Can Vietnam Move to Inflation Targeting?

Nguyen Thi My Hanh(&)

Hanoi, Vietnam
mhanh105@gmail.com

Abstract. This study attempts to investigate the adoption of inflation targeting

framework in Vietnam. This work is done by examining the satisfaction of one
crucial prerequisite of inflation targeting: There is the existence of predictable
and stable linkages between monetary policy instruments and inflation out-
comes. The Johansen multivariate cointegration procedure and Vector Error
Correction Model (VECM) approach are used to check the relationship between
monetary policy instruments and inflation, and the findings point out that there
exists the existence of stable and predictable linkage between monetary policy
instruments and inflation in Viet Nam. However, the relationship is too weak.
As a result, Viet Nam is yet a candidate for adopting inflation targeting
framework.

Keywords: Monetary policy
Inflation targeting framework
Johansen test

1 Introduction

The State Bank of Vietnam focuses on many targets of monetary policy while other
central banks in the world only focus on price stability. From 2012 until now, inflation
of Vietnam is always controlled at one single digit. However, there is no pledge of The
State Bank of Vietnam to stabilize prices in the long term, and as a result, inflation
expectations which are important for price stability could not be anchored. It means that
inflation might occur any time, which threaten the economy very badly like what
happened in the period 2020–2025.
At the moment, Vietnam is choosing to manage exchange rate, or in other words,
exchange rate is seen as a nominal anchor for Vietnam’s monetary policy. According to
the impossible trinity theory, it is impossible to have all three of the following at the
same time: a fixed foreign exchange rate, free capital movement and an independent
monetary policy. This theory has explain why many countries which chose exchange
rate as a nominal anchor for monetary policy got into difficulties in conducting mon-
etary policy such as Thailand, South Korea, Indonesia in 1997 Asian financial crisis.
After that, these countries have moved to inflation targeting policy.
The World Bank’s research group has suggested that Vietnam should choose only
one nominal anchor for monetary policy: exchange rate or inflation because it seems
impossible to achieve two targets at the same time. In order to modernize monetary
policy framework, The State Bank of Vietnam is aiming to inflation targeting monetary
policy in the period of 2011–2020. That’s why the author will examine whether
Vietnam could adoption inflation targeting policy in this paper.

© Springer Nature Switzerland AG 2019

V. Kreinovich et al. (Eds.): ECONVN 2019, SCI 809, pp. 1052–1061, 2019.
https://doi.org/10.1007/978-3-030-04200-4_76
 Can Vietnam Move to Inflation Targeting? 1053

2 Inflation Targeting

Inflation targeting was firstly adopted in New Zealand since 1990. Then it was soon
followed by Canada, Australia, the United Kingdom, and Sweden. Recently, several
emerging economies have also moved in this direction, including Chile, Mexico, and
Brazil. Also, many other central banks are now considering the applicability of this
regime to their countries because this scheme seems to lack some of the disadvantages
of alternative monetary policy regimes. For example, monetary targeting policy has a
number of benefits, but it is only applied if there is a correlation between the chosen
aggregate and nominal income. However, in reality, this relationship does not exist.
The adoption of monetary aggregates as the target variable for monetary policy is
therefore incapable, and many countries decided to search for an alternative nominal
anchor.
Exchange rate is considering as the other popular nominal anchor, which is often
chosen by small, open economies. Such an anchor may take the form of a fixed
exchange rate regime or a crawling peg with well-defined rules for the crawl. However,
the recent financial crises in emerging economies have shown fixed exchange rate
regime can suffer break since this regime can result in systemic banking and financial
crises and broadly affect the level of economic activity. Finally, the failure of choosing
monetary aggregate or exchange rate as a nominal anchor for monetary policy leads
many countries consider adopting inflation targeting.
Nowadays inflation targeting framework has become popular in the world, how-
ever, it can be difficult to define what inflation targeting is (Rogoff 1985; Loayza and
Soto 2001; Angeriz and Arestis 2008; Freedman and Laxton 2009a; Ncube and Ndou
2011). This study will use the definition of Mishkin (2007). He defined that inflation
targeting monetary policy is a monetary policy strategies, including 5 elements: (1) the
public announcement of medium-term numerical targets for inflation; (2) an institu-
tional commitment to price stability as the primary, long-run goal of monetary policy
and a commitment to achieve the inflation goal; (3) an information-inclusive approach
in which many variables (not just monetary aggregates) are used in making decisions
about monetary policy; (4) increased transparency of the monetary policy strategy
through communication with the public and the markets about the plans and objectives
of monetary policymakers; (5) increased accountability of the central bank for attaining
its inflation objectives.
In practice, inflation targeting framework requires a quantitative statement as to
what the inflation rate is consistent with the pursuit of price stability in the long run.
Price stability is the primary goal of inflation targeting monetary policy. In addition, the
target series needs to be defined and measured accurate, timely and easily under-
standable by the public. The central bank makes clear about the main tasks and criteria
of monetary policy so that it makes monitoring inflation targeting simpler for the
public. The more transparent the monetary policy become, the more credibility the
central bank have. As a result, the central can easily pursue the inflation target.
Generally, one may distinguish between three types of inflation targeting: full-
fledged inflation targeting (FFIT), inflation targeting lite (ITL), and electic inflation
targeting (EIT) (Truman 2003). FFIT countries such as Sweden, the UK, Norway,
1054 N. T. M. Hanh

Czech Republic, Australia, New Zealand, and Canada concentrate on financial stability,
the development of financial markets along with the level of credibility (medium to a
high degree). These countries cannot attain and maintain low inflation rate without a
clear commitment to inflation targeting so that they are forced to sacrifice output
stabilization to the various degree. Meanwhile, EIT countries such as European Central
bank, Denmark, Switzerland, Japan and the US can pursue the output stabilization
objective together with price stability. However, ITL is at work in emerging economies
that have a low credibility. It is regimes where central banks announce a broad inflation
objective, but owing to their relative low credibility, they are not able maintain inflation
as the foremost policy objective (Stone 2003).

3 Prerequisites for Inflation Targeting

The experience of successful inflation targeters points to several prerequisites for

adoption of an inflation targeting framework. Firstly, instrument independence must be
granted to central banks in order to fulfill their mandate of ensuring price stability. In
other words, central banks must have the freedom to adjust its instruments of monetary
policy, which helps achieve the objective of low inflation. Also, instrument indepen-
dence means that the central bank is not constrained by the need to finance the gov-
ernment budget.
Secondly, the central bank must have an effective monetary policy instrument
which has a relatively stable relationship with inflation. From the experience of
countries which have success in adopting inflation targeting framework, they choose
indirect instruments, such as short-term interest, rather than direct instruments, such as
credit controls.
Thirdly, accountability and communication with the public are essential require-
ments for adopting inflation targeting framework. In particular, central banks publish
periodically an inflation report, which updates its views of future prospects and
explains its policy actions. That the public understand about how the central bank
operates would help anchor inflation expectations. In case targets are breached, the
central banks need to issue an open letter which explains causes of the breach and how
and when inflation returns to tolerated levels. Moreover, to improve central banks’
transparence, meeting minutes of monetary policy committees are published so that all
decisions of central banks become completely clear to the public.
Fourthly, countries wishing to move to inflation targeting need to focus on several
technical issues. For example, central banks need to make decisions to choose a
suitable price index (underlying or core measure of inflation) and to specify the
inflation target as a point, as a band, or as a medium-term average as well as the time
horizon for meeting the target. In addition, forecast technique is a key element of a
successful inflation targeting policy.
 Can Vietnam Move to Inflation Targeting? 1055

4 Is Vietnam Fulfilled the Prerequisites for Inflation

Targeting?
4.1 Brief Theoretical Framework
Countries adopting inflation targeting framework normally use interest rates as the
nominal anchor to control inflationary pressures. If inflationary exceeds the target rate,
central banks would raise the interest rate to slow down the economy, and subsequently
lower the inflation to get it closer to the target, vice versa. Meanwhile, real exchange
rate incorporates external sectors in the domestic inflation process. It directly affects the
cost of imported intermediate inputs to the production process on the inflation measure.
Moreover, higher interest rate generally leads to a stronger currency that reduces
international competitiveness, impacts export performance, and leads to relatively
greater import penetrations. In other words, higher interest rate pushes up the value of
currency and restrains economic activity, hence influences the aggregate demand and
consumer price. Additionally, many studies have shown relationship between money
supply and inflation (Lucas 1996). However, monetary targeting framework is not a
good choice for several central banks such as the United States, Canada and the United
Kingdom (Mishkin 2000). Then, these countries have switched to inflation targeting
framework.

4.2 Model Specification and Data

Blejer and Leone (1999) have stated that one crucial prerequisite condition for inflation
targeting is that there is a relationship between inflation and monetary policy instru-
ment. As a result, this paper only focuses on the second prerequisite condition, or in
other words, the question whether The State bank of Vietnam has an effective
instrument policy will be answered.
To analyze the relationship between inflation and monetary policy instrument,
Consumer Price Index (CPI) is chosen as a variable representing inflation and three
variables such as Money supply (M2), Exchange rate (ER) and Interest rate (R) are
chosen as variables representing monetary policy instruments. This data set is collected
from International Financial Statistic. The paper employs quarterly data during 2000:
Q1 to 2016:Q4. The Johansen cointegration test and The Vector Error Correction
Model approach are applied on macroeconomic data. The model can be simplified as
follows:

f(CPI) = f(M2, ER, R)

where CPI = Consumer Price Index (as proxy to measure inflation rate)

M2 = Money supply
ER = Exchange rate
R= Interest rate
1056 N. T. M. Hanh

4.3 Empirical Results

+ Unit root test:
Unit root test is performed to determine whether the individual series are stationary
or not since the usage of non-stationary data can generate spurious regressions. In this
case, Dickey-Fuller test is used. The null hypothesis is that each variable has a unit
root, and the alternative hypothesis is that the series are stationary (or each variable has
root outside unit circle).
In order to check for stationary condition of the above variables, this paper uses t-
statistics critical values. The results suggest that all variables are stationary at level.
Since these variables are integrated of order one, the cointegration test can be applied
(Table 1).

Table 1. Dickey-Fuller test for unit root

Variables Test-statistic 5% critical value p-values
CPI −3.863 −2.917 0.0023
R −5.511 −2.917 0.0000
M2 −8.452 −2.917 0.0000
ER −8.883 −2.917 0.0000
Source: Author’s computation using Stata 11

+ Optimal lag selection:

Before running cointegration test, the optimal lag length should be decided because
the weakness of Johansen test is sensitive to the lag length. The result suggests that the
optimal lag length is 3 (Table 2).

+ Johansen tests:
Both Johansen’s trace statistic test and Johansen’s maximum eigenvalue statistic
are applied. In particular, trace statistic test tests the hypotheses:

Table 2. Optimal lag selection

Selection-order criteria Number of obs = 64
Sample: 5–68
Lag LL LR df p FPE AIC HQIC SBIC
0 −1266.21 2.0e+12 39.6941 39.7473 39.8291
1 −935.024 662.38 16 0.000 1.1e+08 29.8845 30.1103 30.5191
2 −904.879 60.29 16 0.000 7.0e+07 29.4025 29.8809 30.6168
3 −888.156 33.445 16 0.006 6.9e+07* 29.3799* 30.0709 31.134
4 −877.802 20.707 16 0.190 8.5e+07 29.5563 30.46 31.8501
Endogenous: CPI M2 ER R
Exogenous: -cons
Source: Author’s computation using Stata 11
 Can Vietnam Move to Inflation Targeting? 1057

H0 ðrÞ : r0  r and the alternative H1 ðr0 Þ : r0 [ r

Where r is the number of cointegrating vectors under the null hypothesis. Johan-
sen’s trace statistic is a join test where the null hypothesis H0 examines that the number
of cointegrating vectors is less than or equal to r against the alternative hypothesis H1
which assumes that there are more than r.

Johansen also derives the Eigenvalue statistic for the hypotheses:

H0ðrÞ : r0 ¼ r and the alternative H1 : r0 ¼: r þ 1

Where r is the number of cointegrating vectors under the null hypothesis H0. The
Johansen’s maximum eigenvalue statistic examine tests its null hypothesis that the
number of cointegrating vectors is r against an unspecified or general alternative which
states that the number of cointegrating vectors is (r + 1) (Table 3).
Both test support for cointegration at a 5% significance level. Also, the presence of
two cointegrations among the variables in Vietnam recommends a long-term rela-
tionship among them. The result of the below table will show the relationship
(Table 4).

Table 3. Johasen tests for cointegration

Trend: constant Number of obs = 65
Maximum rank parms LL Eigenvalue Trace statistic 5% critical value
0 36 −932.56032 64.2383 47.21
1 43 −916.13772 0.39669 31.3927 29.68
2 48 −907.00589 0.24496 13.1290 15.41
3 51 −901.44838 0.15718 2.0140 3.76
4 52 −900.44138 0.03051
Sample: 4–68 Lags = 3
Source: Author’s computation using Stata 11

The relationship is also presented by equations as follows:

e1t ¼ CPI þ 0:028ER þ 184R  1244 or CPI ¼ 1244  0:028ER  184R þ e1t

e2t ¼ M2 þ 0:0029ER þ 8:09R  113 or M2 ¼ 1130:0029ER8:09R þ e2t

Next, the study performs the impulse response function as an additional check. The
impulse response test is to assess the responsiveness of the dependent variables in the
VECM to shocks to each of the variables (Brooks 2008). There is a surprising result
that CPI totally does not respond to shocks from M2, R and ER. It has found that
though there exists a relationship between CPI and other variables, this relationship is
too weak. This finding is also supported by variance decomposition analysis (Table 5).
The changes of CPI are not explained by the changes of M2, ER and R (Fig. 1).
1058 N. T. M. Hanh

Table 4. Cointegrating equations

Table 5. Variance Decomposition analysis

Step (1) (2) (3) (4) (5) (6)
fevd fevd fevd fevd fevd fevd
0 0 0 0 0 0 0
1 1 .153557 .261927 8.9e-06 0 .846443
2 .898911 .133765 .442961 .000375 .10082 .855074
3 .845566 .132758 .495815 .004324 .150102 .845631
4 .819083 .13479 .501762 .018166 .167543 .834235
5 .802622 .134363 .497942 .041073 .175314 .827964
6 .79072 .131925 .493172 .066939 .179956 .826216
7 .781533 .128396 .489283 .088708 .183 .827032
8 .774245 .124348 .486286 .1033984 .185136 .829023
9 .768269 .12017 .483808 .113854 .186775 .831367
10 .763242 .116903 .481597 .120136 .188109 .833663
(continued)
 Can Vietnam Move to Inflation Targeting? 1059

Table 5. (continued)
Step (7) (8) (9) (10) (11) (12)
fevd fevd fevd fevd fevd fevd
Step (7) (8) (9) (10) (11) (12)
fevd fevd fevd fevd fevd fevd
0 0 0 0 0 0 0
1 .019887 .002496 0 0 .718187 .018525
2 .080958 .062323 .000027 .01103 .473383 .0361
3 .157479 .091819 .000794 .019826 .345085 .037717
4 .202243 .105104 .001854 .024556 .291473 .035614
5 .22391 .110366 .002728 .027109 .267469 0.36367
6 .23511 .111405 .003635 .02823 .255861 .037169
7 .241925 .111739 .004648 .02846 .250045 .036964
8 .246781 .113347 .005706 .028273 .247017 .036139
9 .250673 .11689 .006735 .027937 .245316 .035132
10 .253998 .122242 .027565 .027565 .244244 .034149
Step (13) (14) (15) (16)
fevd fevd fevd fevd
0 0 0 0 0
1 0 0 0 .978969
2 .000243 .000131 .002698 .901202
3 .003539 .001785 .001621 .86614
4 .011519 .00642 .004521 .841116
5 .019296 .010564 .010679 .812195
6 .02569 .013628 .015857 .784487
7 .030819 .016112 .018747 .762589
8 .034913 .018357 .019917 .74653
9 .03822 .020525 .020203 .734124
10 .040956 .02268 .020161 .723474
(1) Irfname = irf, impulse = CPI, and response = CPI
(2) Irfname = irf, impulse = CPI, and response = ER
(3) Irfname = irf, impulse = CPI, and response = R
(4) Irfname = irf, impulse = CPI, and response = M2
(5) Irfname = irf, impulse = ER, and response = CPI
(6) Irfname = irf, impulse = ER, and response = ER
(7) Irfname = irf, impulse = ER, and response = R
(8) Irfname = irf, impulse = ER, and response =M2
(9) Irfname = irf, impulse = R, and response = CPI
(10) Irfname = irf, impulse = R, and response = ER
(11) Irfname = irf, impulse = R, and response = R
(12) Irfname = irf, impulse = R, and response = M2
(13) Irfname = irf, impulse = M2, and response = CPI
(14) Irfname = irf, impulse = M2, and response = ER
(15) Irfname = irf, impulse = M2, and response = R
(16) Irfname = irf, impulse = M2, and response = M2
Source: Author’s computation using Stata 11
1060 N. T. M. Hanh

Fig. 1. Impulse response function (Source: Author’s computation using Stata 11)

5 Conclusion

Although the above results have pointed that there exists a relationship between
inflation and monetary policy instruments in Vietnam, the relationship is very weak.
Moreover, from impulse response function analysis, it seems that monetary policy
instruments have no influence on CPI. In other words, Vietnam has not yet satisfied the
crucial condition for adopting inflation targeting framework.
According to IMF, Vietnam is pursuing the exchange rate targeting policy. This
choice is suitable for many developing countries when their economies depend on
export sector. These countries focus on keeping their currency stable so as to boost
exports. However, this practice makes controlling inflation more complicatedly if the
country cannot control foreign inflows. This situation happened in Vietnam during
2007–2011. At that time, Vietnam received a large amount of foreign capital after
becoming a member of World Trade Organization. As a consequence, inflation rate hit
double digits, which led to a difficult period for policy makers. Additionally, the
impossible trinity theory indicated that one country cannot meet the three targets at the
same time: a fixed exchange rate, a free capital movement and an independent mon-
etary policy. Therefore, it is advisable that Vietnam should consider inflation targeting
policy as a solution.
 Can Vietnam Move to Inflation Targeting? 1061

When moving to inflation targeting framework, there are many countries which
have not satisfied all necessary conditions. Therefore, if Vietnam highly appreciates the
advantages of inflation targeting framework, the country is still able to move to a new
monetary policy framework in the future.

References
Angeriz, A., Arestis, P.: Assessing inflation targeting through intervention analysis. Oxf. Econ.
Pap. 60(2), 293–317 (2008)
Blejer, M.I., Leone, A.M.: Introduction and overview. In: Blejer, M.E., Ize, A., Leone, A.M.,
Werlang, S. (eds.) Inflation Targeting in Practice; Strategic and Operational Issues and
Applications to Emerging Markets Economies. IMF (1999)
Brooks, C.: Introductory Econometrics for Finance. Cambridge University Press, Cambridge
(2008)
Freedman, C., Laxton, D.: Why Inflation Targeting? IMF Working Papers, pp. 1–25 (2009a)
Johansen, S., Juselius, K.: Maximum likelihood estimation and inference on cointegration-with
application to the demand for money. Oxford Bull. Econ. Stat. 52, 169–210 (1990)
Loayza, N., Soto, R.: Ten years of inflation targeting: design, performance, challenges. Banco
Central de Chile (2001)
Lucas, R.E.: Nobel lecture: monetary neutrality. J. Polit. Econ. 104, 661–682 (1996)
Mishkin, F.S.: From monetary targeting to inflation targeting: lessons from the industrial
countries. National Bureau of Economic Research (2000)
Mishkin, F.S.: Monetary Policy Strategy. MIT Press, Cambridge (2007)
Ncube, M., Ndou, E.: Working Paper 134-Inflation Targeting, Exchange Rate Shocks and
Output: Evidence from South Africa (2011)
Rogoff, K.: The optimal degree of commitment to an intermediate monetary target. Q. J. Econ.
100(4), 1169–1189 (1985)
Stone, M.R.: Inflation targeting lite. International Monetary Fund (2003)
Truman, E.M.: Inflation Targeting in the World Economy. Peterson Institute (2003)