Page 260 - Special Topic Session (STS) - Volume 3

STS543 Veronica B. B. et al.
(34) measures were classified as being neutral, largely pertain to changes in
reportorial requirements. On balance, the BSP implemented more tightening
than loosening measures. In particular, majority of the tightening measures
were capital- and liquidity -related measures for Basel III compliance while
those of loosening measures were for currency-related instruments and were
implemented in connection to the liberalization of the BSP’s foreign exchange
framework starting in 2007. Similarly, there were more resilience-based
instruments (at 56.3%of the total instruments) adopted compared with
cyclical-based instruments (at 43.7%) from the first quarter of 2014 to the
fourth quarter of 2017. Of the total measures adopted, 44.3% were tightening
measures, 41.8% were loosening and 13.9% were neutral measures.
Measures of monetary policy actions. This database compiles and updates
monetary policy actions by the BSP based on Bayangos (2017) database to
include Term Deposit Facility (TDF) rates under the Interest Rate Corridor (IRC)
system introduced in June 2016. This database is an index of both tightening
and loosening policy actions using a four-quarter window based on the
estimates. Similar to previous specifications, for each of the central bank
official policy rate, a dummy variable is assigned to a value of positive one (1)
if the hike in policy rate is accompanied by a rise in TDF rates; hence, the
monetary policy stance is tight; 0, otherwise, or when the reduction in policy
rate is accompanied by a drop in TDF rates; hence, the monetary policy stance
is loose. The database also includes a measure of the intensity of monetary
policy actions by considering the number of times a policy is implemented.
Taking the average of these measures is also used.
Vector of controls. This dataset includes macro-financial indicators and
bank-specific characteristics used in the study. These include changes in real
Gross Domestic Product (GDP), inflation, real overseas Filipino remittances,
monetary policy rate, TDF rate, bank lending rate, neutral rate of interest rate,
output gap, bank credit to GDP ratio gap, nominal peso-dollar rate, real
effective exchange rates. The bank-specific characteristics in the dataset
include the size of a bank (or total resources in real terms), liquidity ratio
defined as liquid assets relative to total assets, capital ratios using capital
adequacy ratio and Common Equity Tier 1 ratio to total assets, funding
composition using outstanding deposits relative to total liabilities, profitability
of banks using real net interest income, and quality of bank loans using
nonperforming loans, non-performing assets and non-performing coverage
ratio.
Estimation method. In this study, the parameters in the models are
estimated using unbalanced panel Generalized Method of Moment (GMM)
that is a more appropriate empirical methodology to address the endogeneity
between real bank loan commitments and non-performing loans with bank-
specific characteristics and macroeconomic indicators. To handle cross-section

249 | I S I W S C 2 0 1 9