Category: Regulatory Capital

CIBC (Stock symbol CM … I can never quite decide how to present it!) has released its Second Quarter 2008 Report and Supplementary Package, so it’s time to recalculate how much room they have to issue new preferred shares – assuming they want to!

Step One is to analyze their Tier 1 Capital, reproducing the prior format:

Next, the issuance capacity (from Part 3 of the introductory series):

and the all important Risk-Weighted Asset Ratios!

TD has released its Second Quarter 2008 Report and Supplementary Package, so it’s time to recalculate how much room they have to issue new preferred shares – assuming they want to!

Step One is to analyze their Tier 1 Capital, reproducing the prior format:

Next, the issuance capacity (from Part 3 of the introductory series):

and the all important Risk-Weighted Asset Ratios!

The reported Assets-to-capital multiple reflects that goodwill is deducted from total capital (the denominator) AND FROM TOTAL ASSETS (the numerator); given TD’s huge goodwill, this makes rather a difference!

BNS has released its Second Quarter 2008 Report and Supplementary Package, so it’s time to recalculate how much room they have to issue new preferred shares – assuming they want to!

Step One is to analyze their Tier 1 Capital, reproducing the prior format:

Next, the issuance capacity (from Part 3 of the introductory series):

and the all important Risk-Weighted Asset Ratios!

BMO has released its Second Quarter 2008 Report and Supplementary Package, so it’s time to recalculate how much room they have to issue new preferred shares – assuming they want to, in this environment!

Step One is to analyze their Tier 1 Capital, reproducing the prior format:

Next, the issuance capacity (from Part 3 of the introductory series):

and the all important Risk-Weighted Asset Ratios!

Assiduous Readers will recall the post on the Assets to Capital Multiple and my correspondence with Royal Bank’s Investor Relations department:

Well! This is interesting! According to these very, very rough calculations, RBC has an Assets-to-Capital multiple of 23.3:1, which is both over the limit and well above its competitors. This may be a transient thing … there was a jump in assets in the first quarter:

RBC: Change in Assets From 4Q07 to 1Q08
Item	Change ($-billion)
Securities	+6
Repos	+12
Loans	+8
Derivatives	+7
Total	+33

I have sent the following message to RBC via their Investor Relations Page:

I would appreciate learning your Assets-to-Capital multiple (as defined by OSFI) as of the end of the first quarter, 2008, and any detail you can provide regarding its calculation.

I have derived a very rough estimate of 23.3:1, based on total assets of 632,761 and total regulatory capital of 27,113

Update, 2008-04-17: RBC has responded:

Thank you for your question about our assets to capital multiple (ACM). In keeping with prior quarter-end practice, we did not disclose our ACM in Q1/08 but were well within the OSFI minimum requirement. Our ACM is disclosed on a quarterly basis (with a 6-7 week lag) on OSFI’s website. We understand this should be available over the next few days. Below is an excerpt from the OSFI guidelines outlining the calculation of the ACM. We hope this helps.

The supplied excerpt from the guidelines was:

From OSFI Capital Adequacy Requirements (No. A-1)

1.2. The assets to capital multiple

Institutions are expected to meet an assets to capital multiple test. The assets to capital multiple is calculated by dividing the institution’s total assets, including specified off-balance sheet items, by the sum of its adjusted net tier 1 capital and adjusted tier 2 capital as defined in section 2.5 of this guideline. All items that are deducted from capital are excluded from total assets. Tier 3 capital is excluded from the test.

Off-balance sheet items for this test are direct credit substitutes1, including letters of credit and guarantees, transaction-related contingencies, trade-related contingencies and sale and repurchase agreements, as described in chapter 3. These are included at their notional principal amount. In the case of derivative contracts, where institutions have legally binding netting agreements (meeting the criteria established in chapter 3, Netting of Forwards, Swaps, Purchased Options and Other Similar Derivatives) the resulting on-balance sheet amounts can be netted for the purpose of calculating the assets to capital multiple.

Under this test, total assets should be no greater than 20 times capital, although this multiple can be exceeded with the Superintendent’s prior approval to an amount no greater than 23 times. Alternatively, the Superintendent may prescribe a lower multiple. In setting the assets to capital multiple for individual institutions, the Superintendent will consider such factors as operating and management experience, strength of parent, earnings, diversification of assets, type of assets and appetite for risk.

1. When an institution, acting as an agent in a securities lending transaction, provides a guarantee to its client, the guarantee does not have to be included as a direct credit substitute for the assets to capital multiple if the agent complies with the collateral requirements of Guideline B-4, Securities Lending.

I’ve been checking the OSFI disclosures page fairly regularly and today was rewarded with the actual data. As of 1Q08, Royal Bank had:

Well! Here’s a howdy-do! I’ve been puzzling for some time as to how my approximate calculation could be so different from the sub-20 multiple that I assumed was actually reported! I’m glad to see that backs of envelopes still serve some purpose.

I have sent the following query to Royal’s Investor Relations department:

Sirs,

I am most interested to learn that your ACM was 22.05 as of 1Q08.

When did you seek approval from OSFI to exceed 20.0, and what was the rationale for exceeding the normal guideline?

Sincerely,

Incidentally … PrefBlog’s Scary Number Department recommends a glance at the “BCAR Derivative Components” figure. RY has nearly 3.4-trillion in interest-rate swaps outstanding and nearly 414-billion in credit swaps. The total notional for all derivatives is about 5.3-trillion.

Given that RY’s Total Capital Ratio (based on Risk Weighted Assets) is close to that of the other banks, the implication is that RY has greater total exposure with a small average risk weight. I’ll try to have a look at this shortly.

Update: I have also sent an inquiry to OSFI:

I was most interested to learn that Royal Bank had an Assets-to-Capital ratio of 22.05 as of the 1Q08 filing.

It is my understanding that the general maximum allowed by OSFI for this ratio is 20.0, which may be increased to 23.0 upon prior application to the Superintendant.

Is this correct? If so, then:

(a) When did Royal Bank apply to have the maximum increased?

(b) On what grounds did the Superintendant allow the increase?

(c) Were any terms, conditions, or time limits attached to the approval?

Sincerely,

Update, 2008-6-5: I have not received a response from RY Investor Relations. I have received a reply from OSFI:

Thank you for your e-mail of May 26, 2008, concerning the assets to capital ratio for banks.

You are correct that under the assets to capital test, total assets should be no greater than 20 times capital, although this multiple can be exceeded with the Superintendent’s prior approval to an amount no greater than 23 times. Further information on this ratio can be found in Section 1.2 of the Capital Adequacy Guideline (http://www.osfi-bsif.gc.ca/app/DocRepository/1/eng/guidelines/
capital/guidelines/CAR_A_e.pdf).

In response to the second part of your enquiry, you may wish to note that pursuant to section 22 of the Office of the Superintendent of Financial Institutions Act, any information that is obtained by the Superintendent regarding the business or affairs of a federally regulated financial institution (FRFI), or regarding persons dealing with the FRFI, or any person acting under the direction of the Superintendent, is to be treated as confidential and may not be disclosed to a third party.

Thank you for taking the time to write to this Office with your questions.

I was reminded of this topic recently … and reminded some time ago of the perception of relative risks when investing in preferred stock. A stockbroker type was explaining to me that he would never buy bank preferreds due to the risk of default … I pointed out that, while always possible, the banks would surely cut or eliminate their common dividend well in advance of their preferred dividend.

He was flabbergasted … “Do you really think they would cut their dividend? That would be terrible!”

In other words, he very calmly accepted the idea of a default on the perpetuals, but could not conceive of a situation in which a bank would cut its common dividend.

Recently, the poster-boy for credit excesses, Citigroup, cut its common dividend 41% while the preferred dividends just kept on chugging along – even increased in total, as they have raised a lot of capital via preferred offerings … presumably to investors who figured they wanted their interim dividends for the next five-years-odd to be preferred!

So anyway, I was thinking about this a little more and did a little digging … through the RBC Annual Report for 2007:

During 2007, we continued to return capital to our shareholders through dividend increases and share buybacks, delivering a total shareholder return of 16 per cent.
…
For several years, we have made it a management priority to ensure current success was reinvested to fund future growth. This approach allowed us to deliver relatively solid shareholder returns in 2007 while returning capital through increased dividends and share buybacks. We raised dividends twice in 2007 for a total increase of 26 per cent, and we repurchased 11.8 million common shares. Our capital position is strong with a Tier 1 capital ratio of 9.4 per cent, comfortably above our target of greater than 8 per cent.

Share Buybacks are analytically equivalent to dividends – and buyback-suspensions are the easiest way to halt a decline in capital ratios. But what sort of proportion do they make? I’ve had a preliminary look at this via RBC’s Annual Reports for 2001, 2004 and 2007:

So, this is all pretty rough, it’s only one bank (a strong one!) and it’s taken over a period in which the bank examined hasn’t had anything particularly horrible happen to it. Still, it’s interesting to find that about 1/3 of the total capital returned to common shareholders has been in the form of buybacks rather than dividends … and, as the experience of 2005 shows, the buybacks can be cut quite easily.

Assiduous Readers will remember that in the post Bank Regulation: The Assets to Capital Multiple, I performed some very rough calculations for Royal Bank and got a somewhat surprising result – a figure in excess of the OSFI standard limit.

These Assiduous Readers will also remember that I got put in my place pretty smartly by the Royal Bank Investor Relations Department:

Thank you for your question about our assets to capital multiple (ACM). In keeping with prior quarter-end practice, we did not disclose our ACM in Q1/08 but were well within the OSFI minimum requirement. Our ACM is disclosed on a quarterly basis (with a 6-7 week lag) on OSFI’s website. We understand this should be available over the next few days. Below is an excerpt from the OSFI guidelines outlining the calculation of the ACM. We hope this helps.

I hadn’t been aware that OSFI reported data … but they do! There is an excellent bank data resource page on the site that I found through the banks section of the sitemap.

According to OSFI, the assets-to-capital multiple for Royal Bank as of Q4-2007 was 19.94. It’s a good thing nobody sneezed at the time, or they would have been over-limit! I am all agog for the Q1-2008 data to be published so I can determine what Royal Bank’s standards for “well within the OSFI minimum requirement” mean. In the meantime, I have added a link to the reporting page to the blogroll, under the heading “Online Resources”.

Fascinating.

I’ve never been a fan of Banks’ subordinated debt, on the grounds that, while you get paid for term of T, you are taking the risk of a term of T+5 … and the penalty rate of interest that normally gets paid on the last five years means that the only time you will actually have exposure to the T+5 paper is when the issuing bank is under stress, and you’d really, really, not have this exposure. Or, at least, not at your going-in, term T price. Note that Sub-Debt comes in two levels; the ultra-cool European “LT2” means the same thing as the coldly-technical North American “Tier 2B”.

An Italian bank has just allowed its sub-debt to go to T+5:

The market is at risk because one borrower has broken ranks. Credito Valtellinese Scrl ignored the April 30 call date on its 150 million euros ($236 million) of notes. As a result, the bank will pay a penalty interest rate of 160 basis points more than money-market rates — higher than the 100-basis-point premium it paid for the past five years, though still lower than it would probably pay to refinance.
…
There’s another twist to the tale. Because exercising the call option messes with a bank’s capital structure, the repayment has to be sanctioned by regulators. Banca Antonveneta SpA, an Italian lender bought by Banca Monte dei Paschi di Siena SpA last year, has asked the Bank of Italy to agree to its repaying 450 million euros of notes at the April 23 call date.

The authorities, though, may be reluctant to allow firms to weaken their finances in the current environment. Credito Valtellinese’s heresy may swiftly become commonplace.

TD Bank has a good on-line summary of their sub-debt [although some of it is actually Innovative Tier 1 Capital]. Most of their issues carry a penalty rate of BAs+100, although their most recent issue has the T to T+5 portion at BAs+200.

CIBC’s penalty rate is BAs+100.

Scotia’s penalty rate is BAs+100.

RBC’s penalty rate is BAs+100

BMO does not provide a convenient listing … an issue from last year has a penalty rate of BAs+100, while one from last month is CDOR+250

National’s penalty rate is BAs+100

Who knows? We might be in for some interesting times!