Just like with any contact, before you sign a contract to buy a bond, make sure you read the fine print
Bonds are supposed to provide investors with a high level of security. Yet over time, the concept of making payments when payments are due has been compromised by investment bankers, who cleverly knit new kinds of bonds. The result is that if you buy shares in any company, one share is almost always the same as another. That is not true for bonds.
One bond issue may be a senior credit backed by designated capital, perhaps an office building. Another may be a promise by a deadbeat company to pay interest on the subordinate issue after everybody else has been paid.
Read the fine print
There is a general rule to follow when buying bonds.
It requires that you read the covenants for each bond issue. The covenants, really the rules for payment, can be had by asking an investment dealer for the offering documents. Or, if you prefer, or, if your investment dealer is amazed that anybody would ever ask to see the covenants, they can be found at www.sedar.com, the website for all Canadian securities documents. The American equivalent is www.sec.gov.shtml.
Government of Canada bonds have a covenant that can be expressed in one sentence: Holders will be paid a specific amount of money or, in the case of inflation-linked Real Return bonds, a precisely described amount of money on specific dates and the bond will be refunded at a specific date. There is nothing else to say or that can be said.
Most sovereign bonds have this concept, though when you buy a U.S. Treasury bond you have to consider whether the Treasury, often caught in Thelma and Louise crisis when it is about to drive off the monetary cliff, will have the money to pay you when the time comes.
As bonds get riskier, their covenants grow longer and more complex. The basic question — Will interest and principal be paid on time? — has given rise to the credit rating industry, dominated by Standard & Poor’s, Moody’s Investor Service, Fitch Ratings and Canada’s own DBRS, formerly known as the Dominion Bond Rating Service.
Bond analysis draws fine financial minds. It is not the guesswork of stock price forecasting. Rather, it is detailed legal, economic and accounting work.
Yet the industry is criticized for the fundamental flaw in its structure. The fat fees bond raters earn are paid by the companies whose bonds they rate. That is a profound underlying conflict of interest that each rating company denies. But the U.S. Department of Justice isn’t buying the denials.
In the U.S., on February 4 of this year, Justice filed civil fraud charges against Standard & Poor’s, accusing the firm of inflating the ratings of mortgage investments and setting them up for a collapse when the financial crisis began in 2008. A dozen states have joined the action, the first major prosecution of the industry. The problem is embedded in statistical models used by the industry to assess the mortgages, which were packed into synthetic bonds.
Here we need to return to the nature of a debt. A mortgage, like a bond, is a promise to pay. The mortgage is backed by real estate. If the mortgage were issued to dubious borrowers nicknamed NINJAs for “no income, no job, no other assets,” as they were in the years leading up to the 2008 mortgage crisis, they would probably be walkaways, that is, backed by abandoned properties whose owners, under U.S. federal law, could just shut the door and leave with nothing but a bad credit rating.
Under American law, creditors could seize or sue, but not both.
In Canada, our creditors can seize properties whose owners are in arrears and sue for any remaining debt, which is at least part of the reason our home loan abandonment rate is about one-third of one per cent compared to the U.S. dereliction rate that averaged 30 per cent in parts of Florida, Nevada and California.
Crummy loans that major lenders should have known were unlikely to get to the finish line fully paid were dissected into payments due at consecutive dates. The principal was then further dissected into the early repayment risk (there is no penalty for prepayment in the U.S.) and residual risk. An all of these pieces were then glued into new bonds, thus the name synthetic, and sent to the raters for their opinions.
The raters went to work with statistical models based on the concept that even if all eggs eventually rot, they won’t all do it at the same time nor all in the same basket. So bond payments from Vermont were blended with other payments due from Alaska and Nevada and so on. The packages were then turned into tiers, usually 36 of them, so that the top tiers got paid first and the second tiers got paid after the first, and the third after the second and so on. The top tiers got AAA ratings because, of course, even if the structures were built on garbage, any money that came in did go to the top tier.
What the raters did not count on was that mortgage fraud was systematic in the U.S. with almost anybody able to qualify for home loans on the theory that even if the borrower did not pay, there was still solid brick and mortar behind the loan. The mortgages, duly packaged and with what were often 200-page filing documents filled with denials of recourse (the covenants), were sold by lenders to investment banks. Investment banks then peddled them as structured finance products to pension funds and individual investors.
In order to keep the prices of the packaged mortgages high, in spite of admissions in the covenants that the innards of the mortgages might be rotten, the rating agencies used ever more esoteric ratings models.
The lowest levels of the structured mortgages were, because of the designs, unlikely ever to be paid. So the investment bankers took these bottom rung tranches, called “toxic waste” in the industry, and rebuilt them into new mortgage ladders with the top tiers getting paid first and the middle tiers after the top one and so on. And from the toxic levels of those, new third-round structures were built.
The U.S. Department of Justice case turns on the rating system and common sense. The structured finance industry said that you can indeed make a silk purse out of a sow’s ear. S&P evidently knew it was in a troubled market. Emails published in the New York Times on February 5 quote an S&P executive as saying, “this market is a wildly spinning top which is going to end badly.”
Canada had its own version of the subprime crisis in another field of structured finance called Asset Backed Commercial Paper (ABCP). In the mid-2000’s, investment banks figured out a way to make loans with very little risk of non-payment.
Just as mortgages were turned into assets that could be sold as stocks or bonds, through ABCP general car loans and even MasterCard receivables could be securitized and sold to the public. ABCP became a $32-billion market, but when the U.S. subprime crisis brought every structured product into question, investors began to ask for their money back.
The problem was that big investors like the Caisse de depot et placement du Quebec had a lot of money in ABCP. The Caisse’s stake was $12.6 billion, for example. One can presume that Caisse’s own investment staff read the weasel words in the covenants. The language became vital, for the packages of credit card receivables and auto loans and mortgages were supposed to have liquidity provided by chartered banks.
If an investor wanted his money back, the bank would provide it. That, at least, was the idea.
The Canadian ABCP market, part of the global system, required banks in the deals to provide money to pay back holders in the event of a “general market disruption.”
When the subprime crisis hit and the banks were called on the pony up money to nervous ABCP holders, they balked. The banks, which included global giants like Deutsche Bank and HSBC, said the disruption was not “general.”
Lawyers went to work and made a killing. The Office of the Superintendent of Financial Institutions, which had ruled that banks did not have to provide liquidity for paper that they had not issued, protected bank shareholders and the banking system. The big credit raters like S&P and Moody’s which had said that the vagueness of liquidity provisions prevented them from giving any ratings at all, came away with their integrity intact. Only DBRS, which had rated the Canadian ABCP, had to answer for its work.
In the end, individual holders of ABCP got paid. The lesson for any bond investor is to read the fine print in the covenants, believe what it says, and don’t invest in anything incomprehensible or doubtful. †