Tim Brown: Risky to put all your eggs in AAA basket

Ratings are like lamp-posts, they can shed light on the risks so well-informed choices can be made but they are not a guarantee of a successful outcome. Photo / Getty Images

Ratings are a help but we should be doing our own research and not relying on them, writes Infratil's Tim Brown.

Late last year New Zealand investors bought $325 million of 25 year bonds issued by IAG, an Australian listed insurance group. It was the largest NZ domestic corporate bond issue of 2011. Few of the investors in these bonds, or their advisers, would have understood much about the repayment risk. Analysis of the credit would have started and stopped with one sentence: "Standard & Poor's issue rating A-."

Students of the 2007-2008 Global Financial Crisis know that rating agency mistakes, and errors of interpretation by those relying on the rating, contributed to vast amounts being lent which will never be repaid.

John Paulson, the most successful speculator against the US mortgage bubble, bet that AAA rated bonds would default and made a profit of $30 billion. He is portrayed in books and articles as someone who did a lot of his own research and who used common sense when betting that the ratings emperors were nude.

Recently Vanity Fair contributor Michael Lewis wrote about the buyers on the other side of Paulson's trade, in particular German savings institutions. The German perspective "we were following the rules, AAA bonds did not default." They were wrong. On the estimated $600 billion of US "structured" bonds sold in the two years prior to the crash, losses have ranged between 67 per cent for highly rated and 95 per cent for lower rated bonds (i.e. investors got back 33 cents in the dollar on highly rated bonds and 5 cents on "junk"). German losses are estimated at over $200 billion.

Ratings did not actually make anyone lend to borrowers who were not going to repay, but they contributed. "Risk comes from not knowing what you are doing' said Buffett, and the ratings agencies allowed investors to not know, but to think they did thanks to the "AAA" stamp from S&P, Moody's or Fitch.

The rating agencies have come through this and still command respect. When S&P lowered the USA credit rating, the importance was signalled by the near hysterical response of US authorities. When France's AAA was recently impugned by S&P the French Finance Minister raged that Great Britain was in worse shape and should drop first.

Ratings are not a problem in their own right, but they create a problem if misused. Like a lamp post, they should be used for illumination rather than support. When they shed light on risk they enhance knowledge and improve decision making. But those who invested in the IAG bonds purely on the basis of "A-"were leaning on the lamp post.

However, while bond-buyers may have seen no further than "A-", their financial advisers have spotted something else, even without reading the ratings report. A financial adviser who recommends a bond is first and foremost concerned about their own liability: "If it goes wrong what come-back have investors got to me?" The assumed wisdom is that recommending a well rated bond provides bulletproof protection - "I wasn't negligent, blame the rating agency."

Adviser concern about the liabilities that could arise from recommending bonds which perform poorly reflects the regulatory regime being developed by the Financial Markets Authority. Risk is messy, especially for regulators. In this instance the FMA have broadly offered advisers two courses of action to ensure a future defence against any accusations of negligence.

One alternative is for the intermediary to analyse the bond to ensure their advice is suitably informed. This means undertaking research and providing suitably intelligible advice, which will still leave the door open to subsequent criticism and penalty if the bond fails ("your research/advice was poor"). The alternative way for an adviser to limit future liability is to recommend bonds that are rated, all blame can then fall on the rating agency if investors later experience loss. Whether investors will receive the best possible advice from advisers intent on minimising their own residual liability is problematic.

Last year in New Zealand, excluding the government, $5500 million of medium term bonds were issued. A large amount of funding and a large amount of savings. Fifty five per cent of this sum was raised by local banks who were finding it relatively hard to raise term debt internationally, 25 per cent went to local government or Crown owned entities, 13 per cent to rated corporates (including IAG) and 6 per cent to unrated corporates.

For those interested in the NZ capital markets there are several messages from this activity. The first is that bonds are a very important source of funding for NZ borrowers and a very important source of financial assets for lenders.

To date the recent regulatory developments have added costs, have changed how things are done, but have had little obvious impact on what is being done. The question mark is over how regulation, which encourages intermediaries to minimise their own risk, will influence behaviour over the medium term. There are two bad possible outcomes.

If lenders only buy securities which "tick the boxes" they will end up, like the Germans, following the rules over the cliff. The more intermediaries believe they are protecting themselves by offering rated bonds, without even reading the rating report, the more they will be turning their backs on Buffett's "Risk comes from not knowing what you are doing." The other consequence of intermediaries seeing ratings as a short-cut to minimising their liabilities is that the bond market will close for emerging NZ companies.

NZ has a plethora of medium-sized unrated companies which were able to issue bonds. This important source of medium-term debt will struggle in a regulatory environment where arrangers and intermediaries see a credit rating as protection against dissatisfied clients who have bought bonds which perform poorly.

The Germans are not the only people to have put too much reliance on ratings. Over the last decade the NZ corporate bond market has not provided evidence that credit ratings make for better returns. South Canterbury had an "investment grade" credit rating. Credit Agricole issued $200m of "investment grade" rated bonds which recently traded at 40 cents in the dollar.

Several poorly performed structured credit deals had good ratings. On the other hand, Infratil, Fletcher Building, Sky City are three examples of successful NZ companies which have made extensive use of unrated bonds to provide medium term funding. Whether they, and the companies of tomorrow, will continue to be able to use this market to provide funding is not clear given the current regulatory environment.

Regulation is intrinsically about stopping things happening, bad regulation stops more than it should.

Tim Brown oversees the Infratil group debt capital market activities. Infratil and Z Energy were the only issuers of unrated corporate bonds in NZ last year.