Chính sách tiền tệ lạm phát mục tiêu và khả năng áp dụng ở Việt Nam ttta

MINISTRY OF TRAINING AND EDUCATION THE STATE BANK OF VIETNAM
BANKING UNIVERSITY OF HO CHI MINH CITY
----------oo0oo----------

NGUYEN THI MY HANH

INFLATION TARGETING STRATEGY
AND ITS APPLICABILITY IN VIETNAM

SUMMARY OF DOCTORAL THESIS
MAJOR: FINANCE – BANKING
CODE: 9.34.02.01
SUPERVISOR: ASSOC.PROF.DR. LE THI MAN

HO CHI MINH CITY – 2019

1


SUMMARY
For a long time, the State Bank of Vietnam has chosen inflation targeting as its main
monetary policy framework. However, so far, the transition to the inflation targeting
mechanism is still a big challenge for the central bank of Vietnam. In this thesis, the author
will provide empirical evidences to show that the bottleneck in conducting inflation targeting
monetary policy in Vietnam is ineffectiveness of using interest rate tool.
The central bank has the right to choose monetary policy instruments, but an important
condition for the central bank to succeed in applying the inflation targeting mechanism is that
there must exist a relationship between monetary policy instruments and inflation. In this
thesis, the author will examine the relationship between the interest rate tool (the main
monetary policy tool often used when the central bank applies target inflation policy) and
inflation. In addition, the thesis will determine whether the State bank of Vietnam use interest

rate instruments towards stabilizing inflation or exchange rate.
The author has set up a vector regression model of 5 variables: inflation, interest rate,
exchange rate, domestic aggregate demand and foreign aggregate demand. The model is
essentially a reflection of the monetary policy transmission mechanism of the interest rate tool.
There are many different types of interest rates, however, the interest rate that represents the
central bank's interest rate tool is policy rates. In Vietnam, there are two types of policy interest
rates: refinancing rate and interest rate. Research results from the thesis show that operating
the interest rate instruments including refinancing rate and discount rate to control inflation in
Vietnam is not really effective because the relationship between inflation and interest rate
instruments is very weak. In addition, the research results show that the policy interest rate tool
in Vietnam has a great impact on exchange rates. This suggests that it is possible that in the
past, Vietnam has used interest rate tools to protect the exchange rate rather than to stabilize
prices. This also means that Vietnam has not yet satisfied the two basic conditions to be met
before applying the inflation targeting monetary policy set by the International Monetary Fund.

2


CHAPTER 1. INTRODUCTION
1.1.The necessity of the thesis
In a survey of the IMF's International Monetary Fund for 88 non-industrial countries,
more than half of these countries expressed their desire to move to inflation targeting
framework and Vietnam is also included in the group of potential countries that can apply
inflation targeting (Batini and Laxton, 2006). The State Bank of Vietnam’s intention of moving
to inflation targeting is clearly reflected in the development plan of Vietnam's banking industry
to 2010 and development orientation to 2020 in Decision No. 112/2006 / QD-TTg dated
24/5/2006 of the Prime Minister: “Implementing monetary policy based on the regulation of
money amount; at the same time, building necessary conditions to gradually move to the
implementation of monetary policy on the basis of interest rate regulation. To create necessary
conditions for after 2010, the State Bank of Vietnam will move to inflation targeting

framework”. After 2010, nearly three-quarters (3/4) of the countries assessed by the IMF as
potential countries to move to inflation targeting mechanism have fulfilled their wishes.
Meanwhile, the year of 2020 is coming but Vietnam is still on the way towards inflation
targeting policy. Perhaps there are many difficult issues that the State Bank of Vietnam has not
yet solved to move towards the transition to inflation targeting mechanism.
According to the IMF research team, there are two basic conditions that a country needs
to satisfy when applying the inflation targeting mechanism (Debelle et al., 2018). The first
condition is that the central bank is independent in conducting monetary policy. It is not
necessary that the central bank is completely independent of the Government in conducting
monetary policy. The central bank only needs to be independent in selecting tools to run
monetary policy towards controlling inflation. In addition, the management of monetary policy
of the central bank must be independent of fiscal policy. In other words, the central bank must
not allow fiscal dominance, which means that the central bank cannot fund the government
budget deficit because it can lead to to increase inflation and reduce the effectiveness of
monetary policy. Moreover, not letting fiscal policy dominate monetary policy means central
bank does not print money to generate revenue for the budget. To do this, the central bank
needs to have revenues independent of the state budget. The second condition for the central
bank to be satisfied to apply the inflation targeting framework is that the central bank prioritizes
focusing on the most important goal of controlling inflation, instead of focusing on many other
goals such as exchange rate stability, economic growth, employment rates.
The two basic conditions mentioned above seem simple but why so far Vietnam has
not been able to apply the inflation targeting policy? Regarding the first condition, there have
3


been empirical studies proving that Vietnam does not allow the phenomenon of financial
repression because the budget deficit in Vietnam is not the cause of inflation (Nguyen &
Nguyen, 2010; Hoang, 2014). Therefore, it is likely that the problem that Vietnam is facing is
the independence of monetary policy tool to achieve inflation control target. Specifically,
Debelle (2001) analyzes the common obstacles facing emerging economies when applying

inflation targeting mechanisms in these countries, the relationship between monetary policy
instruments and inflation is often unclear. If central banks have adopted the inflation targeting
mechanism, which often uses the indirect tool of interest rates to conduct monetary policy,
central banks in many developing countries still use direct tools followed by issuing
administrative orders serving monetary policy management such as credit limit, deposit rate
ceilings. In order to make it clearer about the independence of monetary policy instruments in
Vietnam, in this study, the author will focus on analyzing the relationship between interest rate
tools in Vietnam and inflation to see whether this tool is really effective in controlling inflation
like the countries that apply inflation targeting. Besides, the thesis also considers whether
Vietnam satisfies the second condition or not? According to the Vietnam 2035 overview report
(World Bank, 2016), the State Bank of Vietnam has too many goals in conducting monetary
policy, while many central banks in the world are allowed to assign pricing tasks and report
implementation results to the government. Since 2012, inflation in Vietnam has always been
controlled at a single-digit level; however, due to the lack of commitment of the State Bank on
price stability in the long term, the expectation of price increases always exist and affect the
sustainable development of Vietnam's economy. Therefore, the World Bank team has proposed
that Vietnam should choose only one main goal of monetary policy: stablize exchange rate or
focus on inflation targeting. In the Banking Sector Development Strategy from 2025 to 2030,
the Prime Minister's Decision No. 986 / QD-TTg of August 8, 2018, clearly affirms that the
objective of controlling inflation is priority. For a developing country like Vietnam, exchange
rate stability is still an important issue to consider before Vietnam the inflation targeting
framework since scientific research has proven that exchange rate is an important factor
affecting inflation in Vietnam (Nguyen & Nguyen, 2010). Will the floating exchange rate
regime as managed in Vietnam last time affect the target of inflation control of monetary
policy? In this thesis, the author will examine whether exchange rate stablity or price stability
is the priority of Vietnamese monetary policy.
There are many studies on adopting inflation targeting monetary policy in Vietnam
such as studies by To Kim Ngoc (2012), To Kim Ngoc and Nguyen Khuong Duy (2012), Ly
Ba Hong Son and Nguyen Van Truong (2012), Nguyen Xuan Hung (2008), Nguyen Xuan
4

Hung (2009), Nguyen Phuc Canh (2013), Nguyen Thanh Nhan (2010), Do Thi Duc Minh
(2003), Nguyen Ngoc Thach (2013), To Kim Ngoc and Le Tuan Nghia (2012). However, these
studies have not yet analyzed in depth the independence of monetary policy instruments in
Vietnam nor assessed the role of exchange rate in transition to inflation mechanism in Vietnam.
Hence, in order to further clarify Vietnam's satisfaction with the necessary conditions to apply
the inflation targeting policy by quantitative method, the author chooses the topic "Inflation
targeting strategy and its applicability in Vietnam”. Empirical results from this thesis will also
be the scientific evidences to make necessary recommendations for Vietnam so that the country
could apply the inflation targeting framework in the future.
1.2.. Objectives of the study
The first research objective: The central bank has the right to choose monetary policy
instruments but the important condition for the central bank to succeed in applying inflation
targeting framework is that the relationship must exist between monetary policy instruments
and inflation. Therefore, the author will examine the relationship between monetary policy
instruments (particularly the interest rate tool - the tool often used when the central bank applies
inflation targeting monetary policy) and inflation.
The second research objective: The thesis will determine whether the priority goal of
the State Bank of Vietnam in the conduct of monetary policy is to stabilize inflation or stabilize
exchange rate.
1.3. Research question
The research question is based on two research objectives of the thesis as follows:
The first research question: Is there a relationship between policy rates and inflation
in Vietnam?
The second research question: Is the current setting of interest rate instruments in
Vietnam geared towards controlling inflation or stabilizing exchange rates?
1.3. Research Method
The thesis uses both quantitative and qualitative research methods. To answer the two
research questions, the author has set up a vector regression model including 5 variables:

inflation, interest rate, exchange rate, Vietnam industrial production index and United States
industrial production index. The model is a reflection of the monetary policy transmission
mechanism of the interest rate tool. There are many different types of interest rates, however,
the interest rate that represents the central bank's interest rate tool is policy interest rates
(interest rates that are determined by the central bank and published to provide a basis for
various types of interest rates in the market). In the context of Vietnam, the central bank decides
5


and announces two types of policy interest rates: refinancing rate and discount rate. Because
there are 2 types of interest rates acting as policy interest rates, the author built 2 models to test
the role of these interest rates. Because the data series in the study all stopped at the first
difference, the author applied the vector error correction model (VECM) in data analysis. The
analysis results from variance component analysis will show whether the interest rate
instrument in Vietnam has a real impact on inflation or on exchange rate. All data is processed
by using STATA 14 Software. In combination with qualitative analysis, the author will assess
the consistency of quantitative research results with the current situation in Vietnam.
1.4. Object and scope of the study
The main research object of this thesis is monetary policy framework and policy interest
rates in Vietnam.
The dissertation's research period starts from January 2011. This is the time to start
applying the State Bank Law 2010. The data in the thesis is collected monthly from January
2011 to June 2018.
1.5. The structure of the thesis
The thesis is designed with 5 chapters as follows:
Chapter 1. Introduction
Chapter 2. Literature review
Chapter 3. Model and research method
Chapter 4. Empirical results
Chapter 5. Conclusions and recommendations

1.6. Research contributions
The study of inflation targeting is not a new research topic in Vietnam. However, in the
current period, the research results from this thesis have proved that the bottleneck in moving
to inflation targeting in Vietnam is the interest rate tool. Recently, the State Bank of Vietnam
has well performed its duties in achieving goals and fulfilling tasks set by the Government and
the National Assembly. However, the State Bank of Vietnam still needs to improve and
modernize the way of conducting monetary policy. The empirical results from this thesis is
helpful for policy makers in Vietnam when it provides evidences explaining why Vietnam
could not apply inflation targeting at the moment as well as recommendations that would be
useful for Vietnam in the future.
1.7. New findings of the thesis
There are many research projects on inflation targeting in Vietnam. Compared to
previous studies, this thesis found 3 new findings:
6


- The thesis provides empirical evidence that using interest rate tools towards
controlling inflation in Vietnam is not really effective when the relationship between interest
rate and inflation is very weak.
- The research results also show that policy rates in Vietnam have a great impact on
exchange rates and the impact of interest rates on exchange rates is much larger than its impact
on inflation. It could be that Vietnam has used interest rate tools to protect the exchange rate
in recent years, or in other words, Vietnam has violated the main conditions of applying
inflation targeting monetary policy.
- This result shows that the level of transmission of the exchange rate into inflation is
very small. In other words, the explanation of the exchange rate on the change of inflation in
this study is very small. Therefore, Vietnam can continue with the current regime of floating
exchange rate management, and at the same time, expand the fluctuation range of the exchange
rate so that the country could apply the inflation targeting framework in the future.

7


CHAPTER 2. LITERATURE REVIEW
2.1. Monetary policy
2.1.1. Concept
Scientists offer different perspectives on monetary policy. Mishkin (2011) argues that
monetary policy is the process of managing money supply of the central bank towards an
interest rate level to achieve the goals of monetary policy such as inflation control, exchange
rate stabilization, full employment or economic growth. In Vietnam, Nguyen Van Tien (2009)
defines monetary policy as macroeconomic policy, through its tools, the central bank actively
changes the supply of money or interest rates (inter-bank interest rates) to achieve its set socioeconomic goals.
According to Vietnam's State Bank Law 2010, the definition of monetary policy
(monetary policy) is as follows: “National monetary policy is monetary decisions at the
national level of competent state agencies, including the determination of the target of
stabilizing the value of money expressed by the inflation target, the decision to use tools and
measures to realize the set objectives. ”
2.1.2. Monetary policy target system
2.1.2.1. The final goal
Each country can set many goals for the economy such as high economic growth, full
employment of inflation, low inflation, export growth ... Depending on the economic
characteristics, each country will have its own monetary policy. However, according to
Mishkin and Eakins (2006), the ultimate goal of monetary policy is usually directed to the main
goals: price stability, increase in employment, and economic growth. In addition, monetary
policy also aims to stabilize interest rates, stabilize exchange rates and stabilize financial
markets.
2.1.2.2. Intermediate target
The impact of monetary policy on final goals usually takes a certain amount of time.
Policy latency can range from 6 months to 2 years. Because of such a delay, the central bank
needs to identify intermediate targets before reaching the final goal. If you wait until the

economy shows signs and then react, it will be too slow.
The criterion for choosing an intermediate target is observable and measurable,
controllable, and is closely linked to the final goal. According to that criterion, the targets often
used as intermediate targets are total money supply (M1 or M2 or M3) or interest rates.
However, the central bank cannot choose both indicators as intermediate targets. This means
that the central bank can only choose one of the two criteria. If the central bank chooses the
8


money supply target, the target of interest rate removal must be removed or if the central bank
chooses the interest rate target, the money supply control target must be removed.
2.1.2.3. Operation target
Intermediate targets may have a direct impact on the ultimate goals of monetary policy
but cannot respond immediately when the central bank adjusts monetary policy instruments.
The operational objective is the variable that is directly affected by monetary policy
instruments and signals the view of monetary policy (tightening or easing). The criteria for
selecting the operational objectives are the same as the criteria for selecting the intermediate
targets. Only the criteria selected will be related to the intermediate target. Therefore, the choice
of activity target will also depend on whether the central bank chooses money supply or interest
rates as an intermediate target. With these criteria, the variables selected by the central bank as
operational targets include two types: (1) reserves including base or intermediate banks, (2)
interest rates such as interbank interest rates, treasury bill rates and open market interest rates.
The quick response of these indicators helps the central bank to check the soundness of
daily monetary policy decisions. Therefore, the ability to influence the final goal (mainly
inflation) is also the key factor for the effectiveness of any criteria selected as the operating
target. The relationship between inflation and interest rates is clearer than the relationship
between inflation and reserves, so most central banks in the world have used the short-term
interest rate target as their operational goal of monetary policy. This is an important theoretical
basis for building a research model in chapter 3.
2.2. Monetary policy inflation targets

2.2.1. The introduction of inflation targeting framework
In order to effectively operate monetary policy, the central bank needs to choose a
nominal ‘anchor’ for monetary policy, such as exchange rates, money supply or inflation. This
nominal anchorage will 'tightly tie' expectations of inflation, preventing prices from rising or
falling too quickly, thereby stabilizing the value of the local currency.
In the 1980s, the total amount of money as a nominal anchor for central bank monetary
policy failed as inflation continued to rise. Developed countries must apply tight monetary
policy, causing nominal interest rates to rise, leading to a suspension in economic activities.
More and more people are distrusting the central bank's monetary policy, even though the
central bank is committed to trying to keep inflation at a low level. This makes inflation
expectations higher and higher. Therefore, countries need to find a new anchor for monetary
policy. In 1989, New Zealand became the first country to target inflation as an anchor for

9


monetary policy. After that, many countries around the world such as Australia, Canada,
Finland, and Spain have applied the inflation targeting monetary policy framework.
2.2.2. Monetary policy concept of inflation targeting
According to Mishkin (2001), this policy relates to five main components:
(1) Transparent of medium-term goals;
(2) Monetary authorities should have a commitment to price stability as a key longterm goal of monetary policy implementation;
(3) Information strategy including many variables (not only total money supply or
exchange rates) used for establishing policy instruments;
(4) Transparent information about monetary policy through communication with the
public and the market about plans and targets set by policy makers;
(5) Strengthen the accountability of the central bank in trying to achieve the set goals
of monetary policy.
2.3. Experiences in adopting inflation targeting of some central banks in the world
2.3.1. Reasons why central banks choose to inflation targeting

Many central banks have accepted the mechanism of inflation targeting policy when
monetary targeting framework failed. The new monetary policy framework is also consistent
with the scientists' view of monetary policy, as follows:
Firstly, Milton Friedman and Edmund Phelps have demonstrated that high inflation
does not go hand in hand with high growth rates and high employment rates. By accepting this
view, policy makers no longer view monetary policy as a management tool in the short term,
but instead focus on the medium-term goal of price stabilization, which is also considered
central target of inflation targeting policy.
Secondly, scientists as well as policymakers realize that the positive side of keeping
inflation low; and that positive side is more beneficial than accepting high inflation to keep
high economic growth. Therefore, price stabilization has become the main goal of central
banks.
Thirdly, there are many scientific studies on monetary policy that have emphasized the
importance of inflation expectations. Monetary policy decisions will influence decisions of
economic factors. If the inflation rate is not fixed, people, businesses ... will expect the future
inflation to continue higher. Therefore, the target inflation policy is considered as an effective
measure to anchor inflation expectations, thereby, keeping inflation at a low and stable level.
2.3.2. Conditions given at the time of application of inflation targeting framework

10


According to Freedman and Laxton (2009), not all countries must have sufficient
prerequisites to apply inflation targeting mechanism. The reality of countries that have applied
inflation targering shows that many countries only meet some conditions but can still apply the
inflation targeting monetary policy framework successfully.
2.4. Review of related studies
2.4.1. International research
•

Impact of applying inflation targeting monetary policy to the economy:
There are many empirical studies that have demonstrated that the national

macroeconomic situation is improved after the application of inflation targeting monetary
policy (Bernake et al., 1999; Corbo et al., 2001; Neumann. and Von Hagen, 2002). The most
evidence is that after applying the inflation targeting policies, these countries have low and
stable inflation rates. However, there are differences between countries that apply inflation
targeting and countries that do not apply inflation targeting in two groups: the group of
countries belonging to the Organization for Economic Co-operation and Development (OECD)
and Emerging economies. Specifically, according to the studies of Ball and Sheridan (2005),
Lin and Ye (2007), Angeriz and Arestis (2008), Mishkin and Schmidt-Hebbel (2007), for
countries under the Organization for Cooperation and Development, there is no significant
difference in the economy (OECD), the difference in inflation and the level of fluctuations in
inflation are insignificant between the country applying the target inflation and the country
applying the target inflation. Meanwhile, in the emerging economies group, the countries that
applied the inflation targeting policy had significantly lower inflation rates than those that did
not apply inflation targeting policy. This has been demonstrated in the studies of Batini and
Laxton (2007), Gonϛalves and Salles (2008), and Lin and Ye (2009).
The impact of the global financial crisis of 2008-2009 is a great shock to many
economies. However, with countries applying the inflation targeting policy, the ability to cope
with the crisis is better than the countries that do not apply the inflation targeting monetary
policy. Although, according to the research of Ball and Sheridan (2005), there is no difference
in inflation and the growth rate between the countries applying the target inflation and those
not applying the target inflation in the group. OECD countries, however, during the global
crisis, countries that apply target inflation have a stable (less fluctuating) inflation rate than
countries that do not apply target inflation (Fouejieu, 2013). In a larger study, öztṻrk et al.
(2014) demonstrated that developed countries and developing countries applying the inflation
targeting monetary policy have the ability to overcome the 2008-2009 global financial crisis

11

compared with non-applicable countries. The study also suggests that central banks should
consider applying inflation targeting to overcome economic crisis.
It is worth noting that there are no studies showing that the application of inflation
targeting will be detrimental to growth, output, and employment in developed countries and
developing countries. No country has given up on inflation targeting after it has been applied.
•

Conditions for applying inflation targeting monetary policy:
According to International Monetary Fund (2005), the prerequisite conditions for

applying inflation targeting monetary policy are institutional independence of the central bank;
technical infrastructure development in terms of forecasting, building models and databases;
deregulated prices; not too sensitive to prices of basic commodities and exchange rates; low
dollarization rate; healthy financial system (especially healthy banks) and developed financial
markets.
To assess the role of prerequisite conditions before the application of inflation targeting
monetary policy, Batini and Laxton (2007) conducted a survey with 21 central banks that
applied inflation targeting policy and 10 central banks have not yet applied the inflation
targeting policy in emerging economies. The survey is sent to the central banks of countries
where inflation targeting is focused, with following questions: how has the target inflation
policy been developed and implemented? How does the central bank develop a communication
policy with the public? How the central bank has changed before and after the application of
inflation targeting policy? The survey results show that no country applying inflation targeting
monetary policy fully meets the prerequisite conditions when applying this policy. In addition,
the research results also show that emerging economies do not necessarily meet the
institutional, technical, and economic prerequisites before applying inflation targeting. Instead,
the success of targeting inflation targeting monetary policy depends on the commitment of the
governing body and the ability to plan for institutional changes after it has been declared to

appply inflation targeting.
Another study by De Mello (2008) on the actual situation of applying inflation targeting
monetary policy of Brazil, Chile, Czech Republic, Indonesia, South Africa and Turkey has
concluded that: When countries applies inflation targeting, there are many prerequisites that
these countries cannot achieve. However, the author pointed out that failing to meet the
monetary conditions before applying inflation targeting does not affect the application of
inflation targeting monetary policy of these countries if they always make efforts to improve
those conditions.

12


In another study by Freedman and Otker Robe (2009) on the experience of countries
that have applied inflation targeting, the study has shown that these countries do not support
the view that it is necessary to apply a long list of prerequisite conditions to the application of
inflation targeting monetary policy. However, in their experiences, there are a number of
important factors to implement the inflation targeting mechanism in a feasible manner as
follows: (1) price stabilization is the most important goal of all other monetary policy
objectives, (2) no fiscal dominance (fiscal dominance), (3) monetary policy instrument
independence, (4) high domestic consensus on inflation (5) a clear understanding of the
monetary policy transmission mechanism and the ability to affect short-term interest rates, (6)
a healthy market and financial system. These are considered factors leading to success in
applying inflation targeting policy.
In terms of prerequisite conditions for emerging economies, it is argued that these
countries do not have the institutional and technical conditions to adopt inflation targeting
policies. According to the research of Eichengreen, Masson, Savastano and Sharma (1999),
they believe that these countries lack the technical conditions and the independence of central
banks. The results of this study also gained agreement from some other studies such as: Agenor
(2000); Carare, Schaechter, Stone and Zelmer (2002); Masson, Savastano and Sharma (1997).
Therefore, the researchers recommend that countries should keep the old monetary policy

framework when the inflation targeting policy is not fully met. However, the central banks of
these countries should strive to achieve the conditions to apply inflation targeting. Research on
the experience of applying inflation targeting to emerging economies also includes studies of
Truman (2003), Jonas and Mishkin (2003), Debelle (2001), and Amato and Gerlach (2002).
2.4.2. Domestic research
Among the studies on the application of inflation targeting monetary policy in Vietnam,
To Anh Duong et al. (2012) is the most notable one. This study begins by reiterating the overall
goals of monetary policy. Next, the empirical research on the application of inflation targeting
policies of developed countries as well as developing countries to draw lessons for Vietnam.
Through an overview of monetary policy management in Vietnam in the period of 2000-2010,
the research team made a comment on the application of inflation monetary policy. Although,
Vietnam is not eligible to pursue inflation targeting policies, but based on the experience of the
previous countries, it is not necessary to converge all the conditions of applying inflation
targeting. Emerging economies can still successfully apply inflation targeting policy. On that
basis, the research team proposed the necessary conditions and roadmap for Vietnam to apply
inflation targeting monetary policy.
13


In addition, some articles mentioned the framework of inflation targeting policy, the
experience of applying inflation targeting monetary policy of countries around the world, the
conditions for applying the regulatory mechanism to implement inflation targeting monetary
policy in Vietnam such as studies by To Kim Ngoc (2012), To Kim Ngoc and Nguyen Khuong
Duy (2012), Ly Ba Hong Son and Nguyen Van Truong (2012), Nguyen Xuan Hung ( 2008),
Nguyen Xuan Hung (2009), Nguyen Phuc Canh (2013), Nguyen Thanh Nhan (2010), Do Thi
Duc Minh (2003), Nguyen Ngoc Thach (2013), To Kim Ngoc and Le Tuan Nghia (2012) ...

14

CHAPTER 3. MODEL AND RESEARCH METHOD
3.1. Research model
In a closed economy, monetary policy decisions affect inflation through interest rates.
Increasing interest rates will make households and businesses consume and invest less. Rising
interest rates also cause banks to tighten credit, making individuals and businesses spend less.
When the economy spends less, the demand for production is less and there is less pressure on
prices (due to falling demand).
In an open economy, exchange rates is another channel through which monetary policy
decisions affect inflation and inflation expectations. While other factors remain constant, the
central bank decides to tighten monetary policy through an increase in policy rates, this may
increase the value of the local currency. Specifically, the increase in interest rates makes
attracting more inflows of investment capital from other markets into the domestic market.
Foreign investors transfer capital to the domestic market, so there will be more demand for
local currency. The increase in demand for the local currency made the local currency
appreciate. Assuming all other factors remain constant, the domestic currency will appreciate
until the difference between domestic and foreign interest rates is equal to the appreciation of
the local currency. The relationship between interest rates and exchange rates in this case is
expressed by the theory of interest rate parity as follows:
rt - rft = et - E(et + 1) or rt = rft + et - E(et + 1)
Inside:
rt is the domestic interest rate
rft is the foreign interest rate
et is the present value of 1 unit of local currency in foreign currency
E(et + 1) is the expectation that the value of 1 unit of local currency is calculated in
foreign currency
Any change related to appreciation or depreciation of the local currency not only has a
significant impact on the import price of finished goods, but also changes in the price of the
first imported goods to serve domestic production activities, indirectly increasing the cost of
domestic production and business activities. When the local currency is devalued, it will cause
higher import prices, moreover, it will also increase marginal costs and lead to an increase in

the price of domestically produced goods. This is called the direct effect of exchange rates on
prices. In addition, when the domestic currency depreciates, the price of exported goods will
be cheaper than the foreign market and more expensive than domestic goods. Therefore, the
increasing demand (for foreign markets) and production (for the domestic market) increase the
15


production pressure on this item, causing the price of labor and the related domestic materials
to produce this item increased. This is the indirect effect of exchange rates on domestic prices.
Central banks that conduct inflation targeting monetary policy usually choose the
interest rate tool as the main instrument (as analyzed in chapter 2). The operation of interest
rate tools will be difficult if the central bank wants to stabilize exchange rates and control
inflation. Without pressure of exchange rate stabilization, the management of interest rate tools
towards stabilizing inflation will be more effective. If inflation is at a high risk, the central bank
will raise interest rates. From the formula of the theory of interest rate parity, the value of the
domestic currency will decrease or the exchange rate will increase. If the central bank wants to
keep the exchange rate stable, that is, do not allow the exchange rate to rise, then raising interest
rates will not help the central bank to achieve this goal. Therefore, the central bank can only
choose one of two goals: stabilize the exchange rate or stabilize inflation.
Figure 3.1. Transmission mechanism of interest rate tool
Market rate

Policy rate

Aggregate
demand

Exchange rate

Output gap

Inflation

Exchange rate passthrough into inflation

Based on the above analysis, in order to evaluate the effectiveness of the interest rate
instrument in operating towards controlling interest rates, and at the same time answering the
two research questions in chapter 1, the author uses the vector regression model that includes
the following variables: interest rate, exchange rate, aggregate demand and inflation. In
addition, due to the indirect effect of exchange rates on inflation through domestic and foreign
aggregate demand, the model separates the effects of domestic and foreign aggregate demand
on inflation. Therefore, the research model can assess the level of transmission of the exchange
rate pass-through. The lower the level of transmission of the exchange rate into inflation, the
easier it is for the central bank to conduct policy towards controlling inflation (Choudhri,
Faruqee & Hakura, 2005).

16


3.2. Research variables
The research model is indeed a model that reflects the transmission mechanism of
monetary policy through the interest rate tool. This model has also been applied to research in
Tunisia country (Boughrara, 2007). The research model consists of 5 variables that need to be
collected as follows:
• Interest rate variable: There are many types of interest rates in an economy,
however, the interest rate chosen in this model is the policy interest rate chosen and decided by
the central bank as the basis for other types of interest rates in the market. For Vietnam, there
are 2 types of interest rates decided by the State Bank of Vietnam : refinancing rate and discount
rate. Then, the author will run 2 models with 2 different interest rates. Model 1 uses refinancing
rate and model 2 uses discount rate.

• Exchange rate variable: The nominal exchange rate will be chosen to represent the
exchange rate variable in the model.
• Inflation variable: The inflation index used in the study is the Headline Inflation.
• Domestic aggregate demand variable: The index of Vietnam's industrial production
is chosen to represent the aggregate domestic demand.
• Foreign aggregate demand variable: The study uses additional variables of US
industrial production index to represent total foreign demand.
3.3. Research hypotheses
The research hypothesis is based on predicting the answers to the two research
questions set out in chapter 1 as follows:
• The first research hypothesis: In Vietnam, there exists a relationship between
interest rate tools and inflation.
• The second research hypothesis: The State Bank of Vietnam use interest rate
instrument only towards controlling inflation.
If the research results from the model are consistent with the two hypotheses above,
Vietnam satisfies the conditions for applying inflation targeting monetary policy. In the
opposite case, Vietnam has not yet satisfied the conditions to apply inflation targeting monetary
policy.
3.4. Research data
Data on research variables are collected on a monthly basis, from January 2011 to June
2018. The research data in the thesis is extracted from the IFS data source of the International
Monetary Fund. , the data source of the State Bank of Vietnam and the ARIC data source of
Asian Development Bank. Below is a summary of the variables used in the thesis.
17


Table 3.1. Summary of research variables

Variables

Symbol

Source

Refinance rate

R1

SBV

Discount rate

R2

SBV

Nominal exchange rate

EX

IFS

INFLATION

ARIC

Vietnam Industrial production index

IPVN

IFS

United States Industrial production index

IPUS

IFS

Inflation

3.4. Data analysis method
Because data series is integrated of order 1, cointegration phenomenon could occur.
Therefore, test of cointegration is performed by Johansen method (on Stata 14 software).
The above procedure is performed as follows:
(1) Stationary test of data series,
(2) Determination of optimal lag and cointegration test,
(3) Vector error correction model (VECM),
((4) Test the fit and stability of the model,
(5) Shock analysis,
(6) Forecast error variance decomposition.

18


CHAPTER 4. RESEARCH RESULTS
4.1. Research results from the first model
4.1.1. Statistical results
The statistical results of the variables used in the model were calculated from Stata
software as follows:
Table 4.1. Statistics of variables in the first model

Variables

Mean

Standard

Minimum Maximum

Number of

deviation

Value

value

observations

R1

8.2

2.84

6.25

15

90

EX

21618.81

812.5778

19497.81

22845.52

90

IPVN

106.9688

31.79286

50.53

195.56

90

IPUS

108.6936

3.048351

101.47

116.23

90

6.2789

5.9055

-0.00221

23.0165

90

INFLATION

(Source: Results from Stata software)
4.1.2. Test the stationary of the data series
To check the stationary of the data series, the author performs Dickey-Fuller test and
results as follows:
Table 4.2. Dickey-Fuller test for the first model

Variables

t- statistics
I(0)

I(1)

R1

-0.565

-6.710***

EX

-2.337

-9.026***

IPVN

-2.007

-9.549***

IPUS

-0.688

-7.452***

INFLATION

-1.053

-3.559***

1% critical value

5% critical value

10% critical value

-3.525

-2.899

-2.584
(Source: Results from Stata software)

For the collected data series, the Dickey Fuller test results indicate that the data series
are not stationary at level 0 (because the statistical value t is 10% greater than the significance

19


level). However, the first-order the data series are stationary, or data series are integrated at the
first order, I (1), within 99% confidence interval (because the statistical value t is less than 1%
significance level).
4.1.3. Determine optimal latency and cointegration test
4.1.3.1. Determine the optimal latency of the data series
The result of the optimal latency selection for the model is done on Stata 14 software
as follows:
Table 4.3. Optimal latency of the first model

Lag

LL

LR

AIC

HQIC

SBIC

37.8123

37.8697

37.955

0

-1620.93

1

-1014.86

1212.1

24.299

24.6436*

25.1552*

2

-978.434

72.848

24.0333*

24.605

25.603

3

-957.245

42.378

24.122

25.0408

26.4051

4

-933.044

46.801

24.1592

25.3652

27.1558

(Source: Results from Stata software)
If the optimal latency is selected based on the AIC criteria, the optimal latency of the
model is 1. In other words, the current value of the current variable is affected by the lag
variables of the previous month.
4.1.3.2. Test of cointegration
After determining the optimal latency, the author performed cointegration test by
Johansen method. The results from Stata 14 software are as follows:
Table 4.4. Johansen test results for the first model

Ho: rank £ r

Eigenvalues

Trace statistic

Critical value

r=0

.

97.1386

68.52

r=1

0.42335

48.1421

47.21

r=2

0.24316

23.3470*

29.68

r=3

0.19415

4.1355

15.41

r=4

0.04480

0.0562

3.76

(Source: Results from Stata software)
According the table above, we start with the hypothesis H0: r = 0 (line 1) is rejected (by
trace statistic> p_value or 97.1386> 68.52) at the 5% significance level. Therefore, the test
continues with the hypothesis H0: r = 1. This hypothesis is further rejected (by trace statistic>

20


p_value or 48.1421> 47.21) at the 5% significance level. Therefore, the test continues with the
hypothesis H0: r = 2. This hypothesis is accepted (by trace statistic at the 5% significance level. This indicates that there are two cointegration relationships
between the variables INFLATION, R1, EX, IPVN and IPUS.
4.1.4. Testing the stability of the model
Firstly, the research conducted tests of AR characteristics to ensure the stability of the
model. The result of testing of AR characteristic test shows that all solutions of the
characteristic polynomial are in the unit circle, reflecting the VECM model to ensure stability
and sustainability.
Figure 4.1. Results of testing polynomial characteristic AR

-1

-.5

Imaginary
0

.5

1

Roots of the companion matrix

-1

-.5

0
Real

.5

1

The VECM specification imposes 3 unit moduli

(Source: Results from Stata software)
4.1.5. Results of shock analysis and variance decomposition
To further examine the relationship between inflation and monetary policy instruments,
the author continues to analyze shocks through the impulse-response function with the option
of a cycle length of 24 months (Figure 4.1). The surprising result is that inflation seems to be
completely unresponsive to shocks from refinancing rates (R1). Meanwhile, the exchange rate
reacted strongly to interest rate shocks. Thus, the adjustment of refinancing interest rate may
not affect inflation much but affect exchange rate.

21


Figure 4.1. Analysis results of refinancing interest rate shock (R1)

(Source: Results from Stata software)
The variance decomposition results also show that the refinancing interest rate tool
contributes very little to the explanation of changes in inflation. In the 12th period (the end of
the first year since the interest rate shock), inflation changes were only explained by 2.6% due
to the change of refinancing rate. In the 24th period (at the end of the second year since the
interest rate shock), the level of explanation for the change in inflation from the change of the
refinancing rate was reduced (only 1.6%). This is evidence that the relationship between the
refinancing interest rate tool and inflation in Vietnam is very weak.
In addition, a striking result from the variance result, the exchange rate also contributes
very little to the change of inflation. Specifically, the level of explanation of the exchange rate
is 2.1% (period 12) and 3.3% (period 24) respectively. This is a good result because it shows
that the level of transmission of the exchange rate into inflation is very small. This is the
evidence that the State Bank of Vietnam can focus on improving the efficiency of operating
the interest rate tool towards controlling inflation without fear of the impact of exchange rates.
Moreover, when the transmission rate of the exchange rate into inflation is low, the State Bank

22


of Vietnam may continue to widen the fluctuation band of exchange rates to affirm that the
State Bank of Vietnam is prioritizing price stability target.
Table 4.7. Results of variance decomposition

(Source: Results from Stata software)

4.2. Research results from the second model
4.2.1. Statistical results
The statistical results of the variables used in the model were calculated from Stata
software as follows:

23


Table 4.8. Statistics of variables in the second model

Variables

Mean

R2

6.2666

Standard Minimum Maximum

Number of

deviation

value

value

observatios

3.004398

4.25

13

90

EX

21618.81 812.5778

19497.81

22845.52

90

IPVN

106.9688 31.79286

50.53

195.56

90

IPUS

108.6936 3.048351

101.47

116.23

90

-0.00221

23.0165

90

INFLATION

6.2789

5.9055

(Source: Results from Stata software)
4.2.2. Test the stationary of the data series
To check the stationary of the data series, the author performed Dickey-Fuller test and
results as follows:
Table 4.9. Test the data series stop properties of the second model

Variables

T- statistics
I(0)

I(1)

R2

-0.868

-8.267 ***

EX

-2.337

-9.026***

IPVN

-2.007

-9.549***

IPUS

-0.688

-7.452***

INFLATION

-1.053

-3.559***

1% critical value

5% critical value

10% critical value

-3.525

-2.899

-2.584

Note: ***: Significance level 1%; **: Significance level 5%; *: 10% significance level
(Source: Results from Stata software)
For the collected data series, the Dickey Fuller test results indicate that the data series
are not stationary at level 0 (because the statistical value t is 10% greater than the significance
level). However, the first-order the data series are stationary at level 1, or they are integrated
at the first order, I (1), within 99% confidence interval (because the statistical value t is less
than 1% significance level).

24


If the data series R2, EX, IPVN, IPUS and INFLATION are not stationary at level 0, it
can lead to spurious regression phenomenon. However, the first-order data series are stationary
at 1% significance level, so it meets the requirement of cointegration test.
4.2.3. Determine optimal latency and cointegration test

4.2.3.1. Determine optimal latency of the data series
The result of calculating the optimal latency selection for the model is done on Stata 14
software as follows:
Table 4.10. Determine the second latency optimal model

Lag

LL

LR

AIC

HQIC

SBIC

37.716

37.7735

37.8587

0

-1616.79

1

-1009.73

1214.1

24.1799

24.5244

25.036*

2

-971.771

75.927

23.8784*

24.5101*

25.448

3

-947.326

48.89

23.8913

24.8102

26.1744

4

-922.862

48.929

23.9038

25.1097

26.9003

(Source: Results from Stata software)
If the optimal latency is selected based on the AIC criteria, the optimal latency of the
model is 2. In other words, the current value of the current variable is affected by the lag
variables of the previous 2 months.
4.2.3.2. Cointegration test
After determining the optimal delay, the author conducted co-linking according to
Johansen method. Calculation results from Stata 14 software are as follows:
Table 4.11. Johansen test results for the second model

Ho: rank £ r

Eigenvalues

Trace statistic

Critical value

r=0

.

63.3987*

68.52

r=1

0.42335

31.3460

47.21

r=2

0.24316

13.0613

29.68

r=3

0.19415

2.5351

15.41

r=4

0.04480

0.0387

3.76

(Source: Results from Stata software)
Based on the results of the table above, we start with the hypothesis H0: r = 0 (line 1)
accepted (by trace statistic indicates that there is no co-association between the variables INFLATION, R2, EX, IPVN

25