Why Bonds Are
Riddles, enigmas and other investment mysteries
By Jim Reber, President & CEO, ICBA Securities
A best-selling book from the 1990s, “Why Things Are,” took a humorous look at curiosities originating from everyday existence, with the modest goal of answering “every essential question in life.” This column has a more ambitious purpose: to clarify a few of the phenomena surrounding your investment in bonds. So, put on your seat belts as we head out for a ride on the fixed-income highway.
Why do continuously callable bonds have a higher yield than bonds with only one call date?
An investor in a callable bond actually does two things simultaneously when he or she makes the purchase: 1. the buying of the bond from the issuer, and 2. the selling of the right to prepay the debt early to the issuer. For bonds with multiple call dates, the issuer pays something for each right. In aggregate, the value of all the call options as of the issue date is reflected in the higher yield compared to a non-callable bond, or one with one only call date.
Why do mortgage-backed securities priced at 100.00 have different yields than the stated coupon?
For this answer, we have to do a two-step move. First, let’s review what bond-equivalent yield (BEY) means. The yield on a bond has a standard calculation that assumes semi-annual interest payments in arrears (meaning at the end of the period), and principal at maturity date. For example, if you buy a new two-year agency bond, your first interest date will be exactly six months from original settlement date.
Mortgage-backed securities (MBS) differ in two ways. First, some interest payments are received monthly. That’s good from a BEY standpoint. However, MBS also have this nice little convention known as “delay days” before any principal or interest are paid. That’s bad for BEY purposes. Most mortgage securities have delay days ranging from 45 to 85 days. Factoring in the delay of both principal and interest over a period of time probably means the bond-equivalent yield for a mortgage will be less than the coupon, at par.
Why do tax-free bonds have lower price volatility than taxable bonds?
Another popular question. The best way to illustrate this is to take a hypothetical taxable bond, like a 10-year Treasury note, and compare it to a 10-year tax-free bond. At the moment, the 10-year Treasury yields (conveniently) 1.50%, and 10-year high-quality munis yield about 1.65% (for a tax-equivalent yield of 2.05%).
If rates rise 100 basis points (1.0%), the Treasury will then yield 2.50% and its price will decline 8.7%. The muni will only have to drop by 6.9% for its tax-equivalent yield to get to the 3.05% that the market will likely require. It’s worth noting that the two bonds’ volatilities have become more alike since tax reform at the end of 2017. Stated another way, muni price volatilities have increased as tax rates have fallen.
Why is the duration of a bond shorter than its average life?
The most commonly used measurement for price volatility of a fixed-rate security is duration. It is the weighted average period of time to receive all the principal and interest on a bond. Average life, on the other hand, is a tool used in many cases to forecast cash flows on mortgage securities and is the weighted-average time to receive principal only. Since most MBS are back-loaded with principal on their paydown schedules, average lives are almost always longer than duration for a given bond.
Duration is useful for estimating how much a bond’s value will change given a change to market rates. A duration of 3.0 means that the price will decline about 3%, given a 1% increase in the market yield. Concepts like effective duration and convexity also factor into this discussion, but that’s another story for another day.
Why are bankers more inclined to sell at gains than losses?
I’m no shrink, but there is an element to behavior that psychologists refer to as “compartmentalization.” It seems to me that taking a loss is seen by the seller as admission of a bad purchase. In fact, the overwhelming majority of losses realized are market-driven, and I think we agree that bankers aren’t market-timers. So, by being more inclined to take gains, portfolio managers can neatly stack their aggregate sales in the “win” column.
There are just two problems with this practice: selling the winners means your portfolio yield is probably going down, and you’ve triggered immediate tax liability. I would suggest, especially in a robust earning year like the present, that a bank work with its tax accountant and brokers to devise an income-deferral strategy. Higher yields mean more future income. And that falls into the “long-term vision” compartment.