Rating Agencies and Bond Mispricing

As reported in March, over the coming months this site will report results of a real-time experiment in high yield bonds.

Casting an eye over potential candidates for the High Yield Portfolio, my attention was caught by two rather curious features of the Co-operative Bank’s 11% Subordinated Loan Notes.

These notes were issued in December 2013 as part of a major refinancing exercise to save the bank from the threat of extinction. The funding raised was used to repay the Bank’s preference shares and perpetual subordinated bonds, which were redeemed at a significant discount. The Notes have two maturities – one repayable on December 20th 2023 (EPIC 42RQ), offered at a price of 120 on April 9th, and the other maturing on December 20th 2025 (42TE), offered at a price of 127.

Here’s the first curious feature. At the offered prices, the 2023 Notes have a yield to maturity (YTM) of 7.8%, compared to a YTM of 7.3% on the 2025 Notes. The implication would seem to be that the Co-operative Bank has a negative expectation of default between 2023 and 2025. But that is a logical impossibility.

But it is not just the relative yield that looks odd. The absolute yield is high, indicating a high risk of default. It is as if the December 2013 refinancing never took place. Or, perhaps more accurately – since, as a matter of historic fact, it did take place – it is very likely to fail.

How has the bond market arrived at such a conclusion?

My suspicion is that it is with the help of a couple of leading ratings agencies.

Fitch awards the bank a “B” rating, which “indicates that material default risk is present…Financial commitments are currently being met; however, capacity for continued payment is vulnerable to deterioration in the business and economic environment.” Moody’s assigns a Caa2 rating – “Obligations rated Caa2 are judged to be of poor standing and are subject to very high credit risk.”

But is this fair? Neither Fitch nor Moody’s have proved exactly infallible in the past, having failed to identify financial stress in a number of major banks during the run-up to the financial crisis of 2007/08. For that reason, I prefer to apply my own forecasting model of financial stress, as set out in the fourth chapter of The Golden Guinea. I benchmarked the Co-op Bank’s final accounts for the year to December 2014, published a couple of weeks ago, against Northern Rock’s accounts for the year to December 2006, applying the seven leading indicators of bank failure set out in The Golden Guinea:

1. Extremely rapid growth of its loan portfolio. The value of Northern Rock’s portfolio more than tripled between 2003 and 2006, suggesting deteriorating loan quality. By contrast, the Co-op Bank’s loan portfolio has fallen by a fifth over the past four years, from £34 billion to £28 billion, suggesting tighter loan control.

2. Increasing loan to value (LTV) ratios. In Northern Rock’s case, the bank advanced mortgages with a value of 100% or more of the properties against which they were secured. The days of 100% mortgages are now long over, and the Co-op bank is lending on far more prudent LTV ratios.

3. Falling interest margins. Northern Rock’s net interest margin – the difference between the interest earned on its loans and the interest it had to pay on its own funding as a percentage of its total loan portfolio – narrowed from 1.53% in 2003 to 0.98% in 2006. Applying the same formula, I calculate the Co-op Bank’s net interest margin as 1.28% for the year to December 31st 2014, up from 0.94% the year before and moving in the right direction.

4. Falling administrative and staff costs as a proportion of the total loan portfolio. This suggests less thorough due diligence of new loans and less intensive monitoring of existing loans. Northern Rock’s staff costs fell significantly as a proportion of its loan portfolio in the lead up to its financial collapse. By contrast, the Co-op Bank’s staff costs rose from 0.92% of its loan portfolio in 2013 to 1.20% in 2014. Unless the bank’s staff are making cups of tea all day (which hopefully is not the case), this suggests that greater resources are being committed to loan monitoring and appraisal.

5. Rising leverage and a falling equity buffer. Northern Rock’s equity fell from 3.6% of its gross assets in 2003 to 1.8% by the end of 2006. By contrast, the Co-op Bank’s shareholders funds have increased from 3.7% of its gross assets at December 31st 2012 to 5.4% at December 31st 2014.

6. Increased dependence on wholesale funding, and reduced support from customer deposits. Northern Rock’s wholesale funding increased from 35% of its gross assets in 2003 to 64% by the end of 2006, leaving it dangerously exposed when the wholesale funding markets froze in July 2007. The Co-op Bank relies on wholesale funding for less than 10% of its total funding, with customer deposits accounting for more than 75%, a much more stable funding structure.

7. Declining cash and liquid reserves. Northern Rock’s cash reserves were less than 1% of its gross assets at December 31st 2006, leaving it desperately short of ready funds to meet customer withdrawals and the loss of access to the wholesale funding market in July 2007. By contrast, the Co-op Bank’s cash reserves represented 12.7% of its gross assets at December 31st 2014, with the Bank holding £4.8 billion of cash within its gross assets of £37.6 billion.

You will notice that a bank’s bad debt provisions are not among the leading indicators that I have identified for bank failure. Banks facing financial distress are very likely to “extend and pretend” – extend further credit to borrowers in difficulty, and pretend that their loans will be repaid. It turns out that bad debts are a lagging rather than a leading indicator of bank failure.

Given these statistics, how can Fitch and Moody’s have assigned an investment grade rating to Northern Rock in 2006, while assigning a junk bond rating to the Co-operative Bank nine years later?

My guess is that they are looking mainly through the rearview mirror, basing their assessment on the Co-op Bank’s historical record under its previous, now departed, management team, rather than on its position and prospects under new owners and managers. My impression is that the bank’s chief executive, Niall Booker, who took up his post in June 2013, is a steady hand on the tiller, steering his stricken vessel to calmer waters after a period of turbulence.

Only time will tell whether I am right, or whether the ratings agencies more pessimistic view is correct. In the meantime, the Co-op bank’s 11% 2023 Notes become the second investment of the high yield bond portfolio.

Issuer: Co-operative Bank plc
Coupon: 11.0%
Coupon Payment Dates: Quarterly on March 20th, June 20th, September 20th and December 20th
Maturity date: December 20th 2023
Minimum Denomination: £100
Offer Price: 120
Running yield (at Offer Price): 9.1%%
Yield to maturity (at Offer Price): 7.8%
Nominal Amount: £1,000
Cost: £1,207.17 (inclusive of £7.17 accrued interest) + £11.75 commission = £1,218.82

Permanent link to this article: http://www.goldenguinea.com/rating-agencies-and-bond-mispricing/

1 comment

    • Matt on December 17, 2015 at 8:12 am
    • Reply

    Is the reason for the differential not the fact that these are bonds from two different issuers? The 2025 is coop group, so convenience shops, funeral care etc and the shorter one is the bank of which the group is no longer a majority shareholder. I think this is the reason for the difference in yield and credit

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.