Saturday, September 30, 2017

Capital Requirements for Large Banks


More on the proposed changes to bank regulatory capital requirements here from the Peterson people.

Article is anti-proposal but there is still some good confirming data there regardless.


How much would the Treasury's proposed exemptions from the exposure base reduce the effective capital requirement for the US G-SIB banks? 
One indicator is that the Treasury considers that "Large U.S. banks hold nearly 24% of their assets in high-quality liquid assets such as cash, U.S. Treasury securities, and agency securities".[10] 
More specifically, at the end of 2016 commercial banks had total assets of $16.07 trillion, of which Treasury and agency obligations amounted to $2.43 trillion and cash (including reserves at the Federal Reserve) stood at $2.21 trillion.


So there are about $4.5T of risk free assets there tying up about 10% of that (assume LR run at about 0.1) as regulatory capital; or i.e. perhaps as much as $450B which would become available to the system if this regulatory reform is promulgated by the Executive Branch.

Also of course no time domain analysis here in the article of how these bank assets got there in the first place and what those actions caused.



23 comments:

Dan Lynch said...

"there are about $4.5T of risk free assets there tying up about 10% of that (assume LR run at about 0.1) as regulatory capital"

Are you saying we need even more private debt?


Matt Franko said...

I'm saying this reform could be considered making an additional $450B available to the depository institutions ...

It will have a meaningful effect in some way just as the establishment of these assets did in causing the GFC and the shitty last 8 years...

May foment the reciprocal to the last shitty 8 years....

André said...

Again, the problem is not the required capital for treasuries. Actually, it doesn't make much sense to require capital for that.

The problem is Internal Ratings Based Approach - where banks are allowed to develop their own empirical model to quantify required capital for credit risk. It's always an error to let banks self regulate.

That discussion about treasuries capital requirements are a distraction. They are not an real issue.

MRW said...

André,

What is the Internal Ratings Based Approach, and who is subject to it?

MRW said...

André,

Don’t bother. I found it in wikipedia.
https://en.wikipedia.org/wiki/Internal_ratings-based_approach_(credit_risk)

André said...

MRW,

For capital requirement purposes, banks choose either a standard approach (regulator defines risk factors that will define the risk-adjusted capital requirements) or the IRB approach (banks define their risk factors).

But if a bank chooses the IRB approach, it needs to have adequate internal rating systems and governance, and prove it to regulator. The regulator/supervisor may come to the conclusion that the bank is not prepared for the IRB and don't allow its adoption.

Because banks where exaggerating on the low capital requirements (that's what it happens when you let them self regulate) the Basel Committee created the Leverage Ratio.

The Leverage Ratio is similar to the capital ratio, but there are no Risk Weighed Assets and no model at all. It's just capital to total assets ratio.

The capital ratio requirement is usually 10,5% for most banks/jurisdictions, while the Leverage Ratio requirement is much lower (3%). Banks should hold enough capital to be compliant to both requirements.

So, in some banks, the capital ratio requirement is binding, and the leverage ratio is not. But some banks adpt very generous risk weights, so the leverage ratio is binding.

Treasures are usually considered risk free assets and so receive a 0% risk weight in the capital ratio requirements.

In the Leverage Ratio requirement, all assets are considered in full - it's like a 100% risk weight for all assets. Including treasuries. A bank that holds a lot of treasuries will usually have a very high Leverage Ratio - which I agree is a distortion. The regulator should not "punish" the bank for holding a lot of safe assets like treasuries.

Matt Franko said...

I assume that their internal controls are more conservative than the govt ones....

Iow if govt says risk based Leverage should be 7% they would run it above that at 10% to provide a margin of safety....

This was also a factor in the GFC... they went no bid for any more assets as the Fed was stuffing them with these assets and they had no capital for the he risk assets....

This guy I know had his 10 year floor planning line pulled by SunTrust and GE and he lost his dealership nobody else would do it... in 2009 not 2008....

We get this regulatory mod and it is going to help immensely....

OTOH the Fed is going to start to lower the assets by 10b/mo starting next month when they start to let their portfolio run off... so help is on the way either fast or slow...

Matt Franko said...

"he problem is not the required capital for treasuries"

I don't know how you could say this it's like you've never had any mathematical training at all.... it's like you're saying "it wasn't a fuel vapor explosion that took down TWA 800 it was fraud!"... it makes absolutely zero technical sense... none....

If on Sept 1 you posses 800b reserve assets and then by Oct 1 you have 2.3T of reserve assets you need to come with 150b in additional capital pronto...

Tom and Warren say that is very easy to do which is a joke statement...

The correct response is zero bid on ANY more assets....

Tom Hickey said...

In the Leverage Ratio requirement, all assets are considered in full - it's like a 100% risk weight for all assets. Including treasuries. A bank that holds a lot of treasuries will usually have a very high Leverage Ratio - which I agree is a distortion. The regulator should not "punish" the bank for holding a lot of safe assets like treasuries.

The BIS is dealing with central banks in general. All public debt is not equal in risk. Even the US Congress is toying with strategic default, but countries whose central banks are not the ultimate currency issuers are not default risk-free.

It's probably a matter politics and transaction cost. Rather than rely on the rating agencies to rate public debt across countries, they impose a general rule that covers the riskiest situations. But even then the US is now a wild card in the pack. Used to be king, now it is the joker.

Tom Hickey said...

Matt, you could also look at it as forcing the banks to hold more capital instead of putting the insolvent ones into resolution, which would involve nationalize the TBTF's since they could not be turned around over a weekend without an influx of government-provided capital, which requires an act of Congress. The Fed choose to pay IOR as partial recapitalization over time, but they could not do the whole thing.

This, I believe, is the common understanding.

Dan Lynch said...

"he lost his dealership nobody else would do it... in 2009 not 2008.... "

Why do we need dealers? Why not eliminate the parasitic middle man and buy direct from the factory online?

Oh wait, dealers are a government protected monopoly.

Since home loans and student loans are backed up by the government, why not eliminate the parasitic middle man and get loans directly from a USPS postal bank?

That's what we should be talking about.

André said...

"I don't know how you could say this it's like you've never had any mathematical training at all.... it's like you're saying "it wasn't a fuel vapor explosion that took down TWA 800 it was fraud!"... it makes absolutely zero technical sense... none...."

I couldn't understand a word you said. I don't know what TWA 800 is, and I have trouble in understanding what some sort of vapor explosion has to do with the capital ratios. I also don't know what "zero bid on any more assets" mean.

If a bank would hold all its assets in treasuries, I believe it would be the safest bank in the world. As a regulator/supervisor I don't think you would have to bother to protect the debtholders, because the bank would be very safe.

On the other hand, if all assets are made of risky credits, as a supervisor, I would require the bank to hold a lot of capital, to protect the bondholder.

Of course, every business has risks, so you should always require some capital to protect the "unqualified" bondholders...

NeilW said...

"If on Sept 1 you posses 800b reserve assets and then by Oct 1 you have 2.3T of reserve assets you need to come with 150b in additional capital pronto..."

Or the regulator reduces the percentage - since the regulator is the one adding the reserve assets in the first place.

They have to go somewhere, so the capital has to be there at the percentage overall.

In reality banks will just shed a load of depositors and force them to go elsewhere.

André said...

"All public debt is not equal in risk"

Agreed. As we all know, sovereign debt of Japan is different from the "local" debt of Spain.

But the BIS and the world in general don't know that.

On the other hand, the new IFRS9 standard requires that banks calculate impairment provisions for credit risk - even for government bonds. That would discount the assets, taking into account the credit risk. But capital requirements would still be at 0% or almost 0% risk weight.

Matt Franko said...

They did TARP Tom which provided like 350B capital.... temporary...

They wouldn't have had to do it at all if they didn't force $1.5t of assets on them in like a month ... so they would have "loanable funds!"... they're f-ing morons...

André said...

Oh, I think I understood what you said.

I didn't mean that the new leverage rate rule wouldn't affect banks. It certainly will affect some banks.

I meant that this sort of new regulation is unproductive. That's bad regulation, and it doesn't matter whether treasuries will have 0% or 100% risk weight, it would still be bad regulation. If you think about it, it makes more sense to have 0% risk weight on treasuries (which is what bankings are arguing for) - but it would still be bad regulation anyway.

The regulators/supervisors should simply prohibit self regulation. Banks shouldn't be allowed to calculate their own risk weights. Problem solved.

Matt Franko said...

Andre,

Right now, the CR is not the operative 'set point', it is the LR as the current CB policy is resulting in $4.5T of risk free assets... this % of risk free assets is UNPRECEDENTED... ie they never did $3.5T of QE before in the history of the CB...

From the perspective of the CR, the system is way over capitalized its nothing to worry about right now...

If you put $10B of risk free assets onto a member depository institution in a short time and that institution cannot reasonably raise the additional required $1B of required regulatory capital then guess what that makes the condition of the institution? INSOLVENT....

CB: "Hey! Here are another $1.5T of deposits for you to lend out!"

Institution: "Hey! I dont have the required $150B of capital to support that!"

CB: "Hey! you're insolvent!!!"

They are all douche bag, unqualified, incompetent morons...

Tom Hickey said...

They wouldn't have had to do it at all if they didn't force $1.5t of assets on them in like a month ... so they would have "loanable funds!"... they're f-ing morons.

The basis of the GFC was the Fed screwing up by adding reserves and not the dodgy loans, corrupt securitization, tripping of systemic risk as parties lost trust in each other?

There's a book in that Matt. Guaranteed to be a best seller if you can make a plausible case. Just connect with a ghost writer and get an agent to shop a book proposal around.

Tom Hickey said...

@ André

The regulators/supervisors should simply prohibit self regulation. Banks shouldn't be allowed to calculate their own risk weights. Problem solved.

The banks are very powerful. If they are having trouble getting this done, the regulators see a good reason. I doubt the BIS would ever permit the banks to self-regulate without including their safe assets in the mix, and the transaction cost is high in imposing outside regulation. Banking is relatively low-cost since transaction costs are kept down with internal regulation and banks'd doing their own credit analysis with minimal supervision. This is a reason there are private banks.

André said...

In Brazil, there is not a single bank that uses the Internal Ratings Based approach for credit risk capital requirements. So they do not self regulate in terms of capital requirements for credit risk.

They all use the standard model. Leverage Ratio is not a problem, because the traditional capital ratio will probably always be binding.

And they are still recognising record profits each trimester...

Matt Franko said...

I'm waiting to test the reversal of my hypothesis...

I think if they put this Warsch in there its a high chance ... as he is a heavy monetarist...

If I can make enough munnie on the reversal I might have more time eventually to try to write it up better...

Matt Franko said...

"not the dodgy loans, corrupt securitization, tripping of systemic risk as parties lost trust in each other?"

That conclusion can only be reached thru political bias...

If you look at it objectively/technically, you come to the correct conclusion...

Dogma dispensed within the context of material systems can only be overcome with science/math... not more opposite dogmas...

Tom Hickey said...

Matt, you are sitting on a bombshell if correct. You really need to take some time and write it up making a case. I admit that I am skeptical, but give it shot.

Of course, you can keep it proprietary and trade on it, too. That's a way of confirmation, too.