DBS (Development Bank of Singapore)

[donmihaihai]

Hi donmihaihai,
A bank capital structure is broadly classified into Tier 1 and Tier 2. 
Under Tier 1 capital, the components are Common Equity Tier 1 (CET1) and Additional Tier 1 (AT1).
CET1 consists mainly of common shares equity plus retained earnings and minority interests.
AT1 consists of perpetual bonds with no maturity.
Under Tier 2 capital, it consists of unsecured bullet bonds of max maturity of 5 years.
If a bank collapses, Tier 1 capital investors would be wiped out first, followed by Tier 2 investors, lastly the depositors.
A bank would have to incur cost for its capital. Eg, coupons for bonds issued. DBS does not earn interest income from its $11b excess capital. On the contrary, it needs to pay money for them. DBS bonds and perpetual bonds coupon ranges from 3% to 5+% depending on the denominated currency. However, a ball park 4% funding cost for AT1 and Tier 2 bonds is reasonable.
Henceforth, if DBS was to return all excess $11b capital, it would save $500m of funding cost annually which would boost bottomline. This $500m savings is not an one-off but is recurring in nature. For context, that would be a 5% earnings growth on a $10b profit. 

As of 31 December 2024, shareholders funds and total equity around 62B, CET1 capital @ 54B, Tier 1 capital @ 56B and total capital(Tier 1 & 2) @ 59B. I am not sure where to look at based on your comments. I am not an expert on bank, just leave it until I have a better understanding.
As for there is no income for those excess capital, since assets = equity +liabilities, any capital retained or borrowed will has the same amount of cash sitting on the asset side or when the cash is being used to generate income, turned into productive assets other than cash. Look at the assets side of DBS B/S, I got to assume even a reasonable run bank would not has cash sitting around not generating income. So pretty sure a "world class" bank like DBS would do better.
So, if the excess is from capital side, any reduction means reduction of income and if the excess is from the liabilities side, then reduction of liabilities will result in reduce income as well.

Hi donmihaihai,

Please take a look at asset side of B/S.

First item "Cash and balances with central banks - $50b". Out of this, some cash earns no income at all. Some of it earns just a little interest from central bank. This is liquid cash. 

Second item "Government securities and Tbills - $70b". This will earn income from SG bonds and Tbill rates.

First and second items are components of High Quality Liquid Assets (HQLA). Other components are highly rated corporate bonds and covered bonds. HQLA is used to calculate Liquidity Coverage Ratio (LCR). DBS LCR for all-currency and SGD are 144% and 297% respectively, well above the regulatory minimum requirement of 100% for both all-currency and SGD.

Assume DBS has $11b excess capital and wish to return them over a period of 5 years. That would be $2.2b per year. It could simply use the non-yielding money from "Cash" rather than yielding money from "Government securities and Tbills" or other yielding items from assets side of B/S. For perspective, $2.2b is just a fraction of DBS HQLA and would hardly make a bent on LCR.

DBS could use $2.2b non-yielding money from "Cash" to redeem expensive AT1 bonds from liabilities side. Thus assets and liabilities sides offset each other in B/S. Assume AT1 bonds coupon is 4%, this would save DBS $88m every year and bump up bottomline and ROE.

Alternatively, DBS could return $2.2b as dividend. "Cash" and "Equity aka Shareholders' Funds" offset each other in B/S. This would bump up ROE even more than the example stated above. However, the downside is Book Value would drop as well, PTB would rise. DBS could also do a mix of both, return capital from Equity and redeem expensive AT1 bonds concurrently. 

Maybe I have mistaken you but I don't really understand why you insist if DBS return capital to shareholders or redeem AT1 bonds, DBS would make a loss and impact profits.

As what you have said, "a world class bank like DBS would do better". I concur.

It would impact profits, but I never say DBS would be loss making. Read back. 

When I wrote cash siting around not generating profit, I meant excess cash, not just cash. Bank operates with a certain level of cash, but these is not excess cash. These are for operating requirements. If there weren't, then try use your digibank transfer money out, you would trigger a run straight away. Also, Piyush Gupta salary has to be paid in cash, not some securities. There is also regulator requirement to maintain certain level of cash, if i am not wrong. 

I don't know how much cash DBS needs for these 2 reasons but I assumed any reasonably run bank would know what to do, let alone a "world class" bank. Or you mean DBS is so profitable that they can afford sitting on excess cash for years waiting to return it to shareholder. Maybe, but all DBS shareholders would hope you are wrong.  

I agree with you that if there is excess capital, it will be in those short term securities.

[jfc18]

Hi donmihaihai,

Thank you for your sharing.

Just to clarify, what you meant is DBS must keep sufficient level of cash for operation purposes (which I totally agree), and any form of capital return to shareholders must be from the excess cash which should yield some income, eg 3% of interest income from cash parked in central bank.

Let’s use the same $2.2b capital return example for the below two scenarios.

According to you:

1) If DBS uses $2.2b cash to redeem 4% coupon AT1 bond, the net saving is only 1% since we need to offset the loss of 3% interest income. Therefore in this scenario DBS profit is "impacted" as it is only up a marginal $22m from 1% net cost saving of AT1 bond coupon instead of the full 4% at $88m.

2) If DBS uses $2.2b cash as special dividend payout to shareholder, this capital return from equity is in fact “loss making”. Because DBS would lose $2.2B x 3% of interest income. All thing else being equal, DBS profit would be down $66m if we make a “before and after capital return” comparison. Therefore in this scenario, it is “loss making” because it brings down profit.

Did I interpret you wrongly on the above?

The essence of your argument is: if DBS conduct a capital return exercise, the money used must come from excess cash which currently yields some interest income. Therefore the loss of this interest income will “impact profit" in scenario 1, and “loss making” of $66m in scenario 2.

Just want to make sure I get you correctly and avoid confusion.

Thank you.

[jfc18]

After a breakfast with Jack Ma in 2014, Piyush Gupta was convinced big tech would eventually take over banking sector. He thought its better to disrupte ourselves now rather to be disrupted by others later. Hence in 2014, DBS digital transformation took off. It even came out with an quirky techy acronym, GANDALF, from Lord of the Rings.

G - Google
A - Amazon
N - Netflix
D - DBS
A - Apple
L - LinkedIn
F - Facebook

Piyush Gupta wants DBS to be in the middle of big techs. Now this shift is of paramount importance. With GANDALF, DBS reference is not the JPMs or Citis anymore. He has set his sights firmly on techs. In order to survive and thrive, DBS most be digitalised to the core. Since 2014, approx $1b is spent on tech yearly. Back then 80% of tech ops is outsourced. Now most tech ops are in house. DBS currently has more technologists than bankers in its organisation. 

To have a deeper understanding how DBS digital transformation has created greater shareholder value. Please to go DBS Investor Day 2023 website, https://www.dbs.com/investorday/index.html. There are dozens of slides and videos which are extremely informative.

hi jfc18,

You sound like an IR officer from DBS.

Jokes aside.

I am generally sceptic when a bank talks about digital transformation (whatever it means) but its basic online banking services break down. It was like SGX talking about digital transformation but its equity/derivative trading broke down some time back. Nonetheless, I believe both companies may have realized that fail-safe backend is more transformative than all the fanciful features we see on their mobile app or the various ML (whatever it means) they are executing.

Putting aside my skepticism, it is a fact that DBS is in the middle of a "transition". I define "transition" here as companies carrying out big value-accretive moves - like Keppel/Capitaland Investment wanting to double their AUM (and reduce their capital footprint) or hospitality firms like Ascott/Mandarin Oriental focusing their business models from ownership to mgt contracts/franchise. In DBS case, they are determined to have higher/maintain ROEs by returning capital (especially the excess ones). And when executed right, Mr Market rightfully recognizes it and gives it a premium valuation. After all, Mr Market weighs things correctly after some time.

Hi weijian,

I am NOT a DBS IR Officer lah!  

In my first post, I have already said I own DBS shares and my views may be biased. However, I also own shares of OCBC, and UOB through Haw Par. Since this thread is about DBS, I just keep my writings on DBS and try to be objective.

DBS digital transformation is quite something. It has helped to win more deposits, lower cost to income ratio, lower credit cost, scale up high ROE businesses, improve margins, and many more. The transformation has yielded impressive tangible results. Investors should cut through the fog and look past the few sporadic banking system outrages. 

I know I risk of sounding like a DBS IR Officer again. Apart from winning many "World's Best" awards from renowned financial publications, DBS transformation journey has also captured the attention of academic institutions. Harvard Business School has made DBS digital transformation into a case study and is being taught in its MBA programme. INSEAD, IESE, IMB, SMU and NTU have also done the same.

Most importantly, DBS ROE and ROA both rank highly among bigger global peers. Zoom out and you will see we really do have a world class bank in our own backyard.

[donmihaihai]

Hi donmihaihai,

The essence of your argument is: if DBS conduct a capital return exercise, the money used must come from excess cash which currently yields some interest income. 

Just want to make sure I get you correctly and avoid confusion.

Thank you.

Yes and only this. 

My views where the excess capital at liabilities or equity is different from you. My views on technology and how good is DBS is different from you but I am keeping them to myself.

[jfc18]

Hi donmihaihai,

The essence of your argument is: if DBS conduct a capital return exercise, the money used must come from excess cash which currently yields some interest income. 

Just want to make sure I get you correctly and avoid confusion.

Thank you.

Yes and only this. 

My views where the excess capital at liabilities or equity is different from you. My views on technology and how good is DBS is different from you but I am keeping them to myself.

Hi donmihaihai,

Thanks for your clarification. Now I understand the fundamental difference between our assessments.

My stand is DBS does not need to use excess cash which yields interest income to fund capital return. It could simply use liquid cash which yields no interest income to do it. DBS net profit is $2.5b per quarter. To fund a $2.2b capital return via dividend or redeem AT1 bond, DBS could just use the most recent one quarter cash earnings to cover it. Hypothetically if DBS announces capital return of $2.2b during Q2 result, this $2.2b would come from cash earnings from Q2. There would be no impact on Q1 vs Q2 balance sheet and profit.

There would be no loss of interest income from assets side at all. Excess cash is untouched. Balance sheet asset side numbers remain the same. Profit would not be "impacted" or "decreased" in a before and after capital return comparison. On the contrary, net profit and ROE would go up. DBS CET1 and LCR are both well above regulatory requirements, it does not need to keep recent cash earnings to beef up capital.

Your assumption that DBS must use excess cash to fund capital return is inherently flawed as it does not take into account ongoing profits generated by the bank. Or you mean DBS is going to be loss making in coming quarters and therefore must use excess cash to fund capital return? To borrow from your exact own words, "Maybe, but all DBS shareholders would hope you are wrong".

[donmihaihai]

Hi jfc18

Not going to response to any further by dingdonging after this and I will be using strong words.

Your assumption that DBS must use excess cash to fund capital return is inherently flawed as it does not take into account ongoing profits generated by the bank. Or you mean DBS is going to be loss making in coming quarters and therefore must use excess cash to fund capital return? To borrow from your exact own words, "Maybe, but all DBS shareholders would hope you are wrong".

Stop putting words in my mouth. I wrote impact profit not loss making. If you don't know the difference, I can't help that.

And yeah, why talk about future profit when DBS is having excess capital right now. laughable.

[jfc18]

Hi jfc18

Not going to response to any further by dingdonging after this and I will be using strong words.

Your assumption that DBS must use excess cash to fund capital return is inherently flawed as it does not take into account ongoing profits generated by the bank. Or you mean DBS is going to be loss making in coming quarters and therefore must use excess cash to fund capital return? To borrow from your exact own words, "Maybe, but all DBS shareholders would hope you are wrong".

Stop putting words in my mouth. I wrote impact profit not loss making. If you don't know the difference, I can't help that.

And yeah, why talk about future profit when DBS is having excess capital right now. laughable.

Hi donmihaihai,

Chill lah. My choice of words is a reflection of what you wrote. You may read back and cross check. Just tongue in cheek responses similar to yours. Please don’t be offended. 

My point is simple. If DBS profit is robust and can cover capital return, it just need to use recent quarter cash earnings to pay dividend announced or redeem AT1 bond during that quarter. It does not need to use excess cash at all. There is no loss of interest income and assets on balance sheet. Profit QoQ will not be impacted as what you have claimed. Only a loss making company or profit less than dividend announced would need to dip into its capital to fund the payout. 

Eg, John plans to spend $10k on a holiday next month August. His monthly net income after expenses is $15k. Does John need to withdraw his fix deposit or sell shares in his portfolio to fund the $10k holiday next month? Laughable.

[weijian]

So, DBS is now a tech company and not a bank? Or it is a lender led by tech (sounds sexier this way)? I think users of DBS services would just hope that outages does not happen anymore.

ChatGPT told me that HBS case study is not paid-research and their articles are behind a paywall. But this does feel like an opportune way to cement the outgoing CEO's legacy

DBS has twice as many technologists as bankers in 2023: Harvard Business School study

Currently, DBS deploys more than 800 AI models across 350 use cases, said Gupta, who expects the measured economic impact of these models to exceed S$1 billion in 2025.

https://www.businesstimes.com.sg/compani...hool-study