The need for income naturally biases a portfolio long. It is difficult to earn income without beneficial ownership of an asset – positive carry trades will almost always be net long, absent major distress or dislocation in the markets. Those who need income to survive must then hope for a bull market. They cannot live well without one, absent an interest rate spike like the late 70s/early 80s. But in order to benefit in that scenario, they had to stay short.
To short bonds with success, you have to identify a tipping point. The one shorting a bond borrows it and then sells it. After that, he has to pay the interest on the bond, and maybe a little more, if the bond is hard to borrow, while he waits for the bond’s price to collapse. If the capital losses to a holder of the bond are not greater than the interest paid, the short loses money. (Yes, he makes some money off of interest on the proceeds from the sale, but let’s ignore that for now.) Bonds mostly have finite maturities; time can work against the short seller as the bond gets closer to maturity, because the bond will mature at par, and he will have to pay the par value.
The same applies to credit default swaps [CDS]. The party buying protection must pay for the protection. He looks for a disaster to happen, but as time elapses, and gets closer to the swap termination date, the odds of making money off of a failure declines.
Thus being short any sort of fixed income, whether through shorting or CDS involves paying money out regularly to support the position, with the possibility of incredible payoffs if default happens within the lifetime of the bond or CDS.
This mindset is the opposite of the way bond managers think. A common way they view things is to maximize expected yield over the expected lifetime of their liabilities. That is a simple way for bond managers at banks, insurance companies, pension funds and endowments to manage their bond assets. It is not so easy for total return mutual fund managers, because they can’t tell with accuracy how patient/jumpy their mutual fundholders will be. Typically, they pick an index of bonds, and mirror the most critical aspects of it — duration, convexity, credit quality, etc. Retail investors don’t care about that but they look at the return series, and analyze whether the volatility is too great or too small for them, and if they have beaten many of their peers.
To a good bond manager, he aims to add risk when he is well compensated for it, and reduce risk when it is not well compensated. That said, many bond managers have dumb clients. They want more yield, because they think that yield is free.
I remember the Chief Actuary of a client insurance firm saying to me, “Why can’t you earn the returns of ARM Financial, General American, Jefferson Pilot, and Conseco? (This was around early 1999.) My response was: “You want to take absurd risks? Not only do these firms take asset risks, they are taking more risks than any large firm that I can find. They take asset risks everywhere. Worse, their liability structures are weak, and their leverage is high. A lot of their liabilities can run at will.”
It was not long before General American and ARM Financial failed. Conseco took a few more years; the acquisition of Green Tree helped kill them. Jefferson Pilot wasn’t as bad as the others, but they sold out to Lincoln National while they could.
It is foolish to be a yield hog. Yet, many institutional investors were yield hogs prior to the crisis. Someone had to buy the CCC junk bonds. Someone had to sell protection in order to receive yield. The investment banks could not manufacture gains for those shorting the mortgage market on their own. There had to be yield hogs that wanted to receive yield in exchange for guaranteeing debts. Given the low interest rate environment that they faced, many parties felt they needed to earn more. AIG in particular offered protection on many bonds in order to suck in extra income so that earnings estimates might be achieved. They were the ultimate yield hog, and like most hogs, they got slaughtered.
As for the one offering protection, he must be sure that there is no tipping point over the life of the swap. Then the extra yield would be safe.
I have more to say on this, but let me summarize for now. The need to earn income biases many bond investors to take too much risk. Repeat after me: “Yield is not free.” It exists because of perceived risks; the great question is whether the perceived risks are underplayed, overplayed, or accurate. The good bond manager looks at the risks versus the incremental yields, and spreads his investments among a mix of good risks.