DBS GROUP HOLDINGS LTD (SGX:D05)
OVERSEA-CHINESE BANKING CORP (SGX:O39)
UNITED OVERSEAS BANK LTD (SGX:U11)
Singapore Banks - Artificial Relationships
What matters most for NPLs is not what you think
- For banking cycles after the GFC, we have trained MKE’s proprietary machine learning algorithms to identify the key variables that affect Singapore banks’ NPLs. From our analysis, it appears that the pace of non-SGD loan growth, domestic inflation and the rate of change in special-mention loans have the strongest influence in setting the direction of NPLs.
- In contrast, interest rates, changes to unemployment and GDP growth – variables that traditionally guide NPL forecasts – seem to have a lesser impact.
- With overseas lending accounting for a large share of loan growth in the past three years, we think NPL risks may heighten as the cycle turns. That said, strong capital and good dividend visibility should provide buffers.
- Remain Positive on Singapore banks, with UOB (SGX:U11) top pick
NPLs: non-SGD loans important, GDP not so much
- We use supervised machine learning algorithms – similar to those employed by social-media and video-streaming services for recommendations – to identify variables that have the highest influence on banking NPLs. Our basket of 17 variables spans micro, macroeconomics and sector drivers.
- Our results indicate that ACU or non-SGD loan growth - particularly business loans - is a key indicator of NPL growth. CPI, changes in special-mention loans as well as changes in property prices also show correlations. Traditional indicators such as GDP, unemployment and changes in SIBOR have weaker correlations.
Credit-charge risks need to be watched
- In the past three years, 63% of incremental loan growth for Singapore banks was fuelled by overseas lending. This implies that increased non- SGD exposure raises asset-quality risks.
- While we forecast 2019-20E credit charges of 25bps/29bps, up from 16bps in 2018, risks are thus likely on the upside. As at 2Q19, OCBC (SGX:O39) had the highest non-SGD loan exposure at 65% of its loans while DBS (SGX:D05) had 60%. UOB had the lowest 53%.
Prefer UOB
- NPL risks notwithstanding, Singapore banks offer some of the highest dividends in the region, together with strong balance sheets. We maintain our POSITIVE outlook.
- UOB’s historically conservative balance-sheet management, high provision coverage and 13.9% CET1 should provide buffers against volatility. Coupled with > 5% dividend yields, UOB is our top pick.
A little help from AI
- We use MKE proprietary supervised machine learning algorithms to find variables that have the highest influence on Singapore banking NPLs. The technique used is Random Forest, which examines correlations through multiple dimensions. Variations of this technique have been used in recommendation engines in social media, video streaming and e-commerce as well as image-classification and fraud-detection systems.
- For our exercise, we ‘trained’ the machine learning algorithm using a basket of 17 variables, starting from 1Q09. Our variables span micro, macroeconomics:
- GDP Growth
- Singapore CPI
- Unemployment
- Residential property price index
- US long term-short term yield spreads
- SIBOR
- DBU SGD Deposits
- DBU non-SGD Deposits
- DBU System Loans
- DBU System business loans
- DBU System consumer loans
- ACU System Loans
- ACU System business loans
- ACU System consumer loans
- Loan utilisation levels
- Loans classified as special mention
- Our results indicate that ACU or non-SGD loan growth - particularly business loans - is a key indicator of NPL growth. CPI, changes in special-mention loans as well as changes in property prices also show correlations. Traditional indicators such as GDP, unemployment and changes in SIBOR have weaker correlations.
- See attached PDF report for detailed analysis.
Thilan Wickramasinghe
Maybank Kim Eng Research
|
https://www.maybank-ke.com.sg/
2019-09-11
SGX Stock
Analyst Report
28.050
SAME
28.050
11.050
SAME
11.050
29.130
SAME
29.130
