What's new

Basel III questions

Thread starter #1
Hi David,

I'm sorry I posted my previous question under operational when it should've been here.

Here's a sentence from Basel III that made me doubt my understanding of securitization :

Regulatory adjustments may also be made to deal with increases in equity resulting from securitization...
Would you please explain/ give an example of a case where securitization increases equity.

Thank you

David Harper CFA FRM

David Harper CFA FRM
Staff member
Hi Hend,

Can you link to the document, sorry I don't recognize the sentence?

Although B2 tends to refer to case of the bank as originator (because by securitization, the bank is transferring assets [exposures] to the SPE). There is a distinction between cash and synthetic securitization, but in a cash securitization, the bank is reducing its balance sheet by selling assets to the SPE; the SPE issues notes to the investors. So, there are (at least) two aspects in which "equity" can be increased:
  • Reduction of the balance sheet: if a bank funds $100 in assets with $90 in liabilties, shareholder equity = 10%. Say the bank needs to maintain equity = 10%. Say bank securitizes $20 in risky assets; sells the assets to the SPE. The SPE has a balance sheet: $20 in risky assets = $20 notes to investors. The $20 received by investors (- fees) pays the bank for the "true sale." Bank uses cash to payoff $20 in liabilities; i.e., funds that supported the assets. Bank balance sheet reduces to $80 assets and $70 liabilities, but 10% equity requires on only $8 in equity. $10 equity is the same, but as an equity requirement bank is +$2
  • There is another possibility: if securization can convert FUTURE cash flows, not booked on the balance sheet, it can create equity by monetizing them (i.e., by converting some FV component into PV cash). For example, if i have a loan out to you, but per accounting I can only book X because Y is uncertain FV, but i can securitize into cash of C, the C-X could be new cash/equity on the B/S.
  • But that is only part of it. To the extent we are referring to economic/regulatory capital (I realize your quote says "equity"), if a securitization can convert risky assets to less risky cash, the risk-weighted exposure drops. This is the classic "regulatory arbitrage" where capital is liberated by using securitization to re-brand the risk weight of an exposure.
Maybe a start ... I don't think i've listed all the possibilities, I think i am missing one ....

David Harper CFA FRM

David Harper CFA FRM
Staff member
Hi Hend, yes I'm not familiar with FMI but the gain-on-sale appears like a variation on #2; e.g., even if the the bank can book (as assets) a future stream of receivables, they would need to deduct provisions for losses, so at best the booked asset would be a "net of" some loss estimate; but securitization could convert that into more cash with a different (lower) loss assumption, and I could see how the securitization could great a gain just by the adjustment in loss assumption. Just musing ... thanks,
hi david, hi hend

talking about securitization, i'm a little bit confused about it, if a bank securitize some of it's assets to SPE in a true sale process, then why it is still receive a risk weights?
since it is no more belong to the bank?
Thread starter #7
Good question, if it does, but I doubt it in true sale, I think it applies in the case of selling to an owned trust.

I'll go research to make sure.

Thanks for bringing it up Ibrahim.
Thread starter #8
This is under Basel II readings,
http://www.garpdigitallibrary.org/download/FRM/free/2012/2012OP Basel_II_Accord.pdf p136

Operational requirements for traditional securitisations:
554. An originating bank may exclude securitised exposures from the calculation of riskweighted
assets only if all of the following conditions have been met. Banks meeting these
conditions must still hold regulatory capital against any securitisation exposures they retain.
(a) Significant credit risk associated with the securitised exposures has been
transferred to third parties....
After this, I think question remains about the part in red. What regulatory capital against securitisation exposures?
You really have to read the whole text. I only read the notes on it.

David Harper CFA FRM

David Harper CFA FRM
Staff member
I agree the red refers to exposures that are not securitized (I'm not current on its status but you've probably read about the controversial want of Dodd Frank to require 5% retention, see section D here of Title IX). A true sale of an entire asset pool, naturally, removes the assets from the originator's balance sheet and accordingly eliminates the bank's need to hold economic or regulatory capital (after all, it's sold, selling is the ultimate in credit risk transfer!). But they weren't/aren't all that simple.

Basel is more concerned, I think, with what I would here call "structured finance" (tranching of liabilities into multiple note types). It frustrates me that terms 'securitization' and 'structured finance' are variously defined, but you can see Basel refers to the tranches (emphasis mine):
539. A traditional securitisation is a structure where the cash flow from an underlying pool of exposures is used to service at least two different stratified risk positions or tranches reflecting different degrees of credit risk. Payments to the investors depend upon the performance of the specified underlying exposures, as opposed to being derived from an obligation of the entity originating those exposures. The stratified/tranched structures that characterise securitisations differ from ordinary senior/subordinated debt instruments in that junior securitisation tranches can absorb losses without interrupting contractual payments to more senior tranches, whereas subordination in a senior/subordinated debt structure is a matter of priority of rights to the proceeds of liquidation.
By characterizing a structure finance, the regs immediately anticipate a scenario where the bank-as-originator might retain some of the exposure (and, of course, they should hold capital against retained exposures like any exposure).

Even then it's not cut and dry: with respect to the traditional securitizations, with the infamous SIVs, the banks did sell the assets and get regulatory relief per a sale ("regulatory arbitrage") while providing liquidity lines to the same structure. So, they basically found a way around the literal true sale. The synthetic securitization is harder still: the bank is retaining the all or most exposures but basically hedging them by purchasing credit protection. Basel went back and increased those requirements; the crisis proved that even if you hedge 100% of your exposure, you've still got basis risk and counterparty risk.
Thread starter #10
I swear I read that and thought I'm going to underline it and ask David! But then , I don't have time to ask all the questions I come across :(:mad: At this point I think we're all asking questions only if it hinders our progress not to know the answers :( No real learning.
Thread starter #11
Hi David,

I know the AIM says we're not expected to memorize specific details such as risk weights, but does that also include detailed additions of Basel III over II?

Specifically I mean :
1- Runoff rates for liabilities and draw down rates of assets wrt LCR calculations, and
2- Types of funding and associated ASF factors, and required stable funding factors wrt net stable funding ratio calculation

Hi everyone. This is a great thread.

My understanding of this stuff seems like it is on the same level with Hend and ibrehim. Certain things just tend to somewhat contradict others: If it is a true sale, then the bank no longer owns the assets, but if it does then that amount must be deducted from capital. I read this somewhere alse and it (at least I thought) was referring to a bank retaining yhe "first loss" trench, as David mentioned above.

True sale: not on your books any more so no need to hold any capital againt these assets.
Retain part of the securitization: might not be a true sale, but for some reason it says we need to hold onto 100% of the value of the first loss trench if we still have it.

Makes me want to pull my hair out. I understand that this is supposed to make us think, but with the Basel readings leaving so many open questions, as Hend mentions, just makes this REALLY frustrating.

Thread starter #13
Hi Shannon,

Thank you for pulling me back to this thread...I just noticed that David didn't answer my last question here :(
I really should keep up with where and what I post :confused:

Love the hair-pulling spirit :) been under its effect for some time now :) I like to think it will prove fruitful in the end

As to Basel readings, I think the concepts don't lend themselves to common comprehension (like mine) from the first reading. It needs to be repeated. AND THAT NEEDS TIME I can't afford right now. We're less than a month away from the exam. So I'm just going to go through its key concepts, then maybe the practice questions will deepen my comprehension of the rest.

Good luck

David Harper CFA FRM

David Harper CFA FRM
Staff member
Hi Hend, I'm sorry, the reason is that I simply did not know (and still do not know). I really have literally no idea about either of your two good questions. In general, the FRM is hard to predict, but the Basel is even harder. I think i meant to try and see if i can suss out some relevant examples, but I frankly don't have time to do that.

(my strongly felt hunch is that the FRM will NOT query your two points, but the last thing i want to do is commit a Type II error in trying to help you; null = will not be tested or you don't need to know)

Even Shannon's last post above, I almost but can't quite identify with the exact discrepancy to which she refers (I read her post carefully now, marinated in it, and i simple don't see the exact problem she's writing about), such that it's hard for me to quickly give help. (I can typically help by broadening/re-framing, but that can take significant time on my part; e.g., my hunch is this is actually a symptom of not quite understanding tranches in a securitization?)

.... but, i do totally agree with the comments about Basel, no doubt. The complexity is exacerbated by the fact that Basel is also a moving target (2 --> 2.5 --> 3.0). I do agree with the suggestion, if it was made, that it is not useful to go to go too deep on detail at the expense of a lack of a superficial comprehension. I would definitely treat Basel "top down" and go deeper only as comfort/time allow (e.g., there is no point is sussing out the securitization retention details if one is foggy on securitization in the first place, imo). Like you, I wish i had a little more time :(
Thread starter #15
Thanks David,
I understand about time.
I'm not gonna be ready in May (as much as I hate admitting that) but I'm going to sit for the exam anyway :(
I feel that this superficial way of studying won't help at all in passing the exam (let alone when having to apply concepts to real work), but it's a bet I'm willing to take for the time being.

Thanks again

David Harper CFA FRM

David Harper CFA FRM
Staff member
Hi Hend, thanks, but just to clarify, I only meant to suggest with regard to Basel (not the exam!) a top-down approach. I don't endorse superficial, I just meant that, in the example above, the detailed understanding (tranche retention in Basel) can be elusive if the superficial (basic securitization) is not first grasped. That's just to clarify, sorry ...

More importantly, GOOD LUCK
Hi David, do you think we'd be expected to know Basel III stuff on a granular level ie. do we have to know the ASF assigned for each asset or liability item for NSFR's calculation? On a tangent, do we have to memorise really cumbersome formulae as well? I remember from Part I that, really tough formulae weren't tested much - does that apply to part II as well? thanks and cheers,


Active Member
Hi David, do you think we'd be expected to know Basel III stuff on a granular level ie. do we have to know the ASF assigned for each asset or liability item for NSFR's calculation? On a tangent, do we have to memorise really cumbersome formulae as well? I remember from Part I that, really tough formulae weren't tested much - does that apply to part II as well? thanks and cheers,
I looked through the 2014 AIMS, LCR is there, NSFR is NOT. So I say - concentrate on LCR but remember about Type II error mentioned by David.


Active Member
Hm, but in AIM 61 (“Basel III: Global Regulatory Framework for More Resilient Banks & Banking Systems(Revised) 2011") it is stated: "Describe changes intended to improve management of liquidity risk including liquidity coverage ratios, net stable funding ratios and use of monitoring metrics". I for once will try to have a basic understanding of NSFR,but probably will not learn every detail regarding this ratio.