“Dimitrie Cantemir” Christian University Knowledge Horizons – Economics Volume 7, No. 3, pp. 1 98–204 P-ISSN: 2069 -0932, E -ISSN: 2066 -1061 © 2015… [616988]

198

“Dimitrie Cantemir” Christian University
Knowledge Horizons – Economics
Volume 7, No. 3, pp. 1 98–204
P-ISSN: 2069 -0932, E -ISSN: 2066 -1061
© 2015 Pro Universitaria
www.orizonturi.ucdc.ro

ECONOMETRIC MODEL REGARDING THE FINANCIAL STABILITY AT THE
MACROECONOMIC LEVEL

Mirela NICULAE1, Mihaela SIMIONESCU2

1 Faculty of Finance, Banking and Accountancy, Dimitrie Cantemir Christian University, Romania, Email: [anonimizat]
2 Institute for Economic Forecasting of the Romanian Academy, E-mail: [anonimizat]

Abstract In this study, a vector -autoregression of order 2 was proposed to explain the evolution of
monetary policy interest rate and consumer index of prices, which is better correlated with the
interest rate than the GDP during 2000:Q1 -2013:Q4. According to Granger causality test for the
stationary data, at 1% level of significance the inflation rate is a cause for the interest rate.The
variation of the logarithm from interest rate in the first period is due only to the ch anges in this
variable. In the second period, 0.63% of the variation in log_ir is due to the changes in log_CPI.
The impact of the inflation increases in time, the contribution of log_cpi arriving till 5.33% in the
10th period. 41.32% of the variation in l og_cpi is due to the changes in log_ir, the influence of
this variable decreasing over time, till 20.64% in the 10th period. The stability of interest rate can
be better ensured by controlling the inflation rate and mentioning it to a stable value Key wor ds:
vector -autoregression,
interest rate, consumer
index of prices, financial
stability
JEL Codes:
C51, E40, G28

1. Introduction
International financial crisis started in 2008 has shown
that there is a need to strengthen shock resistance of a
financial system. In this respect, at the European level,
regulatory authorities and prudential supervision have
developed mechanisms and appropriate instruments,
including the dispute settlement procedures of the situation
of banks whose activity is considered that may adversely
affect the proper functioning of the system. Thus, the
mechanisms and macroprudential policy instruments
affect systemic risk mitigation and ensure financial stability.
To assess financial stability in the Romanian
economy context, we proposed an non -theoretical
econometric model – vector – auto regression model
(VAR). The variables used in the model are the price
index of consumer goods and the inter est rate. VAR
models were proposed to reveal the inter -relationship
between multiple time series. Type VAR models are
frequently used for previewing systems of time series
which are connected together, but also to analyse the
impact of dynamic innovation o n this system of
variables .

2. Literature review
During the last decades there have been three
standard strategy of monetary policy successful in
respect of the provision of effective nominal anchors, respectively monetary aggregates, the exchange rate
and inflation. Frankel (1995) suggested a strategy to be
the most suitable for savings half opened, namely
"dismissed" nominal incomes; however, a major
problem arising from this strategy is that it has not been
put into practice either in the industrialized c ountries, or
the emerging markets (Mishkin and Savastano, 2000).
The process in the context of the opening of national
economies, financial stability has become a
fundamental element of macroeconomic stability,
Having regard to that inputs of capital take advantage
of financial system vulnerabilities to penalize promptly
errors or any other measures nesustenabile on
economic policy. Jaime Caruana (2005), says that “with
all that we have a well structured framework to discuss
and implement monetary policy, o ur thinking with
respect to financial stability is less advanced.”
Haugland and Vikøren (2006) emphasize that, although
it is not very clearly what important considerations
should be given to the financial stability and price
stability, with regard to the application of monetary
policy, “both communication ss well as monetary policy
decisions indicates that financial stability is about to
hold a role better stated in the monetary policy, but may
be due to recognition that financial stability has
consequenc es for future developments at the level of
inflation and of production” (Isărescu, 2008).

Knowledge Horizons – Economics
Volume 7, No. 3, pp. 198–204, © 2015 Pro Universitaria

199 It can appreciate a direct link between the
macroprudential positive policy and financial stability,
such a transparent policy, the potency financial stability
will be solid and credible, by default by touching and
important objective of the monetary policy.
The Agreement Basel III aims to strengthen the
stability banking system, through the use of stringent
standards necessary for improving the ability to
discover to it to absorb shocks of economic and
financial sector, as well as reducing the risk of incident
from financial sector to real economy (Walter, 2010).
Reforms affect the microprudential level, with the aim of
increasing resistance individual banking institu tions to
periods of stress and, respectively, the
macroprudential, with the aim to reduce the frequency
of financial crises. New standards are designed to
improve the ability of the banking to absorb shocks, by
means of a better risk management under a
consolidated governesses and in increased
transparency conditions. (Nucu, 2011)

3. Background research
As a result of recent international financial crisis,
regulatory authorities have drawn up a new legislative
framework, known as the Regulations Basel III, which
entered into force in the year 2014. As follows:
Directive no.2013/36/UE of The European
Parliament and Council since 26.06.2013;
Regulation no. 575/2013 of The European
Parliament and Council since 26.06.2013.
These European regulations have been adjusted
through E.O. no.113/ 2013 and Regulation BNR no.5/
2013. Basically, these normative acts shall provide for:
– consolidation of the equity capital of the own
loan institutions and the assurance of the
banking ac tivity return on investment;
– more rigorous measurement of the risks in all
spheres of activity of credit institutions and
appropriate cover with provisions of the risks
undertaken;
– increase in banking staff responsibility on all the
bearings of competence, so, banking activity
should be one prudent and healthy (Bunescu,
2015).
The European Committee for systemic risk (CERS)
published in 2013 Recommendation concerning
intermediate objectives and macroprudential policy
instruments, which contributes to improving
macroprudential supervision.
In such a context, it is useful the decomposition on
the variant and the impulse response function analysis. The response -impulse function analysis shall scrutinize
the effect of a shock occurred at a given time with in one
of model innovations on present and future values of
endogenous variables. Decomposition on the variant
brings information about relative importance of each
innovation regarding the effect on the variables
dynamics of VAR model.

4. Methodology o f research
The following variables have been chosen, quarterly
data being collected over the period 2000:Q1 -2013:Q4:
monetary policy interest rate, real GDP and index of
consumer prices. The data are provided by the National
Institute of Statistics and Nat ional Bank of Romania.
The data are seasonally adjusted using moving
average method for GDP and spread and Tramo/Seats
methos for the rest of the variables.
The matrix of correlation for all the variables that
have been included in the study with seasonall y
adjusted data was computed. The objective is to
determine the variables that are more correlated with
the interest rate.
Table 1

Correlation matrix of different economic variables
during 2000:Q1 -2013:Q4

Variable IR_SA GDP_SA CPI_SA
IR_SA 1.000000 0.382600 -0.533567
GDP_SA 0.382600 1.000000 -0.842278
CPI_SA -0.533567 -0.842278 1.000000

Source: authors’ computations

As we can see from the previous table, there is a
stronger correlation between interest rate and index of
consumer prices than between interest rate and GDP.
Therefore the VAR model will be constructed with
interest rate and index of consumer prices.
The data were not stationary, being transformed as
it follows: for the consumer price index and interest rate
the logharitm was app lied.

All the lag criteria excepting LogL indicated that the
lag should be 2. For this model all the tests were
checked, resulting that the errors are independent from
the second lag, homoskedastic, following a normal
distribution. The model satisfies th e stability condition.
The results of the tests are presented in Appendix 1.

LOG_IR = 0.8518625812*LOG_IR( -1) – 0.46352629*LOG_IR( -2) + 3.817651185*LOG_CPI( -1) –
7.614673781*LOG_CPI( -2) + 0.2534705446

Knowledge Horizons – Economics
Volume 7 , No. 3, pp. 198–204, © 2015 Pro Universitaria

200

LOG_CPI = 0.011875693 42*LOG_IR( -1) + 0.0009554476811*LOG_IR( -2) + 0.4855293834*LOG_CPI( -1) +
0.4534035819*LOG_CPI( -2) – 0.003963517846

-.2-.1.0.1.2.3
1 2 3 4 5 6 7 8 9 10Response of LOG_IR to LOG_IR
-.2-.1.0.1.2.3
1 2 3 4 5 6 7 8 9 10Response of LOG_IR to LOG_CPI
-.008-.004.000.004.008
1 2 3 4 5 6 7 8 9 10Response of LOG_CPI to LOG_IR
-.008-.004.000.004.008
1 2 3 4 5 6 7 8 9 10Response of LOG_CPI to LOG_CPIResponse to Cholesky One S.D. Innovations ± 2 S.E.

Fig 1 – Impulse -response function in the VAR(1) model

Source: authors’ graph

The variation of the logarithm from interest rate in
the first period is due only to the changes in this
variable. In the second period, 0.63% of the variation in
log_ir is due to the changes in log_CPI. The impact of
the inflation increases in time, the c ontribution of log_cpi arriving till 5.33% in the 10th period. 41.32% of
the variation in log_cpi is due to the changes in log_ir,
the influence of this variable decreasing over time, till
20.64% in the 10th period.

Knowledge Horizons – Economics
Volume 7, No. 3, pp. 198–204, © 2015 Pro Universitaria

201
Table 2

Variance decomposition of the variables

Variance
Decompos
ition of
LOG_IR:

Period S.E. LOG_IR LOG_CPI
1 0.240161 100.0000 0.000000
2 0.303717 99.36997 0.630030
3 0.316368 99.16856 0.831442
4 0.318073 98.26933 1.730673
5 0.324074 96.76949 3.230514
6 0.328574 95.79273 4.207268
7 0.330063 95.22402 4.775979
8 0.330537 94.95244 5.047562
9 0.330949 94.79371 5.206294
10 0.331263 94.66920 5.330803
Variance
Decomposit
ion of
LOG_CPI:
Period S.E. LOG_IR LOG_CPI
1 0.008244 41.32874 58.67126
2 0.008800 36.37102 63.62898
3 0.009949 28.48093 71.51907
4 0.010633 26.69465 73.30535
5 0.011189 24.48150 75.51850
6 0.011523 23.20433 76.79567
7 0.011777 22.21695 77.78305
8 0.011959 21.55076 78.44924
9 0.012105 21.03522 78.96478
10 0.012221 20.64139 79.35861

Source:authors’ computations

According to Granger causality test for the
stationary data, at 1% level of significance the inflation
rate is a cause for the interest rate.

Pairwise Granger Causality Tests

Sample: 2000:1 2013:4
Lags: 2
Null Hypothesis: Obs F-Statistic Probability
LOG_CPI does not Granger Cause LOG_IR 54 3.22154 0.04848
LOG_IR does not Granger Cause LOG_CPI 4.76951 0.01281

Knowledge Horizons – Economics
Volume 7 , No. 3, pp. 198–204, © 2015 Pro Universitaria

202
Therefore, we can conclude that the stability of
interest rate can be better ensured by controlling the
inflation rate and mentioning it to a stable value.

Roots of Characteristic Polynomial
Endogenous variables: LOG_IR LOG_CPI
Exogenous variables: C
Lag specification: 1 2

Root Modulus
0.886190 0.886190
0.476394 – 0.479054i 0.675606
0.476394 + 0.479054i 0.675606
-0.501585 0.501585
No root lies outside the unit circle.
VAR satisfies the stability condition.

VAR Lag Order Selection Criteria
Endogenous variables: LOG_IR LOG_CPI
Exogenous variables: C

Sample: 2000:1 2013:4
Included observations: 54
Lag LogL LR FPE AIC SC HQ
0 128.8897 NA 3.12E -05 -4.699620 -4.625954 -4.671209
1 185.1589 106.2862 4.50E -06 -6.635514 -6.414516 -6.550284
2 202.5328 31.53045* 2.75E -06* -7.130845* -6.762514* -6.988794*
* indicates lag order selected by the criterion
LR: sequential modified LR test statistic (each test at 5% level)
FPE: Final prediction error
AIC: Akaike information criterion
SC: Schwarz information criterion
HQ: Hannan -Quinn information criterion

VAR Residual Serial Correlation LM Tests
H0: no serial correlation at lag order h

Sample: 2000:1 2013:4
Included observations: 54
Lags LM-Stat Prob
1 24.67346 0.0001
2 5.972821 0.2012
3 5.148148 0.2724
4 44.30754 0.0000
5 4.117998 0.3903
6 2.910015 0.5730
7 7.927418 0.0943
8 20.56141 0.0004
9 3.880049 0.4225
10 1.212600 0.8760
11 2.742340 0.6018
12 2.914366 0.5723

Knowledge Horizons – Economics
Volume 7, No. 3, pp. 198–204, © 2015 Pro Universitaria

203 Probs from chi -square with 4 df.

VAR Residual Normality Tests
Orthogonalization: Cholesky (Lutkepohl)
H0: residuals are multivariate normal

Sample: 2000:1 2013:4
Included observations: 54

Component Skewness Chi-sq df Prob.
1 -0.090449 0.073629 1 0.7861
2 -0.151728 0.207192 1 0.6490
Joint 0.280821 2 0.8690

Component Kurtosis Chi-sq df Prob.
1 3.324188 0.236470 1 0.6268
2 2.552937 0.449697 1 0.5025
Joint 0.686168 2 0.7096

Component Jarque -Bera df Prob.

1 0.310100 2 0.8564
2 0.656889 2 0.7200

Joint 0.966989 4 0.9148

5. Conclusions
The construction of VAR model for interest rate and
consumer index of prices is an example of econometric
model for identifying the measures for having a financial
stability. According to Granger causality test for the
stationary data, at 1% level of sign ificance the inflation
rate is a cause for the interest rate. However, the
correlation between GDP and interest rate is quite low
and the policies should be oriented to ensure a low and
stable inflation rate.

References
1. Albulescu, C.T., (2012), Stabilit atea financiară,
politica monetară și coordonarea bugetară în
UEM, Economie teoretică și aplicată, Volumul
XIX (2012), No. 8(573), pp. 77 -88;
2. Baltagi, B.H., (2008), Econometric Analysis of
Panel Data, vol.1, John Wiley & Sons Ltd;
3. Baltagi, B.H., Demetriad es , P. O., & Law, S. H.,
(2009), Financial development and openness:
Evidence from panel data. Journal of
development economics, 89 ( 2), 285 -296; 4. Baum, C.F., (2001), Residual diagnostics for
cross -section time series regression models;
The Stata Journal, Vol. 1, No. 1, pp. 101 –104;
5. Bunescu, P., (2015), Activitatea bancară –
Context, Reglementare și Adaptabilitate,
disponibil la
www.bnro.ro/files/d/Evenimente/2015/R2015021
9PB.pdf ;
6. Campbell, J.Y., Lo, A. W. C., & MacKinlay, A. C.
(1997). The econometrics of financial markets
(Vol. 2). Princeton, NJ: Princeton University
Press;
7. Isărescu, M., (2008), Probleme ale politicii
monetare într -o țară emergentă. Cazul
României, Publicațiile Academiei Regale de
Științe Economice și Financiare, pp. 1 -55;
8. Nucu, A.E., ( 2011), Provocările Basel III pentru
sistemul bancar românesc, Economie teoretică
și aplicată, Volumul XVIII, No. 12(565), pp. 57 –
68;
9. Tomuleasa, I.I., (2013), Macroprudențialitatea și
stabilitatea financiară, Management Intercultural,
Volumul XV, Nr. 3 (29) , pp. 353 -365;

Knowledge Horizons – Economics
Volume 7 , No. 3, pp. 198–204, © 2015 Pro Universitaria

204
10. Triandafil, C.M., (2012), Studii economice.
Politica macro -prudențială și procesul de
convergență la nivelul Uniunii Europene,
Academia Romana, Institutul National de
Cercetari Economice, pp. 1 -22;
11. *** – Implementarea politicii monetare în zona
euro. Documentația generală privind
instrumentele și procedurile de politică monetară
ale eurosistemului aplicabilă de la 1 ianuarie
2012, Banca Centrală Europeană, Jurnalul
Oficial al Uniunii Europene 1 4.12.2011