Phát triển tài chính và hiệu lực của chính sách tiền tệ ttta

MINISTRY OF EDUCATION AND TRAINING
UNIVERSITY OF ECONOMICS HO CHI MINH CITY
--------------

HO THI LAM

FINANCIAL DEVELOPMENT AND THE
EFFECTIVENESS OF MONETARY POLICY

Major: Finance – Banking
Major code: 93 40201

SUMMARY OF PHD THESIS

HO CHI MINH CITY - 2019


The thesis is completed at:
University of Economics Ho Chi Minh City
Supervisor: Prof. Dr. Tran Ngoc Tho
Contradicteur 1: …………………………………………...
…………………………………………………………......
Contradicteur 2: …………………………………………...
……………………………………………………………..
Contradicteur 3: …………………………………………...
……………………………………………………………..
The thesis will be defensed in front of the Academic
Council convened by ………………………………............
……………………………………………………………..
At … … … … …

This thesis can be found at the library: ……………………
…………………………………………………………......
……………………………………………………………..
…………………………………………………………….


1

CHAPTER 1 – INTRODUCTION
1.1.

Introduction

While the relationship between financial development, capital
accumulation and economic growth is clear and gets concerned
by numerous researches, the impact of monetary policy on a real
economy with the role of financial system seems limited.
Financial globalization and integration of financial markets in
different countries have increased the complication in the
environment where national monetary agencies operate. The
appearance of new types of currencies, payment technologies or
substitute financial assets makes it more difficult to determine
money aggregates. Central Banks’ monetary control also
becomes hard to solve. Financial development has changed both
monetary supply and demand, causing policy makers to
encounter obstacles with respect to ability to control capital
flows, manage liquidity, maintain sustainability of exchange
rates and avoid booming cycles in asset markets. From the
perspective of monetary policy, it is worth studying a question of
how such financial development affects the way that Central

Banks implement policies and that such policies are transmitted
to the economies. However, there is little evidence of the effects
that the financial development puts on the effectiveness of the
monetary policy in the current period of time.
Although there are some authors doing research on the same
topic, they still encounter some pending issues. First, the
measurement of the Monetary Policy Effectiveness has not
covered all the objectives of the monetary policy of Central


2

Banks (Ma & Lin, 2016, Carranza, Galdon-sanchez & Gomezbiscarri, 2010). Second, they have not paid attention to factors
affecting the monetary policy effectiveness (Cecchetti, FloresLagunes & Krause, 2006; Olson & Enders, 2012; Olson, Enders
& Wohar, 2012). Third, they have only considered role of some
certain aspects of the concept of the financial development on the
effectiveness of monetary policy (Akhtar, 1983; De Bondt, 1999;
Cecchetti & Krause, 2001, 2002; Georgiadis & Mehl, 2016;
Bernoth, Gebauer & Schäfer, 2017) without considering the
whole role of the financial development.
Starting from both academic and practical perspectives, it is
necessary to do research on the monetary policy effectiveness
and the impact of the financial development on the monetary
policy effectiveness in order to overcome limitations of previous
researches, aiming at producing scientific understandings on the
management of the monetary policy in new context.
1.2.

Objectives

Evaluate impacts of the financial development on the monetary
policy effectiveness.
Specific objectives are: Firstly, testing the Taylor curve theory;
secondly, developing and establishing the efficient frontier of the
monetary policy for countries on the basis of the Taylor curve
theory and measures the monetary policy effectiveness; thirdly,
studying the effects of the financial development (with respect to
various aspects) on the effectiveness of the monetary policy in
the countries of the research sample.
1.3.

Research questions

1.4.

Subject and Scope of study


3

This research studiesthe impact of the financial development on
the monetary policy effectiveness in G-7 developed countries
(including Canada, France, Germany, Italy, Japan, United
Kingdom, and United State) from 1951 to 2017 (depending on
the data availability).
1.5.

Methodologies and data

1.6.

Contributions

1.7.

Thesis’s structure

CHAPTER 2 – THEORETICAL FRAMEWORK AND
LITERATURE
DEVELOPMENT

REVIEW
AND

OF
MONETARY

FINANCIAL
POLICY

EFFECTIVENESS
2.1.

Monetary Policy

2.2.

Monetary Policy Effectiveness and Taylor Curve

Theory

2.2.1.

Taylor Curve Theory

The Taylor curve theory describes permanent trade-off between
output volatility and inflation volatility in implementing the
monetary policy (Taylor, 1979), and it is an important guiding
principle in many studies on the monetary policy (see more,
Taylor, 1994; Fuhrer, 1997; Orphanides et al., 1997; Chatterjee,
2002; Taylor & Williams, 2011; Olson, Enders & Wohar, 2012).
The Taylor curve is considered as the efficient frontier of the
monetary policy, showing the position on which the monetary
policy is optimal towards the inflation volatility and the output is
at lowest level corresponding to central bank’s taste with regards
to whether price stability or business cycle stability is prioritized
(Taylor, 1979; Friedman, 2010).


4

Basic principle of the Taylor curve is based on Central Bank’s
optimizing behavior in implementing monetary policy to
minimize loss to the economy against unexpected effects of
shocks. Loss function measuring weighted total costs of inflation
volatility and output volatility against their target levels:
ℒ = 𝐸[𝜆(𝜋 − 𝜋 ∗ )2 + (1 − 𝜆)(𝑦 − 𝑦 ∗ )2 ]

(2.2)

Where 𝜆 is the central bank’s preference for inflation stability

(0 ≤ 𝜆 ≤ 1).
So as to minimize loss, the central bank has to determine factors
which determine the deviation between real output and inflation
against their targets. A simple economy is affected by two types
of shocks – aggregate demand shock (d) and aggregate supply
shock (s) and these two types require policy response. The
aggregate demand shock causes output and inflation change in
the same direction and aggregate supply shock causes in opposite
direction. Because monetary policy can affect output and
inflation in the same direction, it can completely offset the effect
of the aggregate demand shock. In contrast, the aggregate supply
shock will require monetary agencies to encounter a trade-off
between output volatility and inflation volatility (Taylor, 1979,
Output volatility (σy)

Clarida, Galí & Gertler, 1999; Cecchetti & Krause, 2001).

Efficient Frontier
Inflation volatility (σπ)

Fig. 2.1. The Taylor curve


5

The trade-off relationship is modeled by a downward sloping
curve, convex to the origin on a two-dimensional graph (output
volatility – inflation volatility). The efficient frontier of monetary
policy traces points at which the inflation volatility cannot be
achieved at lowest level without causing any increase in the

output volatility (Taylor, 1979; Cecchetti, FloresLagunes &
Krause, 2006). This efficient frontier is called Taylor curve
(Taylor, 1994; King, 1999; Bernanke, 2004; Friedman, 2010;
Olson & Enders, 2012). Figure 2.1 illustrates the efficient frontier
of monetary policy according to Taylor (1979). If the monetary
policy is optimal, the economy will be on this curve. When the
policy is below optimal level, the economy will not be on this
curve. Instead, the efficient point will be above on the right side
with the fact that the inflation volatility goes beyond other
feasible points. Movements of the efficient points to the frontier
signal that the policy effectiveness is improved.
2.2.2.

Monetary Policy Effectiveness

The monetary policy effectiveness is the capability of central
banks to achieve their objectives, stabilize impacts of the shocks
on the economy and reduce macroeconomic volatility by
implementing monetary policy with available instruments
(Boivin & Giannoni, 2006, Cecchetti & Krause, 2001, Cecchetti
et al., 2006, Mishkin & Schmidt-Hebbel, 2007, Taylor, 1979).
2.2.3.

Factors affecting the effectiveness of monetary policy

2.3.

Financial Development

2.3.1.

Role of financial system in monetary transmission

mechanism
2.3.2.

Financial Development


6

Financial development includes improvement in the functions of
financial system such as (i) aggregating savings; (ii)
appropriating capital for production; (iii) monitoring such
investments; (iv) diversifying risks; and (v) exchanging goods
and services (Levine, 2005).
Financial development involves in competitive and efficient
issues in the financial sector; scope of services provided; variety
of financial institutions in the financial sector; volume of credit
granted by financial intermediaries along with the right to access
financial services and financial stability (Svirydzenka, 2016).
2.3.3.

Measuring Financial Development

The best way to measure financial development is to measure the
degree to which five functions of the financial system are
improved. However, it is a great challenge to have a direct
measurement for the functions. Levine (2005) pointed out that
such empirical variables do not normally measure exactly

concepts coming from theoretical models.
Many studies on the financial development towards economic
growth; inequality or poverty and economic stability have
developed different measures for the financial development
(Svirydzenka, 2016). Some studies focus on the development in
banking sector, some others focus on the development in stock
market, whereas the others compile them in an indicator.
2.4.

Theory of the impacts of the financial development

on the monetary policy effectiveness
Theory of the impacts of the financial development on the
monetary policy effectiveness has been developed very early
with different aspects on money supply, money demand and


7

monetary transmission mechanism (see Gurley & Shaw, 1955,
1967; Vanhoose, 1985; Lown, 1987; Taylor, 1987; Modigliani &
Papademos, 1989; Hendry & Ericsson, 1991; Arestis,
Hadjimatheou & Zis, 1992, etc.).
2.4.1.

Impacts of the financial development on money supply

control
The financial development makes it more difficult to determine
money aggregates. Characteristics of the financial development

such as the continuous appearance of new instruments and
products make the determination and differentiation of
“moneyness” and “liquidity” of the instruments becomes
inaccurate, and thus, it is difficult to define which instrument to
be included in money aggregates. Simultaneously, the
appearance of non-cash payment methods and electronic money
types put impact on the velocity of money circulation, making it
difficult for central banks to control money supply (Akhtar, 1983;
Singh et al., 2008).
2.4.2.

Impacts of the financial development on money

demand
The appearance of financial instruments with floating interest
rate, along with the development of non-cash payment methods,
etc. in a developed financial system cause the elasticity of money
demand to interest rate to change, long-run money demand to
reduce whereas it is hard to predict short-run change.
Accordingly, money demand function is not stable over time,
thus, the size of money demand becomes difficult to be
estimated. Regarding IS-LM procedure, the LM curve becomes
more sloping and more difficult to position (Akhtar, 1983).


8

2.4.3.

Impacts of Financial Development on monetary

transmission mechanism

The monetary transmission mechanism can be modeled into 2
phases as can be seen from Figure 2.3.

Figure 2.3. Factors impacting monetary policy transmission mechanism

Source: Loayza & Schmidt-Hebbel (2002)
First phase: any change in monetary policy will be transmitted
to change in market interest rate and price of other financs output, 𝜋 is inflation, i is interest rate and
oil is oil price, with removal of trends. Ht is a matrix of variance
– covariance, 𝜈𝑡 is a white noise process.
Based on Lee (1999, 2002, 2004); Olson et al. (2012), the
assumption of the trade-off relationship between output volatility
and inflation volatility is tested with the bivariate GARCH–
BEKK model, as discussed by Engle & Kroner (1995). In
particular, I keep track with dynamic behaviors of inflation
volatility and output volatility which are unable to be directly
observed with the estimation of conditional variance of inflation
and output in accordance with the structure model of (3.1), (3.2).
The relationship between ℎ11,𝑡 and ℎ22,𝑡 is center of the Taylor
curve relationship. The assumption of the trade-off relationship
between output volatility and inflation volatility is tested by
testing signs of elements of 𝑏12 and 𝑏21 of B matrix. If these
elements have negative signs, the Taylor curve theory is
determined. The lag of independent variables in the mean
equation is selected on the basis of testing F with the estimation

(3.1)

(3.2)
(3.6)


12

of (3.1) and (3.2) by seemingly unrelated regression (SURs)
beginning with six lags of each variable.
3.2.2.

Data

I use monthly data of 7 countries of the sample for a period from
1951 to 2017 subject to the data availability of each country. The
data are collected from the IMF (International Financial
Statistics) and Federal Reserve Bank of St. Louis’s Database.
The data include Industrial Production Index (IIP), representing
economy output; inflation rate (which means percentage change
in consumer price index in comparison with the same period of
the previous year); short term nominal interest rate is money
market interest rate, representing the

monetary policy

perspective; world oil price which means WTI crude-oil price
with the unit price of USD/barrel, representing supply shock.
3.3.

Results of Taylor curve relationship test

3.3.1.

Result of stationarity test

The result of stationarity test shows that only interest rate
variable integrates at level (1), the other variables integrate at
level (0).
3.3.2.

Taylor curve relationship

Table 3.3 reports the estimation result of the near-VARGARCH-BEKK model. The result shows that all estimates of 𝑏12
and 𝑏21 are negative and most of them are statistically significant
at 5% and 1% in most of the study countries. This indicates that
there exists a signal of the trade-off relationship according to the
Taylor curve theory. However, the fact that how large the tradeoff relationship between output volatility and inflation volatility
varies among countries. This result matches with findings of the


13

trade-off relationship between output volatility and inflation
volatility found in previous empirical studies (Lee 2002, 2004;
Cecchetti & Ehrmann, 2002; Arestis & Mouratidis, 2004;
Cecchetti et al., 2006). Result of Ljung-Box statistics on the
residual and residual square cannot reject standard distribution,
showing the reliability of estimate model and study results.
3.4.

Conclusion

CHAPTER 4 – MEASURING MONETARY POLICY
EFFECTIVENESS
4.1.

Introduction

In this chapter of this thesis, I aim at developing the Taylor curve
for each country in the study period. At the same time, the Taylor
curve is used to evaluate the deviation of the monetary policy in
practice from the optimal monetary policy and to measure the
monetary policy effectiveness with respect to the improvement
of macroeconomic performance.
4.2.

Study methodology

4.2.1.

Model of efficient frontier estimation and monetary

policy effectiveness measurement
4.2.1.1. Structural model
Based on Mishkin & Schmidt-Hebbel (2007), Cecchetti et al.
(2006) and Rudebusch & Svensson (1999), I consider constraint
function for loss function including two linear equations based
on the dynamic aggregate demand and aggregate supply model
as follows:

14

Table 3.3. The estimated GARCH model
C Matrix
Canada
France
Germany
Italy
Japan
UK
US

A Matrix

B Matrix

c11

c21

c22

a11

a12

a21

a22

b11

b12

b21

b22

0.08***

-0.01

0.07***

0.35***

0.01**

0.06***

0.18***

0.93***

-0.004**

-0.01**

0.97***

(8.06)

(-0.75)

(27.51)

(39.82)

(2.39)

(2.66)

(26.61)

(373.77)

( -2.379)

(-2.266)

(815.59)

0.11***

0.01

0.00

-0.07***

-0.003

0.14*

0.24***

0.99***

-0.01***

0.20***

0.97***

(3.21)

(1.35)

(-3.2e-4)

(-2.97)

(-0.84)

(1.35)

(8.32)

(269.13)

(-15.96)

(9.28)

(143.75)

0.63***

0.01***

0.00

0.53***

-0.02***

0.05

0.22***

0.58***

-0.13***

-0.26***

-0.92***

(56.32)

(4.17)

(3.4e-10)

(21.05)

(-3.02)

(0.46)

(19.90)

(43.17)

(-32.04)

(-8.53)

(-368.9)

0.94***

0.02**

0.00

0.46***

-0.001

0.60**

0.37***

0.58***

-0.02***

-0.23*

0.92***

(7.31)

(2.24)

(-2.4e-5)

(7.58)

(-0.12)

(2.42)

(10.46)

(4.68)

(-3.39)

(-1.72)

(64.00)

0.03**

-0.03***

0.0006

0.49***

-0.014**

-0.02

0.25***

0.90***

-0.009

0.004

-0.97***

(2.14)

(-2.85)

(0.03)

(13.09)

(-2.40)

(-0.76)

(7.95)

(68.97)

(-0.56)

(0.11)

(-128.3)

0.7***

-0.003

0.00

0.64***

-0.04***

0.16**

0.25***

-0.24***

-0.03***

-0.12***

0.96***

(38.87)

(-0.54)

(-2.4e-9)

(26.21)

(-13.81)

(2.34)

(46.28)

(-5.02)

(-8.27)

(-3.03)

(786.52)

0.14***

-0.04***

0.00

0.45***

-0.09***

-0.23**

0.27***

0.75***

-0.12

-0.41**

-0.93***

(12.39)

(-3.83)

(-7.9e-6)

(13.96)

(-4.5)

(-2.08)

(10.11)

(12.96)

(-1.41)

(-2.38)

(-27.11)


15
𝑛

𝑦𝑡 = ∑ 𝛼1,𝑗 𝑦𝑡−𝑗
𝑗=1

𝑛

+∑
𝑛 𝛼1,(𝑗+𝑛) 𝜋𝑡−𝑗
+ 𝑗=1
∑ 𝛼1,(𝑗+2𝑛) 𝑖𝑡−𝑗 + 𝛼1,(3𝑛+1) 𝑜𝑖𝑙𝑡−1

𝑛

+ 𝜀𝑗=1
1,𝑡

𝜋𝑡 = ∑ 𝛼2,𝑗 𝑦𝑡−𝑗
𝑗=1

(4.1)

𝑛

+ ∑ 𝛼2,(𝑗+𝑛) 𝜋𝑡−𝑗 + 𝛼2,(2𝑛+1) 𝑜𝑖𝑙𝑡−1
+

(4.2)

𝑗=1
𝜀2,𝑡

The optimal lag is selected in accordance with Schwarz
information criterion (SBC).
4.2.1.2. Developing the Taylor curve
After estimating the structural model for each country, I use
estimate parameters to develop the efficient frontier. As
described in Chapter 2, I develop the efficient frontier by
minimizing the loss function (2.2) in accordance with constraints
imposed by the economy’s dynamic structure.
Equation (4.1) and (4.2) is represented in the form of state space
as follows:

𝐘𝑡 = 𝐁𝐘𝑡−1 + 𝐜𝑖𝑡−1 + 𝐃𝑋𝑡−1 + 𝒗𝑡

(4.3)

In the form of matrix, loss function (2.2) is:
𝐘𝑡 ′ Λ𝐘𝑡

(4.4)

Accordingly, 𝐁 and 𝐃 are matrices of estimate parameters for
aggregate supply and aggregate demand models. Λ is the
weighted matrix attributed to inflation and output volatility.
The problem of policy makers is to choose an interest rate to
minimize (4.4), subject to constraints imposed by (4.3). Given
the quadratic nature of the loss function, the solution for the
interest rate will be linear (Cecchetti et al., 2006; Mishkin &
Schmidt-Hebbel, 2007), which is written as:


16

𝑖𝑡 = 𝐠𝐘𝑡 + Ψ

(4.5)

In particular 𝐠 is a vector of central bank’s response factors to the
change in inflation and output and Ψ is constant term subject to
𝐁, 𝐜, 𝐃 and target value of inflation and output. The equation
(4.5) represents an unrestricted monetary policy rule, in which
the degree of interest rate persistence can be observed by 𝑖𝑡−1

which is a component of 𝐘𝑡 (Rudebusch & Svensson, 1999;
Cecchetti et al., 2006).
The control problem is solved by finding 𝐠 such that:
𝐠 = −(𝐜 ′ 𝐇𝐜)−𝟏 𝐜 ′ 𝐇𝐁

(4.6)

In which, 𝐇 is the solution to the equation:
𝐇 = 𝚲 + (𝐁 + 𝐜𝐠)′ 𝐇(𝐁 + 𝐜𝐠)

(4.7)

With estimate values of parameters in 𝐁 and 𝐜, we can come to
the solution for 𝐇 and 𝐠 for any value of 𝜆.
From the results of the equations above, the optimal variance
value of output and inflation can be calculated one point on the
Taylor curve, which corresponds to one value of given 𝜆, can be
obtained. By repeating the procedure using different values of 𝜆
in the scale of [0; 1], the whole Taylor curve can be found.
With this estimated efficient frontier, I calculate minimum
orthogonal distance between observed fluctuations and their
optimal values on the Taylor curve to represent the monetary
policy effectiveness.
4.2.2.

Data

The data are used as described in Chapter 3.
4.3.

Results and Discussion

4.3.1.

Taylor curve estimated

4.3.2.

Monetary policy effectiveness over time


17

Figure 4.2 demonstrates the effectiveness of monetary policy
changes over time in countries. The monetary policy effectiveness
is measured by the orthogonal distance from the point of inflation
fluctuation and observed actual output to the Taylor curve of each
country respectively. Accordingly, the greater the orthogonal
distance is, the less effective monetary policy is because the
economy is farther from the point of optimal efficiency.
Minimum Distance of Observed Volatilities from Taylor Curve of Canada

Minimum Distance of Observed Volatilities from Taylor Curve of US
0.8

0.7

0.7

0.6

0.6

0.5

0.5
0.4

0.4
0.3

0.3
0.2

0.2
0.1

0.1

0.0

0.0

1965 1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 2013 2016

1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 2015

Minimum Distance of Observed Volatilities from Taylor Curve of France

Minimum Distance of Observed Volatilities from Taylor Curve of Germany

0.6

1.4

0.5

1.2
1.0

0.4
0.8

0.3
0.6

0.2
0.4

0.1

0.2
0.0

0.0

1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 2013 2016

1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 2007 2010 2013 2016

Minimum Distance of Observed Volatilities from Taylor Curve of Italy

Minimum Distance of Observed Volatilities from Taylor Curve of Japan

1.25

2.5

1.00

2.0

0.75

1.5

0.50

1.0

0.25

0.5

0.00

0.0

1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 2015

1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009 2012 2015

Minimum Distance of Observed Volatilities from Taylor Curve of UK
0.6
0.5
0.4
0.3
0.2
0.1
0.0
1967

1970

1973

1976

1979

1982

1985

1988

1991

1994

1997

2000

2003

2006

2009

2012

2015

Figure 4.2. Monetary policy effectiveness over time


18

In all countries, the effectiveness of monetary policy drastically
decreased during global crisis period in 2008. Although during
the period after 2012, the distance to the efficient frontier was
maintained at stable level, the role of monetary policy in
stabilizing output and price still much weakened in comparison
with the period before the crisis in most of the study countries.
The monetary policy effectiveness also changed in accordance
with management practice in the countries with the reduction in
periods impacted by international and domestic shocks.
4.3.3.

Shifts in the Taylor curve

4.4.

Conclusion

CHAPTER

5

–

DEVELOPMENT

IMPACTS
ON

OF

FINANCIAL

MONETARY

POLICY

EFFECTIVENESS
5.1.

Introduction

In this chapter of the thesis, I assess impacts of the financial
development on the effectiveness of monetary policy in G-7
countries. On the basis of the monetary policy effectiveness’s
measurement implemented in chapter 4, I apply the measure to
the degree of financial development established by Svirydzenka
(2016) based on Sahay et al. (2015), then, regression analysis is
conducted to test the relationship between the degree of financial
development and monetary policy effectiveness in addition to
some macroeconomic variables added to the model.
5.2.

Methodology

5.2.1.

Stationarity test

5.2.2.

Cointegration test

5.2.3.

Study model


19

The study model is developed on the background of some
previous studies including Carranza, Galdon-sanchez & Gomezbiscarri (2010); Ma & Lin (2016), as follows:

𝑀𝑃𝐸𝑖𝑡 = 𝛼𝐹𝐷𝑖𝑡 + 𝜷′ 𝒛𝒊𝒕 + 𝛾𝐶𝑅𝐼𝑆𝐼𝑆𝑖𝑡 + 𝜃𝐼𝑇𝑖𝑡 + 𝜀𝑖𝑡
𝑀𝑃𝐸𝑖𝑡 = 𝛼1 𝐹𝐼𝑖𝑡 + 𝛼2 𝐹𝑀𝑖𝑡 +

𝜷′𝟏 𝒛𝒊𝒕

(5.1)

+ 𝛾1 𝐶𝑅𝐼𝑆𝐼𝑆𝑖𝑡 + 𝜃1 𝐼𝑇𝑖𝑡 + 𝜀𝑖𝑡 (5.2)

𝑀𝑃𝐸𝑖𝑡 = 𝛼11 𝐹𝐼𝐴𝑖𝑡 + 𝛼12 𝐹𝐼𝐷𝑖𝑡 + 𝛼13 𝐹𝐼𝐸𝑖𝑡 + 𝛼21 𝐹𝑀𝐴𝑖𝑡 +
𝛼22 𝐹𝑀𝐷𝑖𝑡 + 𝛼23 𝐹𝑀𝐸𝑖𝑡 + 𝜷′𝟐 𝒛𝒊𝒕 + 𝛾2 𝐶𝑅𝐼𝑆𝐼𝑆𝑖𝑡 + 𝜃2 𝐼𝑇𝑖𝑡 + 𝜀𝑖𝑡
Where 𝑀𝑃𝐸𝑖𝑡 represents the monetary policy effectiveness of
country 𝑖 in year 𝑡. It is remarkable that the higher MPE is, the less
effective monetary policy is and vice versa. 𝐹𝐷𝑖𝑡 measures the level
of financial development of country 𝑖 in year 𝑡. In order to examine
impacts of component indicators of financial development including
financial market index (FM) and financial institutions index (FI), I
estimate the model (5.2) with independent variables of major
concern which are 𝐹𝑀𝑖𝑡 and 𝐹𝐼𝑖𝑡 . Similarly, I repeat the estimation
and replaces independent variables which are the specific
component financial development indicators in the model (5.3). 𝒛𝒊𝒕
is a vector of control variables, 𝐶𝑅𝐼𝑆𝐼𝑆𝑖𝑡 is the crisis dummy, 𝐼𝑇𝑖𝑡
is the inflation targeting dummy variable and 𝜀𝑖𝑡 are errors of the
model with mean equal to 0 and following i.i.d distribution.
I use different tests so as to select the best estimate method with
study data including F-test, Breusch-Pagan Lagrangian test and
Hausman test. After performing tests before regression and
selecting estimate method, I estimate the model by the method of
Pooled OLS with respect to the model (5.1) and (5.2) and the
FEM method with respect to the model (5.3). Defect tests after

regression are also applied and in the case the model contains
defects, I fix the defects of the model with GLS.

(5.3)


20

5.2.4.

Data

I use panel data of 7 countries in the period between 1980 and
2016, depending on the data availability. The data are mainly
collected from sources disclosed by World Bank, IMF and
calculated by the author in Chapter 4.
5.3.

Results and Discussion

5.3.1.

Stationarity test

The results report that except MPE and FIE which integrate at
level (1), the other variables integrates at level (0).
5.3.2.

Cointegration test

Testing cointegration on the data table comes to the results that
there is no long term relationship between variables.
5.3.3.

Analyzing correlation matrix and testing multicollinearity

The study results indicate that there is

not serious

multicollinearity phenomenon in the study model.
5.3.4.

Endogeneity test

The test results cannot reject that the estimate model does not
contain endogeneity.
5.3.5.

Autocorrelation test

The results of autocorrelation test of Wooldridge set indicate that,
with H0 assumption, the regression model involving no
autocorrelation at level 1 cannot be rejected in all 3 models.
5.3.6.

Heteroskedasticity test

The results of heteroskedasticity test with Breusch-Pagan test and
Modified Wald test show that in all three models of concern, the

H0 assumption suggests that the model with homoscedasticity is
rejected at significant level of 1%.


21

5.3.7.

Impacts of financial development on monetary policy

effectiveness
FGLS regression results indicate that financial development puts
negative impacts on monetary policy effectiveness with
statistically significant level of 1%.
When examining the impact of the component financial
development index by estimating the model (5.2) and (5.3) with
the gradual addition technique of control variables, the results
show that the development level of financial institutions (FI)
positively impacts, while developing the financial market (FM)
puts a strong negative impact on monetary policy effectiveness.
In particular, the impact of FI is mainly caused by the impact of
the FID index and the impact of FM is mainly caused by the
impact of the FME index, with statistical significance found at
10% for those indices in the model (5.3). These results imply that
the deeper financial institutions are, or the greater the
development of banking and financial intermediaries is, the more
monetary policy will increase the impact on the economy. In
contrast, the more effective the financial market is, the easier it
is for actors to defend against the monetary policy shocks, thus
reducing the effectiveness of policy impacts. The findings of this

study are consistent with the reports of previous studies (see
more, Cecchetti, 1999; Lastrapes & McMillin, 2004; McCauley,
2008; Carranza, Galdon-sanchez & Gomez-biscarri, 2010; Ma &
Lin, 2016).
5.4.

Conclusion

CHAPTER 6 – CONCLUSION AND POLICY IMPLICATION

6.1.

Conclusion


22

Table 5.10. Impacting regressive results of financial development on monetary policy effectiveness
Model (5.1)
FD
GFCF
IT
CRISIS
FI
FM
FIA
FID
FIE
FMA
FMD

FME
cons
Wald
test

(1)
0.009***

(2)
0.009***
-0.002***

Model (5.2)
(3)
0.003
-0.001***
0.001
0.014***

(4)

-0.001
0.006***

Model (5.3)

(5)

(6)

-0.002***

-0.001***
0.001
0.014***
-0.009*
0.005**

-0.004
0.007***

(7)

-0.004***

-0.004**

-0.002

-0.001

0.001

0.005

0.003
-0.007
-0.037
0.002
0.002

0.004*
-0.001

14.02***

77.80***

158.61***

15.09***

80.46***

163.34***

22.72***

(8)

(9)

-0.002***

-0.001***
0.001
0.015***

0.001
-0.009*
-0.042

0.002
0.003
0.004*
0.002

-0.003
-0.009*
-0.036
0.002
0.001
0.004*
0.004

87.20***

170.77***

Note: (t_statistic) in brackets (). *, **, *** represents significance of 10%, 5%, 1%, respectively.


23

In this thesis, with the objective of assessing the role of
development in the financial system towards the effectiveness of
monetary policy, I overcome some limitations of previous
studies, and consider the overall and partial impact the aspects of
the financial development to the effectiveness of monetary policy
in the context of using the monetary policy effectiveness measure
based on the Taylor curve theory. The study was conducted on
the sample of G-7 developed countries in the period between

1951 and 2017. The results from the study show that: (1) there is
a trade-off relationship in the variance of output and inflation,
supporting the Taylor curve theory in the studied countries. (2)
The efficient frontier of monetary policy reasonably built based
on the Taylor curve theory is the difference between countries,
and there is a shift over time. (3) The effectiveness of monetary
policy changes over time, monetary policy tends to be ineffective
in the crisis period and under the impacts of domestic and
international shocks, in contrast, the ability to affect the economy
to achieve central bank's objectives in implementing monetary
policy increases in the recovery periods, in line with actual
happenings. (4) Financial development negatively affects the
effectiveness of monetary policy. In particular, the development
of the financial market (increasing the efficiency of the market reflected by the development of the capital market, the increase
of non-bank financial intermediaries, a variety of financial
instruments and products and payment technology, which
implies the development of a market-based financial system)
reduce the effectiveness of monetary policy. In contrast, the
development of financial institutions (enhancing the depth and


24

scale of financial institutions, implying bank-based financial
system) is a factor contributing to improve the effectiveness of
monetary policy. (5) There has not yet been statistical evidence
of the impact of the inflation targeting regime in operating
monetary policy on improving the effectiveness of monetary
policy in the research period. (6) Disorders and crises in the
economy minimize the role of monetary policy in regulating the

economy.
6.2.

Policy implications

Firstly, policy making agencies in countries can consider the
measurement of the monetary policy effectiveness based on
solving the optimal control problem, according to the Taylor
curve theory.
Secondly, countries require policies so as to enhance the depth,
size and variety of financial institutions. This enables to enhance
healthy competition in the financial market and also create a full
and efficient transmission environment for monetary policy.
Thirdly, central banks need to monitor the development of the
national financial system to make adjustments in making
monetary policy. In particular, it is necessary to anticipate
difficulties in monetary policy planning on a market-based
financial system, thus, central banks should consider using other
support policies or alternative tools flexibly for goals
achievement. In order to ease the pressure on monetary policy in
regulating macro-economy, policy-making agencies need policy
solutions to promote the financial market to develop in a healthy
way so that they can utilize advantages of financial market
towards

economic

growth

and

simultaneously

limiting


25

uncertainties in the financial market, causing spillover effects on
macroeconomic instability and weakening the effectiveness of
monetary policy.
Fourthly,

especially

for

developing

countries,

under

circumstance of weak technical infrastructure, incomplete legal
institutions, central bank of little independence, it is required to
carefully consider between benefits and risks before applying the
inflation targeting regime in operating monetary policy.
Finally, future studies can develop micro-models that focus on
monetary transmission channels in a theoretical model including
the level of financial development which can provide a firmer

direction for next steps in empirical studies.


LIST OF AUTHOR’S PUBLICATION
Thesis title:

FINANCIAL DEVELOPMENT AND THE
EFFECTIVENESS OF MONETARY POLICY

Major:

Finance- Banking

Code:

9340201

PhD Student:

Ho Thi Lam

Supervisor:

Tran Ngoc Tho

1. The inflation–output stability trade-off and monetary policy
2. Macroeconomic
Effectiveness

Performance

and

Monetary

Policy