I will skip any pretense that locking in 0.5% real yield for 30 years is a bargain. It's a lousy deal. The only reason I mention it is that fixed income choices can get even more lousy than this and stay that way for decades. They have been more lousy for about four decades of the 20th century, 1940s through 1970s. Though historical average return from bonds has been 2-3% per year above inflation, bonds during those four decades instead lost 1-2% per year to inflation. To be fair, periods before and, especially, after those four decades have been much kinder to bond investors, but the fact remains that even 0.5% guaranteed real return was desirable for a long time.
If you are (a) under 40-50 years of age and have full 30-year or longer investment horizon, (b) don't mind inevitable significant fluctuations in the value of these 30-year TIPS, and (c) appreciate the portfolio diversification value and long-term (but definitely not short-term!) safety of 30-year TIPS, then it's not unreasonable to buy them even today in tax-sheltered accounts with the firm plan to hold on to them for 30 years.
The tax-sheltered account part is very important. If you hold TIPS in taxable accounts you will be on hook for annual taxes on their growth in value with inflation even though you won't actually see a penny of that growth till the bond matures or you sell it. Do not hold TIPS in taxable accounts unless you really, really know what you are doing. What you should hold in taxable accounts are I-Bonds which today will just keep up with inflation (meaning lose 0.5% annually to the 30-year TIPS) but which are easily redeemable after just one year and have absolutely no chance of even short-term losses. They are TIPS "lite" with a host of very desirable characteristics.
Again, do understand what you are getting and not getting with these 30-year TIPS. You get a chance to lock in for a very long term into returns which are significantly below average but, at the same time, significantly better than worst-case scenario. You should absolutely not count on these 30-year TIPS to guarantee any kind of short-term return. It is very possible for them to plunge in value by 30% or even more over the course of one year. My guess is that they will experience that sort of plunge in value at least once during their 30-year life. But you can count on them to recover their value towards the end of those 30 years and ultimately deliver the promised keeping up with inflation plus 0.5% on top. In addition, they will serve as a diversifier for a portfolio heavy in stocks and non-inflation-adjusted bonds (which is to say, nearly all bonds other than TIPS). It is also conceivable that your 30-year TIPS might even increase in value unexpectedly and provide you with a nice short-term profit -- but you should not count on it. And, again, it's very important to only hold TIPS in tax-sheltered accounts so you are not on hook for annual taxes on their "phantom interest".
You can buy 30-year TIPS direct from brokerages like Fidelity (screen-captured above), directly in your IRA accounts. In many cases such purchases will not even carry commissions.
Thanks so much for all your helpful advice and analyses. I am new to your blog, so I don't know whether you've addressed the following question before, but: What are the differences between holding TIPS directly as individual bonds and investing an equal amount of your IRA funds in a TIPS mutual fund? It seems like the ongoing yield for the TIPS mutual fund will be very low as compared to other investment grade bonds, but you won't necessarily be getting any long-term guaranteed real return due to the turnover in the fund's holdings. Am I missing something?
ReplyDeleteThe mechanics of what happens in a bond fund over N years are quite different from holding an individual bonds for the same N years, but the returns to you, the investor, will likely not be very different.
DeleteThe first difference is that with a bond fund you will be paying an annual management fee. These can range from as low as about 0.1% annually to as high as 1% or higher for really bad funds. You probably already know that you need to always ruthlessly minimize investment expenses and are hopefully investing in bond funds in 0.1-0.2% expense range.
Next comes the question of whether the fee even that low is worth it? For non-US government bond funds the answer is yes, absolutely. It's worth it because the fund diversifies your holdings so if one issuer goes bankrupt you don't even notice it. For bond funds that hold exclusively US federal government debt, this consideration does not apply since they are not diversified across issuers. By the way, don't buy what you occasionally read about diversifying US government bonds with other issuers' bonds "for safety". There is not one bond in the world that would not absolutely implode should the unthinkable happen and the US defaults on its obligations. It would make the 2008-9 meltdown look like a walk in the park. Then it really will be time for gold and maybe guns and ammo too. Fortunately, the likelyhood of US default is extremely remote if for no other reason than that the US can always print more of its money. Hyperinflation is more likely than default (not that I expect hyperinflation either, of course).
So if your TIPS fund does not provide any additional safety, what are you paying your 0.1-0.2% per year for? In today's low yields that may well be 10% of more of the fund's entire return, depending on the average maturity of its holdings! Well you pay it for two things: (1) liquidity and (2) constant maturity over time.
(1) Liquidity means that you can always easily buy and sell shares of your TIPS fund at a relatively low cost. This is not always the case with individual TIPS bonds. Bid-ask spreads on those can exceed 1% meaning that every time you lose full 1% of the trade's value. By comparison, with bond mutual fund you lose nothing when you buy and sell and with a bond ETF you might only lose 0.1% or less (though you also sometimes have to pay commissions). Most TIPS bond funds are also structured to pay out as interest the appreciation of underlying bonds with inflation. With an individual bond, you would only get its headline yield as interest. Its inflation adjustment would increase the principal value that you only realize when you sell or when the bond matures (though you would still be on hook for taxes for this principal growth, which is why I recommend not holding individual TIPS in taxable accounts).
Delete(2) Constant maturity means that a bond fund will strive to always keep average maturity of its holdings at, say, 10 years. As older bonds approach their maturity, the fund sells them and invests the proceeds into new bonds with longer maturities. The end result is that the average maturity stays more or less constant forever. Holding an individual bond to maturity, of course, is very different. You may start off with a 30-year bond. After 10 years you are left with a 20-year bond. After another 18 years you effectively are holding a 2-year bond. And so on.
Whether and how much this constant maturity matters is not easy to answer. As long as the yield curve remains positively sloped with longer maturities yielding more than shorter ones, you can generally expect a constant maturity fund to deliver slightly higher return than an individual bond held to maturity. The caveat is that the opposite is likely to be the case during periods of rising interest rates -- and in my estimation such period is likely ahead of us in the years to come. At least it's more likely than continuing drops in interest rates which has been the case (with some interruptions) for about 30 years now.
Unfortunately neither I nor anybody else in the world has the knowledge of exactly how the yield curve will move over the years. All we can say is that the bond market incorporates all this information into its pricing. In the recent years the bond market's "prediction" of yield curve moves turned out to be rather poor. It's anybody's guess how it will turn out in the coming years.
DeleteComing back to what's relevant to you, the investor... If you have a truly long-term outlook and don't mind holding a TIPS bond for decades till it matures, you can save the 0.10+% annually in expenses by simply holding individual TIPS in your IRA. If you don't have the outlook that long or wish to be able to rebalance often then individual bonds are probably not for you. The constant maturity factor can play out way too many different ways, favorable and unfavorable. I would try to understand the mechanics of it, but not take it into consideration as either positive or negative. Focus on liquidity of a bond fund versus saving annual expense ratio as your primary consideration.
I use a TIPS ETF in my bond allocation portion of my portfolio, along with corporate, junk and BND ETF. After reading your blog the past week I feel I should dump my TIPS( too over priced) buy I bonds-but amounts are limited, move to much shorter term corporate bond funds, avoid treasuries and raise much more cash than I presently have(1%). the outlook seems pretty bleak but what can you do to balance you stock risk. I've been pretty much 50/50. will the TIPS ETF help the balance or even more risky than stocks when held in an ETF? thanks love your ideas.
ReplyDeleteI-Bonds are definitely the best deal going and you should max them out since they essentially start serving double role as emergency cash after a year, in addition to being highly competitive in yield in today's environment and safe from potential rising interest rates.
DeleteThe other moves you mentioned need some thinking through before you jump to them. Yes, treasuries are definitely overpriced relative to corporates and munis, but should another crisis strike, they are also safer. If you dump all your treasuries for corporates, you will increase the risk profile of your overall portfolio. That's not necessarily a bad thing since you are compensated for it in higher yield but you need to realize that there is risk involved there. I would try to do this in combination with lowering portfolio risk elsewhere (e.g. in combination with selling off some stocks and buying bonds instead)
Raising more cash is going the other way... Giving up yield in return for lower risk (from rising interest rates). If you simply plan on holding cash earning more or less 0%, I don't think this is a good idea. But if you can find places to keep FDIC-insured cash while earning 1-2% on it (including CDs with a cheap option to break) then it's a good idea.
Moving to shorter-term corporate bonds from longer-term ones is similar to the cash move... you are giving up yield in return for safety from rising interest rates. It may work out or it may backfire -- I haven't the slightest clue. I would not make such a move without a good reason and I can't think of one now.
As wrote in some other post, I would try to not make moves that change the risk profile of my portfolio when there isn't an obvious reason to do so (and I can't see one now). Instead I would focus on keeping risk profile more or less the same but look for ways to squeeze more yield without taking on new risk. Here are some examples...
- I-Bonds and CDs with option over short-to-intermediate term treasuries/TIPS/corporates or cash
- EE-Bonds over long-term treasuries and maybe even long-term corporates
- In taxable accounts Vanguard "long-term" muni funds over "intermediate-term" corporate funds (nearly same maturity and same yield but munis are of higher quality and tax-exempt)
Thanks that all makes sense. Im essentially using the gone fishin portfolio with higher bond allocation. The essential question is if I keep my portfolio 50/50 stocks and bonds is my interest rate risk going to be too high even with intermediate term ETFs. Btw I cant tell you how helpful your blog has been in the two weeks since I discovered it.
ReplyDeleteYou are very welcome! Glad my blog has helped you.
DeleteAs far as being concerned about interest rate risk with intermediate-term (i.e. maturity/duration of around 6 years) I think it's mostly irrational. Yes, it's very possible, even likely, that somewhere along the way your bonds will drop in value by a few percent. "Worst case" (which is actually best case because you are investing in bonds for yield, not for price stability) scenario your bonds will lose as much as 15% if interest rates revert to around historical averages. And that's not at all likely in my estimation. If and when any drop happens you will immediately be rewarded by higher yields in new investments as well as on re-investment of interest. You will likely recover even a significant loss in value in a year or two. In that "worst-case" scenario it will take maybe 5 years. From that point on you will be enjoying higher yields with no regrets about price drops. Every single intermedite-term bond investor under maybe 75 years of age should absolutely be wishing and hoping that interest rates do in fact spike!
If you are locked into long-term (20+ years) bonds then the interest rate risk really is a concern because price drop potential there exceeds 20-30% in truly adverse scenarios and you might not even live long enough to break even. That's why locking into long-term bonds is a big committment. That is simply not the case with intermediate bonds.
If you are still concerned about it all, then read over my other entries about optimizing fixed income portfolio (e.g. http://www.longtermreturns.com/2012/12/optimizing-fixed-income-portfolio.html ) by avoiding treasuries and TIPS in favor of I-Bonds and CDs. I-Bonds are completely immune from interest rate risk because you can always liquidate them at face value after the 1st year at a small penalty (and no penalty after 5 years). Many CDs come with a similar option to break at a small penalty. Switching out of intermediate-term treasuries/TIPS for I-Bonds and 3-7 year CDs (in IRAs if possible) takes some effort but will likely result in ~1% higher yield with much lower interest rate risk (and equal safety from default).
Furthermore, with the yield curve remaining quite steep your intermediate-term bonds have around 0.20% annual "safety cushion". Meaning that they will not lose any value if interest rates rise only gradually, at around 0.20% per year. As this potential rise happens your bonds are sliding down the yield curve toward maturity in the downward direction. Over a course of the year they will slide around 0.20% down which will offset 0.20% of a rise in interest rates. The end result will be more or less no change in bond's price. The mechanics are slightly more involved but the end result is mostly the same for bond funds.
Bottom line is that all the talk about interest rate risk is majorly overblown for short- and intermediate-term bond investors. The "danger" with today's short and intermediate bonds is not the rising interest rates but the very high likelyhood of not even keeping up with inflation for the duration of the bond. Rising interest rates (back to historical normalcy -- not 20+% hyperinflation or Greece scenarios) should absolutely be hoped for by all bond investors, other than maybe those locked into the very long-term maturities. After the initial drop in bond values rising interest rates will finally bring positive real returns, which is what everyone wants (or at least should want) to see in fixed income space.
Could the availability of leveraged equity etfs at fairly low cost to the average investor be of any use? What if a young investor were to take say 2/3rds of his retirement contribution and buy I bonds and take the other third and buy shares of spxl? You could set a decreasing maximum leverage limit of say 300% of total portfolio when you were 25 to 100% when you were 85 and no minimum leverage limit. Every year or quarter that leverage exceeded the max, you'd sell spxl and buy more bonds. If you could save a total of 15% of income, and you considered worst case the whole spxl bet went south, you'd still end up with -.5 real return on ten percent of your lifetime earnings plus social security to live on - much more than most Americans retire with. The potential for the the "let it almost all ride" spxl bet to pay off huge seems worth considering.
ReplyDeleteA conventional diversified portfolios range of outcomes would be all the way down to worse than just the ,66 i-bond portion only, to just a fraction of what an average 30 year period in equities would produce if one stayed tripled down on the market with 1/3rd of his regular contributions and only protecting the very, very frothiest of returns during ones youth. For the worst case to occur (no money ever from the spxl bet) the S&P would have to have gone down by an average of about 1/2 of a percent per year for 60 years. If that really happened the future would be one different place already, not sure any principal protection/diversification strategy (aside from one that included seeds and ammo) would make much difference. If the S&P goes up by an average of just 1% more than I bond returns (probably half the historical average), and remember you'd be getting a multiple of that on 1/3 of your contribution, you'd have a comfortable retirement and leave a nice estate. If it averaged say 8% (making it in the neighborhood of 20-23% tripled down)- yowser.
I don't think that's a good idea at all -- at least not with daily resetting leveraged ETFs like SPXL. That daily reset feature has very strong potential to result in negative returns even in an upward-trending, but volatile market.
DeleteFor example... consider the market returns from the following sequence trading days: +0.5%, -0.4%, -2.0%, +2.0%
The index will have ended this sequence having advanced by (1.005 * .996 * .980 * 1.02 - 1) = 0.06%. Annualized this would result in close to +4% return
But a daily-resetting triple-leverage product would have resulted in (1.015 * .988 * 0.94 * 1.06 - 1) = -0.08% or about -5% annualized.
So you absolutely can lose money on these resetting leveraged products even if you guess the direction of the market right. Volatility is just as important as the trend.
You can quickly verify that SPXL does not in fact deliver triple-S&P500 returns over a long time even in a strong bull market because of volatility. Since inception of this fund a bit over 4 years ago this fund only delivered only about 2.2 times S&P500 total return (dividends included) despite an incredible bull market with total returns of about 100% for S&P500.
The long-term reality for this product is even worse because of the 0.95% expense ratio which is a very large drag over time.
You may also find this interesting when thinking about leveraged products: http://www.longtermreturns.com/2011/11/leveraged-etf-math.html
There are some non-daily resetting products available. Most practical of those are monthly-resetting ETNs from UBS, but they carry issuer risk.
So right now your saying that the product is up by 120% over the SPY in it's first four years and that balanced against the other things you point out proves it's really, really bad? Even if one uses it in conjunction with drastically increasing ones regular contributions to risk free return over a conventional equity over-weighted portfolio? All of the drags you say are the things to focus on here do actually have to overpower some real life excess returns that unequivocally can be experienced right?
DeleteAre you saying your analysis is definitive? I checked and since 1928 the market has doubled almost 11 times, that's about once every 8 years. My investment time horizon is 60 years. Am I wrong to at least have some trepidation about your advice/ opinion that includes "the market does not move significantly one way or another -- which is most of the time " The market does indeed move significantly. Up. Most of the time. Just my amateur observation.
I looked at the chart and since the absolute market bottom in Mar 2009 SPY is up 220% and SPXL is up 703%, so the vagaries of decay vs. compound interest with these products aren't definitively bad that I can tell for sure. Isn't it true that the further the market price of the non leveraged base index moves away higher from the day you invested the better your leveraged investment performs relative to a non leveraged dollar invested the same day? So far every high has become a distant low in a way way shorter period than when a typical retirement investor usually wants to BEGIN slowly withdrawing their money relative to when they started investing right?
Yes, that's exactly what I am saying. Making conclusions about future returns from daily-resetting leveraged products based on what happened in the past 4 years is a big mistake. The past 4 years have seen an incredible bull market which is not at all representative of historically average 4 years (and history has been very kind to stock investors -- especially US stock investors -- in the 20th century).
DeleteI did run the numbers once a long time ago and found that being in daily-resetting leveraged product would have paid off spectacularly in the past. I don't remember if I used double- or triple-leverage and I don't remember my starting point for that. I also don't think I accounted for the higher expenses.
This is an interesting subject, so I will try to repeat that experiment and make a post about the results. I might use Monte Carlo instead of historical data, since anyway we know history won't repeat exactly. Thanks for the idea and look for the results in my blog.
No, thank you for the thoughtful response! I look forward to more scrutiny of these things, I confess I've been doing the unthinkable and owning the SPXL for going on 2 and a half years. I'll continue to play the gambler role in the discussion. One thing I can't help noticing about the investment advice world is the fact that from the very get go, people are hammered about the risks associated with investing in things that can regularly lose principal, and margin and leverage is almost off the page of discussion for newbies.
DeleteThe big money investors and even the institutions that are seeking growth scour the universe for risky investments and regularly use boatloads of leverage. I read that the too big to fail banks on average were levered up by like 3300% (they outright owned only about 3% of the assets on there books) in mid 2008. They choked and would be history now of course without various and large government supports and mark to market shenanigans, but notice how quickly they are right back to profiting in the billions. It's only leverage that can do that.
Interestingly most individual average investors have literally less than nothing to lose when they start working. They usually start right off by going deep into debt on a home mortgage often while owing big on cars and college loans. Our biggest asset by far is our future earning potential, which pretty much can't be protected by any capital market investing choice at all.
If the venture capital and private equity folks that have lifetimes worth of principal accumulated to lose are continually bellying up to the most risky investments and borrowing against principal to bet even more, doesn't it seem counter intuitive for someone who barely has a pot to piss in and is probably going to be living mostly off weekly earnings for 80% of their life to be advised to avoid principal risk as much as possible and to be told leverage is certain suicide?
Regular, increasing contributions over 40 yrears while not counting on ANY cash flow during that time, plus the basically almost standard condition of most your peers that are sure to arrive at retirement with nearly nothing saved, to me seems like the perfect opportunity to take some measured chances on big returns with at least a part of your long term contributions.
Yes, lets go ahead and grant that 1/3 of your money may marginally under perform a potential poorer than historically performing market. In the suggested alternative (standard stock/bond/cash diversified, annually rebalanced non-leveraged portfolio) that same unexpectedly poor performing market will have provided you with basically unexpectedly poor performance on half or more of your portfolio anyway, right? A slim prize for basically giving up a chance at some real outsized returns should history come even close to repeating no? If you're talking just a few percentage points difference average compound annual growth rate over 45 years I don't have to tell you the the reward would be astonishing, even on an incrementally smaller regular contribution, the balance of which one could use to fortify the risk free allocation of their investment.
I would agree with what you wrote (except maybe the bank leverage analogy) if you were actually locking in 3x stock market returns over the decades. But that's not what you are doing at all. That daily reset feature has the potential to be very destructive and will hugely deviate from 3x market returns over long time. That deviation can be both positive if the market returns are good (which is what you've experienced over the last few years) and negative -- if market returns are mediocre or worse (which is what would have happened if you decided to use this strategy starting in, say, mid-to-late 1990s).
DeleteIf the market performs very well then your strategy will perform spectacularly. But if the market returns are subdued or very volatile -- even if still positive -- then daily leverage strategy might well end up trailing the market. And over 60 years of such tepid and/or volatile returns your strategy could end up trailing the market even if the market returns are positive overall.
This is not intuitive, so it's very instructive to run the numbers for yourself to get an appreciation for this effect. E.g. if the market somehow ends up delivering zero returns going forward, such as never-ending sequence of $1 loss followed by $1 gain on ^SPX, then your SPXL investment will be reduced to less than half its original value over 60 years.
So the net result is that you end up super-wealthy (instead of merely wealthy as you would with a "normal" investing strategy) if the future is kind, and quite a bit poorer if the future is not. And that last part will be true is despite the seemingly conservative approach of only 1/3 in SPXL and 2/3 in I-Bonds/TIPS.
Again, I'm going to have find some time time to put together the simulations with different means and standard deviations (and accounting for extra expenses -- which I suspect are actually quite a bit higher than 0.95 headline ratio) to see how kind the future has to be for your SPXL strategy to benefit you. Until then we're both talking in hypotheticals.
And it's done... http://www.longtermreturns.com/2013/02/leveraged-etf-simulator.html
Delete