TIPS Ladder II

[austinite]

I read Jazzman's recent post and had questions about starting a TIPS ladder from scratch. Actually selling VAIPX (TIPS Fund) and starting a ladder in a tax deffered account.

If I read Larry correctly I could put 100% of this in one issue that matures in 20 years and do well. Then put future contributions in other rungs?

Am I correct?

Thanks in advance.

[Doc]

If I read Larry correctly I could put 100% of this in one issue that matures in 20 years and do well

I don't think that Larry went that far. He said recently that it is less important to have a ladder for TIPS than nominal Treasuries. However I just read a post of his this AM that said he had only eight year TIPS now and was thinking of selling them because of the low real yield.

At the risk of having this thread "censored" I state the following.

I think Larry at times makes very definitive statements that overstate his thinking.

[AzRunner]

My impression is that while Larry Swedroe is clearly a passive investor when it comes to equity, wrt TIPS he is an active investor since he sees them trading in a very narrow real interest rate band.

That may work well for Larry but may or may not fit your own temperament. Also Larry has access to a bond trading desk that enables him to buy and sell TIPS (and other individual bonds) at competitive prices. This may not be available to the typical Diehard.

My own approach is to use Larry's guidelines for TIPS attractiveness when buying new TIPS for my ladder - more when the real coupon yield is higher, less when it is lower. Then you can also look at the best yield by maturity at the time of purchase. My plan is to then hold each TIP bought at auction until maturity. In my own case, I think buying and selling TIPS may end up costing me more money than I make, so I've decided to be a buy andhold investor. JMHO.

Norm

[Norbert Schlenker]

If I read Larry correctly I could put 100% of this in one issue that matures in 20 years and do well.

Larry will answer for himself but I think there's a real problem with this for a lot of people. The volatility on a 20 year TIPS, which has a low coupon and thus long duration, is quite large. If you can close your eyes and hold to maturity, that's fine. If you're looking at your portfolio in the interim, you may have to stand some pretty fierce price adjustments. If real rates go up 1% and the stated price of that bond is 15% lower than your purchase price, and at the same time other markets are giving great returns (and, make no mistake, the two will happen at the same time), then you have to be able to ignore it.

I would say that a long time horizon mandates that you should do this, i.e. buy the long bond and shut your eyes tight. But can you? With the real yield curve as flat as it is, you're picking up almost no yield at all for going long but you get a lot of price risk. How strong is your stomach?

[larryswedroe]

few thoughts

First, I have suggested that TIPS are LIKELY to trade in a range of about 1.5%-3.5%. Of course could move beyond that but IMO risks are relatively low in that direction,. That means that if real yield gets to 1.5% you might want to be out of TIPS and switch to ST bonds. Reason if real rates go lower, they are not likely to go much lower, yet they could go MUCH higher. And reverse is true in terms of when rates are high--they are not likely to go much higher. And who cares anyway if you can lock in 3.5% risk free basically?

Second, I also suggest a DISCLIPINED PASSIVE strategy with buy and hold ranges, just like a DFA strategy. Similar to buy and hold ranges for size and value or shifting maturity. Yes it is active in a sense, but not changing fixed income allocation and adhering to market valuations----not my guesses on what will happen. But you can argue it is active.

Third, I am suggesting to go longer when rates are higher. Now they are in lower end of range. I had bought 10 year TIPS while ago and now they are eight years and at nice profits. Personally I would not extend at this point and would start to sell if rates drop below 2% and switch to DFA 2 year global or similar fund. And if want to stay with TIPS might even shorten up. Also as noted I do have access to institutional/wholesale pricing so I can shift at lower costs. But market is pretty liquid and transparent so markups should be relatively small. Just don't try to trade in small amounts. ON other hand if real rates rose to say 2.75 or so I would not only add more TIPS but lenghten maturity. And then if they went even higher I would add more and go as long as possible.

I have tables suggesting guidelines with buy and hold ranges for both real yields and maturities. p 226 of my bond book has the shifting allocation table and my new book on alternative investments (next year) will show both.

Decision Table for Allocation Between Short-Term Fixed-
Income and Ten-Year TIPS
Real Yield on TIPS Allocation to TIPS
> 3% 75-100%
> 2.5% < 3% 50-75%
>2%< 2.5% 25-50%
>1.5%<2% 0-25%
<1.5% 0


You of course can create your own table, the important thing is not the specific numbers but the discipline. The more risk averse to inflation the lower the numbers might be in the left hand column.

The table should be used in the following manner. If the real yield on ten-year TIPS were 2.6 percent, then an investor would allocate between 50 and 75 percent of their fixed-income assets to TIPS and 25 to 50 percent to nominal-return instruments. A specific number should be chosen upfront and made part of the investment policy statement.


In order to prevent unnecessary trading costs caused by movements slightly above and then below the targets, the targeted rates should be treated as buy ranges—they should be used to make purchase decisions for new investments. A hold range of perhaps 0.25 percent should be created for sell purposes. For example, let’s assume that the real rate on TIPS fell from 2.6 percent to 2.4 percent. What should our investor do with his current portfolio? Nothing—unless the rate fell to 2.25 percent (2.5 minus 0.25). If new cash was available for investment, however, the new cash could be used to move the allocation toward, or even to, 50 percent (the targeted allocation when the real rate is 2.4 percent. Note that even if there are no trading costs (as might be the case with mutual funds) we would not want to be having to alter positions with every small move in rates above and then below the targeted levels.

Decision Table for Maturity of TIPS
Real Yield on TIPS Allocation to TIPS
> 3% Twenty years
> 2.5% < 3% Fifteen years
>2%< 2.5% Ten years
>1.5%<2% Five years
<1.5% < Five years
> = greater than, < = less than
The two strategies, shifting allocation and shifting maturity, of course can be combined. Doing so will cause you to both increase your allocation to TIPS when real rates rise, and to increase the maturity of your TIPS holdings at the same time. Thus you will be locking in higher real rates for a longer time on a larger percentage of your holdings. When real rates fall you will be both lowering your allocation to TIPS and shortening their maturity. You will, of course, also be taking profits.

Hope that is helpful

[austinite]

Just so I do not think I understand when I do not:

Is the real yield for your purposes on the 10 year (Jan 2017) TIPS 2.160%?

I know duration for TIPS is a very different measure than for normal bonds. Could I dollar weight time to maturity to get a rough measure of the maturity I seek?

Thanks in advance.

[larryswedroe]

Yes the real yield now around 216

I also suggest you totally forget about duration of TIPS< IMO it is basically meaningless number. Dont even think about it.

[BlueEars]

Larry, if you are still tuned in here is a further TIPS question. If real rates move very low and we get to a deflationary environment (or even just fears of deflation) wouldn't it be best to be totally out of TIPS? Especially if you are holding TIPS with a significant inflation adjustment wouldn't deflation (like Japan experienced) be a real problem. Wouldn't it be best to be in nominal bonds which could thrive (as in 1929) in that environment? Seems that although this scenario is unlikely it is another argument for phasing out of TIPS as real rates move lower.

[Prokofiev]

" In order to prevent unnecessary trading costs caused by movements slightly above and then below the targets, the targeted rates should be treated as buy ranges . . ."

Larry,

This is the reason I think VG TIPS fund-VAIPX with an ER of .11 makes sense for many/most investors. Perhaps your access to institutional pricing gives you a better edge. But for most of us, a single trade per year would wipe out the ER of the VG fund. Yes, you cannot control the duration of this fund, but you have already said that duration is not that big a deal for TIPS. I can see having a core holding of TIPS in individual issues, but a shifting allocation strategy requires a fund. I don't see the advantage of a TIPS ladder or individual issues unless you plan to buy-and-hold till maturity. Am I wrong????

[larryswedroe]

Les: First, TIPS do fine in deflation as they cannot go below par---thus will provide positive real return---as long as you don't buy them above par to start with. But even if you buy above par, you still earn the real rate of return---remember in falling prices your buying power rises.

Prokofiev--I agree on the fund, though you can buy at issuance, but would likely need to be buy and hold investor unless can have access to instututional pricing. The advantage of the fund also is that it easily allows you to shift allocations between nominal and real bonds, using my tables or ones you create yourself. Just cannot shift maturity.

[Norbert Schlenker]

I also suggest you totally forget about duration of TIPS< IMO it is basically meaningless number. Dont even think about it.

I disagree totally. A useful distinction can be made between real duration - which measures your real price risk against real rate changes - and nominal duration - which I agree is basically meaningless for TIPS. IMO Larry is seriously underestimating the psychological effect of either rises or falls in real rates (euphoria as rates fall and gloom as rates rise) and the consequent change in investor behavior.

Let me point to something I wrote last year. Make an adjustment because of the Canadian context (inflation indexed bonds in Canada are called real return bonds or RRBs; TD Real Return Bond is the largest mutual fund in the asset class; an RRSP is like a regular IRA) but realize that the same behavioral considerations apply.

LINK

[An update to the prediction in that post: The 2006 return on TD RRB was indeed negative and it - along with every other indexed bond fund in Canada - has now fallen to the bottom of all the fixed income performance tables. Money will be leaving that fund (and the asset class), I'm sure.]

[Edited to add RRSP ~ IRA to minimize confusion]

[larryswedroe]

Norbert
You can certainly disagree but I am not underestimating any psychological effects. In fact if you read my posts you should know that I place a high importance of the psychological risks to investors posed by behavioral issues. In fact wrote a book covering many of those mistakes--Rational Investing in Irrational Times

best wishes

[Norbert Schlenker]

I am not underestimating any psychological effects.

I don't see how you can be so blase about ignoring duration then. If the real yield curve stays flat, then the price volatility of 20 year TIPS is close to double that of 10 year TIPS. That has an effect on investor psychology, magnifying both greed and fear, that can't just be waved away by dismissing duration as meaningless.

I tell my clients that they need to own instruments like this in a diversified portfolio and I tell them that, if they have a long time horizon, they should own long TIPS. But I also tell them upfront that long bonds of any kind, TIPS included, are not for the faint of heart or those who check the value of their portfolios every day, because they will cause behavioral errors going either up or down. Make 15% on these puppies one year and that induces one error; lose 15% another year and that induces another.

To use an analogy, let's talk about nominal bonds for just a minute. You're on record as telling people not to extend term too far with nominals and you're right because now you've got price risk that is a big duration multiplied by a big rate change. The rate changes on TIPS are smaller but they're not zero either, so duration still matters a lot.

[larryswedroe]

Norbert

Firstk, you should not care about the price volatility of TIPS as it is basically irrelevant. You buy them to guarantee a real rate of return and that is what they deliver, regardless of the volatility in between.

Second, what you are IMO making way too much of an issue of is the volatility. Again, see point one. But the volatility of a real bond is almost certainly going to be far less than that of a nominal bond and the reason is simple, nominal rates are more volatile than real rates, which are fairly stable. While we have seen nominal bonds trade say between 3-16 or so real bonds would have probably traded between say 1 and 4%. So how much volatility is there anyway? And again, you should not care.

And again the duration of TIPS is IMO basically irrelevant-for the reasons I gave. You care about duration of nominal bonds because you take the price risk related to inflation, and the longer the term the more the price risk. You don't have that with real return bonds at all. The only risk is real rates rising. But if they do, basically so what? You are still going to get your real return.

[Norbert Schlenker]

Firstk, you should not care about the price volatility of TIPS as it is basically irrelevant. You buy them to guarantee a real rate of return and that is what they deliver, regardless of the volatility in between.

Volatility is not irrelevant under withdrawal and, let's be honest about this, volatility is not irrelevant to the average investor at most times. The crowd on this forum is very resistant to being psychologically affected, but that's 1200 out of 300,000,000.

But the volatility of a real bond is almost certainly going to be far less than that of a nominal bond and the reason is simple, nominal rates are more volatile than real rates, which are fairly stable. While we have seen nominal bonds trade say between 3-16 or so real bonds would have probably traded between say 1 and 4%. So how much volatility is there anyway?

The lower volatility in TIPS is granted, but yields in the 1-4% range mean price changes in the 30% range are not only possible but likely. Indeed, we have just come through five years where TIPS prices changed 30% in real terms (more in nominal terms) but everyone's happy because the price change was in the right direction. Do you honestly think that people are going to be just as happy if prices on this asset class fall 30%?

People get antsy, even angry, when prices on equities fall 30%, but everyone knows going in that equities have that sort of possibility built in. TIPS are marketed as AAA credits from the world's hyperpower with all the inflation risk removed. How many TIPS owners know they can lose 30% in a security like that, and what will they do if it happens to them?

The only risk is real rates rising. But if they do, basically so what? You are still going to get your real return.

Are you telling me you have never had a client or prospect or neighbor tell you how mad they were that they had bought a CD at 5% and now the rate was 6%. How do you think all those people who bought 5 year TIPS at 7/8% or 2% 20 years at par just last year feel today? Their statements probably show values around 95 and, if they ask for a bid from the dealer, there is probably a haircut from there.

What's more likely?

"I'm still getting my 7/8% so I'm not worried."

or

"I can't believe I lost 5% in a year in something that I was told was absolutely guaranteed safe."

[larryswedroe]

norbert

The job of an investment advisor is to explain those risks UPFRONT, not after the fact. Thus if the client understands those risks and is educated on them on a regular basis then when the risks show up it should not be a problem. And of course investors should not invest in any instrument they don't fully understand the risks of.

The problem in my experience is that some advisors don't explain those risks fully if at all.

Of course many investors make mistakes of treating the unlikely as impossible and overestimate their ability to deal with risks

Now as to the other issue in withdrawal, again if you are in withdrawal and will have to be selling some TIPS (not an issue if not having to sell them but can sell other assets) you should not be owning long term TIPS, but shorter ones---that is a maturity issue, not a duration issue. Matching maturity to cash flow needs is basic finance and risk management principal. Again duration of a real bond IMO is basically irrelevant, only for nominal bonds is it important

[Norbert Schlenker]

Now as to the other issue in withdrawal, again if you are in withdrawal and will have to be selling some TIPS (not an issue if not having to sell them but can sell other assets) you should not be owning long term TIPS, but shorter ones---that is a maturity issue, not a duration issue. Matching maturity to cash flow needs is basic finance and risk management principal.

I think you're oversimplifying here. It's quite possible that one will have to make withdrawals from a portfolio that has to be invested for the long term, e.g. an IRA full of TIPS that is subject to RMD but where the beneficiary's life expectancy is still 20 years or more. If the duration of your liabilities is 20 years, the duration of your assets should be 20 years too, which means you should probably own 20 year (or even longer) TIPS. If you don't - if you own short TIPS because you are trying to cash flow match - you have an asset-liability mismatch and have needlessly subjected yourself to real interest rate risk that you could have laid off with one immunizing position.

Again duration of a real bond IMO is basically irrelevant, only for nominal bonds is it important

We're going to have to agree to disagree on this. I am actually amazed that you, who constantly repeats the excellent advice, "Don't make the mistake of treating the unlikely as impossible", can just wave this away because the expected risk in TIPS is smaller than the risk in nominals. Smaller is not zero.

Let me just mention as well one fact that contradicts even the theoretical belief that the expected risk in TIPS is smaller than nominals. I'm not mentioning this to question the belief, because I happen to believe the same thing. But facts are stubborn things and don't always follow theory.

The total return in 1994 on the Canadian equivalent of TIPS was -13.7%. There were no short indexed bonds in those days, so that was total return on something with really high duration, and a fair comparison would be against nominal Canadian long bonds in the same year. Theory (yours and mine!) says the total return on nominals should have been worse than that. In fact, total return on long bonds that year was -7.4%. The theory that we both believe says that's impossible but it still happened.

[larryswedroe]

Norbert

First, the liabilities in retirement are likely to be nominal not real. So right away you have a problem. You are mixing nominal liabilities agains the duration of a real asset---again that is why duration of real bonds basically meaningless. The only ones with real liabilities are likely to be the government with social security tied to inflation and the few remaining pension plans tied to inflation. The rest of us have nominal liabilities.

Second, there is no question that real yields are less volatile than nominal yields. That doesnt mean of course that in any ONE year that the reverse can be true. And again you should not be owning TIPS longer than your maturity needs (unless speculating) so that the volatility doesnt matter. Again. we need to consider the reason we hold fixed income---emergency cash flow needs, stability of cash flow to meet those needs and reduction of overall portfolio risk to level you can leave with.

Third, if you have a twenty year horizon, you match the maturity of the cash flows, so some maturities should be shorter. Thus you should not be holding 20 year TIPS to match a need in say 2 years. Then the volatility matters. But that is just poor risk management. Has nothing to do with the duration of TIPS.

[Norbert Schlenker]

The rest of us have nominal liabilities.

I don't and I will bet that asking people who have thought about the issue in any depth will reveal that they don't believe their long term liabilities, such as retirement income, are nominal. The local crowd is very aware financially. I'll start a poll thread and we'll see what they think.

As for TIPS duration, I see you're not going to be swayed by any fact or argument from me. I hope you never have to explain a repeat of 1994 to anyone.

[mas]

First, the liabilities in retirement are likely to be nominal not real. ... The rest of us have nominal liabilities.

I am not retired, but as far as I can tell I only have 1 nominal liability - a fixed rate mortgage (which hopefully will be gone when I do retire). All the rest of my expenses: food, clothing, entertainment, utilities, insurance, etc... All tend to rise in cost over time (roughly in step with inflation).

Why else are we continually warned about the risks of those of living on a 'fixed' income (i.e. nominal), and how to structure portfolios to allow for inflation adjusted withdrawals.

Am I missing something?

[larryswedroe]

First, Sorry I made mistake on liability side---trying to do too many things at once here. Yes liabilities are real---and that is why you should use real assets to hedge them, or at least asset that reduce inflation risk, if not eliminate it. The fixed mortgage is a nominal liability of course.

Second, Norbert, you are totally missing the point re for example 1994. You don't have to explain it to people IF YOU HAVE EXPLAINED IT TO THEM AHEAD OF TIME. Which is the job of any good advisor. That is no different than explaining the tracking error risk of owning int'l stocks or value stocks or small stocks or commodities. But you cannot, or should not, confuse a strategy with its outcome. A strategy is either right or wrong BEFORE The fact because we don't know which universe will show up. And if a 1994 showed up and TIPS fell in value then the investor should have been prepared for that and more importantly the impact on them should be ZERO. As the TIPS are doing exactly what they are designed to do. And if held to maturity they will deliver what they will have supposed to have done.

I would also add that if they do have a very bad year that just provides opportunity to buy more at higher yields.

[larryswedroe]

Norbert one more thing

The alternative to owning TIPS, and minimizing inflation risk, and don't have the price or duration risk you are so concerned about is to own ST bonds. Now if you do that you take on other risks, like you have floating real rate of return, and that could leave you with negative real returns for very long time. For example beginning in 1933 and until 1951 there were only three years where one month bills had positive returns. And from 33-97 the return was negative for full period. Also from73-82 it was negative.

And because of their shorter duration you should expect lower real returns. Example the 26-05 reutrn on one month bills was 0.6 real. Bit higher on three month bills. Well you can lock in much higher real rates on TIPS.

So while you take on some interim price risk with TIPS, that should not matter as I explained. The alternative, either longer term nominal bonds or shorter term nominal bonds takes on real, not psychological risks, and with shorter term bonds you have lower expected returns and with longer term bonds you have the price risk and the inflation risk.

Again, one should understand the purpose of the investment. And if you do then buying TIPS that match the maturity of the cash flow basically (virtually) eliminates all risks other than your psychological problem which an educated investor should be able to ignore--or the advisor should be able to help him ignore.

Which is why your arguments are unconvincing to me.

[Norbert Schlenker]

Second, Norbert, you are totally missing the point re for example 1994. You don't have to explain it to people IF YOU HAVE EXPLAINED IT TO THEM AHEAD OF TIME.

I agree and I do explain it to my clients ahead of time. I am simply pointing out that, as far as I can tell from what you've written above, you explain it to your clients by referring to theory ("real rates have less volatility than nominal rates so you can't get screwed with TIPS like you can with nominals"). Don't get me wrong: I think that theory is correct too in the long run. I just go further and explain that sometimes the model goes wrong, as it did in 1994, so the theory is at best an approximation.

To go back to your "duration is meaningless for TIPS" once more, let me ask you whether you similarly wave away discussions of possible volatility for other asset classes. Let's take vanilla equities as an example.

When I talk to clients about equities, I warn them about what volatility to expect, how often they are likely to see large drawdowns, and how big those drawdowns can be. I do that because, no matter how bulletproof and comforting an IPS is, when equities get crushed, holders of portfolios of equities get even more comfort from having known it could and likely would happen. I use language like, "This asset mix is likely to suffer annual losses of at least x% y times per decade." If y is bigger than 1, I further warn that them two loss years can happen consecutively, and point out what that means in both percentage and dollar terms. I do this in writing and I do it verbally, watching for acknowledgement that they really understand the risks.

It would surprise me if you didn't do exactly the same sort of thing for portfolios heavy with equities. With diversification, we do all we can to mitigate those risks, making x and y as small as we can for our clients and ourselves, but we can't make either one zero.

IMO exactly the same warning applies to TIPS. You and I both believe that TIPS returns should be less volatile than nominal bond returns but what we believe is irrelevant in the real world. So I give that warning, just as I do for equities. My clients don't for the most part understand investment math - that's part of the reason they come to me - so they don't actually want to know about duration. But that does not entitle me, when giving advice, to ignore the effects of duration when I am trying to quantify risk or to tell people who ask about TIPS duration that it's meaningless. A client might have a long time horizon, may have been told to close her eyes about interim volatility, may have agreed wholeheartedly that it's irrelevant, but you can guess who's to blame when this "as close to risk free investment as you can get anywhere" goes down 15% in a year. "Hey, duration is meaningless with TIPS" will be a meager defence.

That's it. I am guessing that nothing I say will change your mind, nor vice versa. You can have the last word.

[larryswedroe]

Norbert
As I thought I explained several times, we explain all those things up front. In great detail. We even show charts like something we call the risk of diversification (tracking error risk)---to show clients that while small and value stocks have delivered higher long term returns they don't do so every year---and diversified portfolios with tilts can underperform for very long periods---like 8 out of the ten years ending in 98.

So yes we tell them upfront and we carefully explain that a strategy is either right BEFORE The fact or wrong before the fact. And if they cannot live with the plan for long time, like ten years we are not interested in working with them as clients. (of course they could fire us tomorrow for bad service, but they should not fire us if portfolio returns are "poor")

And again, if you design the plan correctly then there is no risk for holding TIPS if you don't make the mistake of buying longer term than the expected expenses. Which is why IMO you are wrong about using duration as it is totally misleading figure. But I guess nothing I will say will change your mind. (:-))

[Ariel]

Attention bond gurus 8)

While we have two very smart bond guys here, let me ask a question. Background: I'm 15 years away from retirement, and well on track for my future needs. In other words, I have little need or inclination for risks. For the bond allocation in my tax-deferred accounts, I have basically three options: Stable Value Fund (TIAA Traditional Annuity), Total Bond Market (Vanguard or CREF), and Inflation Protected Secrities (Vanguard or CREF). I am presently targetting about 1/3 to each class, on the grounds that they are all worthwhile and have distinctive attributes in terms of protection against inflation, deflation, and volatility. How would you recommend I allocate funds among these three possibilities?

[larryswedroe]

Norbert--one last attempt here and then I give up. You have an IRA. You have twenty years until you need cash so you buy a 20 year TIPS to give you a guaranteed real return that will at least meet your anticipated spending needs. In that way you basically eliminate risk. Now in the interim the real yield on TIPS rises. So on a mark to market basis you show a loss. But, from a risk perspective NOTHING has changed. Your risk has not increased and since you don't need the funds you don't have to sell.
Now if you have the alternative situation which you seem to be suggesting because you are worried about duration, so you stay shorter, now when your bond matures and real rates have fallen you do have a change ---you no longer earn the real rate of return needed to meet your spending needs.
So which is the real risk?

This is the same type of analysis that a pension plan faces to meet its NOMINAL liabilities (Not tied to inflation). They should be buying nominal bonds of the same maturity to match their nominal liabilities. And if they don't match maturity they have reinvestment risk (shorter) or inflation risk (if longer). So while going longer with nominal bonds for most individuals takes on risk, going longer for pension plans REDUCES or ELIMINATES risk.

Again, duration meaningless IMO for TIPS.

Ariel--First, I don't recommend using TBM for variety of reasons. First, from portfolio context longer nominal bonds than I would prefer. Second it owns significant portion of MBS with unknown maturities with lose/lose characteristics (while get a small premium they shorten up at worst time and extend at worst time).. Thus I would suggest a good starting point would be something like 50% stable value and 50% TIPS. That way you diversify the real rate risk. I have also suggested a shifting allocation approach. When TIPS real yields are higher, increase the allocation. If they are lower, lower it. Create a table that will help keep you disciplined.

[Norbert Schlenker]

Ariel, since I'd hate to agree with Larry in any way wrt bonds , I think your 1/3:1/3:1/3 is just fine. Larry doesn't like MBS, I don't like them much either (try googling "negative convexity" but understand that you do get paid something for taking it on), but they're probably 10% of the total in that mix, so it's not that big a deal IMO. If there are any equities at all in your asset mix, the percentage is minimal.

I don't know what a TIAA "Stable Value Fund" is but it almost sounds like a money market account. If that's what it is, I would question whether having half your fixed income in something like that doesn't put you in a worse position than having a tenth of your fixed income in mortgage backeds.

Let me give you some real but generic advice, because I'm operating in an information vacuum here. I have no details on either your personal situation or the particular funds. The general rules with fixed income are (a) keep your costs to a minimum and (b) try to match the duration of the liability you're trying to cover.

[larryswedroe]

Norbert


To correct your misinformation in your advice to Ariel

First, one of the probems with TBM is its market cap weighting. Thus when you get big refi boom due to rate drop and big housing market for same reason the % of MBS in TBM rises. And so now it is about 36% of the TBM fund. Not an insignificant percentage. At least one should know what one is investing in and fully understand the risks.

Second, TIAA 's traditional annuity is a stable value fund --it is not a money market account by any means. Typically stable value funds extend durations a bit out on the yield curve and they can still provide the stable value via use of insurance contracts that guarantee principal.

Third, a found somewhat interesting your last bit of advice--to match maturity of liability. Which I fully agree with in general. And the reason of course to do that is to eliminate risk. Of course that basically says duration is meaningless in this case since you have matched the asset and liability and eliminated risk, and thus you should not care about duration.

[Norbert Schlenker]

To correct your misinformation in your advice to Ariel

First, one of the probems with TBM is its market cap weighting. Thus when you get big refi boom due to rate drop and big housing market for same reason the % of MBS in TBM rises. And so now it is about 36% of the TBM fund. Not an insignificant percentage. At least one should know what one is investing in and fully understand the risks.

And 36% times 1/3 is about 10% of the fixed income total, just as I indicated. How did I misinform?

Third, a found somewhat interesting your last bit of advice--to match maturity of liability. Which I fully agree with in general. And the reason of course to do that is to eliminate risk. Of course that basically says duration is meaningless in this case since you have matched the asset and liability and eliminated risk, and thus you should not care about duration.

Au contraire. The fact that you need to match duration in order to eliminate the risk makes considering duration critical, not meaningless.

[Ariel]

Thank you both for your replies.

To reiterate and add a little more background, I'm about 15 years from retirement and still contributing substantially to both tax-deferred and taxable accounts. Target allocation for my tax-deferred accounts is:

20% TIAA Traditional (stable value fund, with contractual minimum rate of 3%, currently 5.25% on new contributions, paying me about 5.5% at present given some well timed allocation shifts)

20% Vanguard Inflation Protected Securities

20% Vanguard Total Bond Market

20% TIAA Real Estate Account (directly owned real estate, little leverage, much more stable than REIT funds)

20% Vanguard International Value Fund

My retirement accounts with future contributions (which should be stable given job security) should fully cover my retirement needs. They represent about 55% of my liquid assets.

In addition, I hold the following allocation in taxable accounts (about 45% of liquid assets), which are likely to be legacy accounts unless I suddenly learn to spend more ...

33% Vanguard Tax-managed Growth and Income Fund

33% 15 individual stocks, all bluechip, megacap, dividend payers, and all with covered calls written using LEAP options

33% Vanguard Insured Long-term Tax-exempt Bonds

I realize the allocation in my taxable accounts might be too conservative and income oriented. I justify that on the grounds that I am conservative (helped me largely avoid hi-tech bubble), I like to earn something tangible (even if I don't spend it all), and I enjoy following the market.

Fire away!

[larryswedroe]

norbert
First, my apologies for not picking that up.

Second sorry to disagree. It simple. What difference does it make if TIPS are yielding 4% or 2% and you are trying to match the expenses say do in ten years. With a 4% yield you have shorter duration, thus you would suggest shorter maturity TIPS I assume. But then you have the reinvestment risk when they mature. With TIPS because they are real investments you should be matching maturity

[Norbert Schlenker]

What difference does it make if TIPS are yielding 4% or 2% and you are trying to match the expenses say do in ten years.

Beats me. I'm not the one suggesting an asset mix change depending on real yields; you are. Defend that position if you can. (Not that I'm lily white on my own account when it comes to this, mind you. )

My position is that you should match durations of assets and liabilities. That's why I objected so strongly to your tossed off "duration is meaningless for TIPS". IMO, there is no other characteristic that is as meaningful. The simplest thing that you can teach anyone with a bond portfolio that is intended to fund an obligation is, "Match duration." We can wrangle about all the other bits that go into bond portfolio management - credit, convexity, taxes, what constitutes proper diversification - for days or weeks or months because it's incredibly complex, but 80% of the problem is instantly solved with two words. Match duration. Duration is not meaningless, not for any bond portfolio, and certainly not for TIPS.

[larryswedroe]

Norbert
Really my last attempt and I give up
First, you have match assets and liabilities then you should not care about duration. Period. The interim volatility is irrelevant as you eliminate risk by the matching. A term called defeasance if you are not familiar with it, And you are the one who says you should care about duration because of the volatility--it is why you raised the issue. So either you care or you don't care. you cannot have it both ways

Second, The liabilities of individuals are basically real liabilities and if you own TIPS you match the real liability with real asset so again by matching the MATURITY of the expense with the maturity of the TIPS you completely eliminate risk, you defease the liability in effect. So again duration is irrelevant. That is EXACTLY the case I gave you with pension plans that defease their long term liabilities (not inflation adjusted pensions) with longer term bonds that match the MATURITY. They don't give a hoot about duration because it is not a risk they care about in the interim.

[benethridge]

Hi, Larry. Can you point me to where you are getting this data, please?

For example beginning in 1933 and until 1951 there were only three years where one month bills had positive returns. And from 33-97 the return was negative for full period. Also from73-82 it was negative...Example the 26-05 reutrn on one month bills was 0.6 real.

Ben

[larryswedroe]

DFA returns program I have access to

[oldman]

http://www.treasurydirect.gov/instit/an ... 062009.htm

This is a link to the CPI report and TIPS. How do you use this chart for deciding what to buy now, if at all?

[Lbill]

One question regarding TIPS, particularly long-maturity TIPS. I understand liability matching - that's essentially what I'm trying to do by holding a 20-year TIPS income ladder. What I don't understand very well is YTM. The quoted YTM of each maturity that I've purchased assumes that the semi-annual interest payments will theoretically be reinvested at the same real interest rate, correct? But if I don't reinvest the interest, or I reinvest it at a lower real rate, then my actual YTM will be lower than the quoted YTM, correct? How can I match liabilities when actual YTM will not actually turn out to be the quoted YTM? And isn't the reinvestment risk much greater for long TIPS, so the the risk of liability mis-matching actually increases with the length of maturity?

[Doc]

One question regarding TIPS, particularly long-maturity TIPS. I understand liability matching - that's essentially what I'm trying to do by holding a 20-year TIPS income ladder. What I don't understand very well is YTM. The quoted YTM of each maturity that I've purchased assumes that the semi-annual interest payments will theoretically be reinvested at the same real interest rate, correct? But if I don't reinvest the interest, or I reinvest it at a lower real rate, then my actual YTM will be lower than the quoted YTM, correct? How can I match liabilities when actual YTM will not actually turn out to be the quoted YTM ? And isn't the reinvestment risk much greater for long TIPS, so the the risk of liability mis-matching actually increases with the length of maturity?

The YTM calculation is based on the present value of a steam of cash flows and makes NO assumption on reinvesting those cash flows. The "reinvesting at the same rate" is a common mis-conception that stems from people trying to equate YTM (a present value term) with Total Return (in this case a future value term). They are equal only in very special cases like "reinvesting the cash flows at the YTM" or a zero coupon investment.

So for your second question your actual total return will be different from the original YTM.

As far as the last question the answer is: it depends. If you are only talking real dollars your interpretation is correct but if you are talking about nominal dollars there will be less of a difference between the nominal YTM and the nominal total return because TIPS cash flows are back loaded and look more like a zero coupon bond then the equivalent nominal bond. (Note: the quoted TIPS YTM are real not nominal.)

As a practical matter if the interim cash flows are not very high (low coupon) it really doesn't matter very much and the longer time doesn't matter a whole lot because the longer payments are discounted more in the calculation.

There are some published calculators that allow you to change the reinvestment rate. I think Morningstar's does and I think TFB has a spread sheet on his website that allow a reinvestment rate for TIPS.

[mas]

I understand liability matching - that's essentially what I'm trying to do by holding a 20-year TIPS income ladder. ... How can I match liabilities when actual YTM will not actually turn out to be the quoted YTM?

Sadly there are not zero coupon TIPS, so anything you do will be an approximation. If the 20yr TIPS ladder does not constitute all of your assets, then just make sure that each TIPS' maturing principal matches your equivalent liability and send all of the TIPS interest payments to your "other" portfolio. The "other" portfolio will really include some extra amount of fixed income based on these payments (higher in the early years, and lower when the ladder is nearly depleted).

[Lbill]

Doc said:

So for your second question your actual total return will be different from the original YTM.

As far as the last question the answer is: it depends. If you are only talking real dollars your interpretation is correct but if you are talking about nominal dollars there will be less of a difference between the nominal YTM and the nominal total return because TIPS cash flows are back loaded and look more like a zero coupon bond then the equivalent nominal bond. (Note: the quoted TIPS YTM are real not nominal.)

Thanks for that explanation. This stuff is not easy to understand (even though I have an I.Q. of 200 - er - 20). Of course, I was actually concerned with total real return and whether or not this matches liabilities. As I understand you, I really don't know what that will end up being because it is partially determined by reinvestment of interest payments over the time horizon of the bond.

Mas said:

Sadly there are not zero coupon TIPS, so anything you do will be an approximation. If the 20yr TIPS ladder does not constitute all of your assets, then just make sure that each TIPS' maturing principal matches your equivalent liability and send all of the TIPS interest payments to your "other" portfolio. The "other" portfolio will really include some extra amount of fixed income based on these payments (higher in the early years, and lower when the ladder is nearly depleted).

That's sorta the approach I took - since I know what the real principal value will be at maturity, I used that number for liability matching. The interest payments that are thrown off will just go "somewhere." Not a very efficient way of doing it but I don't know any better way - like to know if there is.

[Doc]

That's sorta the approach I took - since I know what the real principal value will be at maturity, I used that number for liability matching. The interest payments that are thrown off will just go "somewhere." Not a very efficient way of doing it but I don't know any better way - like to know if there is.

Assuming that you have "other assets" than those needed to meet the future fixed obligations the approach you outline is a very good way to approach the problem. You just think of the TIPS principle as what is need to meet the future obligations and the coupon payment as part of the "other assets." At worst it might cause a negligible shift in you asset allocation but that would be small and could probably be ignored.

This is the approach I use in case we have some meltdown in the financial system like maybe a credit crisis. My TIPS ladder is designed to give me ten years of cash flow using only the principle from maturing bonds to meet my absolute minimum needs (after SS & pension) without selling any other securities. Gee whiz, did this just happen last fall?

[bmb]

The beauty of TIPS is that duration, the interest rate, and the magnitude of future inflation are far less important than with nominal bonds. You know your real return and you know your capital is at almost no risk, which are the most important issues with fixed income investments.
As long as you match the maturities to your distribution needs, you can't really lose, even though you may be sacrificing some potential gain for the inflation protection.

[Doc]

The beauty of TIPS is that duration, the interest rate, and the magnitude of future inflation are far less important than with nominal bonds. You know your real return and you know your capital is at almost no risk, which are the most important issues with fixed income investments.
As long as you match the maturities to your distribution needs, you can't really lose, even though you may be sacrificing some potential gain for the inflation protection.

While I agree with your reasoning "beauty is in the eyes of the beholder." Or maybe that's "bondholder".