Page 59 - Special Topic Session (STS) - Volume 2

STS459 Gan C.P. et al.

The effects of macroeconomic variables on future

credit ratings
Gan Chew Peng ; Pooi Ah Hin ; Ng Kok Haur
1
1
2
1 School of Mathematical Sciences, Sunway University, Malaysia.
2 Institute of Mathematical Sciences, Faculty of Science, University of Malaya, Malaysia.

Abstract
The method based on the multivariate power-normal distribution is used to
analyse the data on credit ratings and macroeconomic variables. Initially the
macroeconomic variable is incorporated, one at a time, into the non-
Markovian model which attempts to predict the credit rating in the next
quarter when the ratings in the present and previous quarters are given. The
effects of macroeconomic variables on future credit rating are found to agree
basically with the corresponding results reported so far in the literature. We
next summarize the effects of the set of macroeconomic variables by a small
number of latent factors, and include the latent factors into the non-Markovian
model. It is found that a small number of latent factors is sufficient to release
the explanatory power of the macroeconomic variables in improving the
estimation of the transition probability of the future credit rating.

Keywords
Credit ratings; Macroeconomic variables; Non-Markovian model

1. Introduction
A credit rating is an evaluation of the credit risk of a prospective debtor,
predicting their ability to pay back the debt, and an implicit forecast of the
likelihood of the debtor defaulting (Kronwald, 2009). Credit evaluation for
companies and governments is generally done by a credit rating agency such
as Standard & Poor’s (S&P), Moody’s, or Fitch. These rating agencies are paid
by the entity that is seeking a credit rating for itself or for one of its debt issues.
Macroeconomic variables play a vital role in determining the rating for a
company’s credit rating. Banks’ credit risks are assumed to be affected by
macroeconomic variables, that is, when the macroeconomic conditions have
been improved, the credit risk will be reduced. Several models have been
developed and the results support the assumption.
Alves (2005), and Shahnazarian and Asberg-Sommer (2008) are the
pioneers in the analysis of the relation between default probabilities and
macroeconomic factors using Vector Autoregressive approach. They found
short-term interest rates, economic growth and inflation to be the variables
with significant effects on default frequencies. Distinguin, Raus and Tarazi
(2006) found that market data helped to obtain more reliable default
48 | I S I W S C 2 0 1 9