Journal of the Asia Pacific Economy

ISSN: 1354-7860 (Print) 1469-9648 (Online) Journal homepage: http://www.tandfonline.com/loi/rjap20

The long-run analysis of monetary policy

transmission channels on inflation: a VECM
approach

Ngan Tran

To cite this article: Ngan Tran (2018): The long-run analysis of monetary policy transmission
channels on inflation: a VECM approach, Journal of the Asia Pacific Economy, DOI:
10.1080/13547860.2018.1429199

To link to this article: https://doi.org/10.1080/13547860.2018.1429199

Published online: 01 Feb 2018.

Submit your article to this journal

Article views: 4

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=rjap20
JOURNAL OF THE ASIA PACIFIC ECONOMY, 2018
https://doi.org/10.1080/13547860.2018.1429199

The long-run analysis of monetary policy transmission

channels on inflation: a VECM approach
Ngan Tran
Department of Banking, Banking Academy, Hanoi, Vietnam

ABSTRACT KEYWORDS
In case of small open economy, the conduct of monetary policy has Monetary policy;
faced obstacles to achieve primary goals of price stability, due to transmission channel;
high vulnerability to external shocks and weak policy frameworks. inflation; VECM
Therefore, this paper aims to analyse the effectiveness of monetary
policy transmission channels in restraining inflation in case of
Vietnam for 2001-2015. The use of a Vector Error Correction Model
yields evidence that credit growth is the key determinant of high
inflation. Additionally, the results suggest the interest rate channel
has a perverse effect on inflation in the long run, which means that
the inflation rate increases with the policy rate. There is also
significant short-run causal relationship from credit growth to
inflation, and from the policy rate to inflation. However, empirical
results fail to confirm the existence of relationship between the
exchange rate channel and inflation in both short run and long run.

1. Introduction
Monetary policy is viewed as the most important force stabilising the price. Specially, econo-
mies with high level of government budget deficit have limits to use fiscal policy.1 The trans-
mission mechanism of monetary policy reflects its impact and orients policy-makers to
adjust it effectively. In some developing countries, there is a lack of standardised model to
measure monetary policy effect because of monetary transmission process uncertainty which
results from changes in economic structure (Kamin, Turner, and Dack 1998). An appropriate
evaluation of transmission process will assist central bank in implementing and fine-tuning
monetary policies. Moreover, since the inflation has been more challenging in case of devel-
oping economies (Fraga, Goldfajn, and Minella 2003; Mishkin 2004), it is necessary to find
out the most effective monetary transmission channel controlling inflation rate.
Since 2007, while other economies in the region have maintained single-digit rates of
inflation, Vietnam has suffered from large volatility of inflation, reaching double-digit
numbers during the period of 2008–2011. Despite the success in controlling inflation
from 2013 to 2015, the inflation expectation is still highly sensitive to the increase in the
price level. As such, the objective of controlling inflation is much more demanding to
achieve a stable macroeconomic environment. In fact, the central bank of Vietnam has

CONTACT Ngan Tran ngantt@hvnh.edu.vn

© 2018 Informa UK Limited, trading as Taylor & Francis Group
2 N. TRAN

implemented a contractionary monetary policy with the primary objective of price stabil-
ity over the past few years. However, highly persistent inflation is still problematic. It is
necessary to investigate the effectiveness of monetary policy transmission channels, and
then to find out the solution to stabilise inflation in the long term.
The rest of the paper is organised as follows. Section 2 presents the theoretical and
empirical literature. Econometric methodology and empirical results are provided in Sec-
tion 3, followed by Section 4 of conclusion.

2. A theoretical and empirical literature review

In theory, there are many debates about the influence of monetary policy on the economy
over time. Most economists believe that monetary policy considerably affects economic
growth and inflation in short term. The model of Fleming (1962) and Mundell (1963)
implies for open economy under pegging exchange rate regime, money supply shortly
will have no influence on economy because of sterilising capital flows. Some following
new classical approach argues that systematic monetary shock has no effect on output
and employment even in short run. For this reason, Bernanke and Gertler (1995) call
monetary transmission mechanism as a ‘black box.’ To explicitly clarify how monetary
policy can explain the economy, Mishkin (2013) suggests the use of a structural model
which captures the channels through which monetary policy decisions are transmitted to
the real economy. There are three main transmission channels of monetary policy, namely
the interest rate, exchange rate and credit channels.

2.1. Traditional interest rate channel

As Mishkin (2013), based on both Keynesian investment saving/liquidity preference
money supply (IS/LM) and aggregate demand/aggregate supply (AD/AS) models, interest
rate channel presents as a key factor of transmission mechanisms. According to the AD/
AS model, expansionary monetary policy means increase in money supply which lowers
the interest rate (liquidity effect).2 Low interest rate will motivate investment and con-
sumption via credit expansion or securities issuance, increasing the aggregate demand.
Normally, State Bank of Vietnam (SBV) usually adjusts discount rate in order to affect
interbank rate which determines lending and deposit interest rates in the economy. More-
over, SBV also decides the interest rate at which they lend to corporations, especially large
state-owned firms. Decrease in lending interest rate offered by SBV will encourage firms
borrow more to invest.

2.2. Exchange rate channel

The effect of monetary policy via this channel is explained in accordance with the interest
rate channel. Mishkin (1996) explains a decline in the real interest rate which makes
domestic assets less attractive relative to these assets denominated in foreign currency.
Investors tend to buy foreign assets, driving demand for foreign currency go up and finally
leading to exchange rate depreciation. The fall in domestic currency value is beneficial to
exports. As net export is a component of aggregate demand, exchange rate can indirectly
 JOURNAL OF THE ASIA PACIFIC ECONOMY 3

affect output and inflation. Notably, it is emphasised that the effectiveness of exchange
rate channel is dependent on interest rate effects.
In Vietnam, since the government does not pursue to liberalise exchange rate and capi-
tal flow, variation in exchange rate does not reflect monetary policy shocks comprehen-
sively. SBV law states that stabilisation of the value of currency is considered to be a
primary goal of central bank apart from inflation control (Camen 2006). To achieve the
goal, SBV annually set a targeted margin of exchange rate fluctuation within a range from
1% to 3%. Sterilisation3 is used to eliminate impacts of capital inflows arising from differ-
ence in interest rate. Hence, exchange rate between Dong and US dollar remains stable in
favour of exports.

2.3. Credit channel

Since Vietnam’s financial system is supposed to be bank-based,4 theoretical explanation
will be focused on bank lending channel. Bernanke and Gertler (1995) explain that tight-
ening monetary policy reducing bank reserves would limit available amount of credit.
Fewer loans are offered by bank system, meaning difficulty in funding investment and
consumption. The relationship between the credit and interest rate channel is very close
as lending rate is the key factor of credit amount.
Hung and Pfau (2009) suggest that the major part of credit is for state-owned enter-
prises (SOEs) regardless of their financial positions. SOEs take advantage of government’s
guarantee to make more debt, especially from large state-owned commercial banks
(SOCBs). In recent years, SOEs become more independent in terms of their activities and
fund financing, because of the restructuring of SOCBs. However, the government still
grants financial support for large-scale SOEs though SOCBs system (Nguyen, Diaz-
Rainey, and Gregoriou 2012). If credit is mainly provided without consideration of con-
ducting monetary policy, the transmission mechanism via credit channel will be muted.
In reality, studies about the transmission mechanism of monetary policy in emerging
markets also focus on these above channels. Though empirical results are varied, they are
all useful in terms of policy amendment and interpretation. Arno
stova and Hurnık (2005)
constructed vector autoregressive (VAR) model to analyse the impact of monetary policy
on Czech Republic economy as a case of small open economy. They suggest that there
was a significant change in impact of exchange rate channel after regime shift from fixed
exchange rate to inflation targeting. Regarding small open economies in East Asia where
own similar economic characteristics to Vietnam, Fung (2002) used VAR model to exam-
ine the response of Asian economies to monetary policy changes. Generally, the results
reported that a tightening monetary policy shock led to a fall in output, prices and money
demand. Nevertheless, responses to tightening shock varied among countries. For exam-
ple, while Korea and Singapore experience a price puzzle (the increase rather than
decrease in price after tightening monetary policy), the others witness a significant and
immediate decline in price. Tang (2006) also utilises VAR model to assess how monetary
policy transmits to Malaysian economy through four channels: interest rate, credit,
exchange rate and asset price. Constrained impulse responses to investigate the relative
importance of each transmission channel indicate that the interest rate is the most pre-
dominant transmission channel. Policy implication derived from the paper encourages
the authority to use indirect instruments rather than direct control over credit expansion.
4 N. TRAN

In fact, findings from VAR model in Vietnam also imply a lack of consensus. Hung and
Pfau (2009) suggest price level rises from the first to the sixth quarter after positive shocks
to interest rate, and then decreases from the seventh to the tenth quarter. Impulse
responses from Bhattacharya (2013) model show that an increase in interest rates causes a
significant rise in inflation for only two first quarters after the shock, implying a minimal
effect of interest rate channel. Meanwhile, nominal effective exchange rate is considered
to be the key factor varying inflation in short term. However, Nguyen and Nguyen (2010)
bring a different conclusion that inflation does not significantly depend on neither
exchange rate nor interest rate in the short time. Similarly, results from the paper by IMF
(2006) shows that the degree of the impact of the exchange rate is modest and incomplete.
In general, there is a lack of conclusive evidence to find out which transmission channel
of monetary policy plays a determinant role to control inflation. The difference between
short-run and long-run effects of monetary policy also needs more considerations. In
addition, the presence of structural break due to the openness after WTO integration and
global crisis 2008 is necessarily captured in the system, since it might cause monetary pol-
icy regime shifts in small open economies (Wilson 2011; Kallberg, Liu, and Pasquariello
2005). To contribute to the literature on the assessment of monetary policy transmission
channels in Vietnam particularly and small open economies generally, this study intends
to analyse the dynamic effects of three main transmission channels of monetary policy by
conducting Vector Error Correction Model (VECM) with a structural break.

3. Econometric methodology and empirical results

A VECM is a restricted VAR model designed to test the cointegration in the system of I(1)
series. The reduced-form of VAR model can be presented as follows:

X
k
Yt ¼ A0 þ Ai Yti þ ØXt þ et (1)
i¼1

where Yt and Yti are both (j £ 1) column vectors with j as number of endogenous varia-
bles, Yt includes observations at time t and Yti includes ith lagged value of endogenous
variables. A0 denotes a (j £ 1) vector of intercept terms, while Xt indicates an (m £ 1) col-
umn vector of m exogenous variables. et is a (j £ 1) vector of disturbance terms.
When I(1) variables are cointegrated, the VAR model can be expressed by first-differ-
enced error correction form.

X
k1
DYt ¼ A0 þ PYt1 þ Ai DYti þ ØXt þ et (2)
i¼1

where DYt is the first difference of Yt . P is coefficient matrix containing information on

long-run relationship among Yt , and the rank P written as P ¼ ab0 reflects the number
of cointegrating relations (r). a and b are both (j £ r) matrix. The coefficient ai measures
the speed of adjustment of ith variable towards the equilibrium, while bi denotes the
long-run estimator of that variable. Matrix Ai contains information on short-run dynam-
ics in cointegrated system. The VECM can deal with I(1) variables without taking the first
 JOURNAL OF THE ASIA PACIFIC ECONOMY 5

difference, when the variables are cointegrated. It therefore preserves multiple cointegrat-
ing relationships in the system and then avoids the issue of misspecification (Enders
2010).
The VECM has been a widely used tool in the analysis of monetary policy. There has
been a large and growing literature of employing VECM to analyse the transmission chan-
nels of monetary policy in small open economy. For example, Mello and Pisu (2010) uti-
lise VECM model to test the significance of bank lending channel in the transmission of
monetary policy in case of Brazil. Their study reaches a conclusion of negative long-run
relationship between loan supply and interbank deposit rate, implying that monetary pol-
icy has a significant impact on credit market by driving the borrowing rate faced by banks.
Bhattacharya, Patnaik, and Shah (2011) take the sample of India as an emerging economy
to find out the most effective channel through which monetary policy can control infla-
tion. The authors suggest one cointegration between the exchange rate and price level.
This means the exchange rate pass-through to domestic prices is statistically significant.
In the same vein, the paper developed by Asongu (2014) estimates the long-run and
short-run effects of monetary policy variables on output and inflation rates by conducting
the VECM and simple Granger causality models. Based on the empirical results, the neu-
trality of monetary policy is confirmed, since it only impacts output in the short run.

3.1. Model specification

The group of endogenous variables Y includes the inflation rate (p), the annual growth
rate of credit (Credit), the exchange rate5 (EX) and policy rate (R).

Y ¼ ½p; Credit; EX; R

The group of exogenous variables X contains the world oil price (Oil), the Fed Funds
Rate (FFR), the dummy variable representing the structural break (D). X affects other
domestic variables Y, but there is no reverse impact of the latter on the former. It is a rea-
sonable assumption of model for small economy.

X ¼ ½Oil; FFR; D

Vietnam economy is vulnerable to change in oil price as a result of oil importing

nation. Rise in oil price can cause cost-push inflation when costs of production and raw
materials are higher. Decrease in supply arising from higher cost of manufacturing can
harm output growth. Moreover, increase in imports price together with high inflation
may devalue Vietnam dong (Nguyen 2014; Le and Nguyen 2011). For those reasons, the
inclusion of oil price as an explanatory factor is reasonable.
The choice of FFR as an exogenous variable helps to capture the effect of US monetary
policy, especially unconventional policy on Vietnam. In theory, the difference in interest
rate between two nations will induce currency carry trade, which affects capital net flow
and subsequently the exchange rate, inflation and output. In fact, quantitative easing has
triggered capital inflows, exchange rate appreciation and high stock prices and credit
boom in Latin and East Asia countries (Barroso, Silva, and Sales 2013; Cho and Rhee
2013).
6 N. TRAN

Figure 1. Time series plot of the variables.

Visual inspection of the data from Figure 1 supports for the presence of structural
break. Variables of p, Credit, EX and R all experienced sharp fluctuations since the begin-
ning of 2007, implying that the data is more likely to include a trend break. This is primar-
ily due to the external shocks arising from WTO integration 2006 and global financial
crisis 2008, which triggered large volatility of capital flows in Vietnam’s economy (IMF
2009, 2010). As such, the dummy variable will take value of 0 for observations prior to
2007, and value of 1 otherwise.

3.2. Data and unit root test

The data sample is available from December 2001 to December 2015 with monthly fre-
quency, since the data before that period is not fully available. Monthly data instead of
quarterly or yearly data is chosen since monthly data provides more observations (169
observations), increasing the degree of freedom. Another reason is that all three monetary
policy transmission channels are adjusted every month in order to keep the economy on
the right track. Therefore, data at monthly frequency is able to promptly capture the
 JOURNAL OF THE ASIA PACIFIC ECONOMY 7

Table 1. A summary of descriptive statistics.

Standard
Variable Description Source Mean Max Min deviations
p The annual percentage change in the cost of IFSa 8.348 28.311 –0.010 6.270
consumption for a basket of goods and services
Credit The annual growth rate of gross domestic credit (end Bloomberg 27.418 57.479 6.744 11.615
of period)
EX Nominal exchange rate announced by the central IFS 4.250 4.340 4.178 0.057
bank (end of period)
R Discount rate offered by the central bank to eligible IFS 7.417 15.000 4.800 2.901
commercial banks and depository institutions (end
of period)
Oil The average of spot prices quoted by West Texas FREDb 1.795 2.126 1.287 0.197
Intermediate
FFR Fed Funds Rate charged for borrowings to maintain FRED 1.434 5.260 0.070 1.746
reserve requirement (end of period)
a
International Financial Statistics (IFS).
b
Federal Reserve Economic Data (FRED).

impact of monetary policy in short term (one to three months). Descriptive statistics for
all variables are provided in Table 1. The exchange rate and oil price are defined in loga-
rithmic form while the others are in percentage.
The unit root test should be carried out before running the model to avoid spu-
rious regression, as Brooks (2008) insists. A variable is considered to be stationary
if the null hypothesis of a unit root is rejected. A time trend should be included,
because the exchange rate and oil price seem to steadily rise over time. Results of
Augmented Dickey Fuller (ADF) and Phillips Perron (PP) unit root test yield a
fairly similar result (Table 2). All variables are integrated of order one I(1),
except the case for FFR which shows an inconsistent conclusion between ADF and
PP test.
Since monthly data may exhibit seasonality which may bias results for the unit root
tests, we use Hylleberg-Engle-Granger-Yoo (HEGY) test to examine the presence of sea-
sonal unit root (Hylleberg et al. 1990). Table 2 indicates that variables are I(1) at the zero
frequency since the null hypothesis of non-seasonal unit root cannot be rejected at 5% sig-
nificance level. First-differenced variables are stationary, and however, differencing can
cause loss of information on long-run relationship between the series, especially when
variables are cointegrated. It is, therefore, imperative to test the cointegration among I(1)
endogenous variables.

3.3. Cointegration and identification tests

The Johansen (1988) offers several ways to determine the number of cointegration vec-
tors, including the methods of the trace statistics, the maximum eigenvalue and informa-
tion criterion. All methods are based on the cointegrating VECM as seen in Equation (2).
In this paper, the main use of VECM is to capture the long-run and short-run influences
of the channels of monetary transmission on inflation. The lag length selection in the
model is based on information criteria with the maximum of 12 lags specified.6 The model
includes three lags as the Akaike information criterion (AIC) and final prediction error
(FPE) suggest, since these two criteria are superior to others in case of small sample (Liew
2004) and in Granger causality tests (Thornton and Batten 1985). Moreover, the Lagrange
8 N. TRAN

Table 2. Results of testing for unit roots.

(1) Traditional unit root test
ADF PP
Variable Levels First differences Levels First differences
p –4.785 –6.055 –2.516 –5.376
(8) (11) (9) (3)
Credit 3.126 –7.920 –2.725 –10.354
(12) (11) (8) (7)
EX –5.150 –19.115 –1.642 –13.360
(5) (0) (7) (4)
R 5.066 –11.184 –2.294 –10.830
(4) (0) (6) (5)
Oil –4.036 –6.918 –1.352 –9.321
(2) (5) (2) (5)
FFR –6.526 –5.859 –1.581 –5.476
(8) (1) (9) (5)

(2) Seasonal unit root test

Variable Non-seasonal 2-month 6-month 12-month

p –2.018 –4.409 12.315 16.927

(7)
Credit –2.287 –3.249 19.880 19.478
(9)
EX –2.537 –3.590 11.307 6.996
(4)
R –2.363 –2.376 7.447 9.803
(7)
Oil 0.008 –2.960 15.041 11.320
(2)
FFR –3.280 –3.265 19.576 20.652
(0)
Note: Numbers in parenthesis indicate the lag length selected by AIC for corresponding ADF and HEGY tests. , , 
indicate that the null hypothesis of the existence of a unit root is rejected at the 1%, 5% and 10% levels, respectively.

multiplier (LM) tests with the null of no serial autocorrelation at lag orders of one and
three are not rejected.
The results of Johansen tests are reported in Table 3. The null hypothesis of trace statis-
tic is that there are no more than r cointegrating vectors (or the rank of the matrix p  r).
Therefore, the null is rejected for r = 0 at 1% significance level, implying that there is only
one cointegrating equation in the system. Apart from the trace statistics, the maximum
eigenvalue statistic method tests the null hypothesis that the number of cointegrating vec-
tors is equal to r (or the rank of the matrix p ¼ r). The null is also rejected for r = 0 at 5%
significance level, suggesting the existence of one cointegrating relationship. In addition,
since all eigenvalues have modulus less than one, the estimated VECM satisfies the stabil-
ity condition and converges towards its long-run equilibrium.

Table 3. Results of testing for the cointegration.

Null hypothesis Eigenvalue Trace statistic Max-Eigen statistic
r=0 0.186980 73.79216 34.15492
r 1 0.134228 39.63724 23.78197
r 2 0.048198 15.85528 8.150714
r 3 0.045621 7.704562 7.704562
Note: , ,  indicate that the null hypothesis is rejected at the 1%, 5% and 10% level,
respectively.
 JOURNAL OF THE ASIA PACIFIC ECONOMY 9

Table 4. Results of identification tests.

p Credit EX R
Long-run exclusion 10.372 8.735 0.114 5.144
[0.001] [0.003] [0.734] [0.023]
Weak exogeneity 4.826 0.717 4.542 2.787
[0.028] [0.396] [0.033] [0.095]
Note: Numbers in bracket are the p values. , ,  indicate that the null hypothesis is rejected
at the 1%, 5% and 10% level, respectively.

To identify the cointegrated relationship, this paper follows the strategy developed by
Mello and Pisu (2010), Alessandro, Marta, and Jo~ao (2006), and H€ ulsewig, Winker, and
Worms (2001). First, the long-run exclusion tests are conducted to check whether a vari-
able included in the VECM can be omitted in the long-run relationship. The rejection of
the null hypothesis means that the variable should be added in the cointegrating equation.
Second, the application of exogeneity test is to indicate which variable is weakly exoge-
nous, or does not respond to the deviations from the long-run equilibrium. If the test fails
to reject the null hypothesis, the variable is weakly exogenous. The results of those tests
provide restrictions imposed in cointegrating equation.
According to Table 4, the variable of exchange rate should be excluded from the cointe-
grated equation, while the others cannot be omitted. This indicates that there is no long-
run relationship between exchange rate and inflation. An exogeneity test fails to reject the
null hypothesis that Credit is weakly exogenous, meaning credit growth does not react to
variations in inflation, exchange and policy rates. This finding seems to be important to
the conduct of monetary policy.
As a result of those tests, the following joint exclusion and exogeneity restrictions are
imposed on cointegrating parameters in Equation (2): H0 : bEX ¼ aCredit ¼ 0. If the null
hypothesis is not rejected, inflation is not affected permanently by the exchange rate, and
credit growth is weakly exogenous. Since the likelihood ratio (LR) tests fail to reject the
null hypothesis of weak exogeneity (x2 ¼ 0:728; p value = 0.694), estimates of the levels
relationship given as follows (t-statistics values are given in parentheses).

pt ¼ 13:096 þ 0:339 :Creditt þ 1:675 :Policyt þ et

ð5:712Þ ð5:676Þ

These parameter estimates show that both credit growth and the policy rate are signifi-
cant determinants of inflation rates (at 1% significance level). Inflation is positively corre-
lated to credit growth, since the increase in the growth rate of credit by 1% will lead to the
corresponding rise in inflation rate by 0.339%. This is similar to the results from previous
studies, such as Bhattacharya (2013) and Camen (2006). However, the study finds out the
significantly positive long-run relationship between inflation and policy rate, which is dif-
ferent from others findings.7 This is also contrary to the New-Keynesian theory of mone-
tary transmission mechanism that increasing policy rate is expected to lower inflation. In
case of developing economies under fiscal dominance, higher interest rates, in fact, result
in higher risk of sovereign defaults and eventually in the devaluation of domestic cur-
rency, which leads in turn to further rise in inflation (Blanchard 2004; Favero and Gia-
vazzi 2004). In addition, due to the high degree of Vietnam’s inflation persistence
(Nguyen and Nguyen 2010), the inflation increase is prone to be faster than expected. The
10 N. TRAN

Table 5. Results of Granger causality.

Short-run Granger causalities Exogenous variables
Dependent variable Dp DCredit DEX DPolicy Oil FFR D
Dp – 7.203 5.785 6.614 1.171 0.048 –0.351
DCredit 10.085 – 11.373 4.363 –2.699 0.147 1.060
DEX 4.732 0.440 – 4.198 0.001 –0.0001 –0.0003
DPolicy 22.949 4.999 3.834 – 0.485 –0.013 –0.145
Note: Numbers in the table represent the x2 values. , ,  indicate that the null hypothesis is rejected at the 1%, 5%,
and 10% level, respectively.

central bank raises the policy rate only when the inflation rate already reaches high level
and keeps moving upward. As a result, the policy rate cannot perform as a proactive
instrument. This might imply a serious challenge to the central bank when it cannot con-
trol the effectiveness of interest rate instrument.

3.4. Short-run dynamics

The short-run effects of Credit, Policy and EX on p are estimated using the matrices A in
Equation (2). The Granger causality is applied to check whether an explanatory variable
short-run Granger causes a dependent variable. For example, the exchange rate short-run
Granger causes inflation ðEX ! pÞ if and only if lagged differences of exchange rate
ðDEXti Þ help to forecast changes in inflation from t ¡ 1 to t ðDpÞ. As such, the Wald-
tests is used to test the joint significance of coefficients on DEXti in equation normalised
in Dp. The rejection of null hypothesis provides an indication of significant short-run
Granger causality from the exchange rate to inflation.
Results of Granger causality statistic for are represented in Table 5. There is an evi-
dence of a short-run bi-directional Granger causality between inflation and credit, and
between inflation and policy rate. This supports the important role of credit growth and
policy rate on high rates of inflation. However, the absence of causality from the exchange
rate to inflation confirms the fact that this channel is ineffective in both short and long
run. Indeed, monetary policy under the fixed exchange rate regime becomes less powerful
than that under flexible exchange rate in regard to inflation control (Cevik and Teksoz
2012). In respect of exogenous variables, the oil price is significantly associated with the
inflation rate at 1% significance level, implying the presence of imported inflation in Viet-
nam economy. The inelasticity of domestic monetary variables to FFR reflects a long
period of the zero-lower bound during and after the global crisis.
Estimated model has passed several stability diagnostic tests against serial autocorrela-
tion (Breusch–Godfrey test) and heteroskedasticity (White test). The Ramsey RESET test
also indicates that the model is well specified. All results of these tests are reported in
Table 6. The parameter stability can be assessed using the cumulative sum of recursive
residuals (CUSUM) and the CUSUM of square (CUSUMSQ). According to Figure 2(a,b),

Table 6. Results of diagnostics tests.

x2 statistic Probability
Serial correlation test 1.787 0.117
White heteroskedasticity test 7.860 0.640
Ramsey RESET test (log likelihood test) 1.140 0.285
 JOURNAL OF THE ASIA PACIFIC ECONOMY 11

Figure 2. (a) Plot of CUSUM test. (b) Plot of CUSUMSQ test.

since the lines of the CUSUM and CUSUMSQ statistic go within the area between the 5%
critical lines, the estimated parameters are stable over the sample period.

4. Conclusion
This paper investigates the effectiveness of three main transmission channels of monetary
policy in controlling inflation in Vietnam for the period from 2001 to 2015. It implements
a restricted VECM to estimate the long-run impacts of the interest rate, exchange rate
and credit channels on inflation. The identification strategy is to use the tests of exclusion
and endogeneity restrictions on cointegrating equations. The results can be summarised
as follows.
First, this study supports the importance of credit channel to inflation in both short
time and long time. Expansionary credit positively affects the inflation rate, implying that
credit restrictions might perform as an effective method to stabilise inflation. Credit is the
key driver of inflation, which raises concern about the level of investment efficiency.
Second, a restriction in the VECM suggests that credit channel, in turn, does not
respond to variations in inflation, and has no link with interest rate channel as well. This
might refer to the issue that the growth of credit could be primarily driven by non-market
factors, namely government interventions.
Third, there is an evidence of causal relationship from policy rate to inflation, which is
contrary to previous Vietnam’s studies. It is remarkable that this channel has a perverse
effect on inflation: an increase in discount rate to lower inflation actually causes higher
inflation. This can be due to the dominance of fiscal policy that hails the effectiveness of
monetary policy. Under the burdens of chronic deficits, rising interest rate threatens pub-
lic sector solvency and then results in domestic currency depreciation. The real deprecia-
tion ultimately accelerates domestic inflation through changes in import prices
(Blanchard 2004). The interest rate is more likely to be a passive channel as changes in
policy rate is too sluggish to curb inflation. In the case of fiscal dominance, fiscal policy
appears as an effective instrument to inflation control, instead of monetary policy. Addi-
tionally, the credit channel outperforms the interest rate channel in terms of controlling
inflation, Open market operations (OMOs) should be, hence, a superior instrument to
discount rate towards implementing monetary policy. Vuong and Tran (2009) give a
12 N. TRAN

recommendation that OMOs should be used at higher intensiveness (e.g. interest rate in
the open market should be cut down/raised more aggressively) to enhance the effective-
ness of OMOs. Moreover, higher development of bond and money market would diversify
OMOs.
Fourth, this paper also postulates that inflation is insensitive to changes in the official
exchange rate in both long run and short run. Indeed, the central bank has lost the chan-
nel since it pursued the fixed exchange rate system. To strengthen the effectiveness of
exchange rate channel, it should take more considerations in respect of monetary regime
shift to more flexible exchange rate regime.
Finally, even though the results in this paper can be much enriched by using more
advanced economic models, this requires larger sample and more complex methodology
to capture the interactive relationships in the system. Within the frameworks of the
VECM, the estimation is proved to be efficient and the estimated results are robust.

Notes
1. During normal periods of economic growth, large budget deficit would constrain increase in
government spending to promote aggregate demand. In Vietnam, government budget deficit
(as percentage of GDP) has increased sharply from only an average of 1.3% in the 2003–2007
period to 3.7% in the 2009–2011 period, becoming the country with the highest state’s budget
deficit in ASEAN (ADB 2014).
2. The liquidity effect raised by Friedman (1969) refers to the ceteris paribus downward trend in
interest rate caused by rise in money supply.
3. As Mundell (1963) and Fleming (1962)’s open economy IS/LM, OMOs (e.g. buying domestic
bonds) will be conducted by central bank to offset capital inflow to defend fixed exchange rate.
4. The ratio of deposits and loans to GDP is increasingly high overtime (both reaching over 90%
after 2008) while that of total domestic currency bonds and corporation bonds are modest
(accounting for less than 20% of GDP), reported by World Bank Statistics.
5. The units of domestic currency per US dollar (VND per USD).
6. According to Enders (2010), the maximum lag length can be chosen by the frequency of the
data. Monthly data can have a maximum lag length of 12.
7. Most empirical studies give a comment that the interest rate channel has only a trivial impact
on Vietnam’s inflation movements.

Disclosure statement
No potential conflict of interest was reported by the author.

Notes on contributor
Ngan Tran is a lecturer at Department of Banking, Banking Academy, Hanoi, Vietnam. She is cur-
rently a PhD candidate at School of Economics, Finance and Marketing at RMIT University, Mel-
bourne, Australia. Her research area is applied econometrics and macroeconometrics.

References
ADB. 2014. “Asian Economic Integration Monitor.” Asia Economic Monitor. Vietnam.
Alessandro, C., M. Marta, and S. Jo~ao. 2006. “Aggregate Loans to the Euro Area Private Sector.”
Quarterly Review of Economics and Finance 46: 2211–2226.
 JOURNAL OF THE ASIA PACIFIC ECONOMY 13

Arno
stova, K., and J. Hurnık. 2005. “The Monetary Transmission Mechanism in the Czech Repub-
lic (Evidence from VAR Analysis).” CNB Working Paper Series 4.
Asongu, S. A. 2014. “Does Money Matter in Africa?: New Empirics on Long- and Short-run Effects
of Monetary Policy on Output and Prices.” Indian Growth and Development Review 7: 142–180.
Barroso, J. B. R. B., L. A. P. D. Silva, and A. S. Sales. 2013. “Quantitative Easing and Related Capital
Flows into Brazil: Measuring its Effects and Transmission Channels Through a Rigorous Coun-
terfactual Evaluation.” Research Department, Central Bank of Brazil Working Papers Series 313.
Bernanke, B. S., and M. Gertler. 1995. “Inside the Black Box: The Credit Channel of Monetary Pol-
icy Transmission.” Journal of Economic Perspectives 9: 27–48.
Bhattacharya, R. 2013. “Inflation Dynamics and Monetary Policy Transmission in Vietnam and
Emerging Asia.” IMF Working Paper 155. Washington: International Monetary Fund.
Bhattacharya, R., I. Patnaik, and A. Shah. 2011. “Monetary Policy Transmission in an Emerging
Market Setting.” IMF Working Paper 5. Washington: International Monetary Fund.
Blanchard, O. 2004. “Fiscal Dominance and Inflation Targeting: Lessons from Brazil.” NBER Work-
ing Paper Series 10389. Cambridge, MA: NBER.
Brooks, C. 2008. Introductory Econometrics for Finance. 2nd ed. Cambridge: Cambridge University
Press.
Camen, U. 2006. “Monetary Policy in Vietnam: The Case of a Transition Country.” Monetary
Policy in Asia: Approaches and Implementation 31: 232–252.
Cevik, S., and K. Teksoz. 2012. “Lost in Transmission? The Effectiveness of Monetary Policy Trans-
mission Channels in the GCC Countries.” IMF Working Paper 191. Washington: International
Monetary Fund.
Cho, D., and C. Rhee. 2013. “Effects of Quantitative Easing on Asia: Capital Flows and Financial
Markets.” ADB Economics Working Paper Series 350. Manila: Asian Development Bank.
Enders, W. 2010. Applied Econometric Time Series. 3rd ed. New York: Hoboken, NJ: Wiley Chiches-
ter John Wiley [distributor].
Favero, C., and F. Giavazzi. 2004. “Inflation Targeting and Debt: Lessons from Brazil.” NBER Work-
ing Paper Series 10390.
Fleming, M. 1962. “Domestic Financial Policies Under Fixed and Under Floating Exchange Rates.”
IMF Staff Papers 9: 369–380.
Fraga, A., L. Goldfajn, and A. Minella. 2003. “Inflation Targeting in Emerging Market Economies.”
NBER Macroeconomics Annual 18: 365–400.
Friedman, M. 1969. The Optimum Quantity of Money. Chicago, IL: Aldine Publishing Company.
Fung, B. 2002. “A VAR Analysis of the Effects of Monetary Policy in East Asia.” BIS Working
Papers 119, 1–12.
H€ulsewig, O., P. Winker, and A. Worms. 2001. “Bank Lending in the Transmission of Monetary
Policy: A VECM Analysis for Germany.” International University in Germany Working Paper
08.
Hung, L. V., and W. D. Pfau. 2009. “A Vector Autoregression (VAR) Analysis of the Monetary
Transmission Mechanism in Vietnam.” Applied Econometrics and International Development 9.
Hylleberg, S., R. F. Engle, C. W. J. Granger, and B. S. Yoo. 1990. “Seasonal Integration and Cointe-
gration.” Journal of Econometrics 44: 215–238.
IMF (International Monetary Fund). 2006. Vietnam: Selected Issues. IMF Country Report. Wash-
ington, DC: International Monetary Fund.
IMF (International Monetary Fund). 2009. “Staff Report for the 2008 Article IV Consultation.” IMF
Country Report 09/110. Washington: International Monetary Fund.
IMF (International Monetary Fund). 2010. “Staff Report for the 2010 Article IV Consultation.” IMF
Country Report 10/281. Washington: International Monetary Fund.
Johansen, S. 1988. “Statistical Analysis of Cointegration Vectors.” Journal of Economic Dynamics
and Control 12: 231–254.
Kallberg, J. G., C. H. Liu, and P. Pasquariello. 2005. “An Examination of the Asian Crisis: Regime
Shifts in Currency and Equity Markets.” The Journal of Business. The University of Chicago Press
78: 169–212.
14 N. TRAN

Kamin, S., P. Turner, and J. V. O. Dack. 1998. “The Transmission Mechanism of Monetary Policy
in Emerging Market Economies: An Overview.” BIS Working Papers 3: 5–64.
Le, V. T., and V. T. T. Nguyen. 2011. “The Impact of Oil Prices, Real Effective Exchange Rate and
Inflation on Economic Activity: Novel Evidence for Vietnam.” Discussion Paper Series 9.
Research Institute for Economics & Business Administration, Kobe University.
Liew, V. K. S. 2004. “Which Lag Length Selection Criteria Should We Employ?” Economics Bulletin
3: 1–9.
Mello, L. D., and M. Pisu. 2010. “The Bank Lending Channel of Monetary Transmission in Brazil:
A VECM Approach.” The Quarterly Review of Economics and Finance 50: 50–60.
Mishkin, F. S. 1996. “The Channels of Monetary Transmission: Lessons for Monetary Policy.”
NBER Working Paper Series 5464. http://www.nber.org/papers/w5464.pdf.
Mishkin, F. 2004. “Can Inflation Targeting Work in Emerging Market Countries?” NBER Working
Paper Series 10646. Cambridge, MA: NBER.
Mishkin, F. 2013. Economic of Money, Banking and Financial markets: Global Edition. 10th ed.
Harlow: Pearson.
Mundell, R. 1963. “Capital Mobility and Stabilization Policy under Fixed and Flexible Exchange
Rates.” The Canadian Journal of Economics and Political Science 29: 475–485.
Nguyen, C. V. 2014. “Impacts of International Oil Price Changes on Vietnam’s Economy – an
Input-Output Study.” Asian Economic and Financial Review 4: 432–439.
Nguyen, D. T., I. Diaz-Rainey, and A. Gregoriou. 2012. “Financial Development and the Determi-
nants of Capital Structure in Vietnam [Online].” Accessed 25 July 2016. http://ssrn.com/
abstract=2014834 or https://doi.org/10.2139/ssrn.2014834.
Nguyen, H. T. T., and T. D. Nguyen. 2010. “Macroeconomic Determinants of Vietnam’s Inflation
2000–2010: Evidence and Analysis.” In Vietnam Centre for Economic and Policy Research. Uni-
versity of Economics and Business, Vietnam National University Hanoi.
Tang, H. C. 2006. “The Relative Importance of Monetary Policy Transmission Channels in Malay-
sia.” CAMA Working Papers 23. Crawford School of Public Policy, The Australian National
University.
Thornton, D. L., and D. S. Batten. 1985. “Lag-Length Selection and Tests of Granger Causality
Between Money and Income.” Journal of Money, Credit and Banking 17: 164–178.
Vuong, Q. H., and T. D. Tran. 2009. “Financial Turbulences in Vietnam’s Emerging Economy:
Transformation over 1991-2008 Period.” In Contemporary Issues in Finance, edited by Jagadee-
sha, and Deene. New Delhi: Excel Books.
Wilson, P. 2011. Challenges for the Singapore Economy After the Global Financial Crisis. Singapore:
World Scientific.
World Bank Statistics. Accessed 25 July 2016. http://data.worldbank.org/country/vietnam.