Four seasons make up a year, but most people fish (for sport) or farm (for a living) only a couple of them, using the slack times for rest and maintenance, for reviewing the past season and planning the coming one. The same practice might be useful to bond investors, faced as we are with interest rates at 40-year lows. Our cycle isn't an annual one, but it is nonetheless a cycle of planting and harvesting, of highs and lows tied to Fed policy and global political/economic events. Splotto touches on this theme of preparedness in another thread, vowing to jump on bonds when (if?) they ever reach 8% again, and I'd like to sketch in more detail some of the thinking that needs to happen between now and then. (Note: In what follows I ignore munis, preferreds, and converts, which are topics for other posts.)The two major rating systems for corporate bonds are Moody's and S&P's, with the former alleged the more conservative, but, for all practical purposes, the two are interchangeable and S&P's all-caps is easier to read and type, so that's what I'll work with here. Using a system of triple to single letters qualified by pluses and minuses, S&P creates a ranking system we all have seen examples of, but it is instructive to repeat in full:AAA AA+, AA, AA-A+, A, A-BBB+, BBB, BBB-BB+, BB, BB-B+, B, B-CCC+, CCC, CCC-CC+, CC, CC-C+, C, C-D NRNote a couple of things: - Treasuries fall outside (and above) such a rating scheme, for being guaranteed against default. Corporates are merely a promise to pay. (Were Treasuries included in the ranking, they'd be AAAA or similar.)- The top category, AAA, isn't qualified by plus/minus. - Notch differences among bonds of the same issuer indicate subordination of issues to more senior ones. (Yes, this point deserves fuller explanation. Hit the books is my suggestion.)- The top four categories are commonly known as “Investment Grade” ratings, and the emphasis is on estimating the ability of the issuer to make timely payments of principal and interest. - The next five categories are considered “Speculative Grades” and are commonly called “Junk Bonds”. Here defaults are assumed to happen in statistically significant numbers and estimating recovery rates --rather than primarily the issuer's ability to make timely payments of principal and interest-- plays a part in assigning a rating. - “D” means the issuer has stopped making payments of principal and interest, not that the bond has no value. - “NR” means “not rated”, which happens for a lot of reasons, some of which the guide books will mention and all of which need to be investigated. It is not a comment on the issuer's creditworthiness.Now, here's where things get interesting. Given the smorgasbord of categories to choose from, how does one put together a bond portfolio?For a start, let's conflate the full list of 27 some categories given above into a more manageable 5:1) Treasuries and AAA's2) AA and A3) BBB and BB4) B and CCC5) CC/C/DConsider what I've done: I've made the very best corporates equivalent in creditworthiness to Treasuries on the assumption that defaults are going to be so rare as to be negligible. (The historical record suggests the number is virtually nil). If you want only the bluest of the blue chips, this is where you'll confine your buying. If you want a bit more yield in exchange for accepting a bit more risk, you'll buy AA/A as well.I've downgraded BBB bonds to junk, which is where I think they belong. At best, BBB is a transitional category and a would-be buyer suffers under two disadvantages: One, the nominal investment rating of the bond means prices are held at a premium compared to default risk, and, Two, any subsequent downgrade means the institutions – many of whom can't own spec-grade bonds—will dump them and prices will plummet. My thinking is this: If you want investment-grade, then buy it. If you want junk, then buy it, but don't screw around with BBB's, thinking you're getting fat yields as well as safety. (You can't have your cake and eat it, too.) The historical record suggests that the defaults rates over 1-year and 5-year periods for AA and A are ~1/10%, which is tiny compared to the hazards stock investors accept without second thought, and which due diligence --in the form of diversification and proper position sizing-- should offer adequate protection against or make any losses manageable, especially since “default rate” and “recovery rate” are two very different things. But risk is risk and if you want to avoid it, then avoid BBB's. However, it is no biggie if an issue in your portfolio fails as long as you were aware of the risk and managed for it properly. Losses are simply a part of investing, and the average recovery rate --even for junk-- is 40%. I've split junk bonds into three groups: BBB/BB vs. B/CCC vs. CC/C/D on the belief and experience they involve quantitatively different research efforts and investment management efforts (which is a topic for another post: “Easy Junk, Hard Junk, and Special Situations”). Now, just because I've cut the field down to five groups, that doesn't mean an would-be bond investor has to confine herself to just one of them, nor does it suggest how they might be mixed and matched, nor how they are to be managed. At the risk of over-simplification, I'd like to present the following portfolio scheme, based on “Selections” (from 1-5) and “Investing Styles” ( Buy-and-Hold Investing, Opportunistic Investing [aka, Trading], and Hedged Investing—which is a term I'll need to explain in another post):.........Limited to......create........focus on......focus on ........Invest-Grade..Diversified..Spec-grade...Special Situations B&H.......yes.....yes.....na.....naTrade.....yes.....yes.....yes.....yes Hedge......na.....na.....yes.....yes[“na” doesn't mean that it can't be done, just that it is redundant, not worth the effort, or inappropriate- IMHO. Also, my apologies for this "table". TMF doesn't translate Word docs well.]As an example of how the typology might be interpreted, let's assume the would-be investor describes herself as a Buy-and-Hold sort of person –-which is certainly appropriate to owning individual bonds-- and wants to build a diversified portfolio by selecting from across the credit spectrum, rather than limiting her buying to just investment-grade bonds. How much of Group1? Group2? Etc.?Here's where the ugly topic of MPT (Modern Portfolio Theory) normally steps forward and takes over, but let's set it aside and do things the Fool's way, from the beginning. The first distinction one has to make is between an arithmetic series (e.g., 1, 2, 3, 4, …), a purely geometric one (2, 4, 8, 16, 32,…). and quasi-geometric ones like 1, 2, 3, 5, 8, 13, 21, … (aka, a Fibonacci series, which I'll argue in another post has great utility for bond investors). To keep things simple, let's assume our would-be builder of a diversified bond portfolio isn't going to mess with special situations (eliminating Group5) and she wants to allocate using a simple arithmetic series: 1, 2, 3, 4, or --more fully, and from the top down: 40%, 30%. 20%, 10%, which –-conveniently –- adds to 100%, thusly: 40% Treas & AAA30% AA/A20% BBB/BB10% B/CCC ----100%Such an allocation would create a fairly high-quality portfolio that offers a bit more total return than a purely investment-grade portfolio over the long haul and other weightings invite consideration. How much money would it take? Ah, now, finally, things get interesting. In normal times, investment-grade bonds are priced at par and discounts obtain as credit quality decreases on something like the following schedule:Treas/AAA... 100AA/A ....... 98BBB/B....... 88B/CCC....... 75CC/C/D...... 10-60 Buying a single bond is possible, but 5 or 10 is the more frequent minimum, and buying more than 1 both lowers the cost of average commission paid and also makes the lot more sellable. It is my preference and practice –- which you will need to match to your own investment situation--to limit junk bonds to 5 bonds/issuer and to allow as much as 15 bonds/issuer for the high quality stuff (with no limits on Treasuries.) If that is the scheme followed, then $100,000 divides itself up this way: Treas/AAA @2-4 positions, @ 10-15 bonds each, @ 100 = ~$40,000AA/A @3-5 positions @ 5-15 bonds each, @98 = ~$29,400BBB/BB @4 positions, @5 bonds each, @88 = ~$17,600B/CCC @ 2 positions, @5 bonds each, @75 = ~$7,500 ----------------------------------------------------~$94,500Obviously, each of the variables can be varied to create very different profiles and very different portfolios, but the strategy of using face value rather than cash value is a self-restricting way to limit risk –-you're buying less of the cheapest stuff, which is the riskiest stuff, but also the highest yielding stuff, meaning, the dollars you do commit are -–potentially-- working very hard for you. Summary: there's lots more that needs to be said about how to build a bond portfolio and how to put this stuff into Excel so that portfolio statistics like avg. YTM, avg. CY, avg. cost, avg. rating, etc. can be calculated and benchmarked, but this much at least puts some ideas on the table. Caveat: Money is the easy part about building a bond portfolio. The hard part is time: - The waiting time it takes for the shopping opportunities to happen. - The research time it takes to verify that they are, in fact, opportunities.- And the management time it takes to ensure that the promise of the now-owned opportunity doesn't die from neglect. Charlie
Best Of |
Favorites & Replies |
Start a New Board |
My Fool |
BATS data provided in real-time. NYSE, NASDAQ and NYSEMKT data delayed 15 minutes.
Real-Time prices provided by BATS. Market data provided by Interactive Data.
Company fundamental data provided by Morningstar. Earnings Estimates, Analyst Ra