REVIEW
of
FINANCE -
November, 2018
27
Research model and research method
Proposed research model
A series of studies have identified the factors
that influence the NPL which is conducted by
Messai and Jouini (2013); Louzis et al. (2010);
Klein (2013); Boudriga (2009). There are two
essential groups of factors that affect NPL: one
group insists of macroeconomic factors and
the other covers bank-specific factors. Klein
(2013) studied factors influencing NPL such as
macroeconomic factors and bank-specific factors
for Central, Eastern and Southeastern Europe in
the period between 1998 and 2011. Messai and
Jouini (2013) analyzed the factors affecting NPL
for banks in Italy, Greece and Spain with 85
banks in the period from 2004 to 2008 which took
into account both macro and micro variables. In
line with Klein’s research (2013), they concluded
that GDP growth, as well as bank profits, was
negatively correlated to NPL. Unemployment,
provisions for outstanding loans and real
interest rates have a positive correlation with
NPL (Messai and Jouini, 2013). Louzis et al.
(2010) investigated the same problem in Greek
banks from 2003 to 2009. Their results showed
that GDP growth, unemployment and lending
rates affected NPL to a great extent (Quadt and
Nguyen, 2016).
In the study conducted by Baholli et al. (2015),
the credit risk model showed the difference in
Albania’s NPL with Italy to help analyze the main
macroeconomic factors that affected NPL in both
countries. The independent variables used in the
theory were considered important to explain their
necessity for the model. Modeling the behavior
of NPL in macroeconomic indicators was used
to compare the Albania and Italy economies.
Situational problems were assumed to test the level
of NPL in the two countries. The model for Albania
was formed with the dependent variable NPL,
while the four independent variables were GDP,
interest rates, inflation, and real exchange rates. For
Italy, the model explained the transformation of
NPL to Italy. Both models explaining the behavior
of NPL inAlbania and Italy showed that the pattern
was statistically significant at the significance level
of less than 1%. At the same time, the explanatory
model of at NPL in Italy was around 99% and 88%
for Albania.
The study by Zribi and Boujelbene (2011)
examined the determinants of CRM in banks in
Tunisia which were referred to as an emerging
country. The model in the paper included the
dependent variable which was credit risk.
Meanwhile, the independent variables consisted
of (1) the matrix of variables belonging to the
bank’s characteristics as ownership characteristics
measured by dummy variables; (2) regulation for
banks: this variable was measured by the dummy
variable. If the banks complied with CAR of Basel
II 8%, their value equaled to 1, the remainder was
0; (3) matrix of macro factors; and (4) the size of the
bank.
In the research conducted by Louzis et al. (2010)
in the Greek banking sector, they used a tabular
data method to examine the factors that influence
NPL for each type of loan.
Godlewski (2004) used ROA as the performance
index. He showed that the impact of bank profits
was the opposite of the level of NPL rates. However,
when using 129 banks in Spain for the period
from 1993 to 2000, Garciya-Marco and Robles-
Fernandez (2008) quoted in the study of Mesai and
Jouini (2013) showed that the higher equity (ROE)
was, the higher risk was. They argued that profit
maximization policies were accompanied by high
levels of risks.
According to an analysis based on previous
researchers’ models of factors affecting bad debt
ratios, the authors establish a quantitative model
with a dependent variable of NPL ratio and
variables which are suitable to the commercial
banks in Vietnam as in the model below:
NPLit = β1+ β2 * CARit + β3 * SIZEit + β4 *
ROAit + β5 * ROEit + β6 * Total loan to total
deposits ratioit + β7 * Provision ratioit + β8 *
Leverage ratioit + β9 * Quick ratioit + ε
NPL = NPL measured by non-performing loan
ratio (NPLR)
CAR = Capital Adequacy Ratio
SIZE = bank size = BANK_SIZE
ROA = Efficiency of bank operation
ROE = Efficiency of bank operation
Total loan to total deposits ratio
Provision ratio
Leverage ratio
Liquidity ratio (Quick ratito)
