A special thank you to Brett Freeze of Global Technical Analysis for his analytical rigor and technical expertise.

There is an old story about a little boy who was such an extreme optimist that his worried parents took him to a psychiatrist. The doctor decided to try to temper the young boy’s optimism by ushering him into a room full of horse manure. Promptly the boy waded enthusiastically into the middle of the room saying, “I know there’s a pony in here somewhere!”

Such as it is with markets these days.

Finding Opportunity

These days, we often hear that the financial markets are caught up in the “Everything Bubble.” Stocks are overvalued, trillions in sovereign debt trade with negative interest rates, corporate credit, both investment grade, and high yield seem to trade with far more risk than return, and so on. However, as investors, we must ask, can we dig through this muck and find the pony in the room.

To frame this discussion, it is worth considering the contrast in risk between several credit market categories. According to the Bloomberg-Barclays Aggregate Investment Grade Corporate Index, yields at the end of January 2020 were hovering around 2.55% and in a range between 2.10% for double-A (AA) credits and 2.85% for triple-B (BBB) credits. That means the yield “pick-up” to move down in credit from AA to BBB is only worth 0.75%. If you shifted $1 million out of AA and into BBB, you should anticipate receiving an extra $7,500 per year as compensation for taking on significantly more risk. Gaining only 0.75% seems paltry compared to historical spreads, but in a world of microscopic yields, investors are desperate for income and willing to forego risk management and sound judgment.

As if the poor risk premium to own BBB over AA is not enough, one must also consider there is an unusually high concentration of BBB bonds currently outstanding as a percentage of the total amount of bonds in the investment-grade universe. The graph below from our article, The Corporate Maginot Line, shows how BBB bonds have become a larger part of the corporate bond universe versus all other credit tiers.

In that article, we discussed and highlighted how more bonds than ever in the history of corporate credit markets rest one step away from losing their investment-grade credit status.

Furthermore, as shared in the article and shown below, there is evidence that many of those companies are not even worthy of the BBB rating, having debt ratios that are incompatible with investment-grade categories. That too is troubling.

A second and often overlooked factor in evaluating risk is the price risk embedded in these bonds. In the fixed income markets, interest rate risk is typically assessed with a calculation called duration. Similar to beta in stocks, duration allows an investor to estimate how a change in interest rates will affect the price of the bond. Simply, if interest rates were to rise by 100 basis points (1.00%), duration allows us to quantify the effect on the price of a bond. How much money would be lost? That, after all, is what defines risk.

Currently, duration risk in the corporate credit market is higher than at any time in at least the last 30 years. At a duration of 8.05 years on average for the investment-grade bond market, an interest rate increase of 1.00% would coincide with the price of a bond with a duration of 8.05 to fall by 8.05%. In that case a par priced bond (price of 100) would drop to 91.95.

Yield Per Unit of Duration

Those two metrics, yield and duration, bring us to an important measure of value and a tool to compare different fixed income securities and classes. Combining the two measures and calculating yield per unit of duration, offers unique insight. Specifically, the calculation measures how much yield an investor receives (return) relative to the amount of duration (risk). This ratio is similar to the Sharpe Ratio for stocks but forward-looking, not backward-looking.

In the case of the aggregate investment-grade corporate bond market as described above, dividing 2.55% yield by the 8.05 duration produces a ratio of 0.317. Put another way, an investor is receiving 31.7 basis points of yield for each unit of duration risk. That is pretty skinny.

After all that digging, it may seem as though there may not be a pony in the corporate bond market. What we have determined is that investors appear to be indiscriminately plowing money into the corporate credit market without giving much thought to the minimal returns and heightened risk. As we have described on several other occasions, this is yet another symptom of the passive investing phenomenon.

Our Pony

If we compare the corporate yield per unit of duration metric to the same metric for mortgage-backed securities (MBS) we very well may have found our pony. The table below offers a comparison of yield per unit of duration ratios as of the end of January:

Clearly, the poorest risk-reward categories are in the corporate bond sectors with very low ratios. As shown, the ratios currently sit at nearly two standard deviations rich to the average. Conversely, the MBS sector has a ratio of 0.863, which is nearly three times that of the corporate sectors and is almost 1.5 standard deviations above the average for the mortgage sector.

The chart below puts further context to the MBS yield per unit of duration ratio to the investment-grade corporate sector. As shown, MBS are at their cheapest levels as compared to corporates since 2015.

Chart Courtesy Brett Freeze – Global Technical Analysis

MBS, such as those issued by Fannie Mae and Freddie Mac, are guaranteed against default by the U.S. government, which means that unlike corporate bonds, the bonds will always mature or be repaid at par. Because of this protection, they are rated AAA. MBS also have the added benefit of being intrinsically well diversified. The interest and principal of a mortgage bond are backed by thousands and even tens of thousands of different homeowners from many different geographical and socio-economic locations. Maybe most important, homeowners are desperately interested in keeping the roof over their head

In contrast, a bond issued by IBM is backed solely by that one company and its capabilities to service the debt. No matter how many homeowners default, an MBS investor is guaranteed to receive par or 100 cents on the dollar. Investors of IBM, or any other corporate bond, on the other hand, may not be quite so lucky.

It is important to note that if an investor pays a premium for a mortgage bond, say a 102-dollar price, and receives par in return, a loss may be incurred. The determining factor is how much cash flow was received from coupon payments over time. The same equally holds for corporate bonds. What differentiates corporate bonds from MBS is that the risk of a large loss is much lower for MBS.

Summary

As the chart and table above reveal, AAA-rated MBS currently have a very favorable risk-reward when compared with investment-grade corporate bonds at a comparable yield.

Although the world is distracted by celebrity investing in the FAANG stocks, Tesla, and now corporate debt, our preference is to find high quality investment options that deliver excellent risk-adjusted returns, or at a minimum improve them.

This analysis argues for one of two outcomes as it relates to the fixed income markets. If one is seeking fixed income credit exposure, they are better served to shift their asset allocation to a heavier weighting of MBS as opposed to investment-grade corporate bonds. Secondly, it suggests that reducing exposure to corporate bonds on an outright basis is prudent given their extreme valuations. Although cash or the money markets do not offer much yield, they are always powerful in terms of the option it affords should the equity and fixed income markets finally come to their senses and mean revert.

With so many assets having historically expensive valuations, it is a difficult time to be an optimist. However, despite limited options, it is encouraging to know there are still a few ponies around, one just has to hold their nose and get a little dirty to find it.

Believe it or not, any domestic bond trader under the age of 55 has never traded in a bond bear market. Unlike the stock market, which tends to cycle between bull and bear markets every five to ten years, bond markets can go decades trending in one direction. These long periods of predictable rate movements may seem easy to trade, especially in hindsight, but when the trend changes, muscle memory can trump logic leaving many traders and investors offside.

If you believe higher yields are upon us in the near future, there are many ways to protect your bond portfolio. In this article, we present one idea applicable to municipal bonds. The added benefit of this idea is it does not detract from performance if rates remain stubbornly low or fall even lower.  Who says there is no such thing as a free lunch?

Munis

Municipal bonds, aka Munis, are debt obligations issued by state and local government entities. Investors who seek capital preservation and a dependable income stream are the primary holders of munis. In bear markets, munis can offer additional yield over Treasury bonds, still maintain a high credit quality, and avoid the greater volatility present in the corporate bond or equity markets.

Munis are unique in a number of ways but most notably because of their tax status. Please note, munis come in taxable and tax-exempt formats but any reference to munis in this article refers to tax-exempt bonds.

Because of their tax status, evaluating munis involves an extra step to make them comparable to other fixed income assets which are not tax-exempt. When comparing a muni to a Treasury, corporate, mortgage backed security, or any asset for that matter, muni investors must adjust the yield to a taxable equivalent yield. As a simple example, if you are in a 40% tax bracket and evaluating a muni bond yielding 2%, the taxable equivalent yield would be 3.33% (2.00% / (1-40%). It is this yield that should be used to equate it to other fixed income securities.

Negative “Tax” Convexity Matters

Thus far, everything we have mentioned is relatively straight-forward. Less well-understood is the effect of the tax rate on muni bonds with different prices and coupons. Before diving into tax rates, let’s first consider duration. Duration is a measure that provides the price change that would occur for a given change in yield. For instance, a bond with a duration of 3.0 should move approximately 3% in price for every 1% change in yield.

While a very useful measure to help quantify risk and compare bonds with different characteristics, duration changes as yields change. Convexity measures the non-linear change in price for changes in yield. Convexity helps us estimate duration for a given change in yield.

For most fixed rate bonds without options attached, convexity is a minor concern. Convexity in the traditional sense is a complex topic and not of primary importance for this article. If you would like to learn more about traditional convexity, please contact us.

Munis, like most bonds, have a small amount of negative convexity. However, because of their tax status, some muni bonds have, what we call, an additional layer of negative tax convexity. To understand this concept, we must first consider the complete tax implications of owning munis.

The holder of the muni bond receives a stream of coupons and ultimately his or her invested principal back at par ($100). The coupons are tax free, however, if the bond is sold prior to maturity, a taxable capital gain may occur.

The table below illustrates three hypothetical muni bonds identical in structure and credit quality. We use a term of 1 year to make the math as simple as possible.

In the three sample bonds, note how prices vary based on the range of coupons. Bond A has the lowest coupon but compensates investors with $2.41 ($100-$97.59) of price appreciation at maturity (the bond pays $100 at maturity but is currently priced at $97.59). Conversely, Bond C has a higher coupon, but docks the holder $2.41 in principal at maturity.

For an uninformed investor, choosing between the three bonds is not as easy as it may appear. Because of the discounted price on bond A, the expected price appreciation ($2.41) of Bond A is taxable and subject to the holder’s ordinary income tax rate. The appropriate tax rate is based on a De minimis threshold test discussed in the addendum. Top earners in this tax bracket pay approximately 40%.

Given the tax implication, we recalculate the yield to maturity for Bond A and arrive at a net yield-to-maturity after taxes of 4% (2.50% + (2.50 *(1-.40). Obviously, 4% is well below the 5% yield to maturity offered by bonds B and C, which do not require a tax that Bond A does as they are priced at or above par. Working backwards, an investor choosing between the three bonds should require a price of 95.88 which leaves bond A with an after tax yield to maturity of 5% and on equal footing with bonds B and C.

Implications in a rising yield environment and the role of “tax” convexity

Assume you bought Bond B at par and yields surged 2.50% higher the next day. Using the bond’s stated duration of .988, one would expect Bond B’s price to decline approximately $2.47 (.988 * 2.5%) to $97.53. Based on the prior section, however, we know that is not correct due to the tax implications associated with purchasing a muni at a price below par. Since you purchased the bonds at par, the tax implication doesn’t apply to you, but it will if anyone buys the bond from you after the 2.5% rise in yields. Therefore, the price of a muni bond in the secondary market will be affected not just by the change in rates, but also the associated tax implications. Assuming the ordinary income tax rate, the price of Bond B should fall an additional $1.65 to $95.88.  This $1.65 of additional decline in Bond B’s price is the penalty we call negative tax convexity.

The graph below shows how +/- 2.50% shifts in interest rates affect the prices of bonds A, B, and C. The table below the graph quantifies the change in prices per the shocks. For simplicity’s sake, we assume a constant bond duration in this example.

It is negative tax convexity that should cause investors, all else being equal, to prefer bonds trading at a premium (such as bond C) over those trading at par or a discount. It is also worth noting that the tax convexity plays an additional role in the secondary market for munis. Bonds with prices at or near par will be in less demand than bonds trading well above par if traders anticipate a near term rise in yields that will shift the par bond to a discounted price.

Summary

Yields have fallen for the better part of the last thirty years, so muni investors have not had to deal with discounted bonds and their tax implications often. Because of this, many muni investors are likely unaware of negative tax convexity risk. As we highlighted in the table, the gains in price when yields fall are relatively equal for the three bonds but the negative deviation in price in a rising yield environment is meaningful. Given this negative divergence, we recommend that you favor higher coupon/ higher priced munis. If you currently own lower priced munis, it may be worth swapping them for higher priced (higher coupon) bonds.

Addendum: De minimis

The tax code contains a provision for munis called the de minimis rule. This rule establishes the proper tax rate to apply to capital appreciation. The following clip from Charles Schwab’s Bond Insights provides a good understanding of the rule.

The de minimis rule

The de minimis rule says that for bonds purchased at a discount of less than 0.25% for each full year from the time of purchase to maturity, gains resulting from the discount are taxed as capital gains rather than ordinary income. Larger discounts are taxed at the higher income tax rate.

Imagine you wanted to buy a discount muni that matured in five years at $10,000. The de minimis threshold would be $125 (10,000 x 0.25% x five years), putting the dividing line between the tax rates at $9,875 (the par value of $10,000, minus the de minimis threshold of $125).

For example, if you paid $9,900 for that bond, your $100 price gain would be taxed as a capital gain (at the top federal rate of 23.8%, that would be $23.80). If you received a bigger discount and paid $9,500, your $500 price gain would be taxed as ordinary income (at the top federal rate of 39.6%, that would be $198).

It is important to note that some bonds are issued at prices below par. Such bonds, called original issue discount (OID), use the original offering price and not par as the basis to determine capital gains. If you buy a bond with an OID of $98 at a price of $97.50, you will only be subject to $0.50 (the difference between the OID price and the market price) of capital gains or ordinary income tax.