to comprehensively look at the theoretical and empirical evidence on I.T
to see if I.T is a really superior monetary policy regime for central banks, then examine
the prerequisites, experiences of inflation targeters and the conditions in Vietnam in
order to give policy implications for Vietnam. The dissertation finds that I.T is not a
perfect regime in both theory and practice, the success of I.T depends on the institutional
and economic conditions in individual country. The most common feature in successful
developing inflation targeters like Chile and Korea is the strict fiscal rule and central
bank independence. However, the institutional conditions in Vietnam shows that
Vietnam does not yet meet the requirements for adopting I.T.
Therefore, the dissertation finds that Vietnam’s monetary policy need
transparency, accountability, coherence and communication first. Then, the dissertation
recommends institutional reforms before adopting I.T or any other monetary policy
2



regime. After the conditions for conducting monetary policy have improved, the
adoption of I.T will depend on political considerations, in other words, whether the
government is willing to target inflation even at the cost of slower economic growth,
exchange rate volatility and high interest rates.

3



CHAPTER 2: THEORETICAL AND EMPIRICAL OVERVIEW ON I.T
I.T starts from the simple premise that the primary goal of monetary policy is to
achieve and maintain a low and stable rate of inflation (Masson et al., 1997, p. 5).
Although I.T became popular in 1990s it actually appeared much earlier. Haldane (1997,
p. 8) notes that price targets can be traced back to the last century – to Marshall (1887)
and Wicksell (1898), Fisher (1911) and Keynes (1923). In the 1990s, there was a wave of
countries trying to adopt I.T as a monetary policy framework. This section will discuss
the theoretical foundations of I.T, and then summarize the global evidence on the
implementation and evaluation of I.T.
2.1. Conceptual framework
2.1.1. Supporting ideas for I.T
So far, there is much research and debate on I.T as a monetary policy framework.
One supporting idea is that inflation is very costly and monetary policy is neutral in the
medium and long run, or monetary policy affects only the price level, and not output and
unemployment. (Friedman, 1967, quoted in Mishkin, 1997, p. 5) famously made the case
that there is no Phillips curve trade-off between unemployment and inflation in the long
run because inflation expectations counter-act the effects of monetary and fiscal policy.
Expectations can shift the Phillip curve upward, and hence there is no long run trade-off
between inflation and unemployment. An unemployment rate below the natural rate of
unemployment cannot be realized with permanently high inflation. In his Nobel prize
address (1967), Milton Friedman says that in the long run, higher inflation leads to
higher unemployment and the Phillip curve may be upward sloped.
Another supporting opinion (Mishkin, 1997, pp. 2-9) is that expansionary
monetary policy can not reduce unemployment whenever it increases above the “full-
employment level” because complicated macroeconomics models can not accurately

predict the impact of monetary policy on aggregate economy because of some problems.
First among these is the long lags associated with monetary policy. Indeed, the effects of
monetary policy are highly uncertain so using expansionary monetary policy to affect
the economy can even be counterproductive. Most politicians focus on the short run and
tend to fall into the trap of over-manipulating policy levers as they always want policy to
4



have immediate impact on the economy. The result is overheating and inflationary
pressure on the economy. Second is the time-inconsistency problem. A common way for
making policy decisions is to assume that expectations are formed at the time the policy
is made. In fact, firms and workers not only change their expectation but also wages and
prices. Thus output may not increase but actual inflation does. In short, these economists
think that I.T as monetary policy framework can be a good choice for central banks.
2.1.2. The model of I.T
In terms of theoretical foundations, I.T is closely associated with the New
Consensus Macroeconomics (NCM), which can be represented in 3 equations ( from
McCallum, 2001; Arestis and Sawyer, 2002, quoted in
Arestis and Sawyer, 2003, p. 3):
(2.1) Y
g
t
= a
0
+ a
1
Y
g
t-1

+ a
2
E
t
(Y
g
t+1
) – a
3
[R
t
– E
t
(p
t+1
)] + s
1

(2.2) p
t
= b
1
Y
g
t
+ b
2
p
t-1
+ b

3
E
t
(p
t+1
) + s
2

(2.3) R
t
= (1- c
3
)[RR* + E
t
(p
t+1
) + c
1
Y
g
t-1
+ c
2
(p
t-1
– p
T
)] + c
3
R

t-1

where
Y
g
is the output gap
R is nominal rate of interest
p is rate of inflation
p
T
is inflation rate target
RR* is the “equilibrium” real rate of interest, that is the rate of interest consistent
with zero output gap which implies from equation (2.2), a constant rate of inflation, s
i

(with i = 1, 2) represents stochastic shocks, and E
t
refers to expectations held at time t.
Equation (2.1) is the aggregate demand equation with the current output gap
determined by the past and expected future output gap and the real expected rate of
interest.
Equation (2.2) is a Phillips curve with inflation based on current output gap and
past and future inflation.
5



Equation (2.3) is a monetary-policy rule. In this equation, the nominal interest
rate is based on expected inflation, the output gap, the deviation of inflation from target
(or “inflation gap”), and the “equilibrium” real rate of interest.

The model suggests that if a central bank can credibly signal its intention to
achieve and maintain low inflation, then expectations of inflation will be lowered and
this means current inflation may be lowered at lower cost in terms of output. Inflation
above the target dictates higher interest rates to contain inflation whereas inflation below
the target requires lower interest rates to stimulate the economy and increase inflation.
In operational terms, expected inflation rate is considered the intermediate target
of monetary policy under I.T.
However, this theoretical model of I.T assumes that inflation is affected by
monetary policy only and does not mention the real shocks to inflation, the fiscal-root
inflation as well as the exchange rate impact on inflation. Masson et al (1997, p.11) also
implies that there are some “imperfect controllability” problems in an I.T regime. The
central bank can not perfectly control the inflation rate due to aggregate demand/supply
shocks, velocity shocks and fiscal policy.
2.1.3. Issues in the implementation of I.T
There are many ideas on the various issues in implementing I.T as a monetary
framework. According to Masson et al (1997, pp. 33-35), the first issue involved is
specification of the inflation target chosen by the central bank. For example, the central
bank must choose a price index, the numerical value of the target, the time horizon (one
period or multi period) and the time path (declining or flat path), the range (tolerance
interval or a single point), and the specification of an “escape clause” in certain
circumstances.
The second issue is that institutions matter. Should I.T be an officially mandated
objective or an operational requirement of monetary policy? What is the best way to
integrate I.T with other macroeconomic policies? Which are the vehicles or means to
improve the transparency and accountability of monetary policy?
The third issue is the trade-off between the credibility and flexibility of central
banks, especially when an inflation target is specified as a range which allows the central
6

bank to have some flexibility in the short run, such as stabilization objectives including
the exchange rate and output as well as financial stability.
Considering all these problems, Bernanke and Mishkin (1997, p. 22) conclude
that I.T is not a rigid rule but rule-like with “constrained discretion” as the nature of I.T
is forward looking. But they conclude that it is too early to say if I.T is a fad or a trend.
The only thing they agree on are the advantages, such as transparency, coherent policy-
making and accountability of moving to an I.T. regime.
2.2. Global evidence on I.T. adoptation
Despite some problems and issues in I.T implementation, many countries around
the world have adopted I.T as a monetary policy regime and customized the approach in
various ways. I.T was first adopted in New Zealand, U.K (U.K), Canada and Sweden in
the 1990s (Allen et al., 2006, p. 4). As of 2010, there were 26 inflation targeters, of
which more than half were emerging market countries (Scott Roger, 2010, p. 4). The
evidence on the success of I.T is mixed. Some evidence says that I.T. improved the
performance of the economy and others say that it was not I.T. that enhanced the
macroeconomic performance but other factors.
2.2.1 Evidence supporting I.T
Neumann and von Hagen (2002) use VAR models with monthly and quarterly
data for six I.T. countries (Australia, Canada, Chile, New Zealand, Sweden, and the
U.K) and three non-IT countries (Germany, Switzerland, and the United States) for two
sample periods: a pre-IT period (1978-92) and a post-IT period (1993-2001) to examine
the volatility of inflation, output gaps, and central bank interest rates under I.T. The
evidence confirms that adopting I.T reduces the inflation rate and limits the volatility of
inflation and interest rates. Thus, I.T has helped the high-inflation countries to achieve a
degree of credibility similar to that of the Bundesbank and the Swiss National Bank. But
among inflation targeters, the U.K has performed best even though its target rate of
inflation was higher than others. However, this does not necessarily mean that I.T is
superior to monetary aggregate regimes, such as the Bundesbank’s approach to monetary
targeting between 1974 and 1998, or even to the strategy of Federal Reserve System

(Fed) in the 1980s and 1990s, which pursued neither monetary nor on inflation targets.
7



A comparision of I.T and non-I.T regimes shows that I.T conformed to the monetary
policy set by the Bundesbank, the Fed and the Swiss National Bank in the 1970s and
1980s. Under the Bundesbank’s former monetary policy concept, the inflation objective
served to anchor medium-run inflation expectations while short-run operations were
guided by an intermediate monetary target. The study concludes that I.T must be seen in
the context of economic, the specific environment and conditions under which it is
implemented.
Allen et al (2006) uses illustrative simulations based on a model calibrated for
typical emerging countries to assess the performance of different monetary regimes. The
study finds that exchange rate and money targeting regimes create higher
macroeconomic variability compared to I.T. Statistical analysis of macroeconomic
variables (inflation, output and their variability) in 13 emerging I.T and 29 emerging
non-I.T countries for the period 1985-2004 also yields some interesting findings.
Macroeconomic performance in emerging countries under I.T has been better than under
other regimes. First, the I.T countries have less economic instability than the others.
Long-run inflation expectations in I.T countries fell from 2.7-2.1 percent, compared with
an increase from 3.1-2.3 percent under an exchange rate peg (Allen et al., 2006, p. 13);
Second, the volatility of nominal exchange rates
1
, real interest rates and reserves in I.T
countries was lower than in non-I.T countries. Although the environment in 1990s was
relatively benign, the performance of I.T and non-I.T are compared in the same period
and I.T outperforms the non-I.T countries.
However, the caveats of the studies were the short period of time since the
introduction of I.T in emerging countries and the difficulty of establishing causality from

I.T to the observed outcomes as there were many other reforms that accompanied the
shift to I.T.
Moreover, Arestis and Sawyer (2003) says that the evidence on I.T practice as
mentioned in the above studies has some limits: (i) the evidence is not convincing as the
environment in the 1990s was very conducive to price stability, so that there is no proof
that I.T outcomes were better than non-I.T outcomes; (ii) the dangers of not adopting I.T

1
Measured by standard deviation of variables
8



never materialized, even though European Central Bank (ECB) and Fed did not move to
an I.T regime; (iii) in a number of countries like New Zealand, Canada and the U.K,
inflation had been “tamed” well before introducing I.T (Mishkin and Posen, 1997,
quoted in Arestis and Sawyer, 2003, p. 21).
2.2.2 Evidence against I.T
There are plenty of studies that give opposite evidence on the impact of I.T on
economic performance. Stephen G. Cecchetti and Michael Ehrmann (1999) study the
trade-off between inflation and output in the short run for 23 countries, including nine
inflation targeters (Australia, Canada, Chile, Finland, Israel, New Zealand, Spain,
Sweden, U.K) in the period 1984-1997, using a structural VAR model. The authors find
that aversion to inflation variability increased during the decade of the 1990s. The
inflation targeters increased their aversion to inflation volatility by more than the
nontargeters, although the difference is modest. For nine inflation targeters, inflation fell
by more than seven percentage points on average while the non-targeters’s average
reduction is only 3.6 percent. But all of these countries suffered higher increases in
output volatility as a result.
Ball and Sheridan (2004) examine the effects of I.T on macroeconomic

performance, using 20 OECD countries during the 1990s, seven of which are inflation
targeters (Australia, Canada, New Zealand, Finland, Spain, Sweden, U.K) and thirteen
are non-inflation targeters. The methodology used is “difference in difference OLS” and
the data are the annualized average rates in the period 1960-2001. The results show that
economic performance is different in individual countries but on average, there is no
evidence that I.T improves performance with regards to inflation, output, or interest
rates. Inflation targeters have caught up with nontargeters, but this convergence was not
caused by I.T. It may be the result of regression to the mean rather than the effects of
I.T. The paper also concludes that I.T is neither beneficial nor harmful.
Ricardo D. Brito and Brianne Bystedt (2010) use a dynamic panel estimator
developed by Arellano and Bover (1995) and Blundell and Bond (1998) to test the
impact of I.T on economic performance. The partial adjustment model uses annualized
average inflation and real output rates of growth in 46 emerging countries, in which 13
9



are I.T in the period 1980-2006. In contrast to previous studies, this methodology tries
to isolate the impact of purely I.T on macroeconomic performance, or the causality of
adopting I.T on economic performance. The study finds that there is no significant
evidence to conclude that the I.T meets its main goal of stabilizing inflation and output
growth in emerging economies. Although they admit that I.T leads to lower inflation,
this relation is weaker than in previous studies. What actually leads to lower inflation in
these developing I.T countries is their decision to aim for lower inflation. The trade-off
between lower inflation and output under I.T is similar in I.T regimes and alternative
monetary frameworks. The finding that a trade-off exists between output growth and
inflation and inflation contradicts the core assumption of I.T.
Yifan Hu (2003) specifies a regression model for a 66-country sample over the
1980–2000 period to examine which factors are systematically related in a country’s
choice of I.T and what is the influence of I.T on economic performance. The paper finds

that economic conditions, structure, and institutional variables are systematically
associated with a country’s decision to adopt I.T. Low GDP growth and high real
interest rates are significantly associated with the choice of I.T. The negative impact of
high inflation on the choice of I.T reflects the fear of losing public credibility of the
central bank: they tend to adopt I.T when inflation rates are low as it is easier to fulfill
the goal. A sound fiscal position benefits the authority when adopting the I.T framework.
A de facto floating exchange rate regime seems a better indicator in describing the
choice of I.T than a de jure floating exchange rate regime.
The paper also carries out a descriptive analysis of 37 countries, of which eight
are industrialized inflation targeters except for Chile (Australia, Canada, Chile, Finland,
New Zealand, Spain, Sweden, and U.K) and 29 are nontargeters in pre- and post-I.T
adoptation for the period 1985-2000. The results show improvements to inflation and
output in both inflation targeter and nontargeter groups. It is not easy to evaluate the
impact of I.T on economic performance by descriptive analysis, but the inflation rate of
the I.T countries dropped more than that of the nontargeters for both the 37 country and
industrial-country-only cases. The results are similar with respect to the output. To
investigate further the impact of I.T on economic performance, the study fits a
10



regression model for 37 countries and finds that I.T at the same time lowers both
inflation and output variability but neither effect is statistically significant.
In sum, the global evidence tells us that the impact of I.T on macroeconomic
performance is controversial. The main reasons for these difference are the variations in
methodology (VAR models, difference in difference OLS models, partial adjustment
models, descriptive analysis and illustrative simulations), the countries selected, the
period used, as well data measurements (monthly, quarterly and yearly). It is therefore
difficult to specify the impact of I.T on economic performance. In fact, the observed
outcomes are different depending on factors specific to individual countries, their

economic culture and institutional conditions. It is not possible on the basis of this
evidence to conclude that I.T is always a superior monetary regime.
2.3. The prerequisites of I.T
As seen above, I.T has been widely adopted but is not successful everywhere
because the impact of I.T depends on some institutional and technical conditions. The
detailed requirements for adopting I.T differ among different studies.
Allen et al. (2006, p. 18) carried out surveys of central banks and created four
groups of technical conditions or prerequisites for adopting I.T for emerging countries as
follows:
(i) Central Bank infrastructure, including data availability, a systematic
forecasting process and models capable of conditional forecasts. These conditions are
important as the central banks need full data sets as well as high quality models to
forecast inflation.
(ii) Health of the financial system, including six indicators to measure the
soundness and development of the banking and financial systems, such as the banks’ risk
weighted capital adequacy ratios, stock market capitalization, depth of the public and
private bond markets, stock market turnover, foreign borrowing and exposure to short
term debt. A sound financial system minimizes conflicts between inflation and financial
stability and assure a good monetary policy transmission mechanism as well as support
for the absorb of debt instruments from the government.
11



(iii) Economic structure is often thought to be an important factor affecting the
success of I.T, such as low exchange rate pass-through, low sensitivity to commodity
prices, low extent of dollarization and high extent of trade openness. It is easy to see that
high exchange rate pass-through can make it harder for the central bank to achieve an
inflation target due to imported inflation. Dollarization can distort money aggregates and
renders domestic interest rates an ineffective policy tools, as borrowers shift into dollars.

Trade openness can force central banks to cope with the huge mobilization of goods,
which makes them more difficult in conducting monetary policy due to the danger of
huge trade deficit and imported inflation.
(iv) Institutional independence can be measured by six indicatiors, such as
absence of fiscal obligations (or fiscal discipline), operational independence, inflation-
focused mandate, favorable fiscal balance, low public debt and central bank
independence.
Most studies agree that independence is the first prerequisite of any central bank
that wants to adopt I.T. But this is not understood as full independence as the central
bank is only independent in terms of instruments to achieve the goal, especially
independent from fiscal policy (Amato and Gerlach, 2001; Masson et al., 1997;
Bernanke and Mishkin, 1997; Allen et al., 2006). Fiscal independence means that there
is no symptom of fiscal dominance, or monetary policy is not severely constrained by
fiscal developments. In other words, borrowing from the central bank and the banking
system for budget purposes must be very small, or the government does not rely on the
central bank to finance its budget shortfalls. Fiscal dominance can create inflationary
pressure when the government abuses this source of revenues from the central bank and
banking system, namely fiscal root inflation, and is left unchecked by the central bank.
Thus, monetary policy will meet difficulties in controlling fiscal-origin inflation and is
forced to follow the fiscal policy (accommodative monetary policy).
In developing countries, fiscal dominance is manifested in reliance on
seigniorage and financial repression (Masson et al., 1997, p. 23). The reliance on
seigniorage can come from a weak and unstable taxation system or abuse by the
authorities of this source of revenue. Financial repression – like interest rate ceilings,
extremely high reserve requirements, and directed credit - is both cause and effect of
12



shallow financial markets. Shallow financial markets limit the ability of the market to

absorb the debt instruments issued by the government, thus forcing the government to
rely on seigniorage.
However, the evidence of Masson et al. (1997) suggests that meeting strict
preconditions may be less important than the sustained pursuit of improvements once I.T
has been adopted. In fact, nearly all the inflation targeters get low scores in these criteria
prior to the adoption of I.T and the conditions improve gradually during the period of
applying I.T (Allen et al., 2006, p. 58). Thus, the improvement in conditions and
institutions is associated with better performance under I.T, and they do not come prior
to the adoption of I.T. Therefore, Allen et al (2006, p. 17) emphasizes these prerequisites
as essential, not desireable conditions.
In addition to these requirements, we must also consider the impact of an open
capital account, which can make it impossible to target inflation using domestic interest
rates. Higher interest rates attract foreign capital (seeking higher returns). The inflow of
foreign capital increases the domestic money supply unless it is sterilized by the central
bank. But this requires selling bonds in the domestic market, which forces up interest
rates even more, and draws in even more foreign capital. If the inflows are sterilized,
interest rates rise to high levels, slowing growth. If the inflows are not sterilized, they
can lead to asset bubbles and inflation. Besides, I.T must be the overriding target and the
main commitment of the monetary authorities (Masson et al.,1997, p. 8), while in the
presence of free capital mobility, a central bank cannot at the same time pursue both
fixed exchange rates and an inflation target.
2.4. Experiences of inflation targeters
2.4.1 Experiences of industrial country inflation targeters
Notwithstanding many technical and implementation issues of I.T as a monetary
policy framework, it has been used popularly in industrial countries, including New
Zealand, Canada, U.K, Sweden, Australia, Finland, Spain, and Israel. New Zealand was
the first country to adopt I.T. in March, 1990. Bernanke and Mishkin (1997, pp. 2-8)
generalize five major features of these inflation targeters, which are mostly similar to the
review of Masson et al (1997, pp. 18-20).
13

First, inflation targets are announced by the government, the central bank or
combination of the two. The central bank pledges to keep inflation at or near the
announced target. The inflation rate to be targeted is often the Consumer Price Index
(CPI), excluding interest cost components, indirect taxes and subsidies, government
charges, and significant price effect changes in terms of trade, food and energy. This is
sometimes called core inflation. The target can be a specific number or range or goal
with an upper and lower band. Therefore, the central bank still has room for short term
stabilization measures for exchange rates and output.
For example, New Zealand and Canada announce target rates as a range (0-2
percent through the five year tenure of the Governor and 1-3 percent through 1998,
respectively), U.K and Sweden announce a specific target with a band, such as 2.5
percent ±1 percent (Masson et al., 1997, p. 19).
Second, together with the announcement of the authorities is the statement that
the central bank will be accountable for the inflation target. In New Zealand, the
responsibility for setting out inflation targets is stipulated in the Reserve Bank of New
Zealand Act of 1989, so there is a link between the tenure of the Governor of the
Reserve Bank and the achieving of the inflation target. In Switzerland, Canada and U.K,
the inflation goal is not stated by law but in statements by the central bank to the public.
Therefore, there are no sanctions on the central bank for missing the target but there will
be loss of prestige and reputation in personal or institutional terms. Bank governors are
less likely to be reappointed if they consistently miss inflation targets.
Third, another feature of these countries is that the inflation target is considered
the overriding goal, or, in other words, this is a committed goal by the authorities. Thus
the role of exchange rates and money growth is reduced, especially in cases of conflict.
Under I.T the money supply is endogenous to the model, and adjustments take place due
to changes in inflation expectations rather than mechanically through the money supply.
The relationship between money growth and inflation is unreliable due to unstable

velocity, and thus the central bank prefers to focus directly on inflation rather than on an
intermediate target such as money growth or credit. In case the magnitude of capital
mobility is huge, I.T is associated with higher exchange rate volatility, as can be seen in
U.K, Sweden and Finland in 1990s (Masson et al., 1997, p. 18).
14



Fourth, the role of forecasting is very important in I.T. The central bank must
pay close attention to a variety of factors that affect inflation, such as exchange rates,
interest rates, money and credit aggregates, commodity prices, wages and asset prices.
Together with the forecast capabilities, the central bank must submit regular reports on
inflation to keep track of inflation movements. For instance, the Bank of England
publishes a quarterly Inflation Report, including the determinants of inflation as well as
the forecast of inflation; the Bank of New Zealand issues reports every six months.
These reports help improve the communication with the public.
The last (but not least) feature of industrial inflation targeters is the independence
of the central bank in these countries. Using the measurement of Central Bank
Independence (CBI) by Cukierman (1992), most of the industrial country inflation
targeters like Sweden, United Kingdom, Australia, Canada, Spain and New Zealand are
listed among the top 10 in terms of the independence of their central banks.
In addition, Masson et al. (1997, p. 20) finds that most inflation targeters adopt
I.T. when the inflation rates are already low (less than 10 percent per year in all
countries) and the credibility of the central bank is relatively high, thus enabling them to
influence public expectations on inflation.
2.4.2 Experiences of developing country inflation targeters
Aside from the industrial country inflation targeters, a wide variety of emerging
countries have adopted I.T as a monetary policy framework. Masson et al., (1997, pp.
21-34) finds that emerging and transition economies have some major problems. First,
most of these developing countries do not meet the prerequisites described above. They

adopted I.T when their macroeconomic and institutional conditions were relatively poor.
As seen in the IMF survey (Masson et al., 1997, p. 26), developing country inflation
targeters have very low CBI index compared to industrial inflation targeters. Similarly,
the seigniorage to GDP index is relatively high, and financial markets are shallower. A
high seigniorage index means that developing countries have high fiscal dominance.
Besides, shallow financial markets are a consequence of financial repression and also an
expression of fiscal dominance in these countries. Financial repression over a long
15



period leads to fragile banking systems in these countries and creates a dilemma for the
central bank between achieving financial stability and other objectives.
The second problem faced by emerging inflation targeters is a conflict of
objectives. Most of these economies are subject to tension between inflation and other
policy goals such as output growth, capital flows and the nominal exchange rates. In
most cases, the problem of conflict of objectives is left unsolved.
The third issue is the specification of the inflation target, such as the choice of
the level/horizon for the inflation target, the choice of an “escape clause” and the
problems of administered prices.
The fourth issue is a high level of dollarization, exchange rate fluctuations and
pass-through effects in emerging I.T countries (Mishkin, 2004, pp. 22-23). Exchange
rate depreciation can lead to inflation due to high import prices (supply side) and greater
demand for exports (demand side). In addition, liability dollarization means that a large
depreciation will lead to an increasing debt burden on domestic firms, deterioration of
balance sheets and financial instability, as the likelihood of bankrupcy becomes bigger
when the domestic debt burdens are so big that firms cannot pay. This phenomenon is
more severe when trade openness and capital mobility are high.
It is important to distinguish between the impact of exchange rates on inflation
and output to see if the shock stems from a sudden stop or a terms of trade shock (falling

export prices and/or rising import prices) to identify the correct monetary policy
response. If the depreciation is due to a pure portfolio shock, leading to inflationary
pressure, the central bank should raise interest rates to control inflation and attract
capital, which will slow domestic investment and consumption. If the depreciation
results from negative terms of trade shock, which lowers export demand and thus
aggregate demand, the central bank should reduce interest rates to cope with the drop in
demand aggregate and as a result capital will flow out. These are dilemmas that
developing countries have to cope with. The problem for developing countries is that
these swings are often very large relative to the size of the economy, and therefore
interest rates must rise or fall extremely sharply to achieve the desired effects. The
impact on the real economy may be large and potentially pro-cyclical.