The question could be generalized to 'funds vs their underlying', because many would-be stock investors ask what amounts to the same question. "Should I do my own stock-picking, or should I hire it out and buy a fund (open-end, closed-end, ETF, etc.)?As the linked article suggests, there's no one, single, right answer with regard to 'bonds vs bond funds'. In fact, they suggest that what makes the most sense is to use both. But it all depends one's means, goals, skills, interests, objectives, and opportunities. Right now, the bond market --to use a technical term-- is very sucky. Prices are high; yields are low; and the Fed is committed to destroying price discovery. So, what's a fellow to do? Obviously, go back and review some basics while waiting for The Big Crash (TBG). I like Fidelity's overview a lot. It's solid, thoughtful, honest, careful. But given that I'm "me", I've got some quibbles with it --that no one else needs to agree with--, specifically, what they say about obtaining "proper diversification". Here's my belief, which is easy to provide evidence for. "When markets are under stress, correlations go to 1.0." Therefore, when supposed "diversification" is most needed, it fades like the morning dew. But what spreading your bets widely does force you to do is to bet small, and that reduction in exposure to specific issuers is what's going to save your butt, not the fact that during stress-free times what your buying might not all wiggle and waggle in synch. In short, if you don't over bet your hand, you can't --better, 'likely'-- won't get thrown out of the game. Fido suggests that if you're going to mess with corps or munis, you're going to need a couple hundred thousand dollars to build a diversified portfolio. I'd say their guess is both too small --a quarter million to half million is better-- and too large by about $190k. Here's why I say "diversification" can be obtained with small money. The diversification needn't be synchronic. Although single corporate bonds can be bought, the position isn't often marketable. So you're better off buying at least 2, or not at all. So let's create a starter bond account and fund it with a modest $10k. That means you can buy just two munis, and the diversification police will want to arrest you for recklessness. But here's where things get interesting. If those two munis are solid double-AAs (or better) and GOs, your risks of default are so miniscule, you don't need the hedge that "diversification" supposedly provides. Therefore, if you're mainly making interest-rate bets --not credit-worthiness bets-- you can bond-pick with tiny money, and the fact you can hold to maturity gives you an edge over a fund with a comparable investment objective. Yeah, they're bigger. A lot bigger. But you can go toe-to-toe with them and --maybe-- even do better, because you're not paying annual fees and expenses. (con't in the next post)https://www.fidelity.com/learning-center/investment-products...
Now, let's mix things up a bit and buy some junk (aka, spec-grade bonds). Triple-BBB corporates --especially the lower tier-- are junk disguised as investment-grade. If you're slumming in this trance, you'd better be "diversifying" as widely as you can, and I wouldn't attempt buying with small money. It's just not worth the risk/effort when you're not going to do better than what a well-managed bond fund will do for you. But consider the double-BB tranche. Not many bonds fall into this category. They've been shorn of their tarnished, invest-grade halo --hence, lost an undeserved price prem-- but their fundamentals aren't so bad that default is a huge worry. (It's real, but it isn't more likely than not, as is clearly the case with double-CCs and lower.)So, here's where things can get interesting. Typically, all spec-grade bonds trade at a discount to par and tend to be fairly near-term. So fundamentals can be forecast with fairly reasonable accuracy, as can default-rates. Therefore, managing your risks with this tranche lends itself to being gamed. "If YTMs are such-and such, how many positions can I lose before I would have been better not buying in this tranche at all?" (In short, you're treating the bond as put and its price is your prem. Pair a low prem with a high return, do it often enough, and you're making decent money.)Let's say that a tolerable loss-number --when positions are equally-sized-- is one in five. Thus, if you own 15 positions, you can expect to lose 3 of them. You just don't know which three, or when? Let's return to our tiny starting bond portfolio and its tiny $10k funding. You like the returns that investing in spec-grade debt could offer. But you also understand probability well enough that if you put on five positions --at 2 bonds each to keep the position marketable-- you run the chance of losing all five of them, because 'average expectations' cannot be applied to small samples. (E.g., it's possible to get two "100-year" floods back to back and then none for the next 198 years.) Same-same with bond default-rates. Across a portfolio of 100 spec-grade positions, the expectation of losing just one in five might be what is experienced. Now, here's the problem. In two decades plus of doing this stuff, I have never seen a time when you could waltz into the bond market a pick up 100 double-BBs. So, even if you had the money, you couldn't put it to work. But our starter bond portfolio doesn't have big money. Also, it can't afford to lose a big chunk of it. Thus, realistically, in order to manage risks responsibly, only a single spec-grade issue at a time could be bought, but with this understanding. Over time --i.e., diachronically-- serial diversification can be obtained, just by continuing to buy in that tranche until the sample becomes large enough that 'average expectations' can be applied retrospectively in a meaningful, responsible manner. (to be con't, because I want to talk about bond funds.)
Best Of |
Favorites & Replies |
Start a New Board |
My Fool |