India vs China - Banks

For this analysis, I will be looking at only the top 5 listed banks in each country. 

 China

The total assets of the Chinese banking industry are ~ USD 49 tn and system wide non-performing loans (NPLs) are at ~1.8%. Classication of NPLs is done using the Expected Credit Loss (ECL) model in IFRS accounting. Loan loss allowance is stipulated at 150% of NPLs. 

There are 23 banks greater than USD 10 bn market cap. For analysis, I will only be looking at the top 5 banks by market cap, which are Industrial and Commercial Bank of China (ICBC), China Merchants Bank (CMB), China Construction Bank (CCB), Agricultural Bank of China (ABC), and Bank of China (BOC). All but CMB are government owned. They constitute ~38% of total banking assets and ~ 54% of total banking profits.

I have summarised the key financial and business metric in the table below (figures in USD bn except avg employee salary that is in USD). 

India

The total assets of the Indian banking industry are ~ USD 2.7 tn and system wide non-performing loans (NPLs) plus restructured assets are at ~7.5%. Classication of NPLs is done using the days past due (dpd) model with loans overdue for more than 90 days being declared non-performing and provisioning starting from 15% and changing with time and collateral. 

There are only 6 banks greater than USD 10 bn market cap. For analysis, I will only be looking at the top 5 banks by market cap, which are HDFC Bank (HDFC), ICICI Bank (ICICI), State Bank of India (SBI), Kotak Mahindra Bank (KMB), and Axis Bank (Axis). All but SBI are privately owned. They constitute ~ 49% of total banking assets and ~89% of total banking profits (shows the absolutely pathetic performance of the smaller banks). 

I have summarised the key financial and business metric in the table below (figures in USD bn except avg employee salary that is in USD). 

The NPLs look higher (and conversely the provisions are also lower) as Indian banks are not yet following IFRS whereas Chinese banks follow IFRS, which gives the management flexibility in accounting for bad loans, whereas in India it is fixed model without much flexibility. Although the NPLs in China look very low 

1. The reporting by Chinese banks is much better and more standardised with relevant data than reporting by Indian banks, which is not standardised and all over the place; to be honest the Indian banks annual reports look more like art projects than financial documents
2. A key observation across both countries is that retail focused banks have done better than corporate focused banks and get a higher valuation multiple (no surprise)
3. Indian banking sector is even more consolidated than the Chinese banking sector
4. Government banks get lower valuation multiples as they are used as a vehicle for impleneting government projects, though government banks in China are at least in a leading position whereas in India they seem to have lost their way. For example in terms of assets there are 2 other government owned banks in India that are bigger than KMB and Axis but their valuation (performance has not been great either) has put them out of the list.
5. Government banks in China seem to be doing much better than their Indian counterparts and holding their own against private competition (on an unrelated note the government control of the banking sector has been shown to be one of the factors for rapid economic growth in Asia in How Asia Works by Joe Studwell. Review at https://www.goodreads.com/review/show/2596750911?book_show_action=false&from_review_page=1). It is a great book and a must read.
6. Chinese bank shares look to be trading quite cheap even accounting for the lower metrics
7. The Chinese economy is highly leveraged, even after accounting for the 5x differential in GDP between India and China
8. Banks in China operate at roughly half the NIM of Indian banks, which is a big difference which accounts for the lower RoE's but it is offset by lower Cost-Income ratios

I would like to just focus on 3 things among all the numbers that are in the table. Ignoring Other Income % as it can be due to different lines of business of the Bank such as insurance, asset management, custodian services, etc. For example CMB has a high percentage of fee income from Trust services which other banks may or may not have.

1. Lower NIM in China than India
• One major reason could be the higher allocation to corporate loans in China, which generally have lower margin than retail loans (for example CMB, which has a higher allocation to retail loans has the highest NIM)
• The government has a big say in the system and since the government owned banks are still dominant they are the benchmark for all competitors
2. Big difference in Cost-Income
• Again to some extent linked to the first point for the difference in NIMs, higher allocation to corporate loans would also have lower Cost-Income due to it being a more head office oriented work than branch oriented work
• Bigger balance sheet size would also give scale advantages and that could also play a big role; avg employee costs are ~ 3x of India in China and the banks have slightly less than double SBI's (it is the worst bank among all in terms of cost) employee count and >5x the balance sheet of SBI, so they must be really efficient in the non-employee costs (employee costs are broadly ~ 50-60% of operating expenses)
3. Low valuations for Chinese banks
• At first glance they definitely look like value buys even though bad loans may be hidden and real NPL numbers may be higher (as has been mentioned in numerous articles)
• Given their size, highly likely loan growth will be slower than in India
• Barring CMB, 5-yr PAT growth has lagged behind asset growth, which has compressed RoE's; Barring CMB, Cost-Income ratio has fallen for the other 4 banks and NIMs and NPLs have also been stable, so this means that the provision coverage has gone up and thats why PAT growth has lagged asset growth 
• Barring CMB, the other 4 banks trade at ~ 6% dividend yield, basically they are valued just on their annual cash flows like a bond and are getting no value for future growth
• Another possible issue could be related to loans to bad banks who in turn buy the bad loans from the banks thereby technically removing the NPLs from the banks books; for example the government set up China Huarong Asset Management Company (Huarong) and initially issued bonds in 2001 for a period of 10 years but the maturity of the bonds has been extended twice in 2011 and 2021 and the current maturiy is 2031. ICBC has a USD 14 bn exposure to Huarong (~ 3% of networth) and the other state run banks would also have exposure to bad banks bonds and this could be a source of stress; Huarong was also in the news recently as it bond prices had fallen and it was at risk of defaulting on its bonds though nothing like that has happened till now; but it is always a risk