TIPS and equities
Average Nominal Annual Return/Average Annual SD
Period 3 Yr, 5 Yr , 10 Yr
Lehman 5-10 Yr Treasury: 5.59%/4.13%, 5.40%/4.24%, 5.97%/4.91%
Lehman TIPS Index: 5.64%/4.87%, 6.67%/5.03%, N/A
TIPS exhibit somewhat higher total returns and volatilities for 3 and 5 year periods (no 10 year available for TIPS). Similar risk/return ratio.
I agree that TIPS have special characteristics that might result in a greater divergence of returns and volatilities from recent past.
Period 3 Yr, 5 Yr , 10 Yr
Lehman 5-10 Yr Treasury: 5.59%/4.13%, 5.40%/4.24%, 5.97%/4.91%
Lehman TIPS Index: 5.64%/4.87%, 6.67%/5.03%, N/A
TIPS exhibit somewhat higher total returns and volatilities for 3 and 5 year periods (no 10 year available for TIPS). Similar risk/return ratio.
I agree that TIPS have special characteristics that might result in a greater divergence of returns and volatilities from recent past.
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard |
|
"You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs
-
- Posts: 16022
- Joined: Thu Feb 22, 2007 7:28 am
- Location: St Louis MO
Lbill
Few points. You have two apples to oranges comparisons going on.
First, you have TIPS index which is close to about a ten year average vs. a 5-10 year index. So the TIPS index will have longer duration
Second, you are comparing a real return instrument with a nominal return instrument. As I said the nominal SD of a real instrument is basically an irrelevant figure.
Third, we have a period of relatively very stable inflation and the unexpected inflation premium in nominal bonds is apparently now negative, providing a one time boost to returns of nominal bonds that almost surely cannot be repeated
Let me conclude again by saying that ST bonds are a good alternative, it is not like say choice between investment grade debt and junk bonds--two entirely different animals.
Few points. You have two apples to oranges comparisons going on.
First, you have TIPS index which is close to about a ten year average vs. a 5-10 year index. So the TIPS index will have longer duration
Second, you are comparing a real return instrument with a nominal return instrument. As I said the nominal SD of a real instrument is basically an irrelevant figure.
Third, we have a period of relatively very stable inflation and the unexpected inflation premium in nominal bonds is apparently now negative, providing a one time boost to returns of nominal bonds that almost surely cannot be repeated
Let me conclude again by saying that ST bonds are a good alternative, it is not like say choice between investment grade debt and junk bonds--two entirely different animals.
Larry, I understood all but the last sentence. Seems the 20yr rate is close to historic average real rates but the 5yr and 10yr are below:larryswedroe wrote:...Les
Yes and I started buying again at 2.20, though for longer maturity. The reason is the first time the TIPS curve was very flat while now it is very steep. Again, valuations matter IMO. Now you are getting paid to take more of that duration risk (though you are not taking inflation risk)
5yr 1.36
10yr 1.68
20yr 2.15
Are you saying that you are buying 20yr TIPS? If so, would you be prepared to hold them very long term should the 20yr rates move up from here?
-
- Posts: 16022
- Joined: Thu Feb 22, 2007 7:28 am
- Location: St Louis MO
Yes, the steep curve is exactly why I am buying the 20 year. And the fact that it is about the long term historical average for real rates on long term government debt, while the other parts of the TIPS curve are well below the historic average.
As I said, I have now about 10% allocation of my available dollars. If rates go much lower I will "take profits" and sell. If they go higher I will start to buy more a bit at a time. What part of the curve I will buy on depends on its shape--where the "sweet spot" seems to be.
Remember with TIPS you don't have the inflation risk that you do with nominal long term bonds. Only the real rate risk, which is nothing more than an opportunity cost. If you hold to maturity you will earn that real rate.
And again IMO people make a huge mistake looking at the NOMINAL SD of a REAL interest rate instrument. Gives totally misleading picture. The NOMINAL SD is important for nominal bonds, not real bonds.
Example, in extreme to make the point. Say you buy a TIPS long term at 2% real yield and the nominal bond is 4%. (roughly today's figures). Now say inflation goes way up so that the nominal bond is yielding 10% but the real yield is unchanged at 2%. The real value of the TIPS is unchanged but it will show a high NOMINAL SD. The real value of the nominal bond will have fallen sharply and that risk is reflected in the high SD.
TIPS have longer duration that nominal bonds because of their lower interest rate (it is only the real part). That means its price will move more for each 1% move in rates. But as you see above the nominal SD is not important, can give a totally misleading picture.
As I said, I have now about 10% allocation of my available dollars. If rates go much lower I will "take profits" and sell. If they go higher I will start to buy more a bit at a time. What part of the curve I will buy on depends on its shape--where the "sweet spot" seems to be.
Remember with TIPS you don't have the inflation risk that you do with nominal long term bonds. Only the real rate risk, which is nothing more than an opportunity cost. If you hold to maturity you will earn that real rate.
And again IMO people make a huge mistake looking at the NOMINAL SD of a REAL interest rate instrument. Gives totally misleading picture. The NOMINAL SD is important for nominal bonds, not real bonds.
Example, in extreme to make the point. Say you buy a TIPS long term at 2% real yield and the nominal bond is 4%. (roughly today's figures). Now say inflation goes way up so that the nominal bond is yielding 10% but the real yield is unchanged at 2%. The real value of the TIPS is unchanged but it will show a high NOMINAL SD. The real value of the nominal bond will have fallen sharply and that risk is reflected in the high SD.
TIPS have longer duration that nominal bonds because of their lower interest rate (it is only the real part). That means its price will move more for each 1% move in rates. But as you see above the nominal SD is not important, can give a totally misleading picture.
.
For another take on real risk and return of TIPS: Chapter 2 of the Handbook of Inflation Indexed Bonds (pg. 9), presents the real returns and real risk of various asset classes as:
I presume cash is taken as T-bills.
Here's another extract from the text.
Robert
.
For another take on real risk and return of TIPS: Chapter 2 of the Handbook of Inflation Indexed Bonds (pg. 9), presents the real returns and real risk of various asset classes as:
Code: Select all
Asset Real Risk Real Return
Equity 20.6 7.2
Bonds 7.1 2.1
Cash 4.2 0.6
TIPS* 4.2 3.0
*For TIPS the reported real risk is estimated and real returns are contractually guaranteed. Other risks and returns are based on Ibbotson 1926-1995.
Here's another extract from the text.
Link to a few pages of the book“There is a nice feature about inflation-linked bonds, which doesn’t apply to any other asset classes. For inflation-linked bonds, future real returns are contractually guaranteed by the US Treasury.
In terms of risk, I have estimated the risk for inflation indexed bonds as 4.2%. If you were looking at real risk and had a time frame that coincides with the maturity of the inflation linked bonds, in essence, the risk would be zero. However, on a mark-to-market basis, ignoring liabilities, inflation indexed bonds do have some risk. It is substantially less than nominal bonds, and I would suggest, even less than the risk of cash. To be conservative, I have used a number equal to the real risk of cash.
Incidentally, for long-term investors cash is not a riskless asset in real or nominal terms. The real return on cash, even though it may be known for the next three months, over time, is quite volatile. During the entire decade of the 1970s cash averaged a negative real return. During the 1980s because of the budget deficits, trade deficits, and a monetary authority that was crushing inflation, the real return on cash was extremely high, almost 5%.”
Robert
.
Larry-
It seems simple to me. TIPS will work better in an unexpected inflationary environment than conventional bonds. If inflation exceeds the difference between the yield of conventional and TIPS, TIPS will win. If inflation does not exceed the difference, then TIPS and conventional bonds of the same duration have similar returns (nominal and real) and volatilities. The correlation between the two over the last 5 years is nearly 1.0. You should buy TIPS as insurance against unexpected inflation in the future. I think that's all there is to it. I just don't get all this stuff about real vs. nominal yields because you can compare TIPS and conventional bonds on either basis. To me, it makes no sense to compare real yields on TIPS to nominal yields on conventional bonds. That's apples and oranges. Now, I agree that it probably does make some sense to think of TIPS as a separate asset class from conventional bonds for purposes of asset allocation, even though their correlation has been extremely high over the 10-year lifetime of TIPS. We don't know what would happen in a highly inflationary scenario because we haven't experienced that in the last 10 years. But we believe TIPS and conventional bonds would diverge significantly - TIPS having positive total returns and conventional bonds have negative returns. Is it any more complicated than that?
It seems simple to me. TIPS will work better in an unexpected inflationary environment than conventional bonds. If inflation exceeds the difference between the yield of conventional and TIPS, TIPS will win. If inflation does not exceed the difference, then TIPS and conventional bonds of the same duration have similar returns (nominal and real) and volatilities. The correlation between the two over the last 5 years is nearly 1.0. You should buy TIPS as insurance against unexpected inflation in the future. I think that's all there is to it. I just don't get all this stuff about real vs. nominal yields because you can compare TIPS and conventional bonds on either basis. To me, it makes no sense to compare real yields on TIPS to nominal yields on conventional bonds. That's apples and oranges. Now, I agree that it probably does make some sense to think of TIPS as a separate asset class from conventional bonds for purposes of asset allocation, even though their correlation has been extremely high over the 10-year lifetime of TIPS. We don't know what would happen in a highly inflationary scenario because we haven't experienced that in the last 10 years. But we believe TIPS and conventional bonds would diverge significantly - TIPS having positive total returns and conventional bonds have negative returns. Is it any more complicated than that?
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard |
|
"You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs
Lbill, I agree with keeping it simple. I think that so far TIPS have outperformed mainly because they are new and investors have been slow to accept them, hence the returns offered have been higher than warranted. This is changing. As more investor become aware of their benefits, TIPS will be primarily a hedge against unexpected high inflation, not a way to consistently outperform nominal bonds. In the 1970s, inflation was high, and Treasuries had very low real returns. This changed in the 1980s - Treasury yields soared and beat inflation substantially. This is probably due to the fact that in the 1970s investors were somewhat surprised by the magnitude of the sudden increase in inflation, and they expected it to continue in the 1980s, but it abated. If TIPS had been available, no doubt they would have significantly outperformed nominal Treasuries and all other investments except commodities in the 1970s but lagged in the 1980s.
For the same reasons, I do not really understand why people would buy short-term TIPS now when the longer term TIPS offer substantially higher yields.
For the same reasons, I do not really understand why people would buy short-term TIPS now when the longer term TIPS offer substantially higher yields.
-
- Posts: 16022
- Joined: Thu Feb 22, 2007 7:28 am
- Location: St Louis MO
Lbill
Basically agree
The issue is this. If you have the same expected return than clearly unless you are more exposed to deflationary risks than inflationary risks you should prefer TIPS. And since most people are not risk neutral, they should prefer TIPS even if there is a risk premium. The only issue is the size of the premium. Today it appears to be a negative premium given ten year TIPS at about 1.7 and ten year bonds at 3.7 and all forecasts of inflation long term I see are at least 2 percent and most higher.
We should expect correlations to be very high when rates are moving because of changes in real rates. We should not expect that to be the case when rates move because of changes in expected inflation. We have had very stable inflation in last ten years, and we know that may not be the case going forward. And certainly there is more risk of upside surprise than downside as with TIPS risk of unexpected inflation is limited to zero while upside of inflation is theoretically unlimited. That of course tells us nothing about the odds of upside or downside, but it does tell us a lot about the risks if either shows up.
Note that we don't believe they will diverge if inflation rises rapidly, We KNOW that will happen.
And clearly TIPS should be treated as separate asset class as their risks are different--they are not subject to unexpected inflation risks while nominal bonds are, thus they are a separate asset class within the fixed income arena.
Also I believe it makes all the sense in the world to compare the real yield on TIPS with the nominal yield on bonds to see what the breakeven inflation rate is--and that allows you to estimate the size of the risk premium, if any. And then you can decide if you are willing to pay the risk premium, or today if you are willing to get paid with a negative risk premium as I believe exists today
Finally, the biggest advantage of TIPS which seems to be overlooked is that because of the elimination of inflation risk you can take more duration risk and get paid for it and when you do go longer with TIPS you don't increase correlation with stocks as you do with nominal bonds.
Basically agree
The issue is this. If you have the same expected return than clearly unless you are more exposed to deflationary risks than inflationary risks you should prefer TIPS. And since most people are not risk neutral, they should prefer TIPS even if there is a risk premium. The only issue is the size of the premium. Today it appears to be a negative premium given ten year TIPS at about 1.7 and ten year bonds at 3.7 and all forecasts of inflation long term I see are at least 2 percent and most higher.
We should expect correlations to be very high when rates are moving because of changes in real rates. We should not expect that to be the case when rates move because of changes in expected inflation. We have had very stable inflation in last ten years, and we know that may not be the case going forward. And certainly there is more risk of upside surprise than downside as with TIPS risk of unexpected inflation is limited to zero while upside of inflation is theoretically unlimited. That of course tells us nothing about the odds of upside or downside, but it does tell us a lot about the risks if either shows up.
Note that we don't believe they will diverge if inflation rises rapidly, We KNOW that will happen.
And clearly TIPS should be treated as separate asset class as their risks are different--they are not subject to unexpected inflation risks while nominal bonds are, thus they are a separate asset class within the fixed income arena.
Also I believe it makes all the sense in the world to compare the real yield on TIPS with the nominal yield on bonds to see what the breakeven inflation rate is--and that allows you to estimate the size of the risk premium, if any. And then you can decide if you are willing to pay the risk premium, or today if you are willing to get paid with a negative risk premium as I believe exists today
Finally, the biggest advantage of TIPS which seems to be overlooked is that because of the elimination of inflation risk you can take more duration risk and get paid for it and when you do go longer with TIPS you don't increase correlation with stocks as you do with nominal bonds.
-
- Posts: 112
- Joined: Sat Aug 16, 2008 2:12 pm
- Location: Colorado
Larry,
The only problem I have with your strategy is that at some real return (say 3-4%) TIPS become the perfect asset class--and a possible substitute for a large portion of equitites in a portfolio. The greater the long duration positions you take between 2 and 3%, the harder it will be, without having to take capital losses, to make a substantial commitment if rates bust through to higher levels.
The only problem I have with your strategy is that at some real return (say 3-4%) TIPS become the perfect asset class--and a possible substitute for a large portion of equitites in a portfolio. The greater the long duration positions you take between 2 and 3%, the harder it will be, without having to take capital losses, to make a substantial commitment if rates bust through to higher levels.
-
- Posts: 3665
- Joined: Thu Oct 18, 2007 11:42 am
Interesting stuff. It is hard for me to comprehend, though, why one should incorporate a "forecast" into valuation or evaluation of this asset class. I've not seen forecasts of any sort play much of a role for Diehards, if I am not mistaken.larryswedroe wrote:Today it appears to be a negative premium given ten year TIPS at about 1.7 and ten year bonds at 3.7 and all forecasts of inflation long term I see are at least 2 percent and most higher.
I believe if you were to purchase TIPS yielding 2.5% and real rates rose to 3.5%, then you would already be "committed" to a real yield of 3.5%, whether you chose to realize it or not.mountainmoney wrote:The only problem I have with your strategy is that at some real return (say 3-4%) TIPS become the perfect asset class--and a possible substitute for a large portion of equitites in a portfolio. The greater the long duration positions you take between 2 and 3%, the harder it will be, without having to take capital losses, to make a substantial commitment if rates bust through to higher levels.
-
- Posts: 112
- Joined: Sat Aug 16, 2008 2:12 pm
- Location: Colorado
I believe if you were to purchase TIPS yielding 2.5% and real rates rose to 3.5%, then you would already be "committed" to a real yield of 3.5%, whether you chose to realize it or not.mountainmoney wrote:The only problem I have with your strategy is that at some real return (say 3-4%) TIPS become the perfect asset class--and a possible substitute for a large portion of equitites in a portfolio. The greater the long duration positions you take between 2 and 3%, the harder it will be, without having to take capital losses, to make a substantial commitment if rates bust through to higher levels.
That's my point. If you had owned ST nominal bonds, instead of long duration TIPS--as a risk reducer in your comprehensively diversified portfolio--you could now take advantage of the higher real yields and receive a 3.5% real YTM on a greater principal amount.
-
- Posts: 3665
- Joined: Thu Oct 18, 2007 11:42 am
Ah, I see. It sounded a little to me as if you were wanting to make that commitment, take advantage, etc. only when you saw those real rates rise. Are you not just talking about timing and guessing about future rate changes? Unfortunately, my best estimate of what 20 year TIPS ought to be paying me for real yield is the one the market today is willing to give me.mountainmoney wrote:That's my point. If you had owned ST nominal bonds, instead of long duration TIPS--as a risk reducer in your comprehensively diversified portfolio--you could now take advantage of the higher real yields and receive a 3.5% real YTM on a greater principal amount.
-
- Posts: 112
- Joined: Sat Aug 16, 2008 2:12 pm
- Location: Colorado
Yes, I am talking about timing. But I'm turning what has been one timing conversation into two. The right hand has been playing a tune about how to handle the fixed income allocation of a portfolio that is dominated by equities and using bonds to reduce the overall risk. And the timing conversation is about expected vs. unexpected inflation and duration risk.
But with the left hand, I'm talking about keeping powder dry for the purpose of an entire portfolio transformation--ala Milevsky and Kotlikoff--by which you--God willing--might be able to invest most of your portfolio in 4% real yield 30 year TIPS and achieve all your financial goals with no risk. Game over.
The first timing game is rather tactical and even if you win--as the earlier dialogue between Larry and SH demonstrates--doesn't really pay you all that much.
Ah, but the second...
But with the left hand, I'm talking about keeping powder dry for the purpose of an entire portfolio transformation--ala Milevsky and Kotlikoff--by which you--God willing--might be able to invest most of your portfolio in 4% real yield 30 year TIPS and achieve all your financial goals with no risk. Game over.
The first timing game is rather tactical and even if you win--as the earlier dialogue between Larry and SH demonstrates--doesn't really pay you all that much.
Ah, but the second...
Last edited by mountainmoney on Tue Sep 09, 2008 11:10 am, edited 1 time in total.
-
- Posts: 3665
- Joined: Thu Oct 18, 2007 11:42 am
Yes, I understand, the day that the real interest rate on a risk-free security equals the oft-quoted SWR.mountainmoney wrote:. . . I'm talking about keeping powder dry for the purpose of an entire portfolio transformation--ala Milevsky and Kotlikoff--by which you--God willing--might be able to invest most of your portfolio in 4% real yield 30 year TIPS and achieve all your financial goals with no risk. Game over.
If we ever see a 4% real rate on TIPS, would not the proponent of such a portfolio transformation want to buy those 4% that very day, regardless of anything else that had happened to the portfolio in the interim? So, yes, I can comprehend the wish to minimize exposure to the very assets (long term TIPS) that would suffer so in getting to 4% real.
It is clear that you would favor a short term FI instrument over 20-year TIPS, if you thought you could get a comparable degree of portfolio diversification and inflation protection you needed from that instrument.
This 4% real, though. Is it something we are likely to see, and thus just a matter of maximizing or preserving assets until that day?
-
- Posts: 112
- Joined: Sat Aug 16, 2008 2:12 pm
- Location: Colorado
4% may be a pipe dream. But we have had mid 3's in the last decade. (I bought a bunch of 30 year I-Bonds at 3% about six years ago, when you could buy a bunch). And the higher the number you can lock in, the more equities you can take off the table. So if you can take your portfolio from 60/40 to 40/60, that's a big win in my book.
What I don't know and what hasn't been discussed here, is what kinds of economic conditions/government policy takes real rates up.
What I don't know and what hasn't been discussed here, is what kinds of economic conditions/government policy takes real rates up.
- Rick Ferri
- Posts: 9703
- Joined: Mon Feb 26, 2007 10:40 am
- Location: Georgetown, TX. Twitter: @Rick_Ferri
- Contact:
When TIPS first came out, they were new, limited in quantity, and not used by institutions. As such, the yield was about 4% when today, given the larger quantity of issues and the number of TIPS mutual funds in the marketplace, the yield would have been less that half that.
Due to this one time phenomenon, going back to 1997 gives an investor an overstated view benefits regarding how TIPS can help a portfolio in the future. I agree with Larry that TIPS will likely help, but don't use that 1.2% number. Larry does talk about this above.
Rick Ferri
Due to this one time phenomenon, going back to 1997 gives an investor an overstated view benefits regarding how TIPS can help a portfolio in the future. I agree with Larry that TIPS will likely help, but don't use that 1.2% number. Larry does talk about this above.
Rick Ferri
Last edited by Rick Ferri on Tue Sep 09, 2008 11:27 am, edited 1 time in total.
- Norbert Schlenker
- Posts: 544
- Joined: Mon Feb 19, 2007 10:06 pm
- Location: The Dry Side of the Wet Coast
You made the reference to Kothari and Shanken, not me. The numbers you quote for standard deviations are not from Kothari and Shanken. Suggesting that the standard deviation of nominal bond returns is five times that of indexed bond returns does not come from Kothari and Shanken. Suggesting that the standard deviation of TIPS is anything in the 2% range does not come from Kothari and Shanken, nor does it come from actual historical performance. You're quoting from your own book, Larry. Where did the numbers come from?larryswedroe wrote:Here is the live data for the US. You stated that TIPS and stocks do not have negative correlation, not even close
3/97-8/08 the quarterly correlation between TIPS and the S&P 500 has been -.342.
And for the 10-year period 1998–2007, the quarterly and annual Don't know what data you are looking at.
On a more general note, past performance is no guarantee of future results. All I suggested is that one should cast their eyes a little further than the 11 year record of TIPS in the US. I submit that you suggested exactly the same thing when
Canada is a developed country. And the record of inflation indexed bonds in Canada does not show what you have presented.you wrote:BTW-we only have about ten years of data with US TIPS. There is however much longer data available with IPS from other developed countries. And it shows what I have presented.
We've been round this before. I agree that there should be a correlation. That doesn't change the fact that what little historical data we have does not show a strong correlation. I'm a pretty simple guy. When I have a model of the world that says "so and so should happen" and in the real world it doesn't, I start considering whether my model has flaws or whether the world is temporarily nuts. I don't just ignore the fact that the two don't jibe.As to the correlation between TIPS and inflation and horizon. That statement is from The Handbook of Inflation Indexed Bonds, Brynjolfsson and Fabozzi, p. 15.
I disagree vehemently here. You can't use different yardsticks. Standard deviations need to be measured in consistent terms for both asset classes. Either you use nominal SDs when comparing or you use real SDs. Using one for nominals and the other for TIPS is apples and oranges.As to SD of TIPS, it makes no sense to look at SD of a real bond and compare it to the SD of a nominal bond as I have stated and explained.
So let's just stick with real SDs. Any plausible model says real SDs should be lower for TIPS. I agree. Actual historical data says real SDs are lower for TIPS. But here's the crucial point: They are not lower by a factor of 5.
I like your argument re Canadian equities. It makes sense to me. As a model. (You knew that was coming. ) The trouble is that it hasn't been borne out. Correlations of Canadian indexed bond returns against the S&P 500 are positive on both annual (0.06) and monthly (0.13) bases. There are diversification benefits to be had from correlations so close to zero, so they're worth buying IMHO.You are basing your comments on the performance of Canadian inflation rate instruments which of course are not TIPS which are US based inflation instruments.
Now there is likely to be a big difference in the correlations with the respective domestic equities. While I am certainly not an expert on Canadian equities my guess is that they are much more heavily dominated by commodity producers than US stocks and thus could logically be expected to have a very different correlation picture to inflation protected securities than would US equities. My guess is the same is true for any major commodity producer.
So it may be that inflation protected instruments in Canada are positively correlated to equities because Canadian equities may be as well because of the commodity component. But the data for the US and TIPS is very clear--the correlations to equities are negative as I have shown.
In the end, we agree that TIPS are a useful asset class, that they are a powerful diversifier for the typical portfolio, and that any plausible model of asset class returns would lead a rational investor to use them. Where I disagree with you is in your presentation of the argument. Unsourced and unverifiable numbers weaken your argument. A claim that similar correlations hold in other developed countries, which can be demonstrated to be untrue, weakens your argument. You already have a good argument. Why bother framing OJ?
Finally, let me quote - perhaps paraphrase - from a very smart man.
Your model says TIPS should be negatively correlated with equities, i.e. it is highly improbable that they are positively correlated. That doesn't make it impossible. What is about TIPS that exempts this particular asset class from your own very good advice?Larry Swedroe wrote:Never confuse the highly improbable with the impossible.
-
- Posts: 112
- Joined: Sat Aug 16, 2008 2:12 pm
- Location: Colorado
-
- Posts: 16022
- Joined: Thu Feb 22, 2007 7:28 am
- Location: St Louis MO
-
- Posts: 16022
- Joined: Thu Feb 22, 2007 7:28 am
- Location: St Louis MO
Tramper Al
I am not incorporating any forecast per se. I am simply observing the market's forecast of inflation and subtracting it from the nominal bond and then comparing it to the TIPS rate to see what the risk premium is estimated at. A high risk premium will lower my demand for TIPS, all else equal and a low risk premium will raise it. No different than for any insurance product you buy. The lower the cost the more the demand. The higher the cost the more you are willing to self insure
I am not incorporating any forecast per se. I am simply observing the market's forecast of inflation and subtracting it from the nominal bond and then comparing it to the TIPS rate to see what the risk premium is estimated at. A high risk premium will lower my demand for TIPS, all else equal and a low risk premium will raise it. No different than for any insurance product you buy. The lower the cost the more the demand. The higher the cost the more you are willing to self insure
-
- Posts: 16022
- Joined: Thu Feb 22, 2007 7:28 am
- Location: St Louis MO
Norbert
Dont have time to address all the issues you raise
But I never said it was impossible that TIPS and equities would be positively correlated. In fact the data shows it is not perfectly negative, just negative. Which means that there is a tendency for it to be negative-
Also the issue should not be about Canadian inflation indexed bonds and US stocks but Canadian inflation indexed bonds and Canadian stocks
Also you are entitled to your opinion on the issue about real SD but I don't see how you can make that claim. It makes no sense IMO. SD is about risk and the nominal SD of a real bond can be totally unrelated to its risk. As the example I gave shows. That is why the study looked at the real SD.
Finally sorry gave the wrong citation on that --corrected it above
Dont have time to address all the issues you raise
But I never said it was impossible that TIPS and equities would be positively correlated. In fact the data shows it is not perfectly negative, just negative. Which means that there is a tendency for it to be negative-
Also the issue should not be about Canadian inflation indexed bonds and US stocks but Canadian inflation indexed bonds and Canadian stocks
Also you are entitled to your opinion on the issue about real SD but I don't see how you can make that claim. It makes no sense IMO. SD is about risk and the nominal SD of a real bond can be totally unrelated to its risk. As the example I gave shows. That is why the study looked at the real SD.
Finally sorry gave the wrong citation on that --corrected it above
Last edited by larryswedroe on Tue Sep 09, 2008 12:42 pm, edited 1 time in total.
-
- Posts: 3665
- Joined: Thu Oct 18, 2007 11:42 am
Hi Larry,larryswedroe wrote:Tramper Al
I am not incorporating any forecast per se. I am simply observing the market's forecast of inflation and subtracting it from the nominal bond and then comparing it to the TIPS rate to see what the risk premium is estimated at. A high risk premium will lower my demand for TIPS, all else equal and a low risk premium will raise it. No different than for any insurance product you buy. The lower the cost the more the demand. The higher the cost the more you are willing to self insure
Thanks for reply. I still don't see the distinction, though. Risk premium defined by break-even spread and a forecast, and concluding premium is currently negative (or very low) because break-even is lower than forecast. I happen to agree with all that, but I try not to consider forecasts, least of all my own!
-
- Posts: 16022
- Joined: Thu Feb 22, 2007 7:28 am
- Location: St Louis MO
Tramper Al
But the market forecast of inflation is important. It sets the price of the nominal bond, along with the risk premium for unexpected inflation
And the literature states that one should prefer TIPS (and I agree) even if risk premium is say 50bp. So if it is zero or negative you should prefer them even more
But the market forecast of inflation is important. It sets the price of the nominal bond, along with the risk premium for unexpected inflation
And the literature states that one should prefer TIPS (and I agree) even if risk premium is say 50bp. So if it is zero or negative you should prefer them even more
-
- Posts: 3665
- Joined: Thu Oct 18, 2007 11:42 am
Thank you. Exactly.bookshot wrote:Perhaps a naive question: So how can one know that the spread of TIPS to nominal bonds as determined by the market is not a more accurate reflection of future inflation than extraneous predictions?
That spread is the sum of (expected inflation + insurance premium). Say that the 10 year spread is 3.58 - 1.64 = 1.94. The efficient bond market's best estimate.
You can bring in an outside number for insurance premium and calculate the expected inflation.
Or you can bring in an outside (consensus of economists?) estimate of future inflation and calculate the insurance premium.
Or you can just give your own best guess and say, "well I know future inflation has got to be more than 1.94, even without adding any premium" etc.
Maybe the confusion stems from Larry saying "market forecast of inflation" as being from some other source (?) than the yield spread itself? What other source could it be? Otherwise the calculation is circular.
-
- Posts: 16022
- Joined: Thu Feb 22, 2007 7:28 am
- Location: St Louis MO
The problem here is that we cannot directly calculate the inflation estimate or the risk premium because there is no futures market on inflation
What we can do is take consensus economic forecasts. You can even take the Feds forecast, or the CBO or the CEA.
That should give you some idea of the size of the premium
What we can do is take consensus economic forecasts. You can even take the Feds forecast, or the CBO or the CEA.
That should give you some idea of the size of the premium
-
- Posts: 3665
- Joined: Thu Oct 18, 2007 11:42 am
Right, that's what I thought you had said in the past. I don't think I'd call that a "market forecast of inflation", though.larryswedroe wrote:The problem here is that we cannot directly calculate the inflation estimate or the risk premium because there is no futures market on inflation
What we can do is take consensus economic forecasts. You can even take the Feds forecast, or the CBO or the CEA.
That should give you some idea of the size of the premium
So, basically, this forecast is used to estimate the premium inherent in the break-even spread, and thus used to evaluate TIPS at a given price.
That was my only point, a few posts back.
Again, I like TIPS, I own TIPS, and I happen to think the bond market is letting them go a bit cheaply at the moment. But that is also a judgment based on a forecast.
Thanks.
-
- Posts: 16022
- Joined: Thu Feb 22, 2007 7:28 am
- Location: St Louis MO
Tramper Al
Agreed
Norbert
Again my apologies on that citation, I checked it from a old early draft that had been changed in the editing and thus cited the wrong study
Handbook of Inflation Indexed Bonds, edited by John Brynjolfsson and
Frank Fabozzi (Hoboken, New Jersey: Wiley, 1999). page 13
That is the correct citation for that study
Agreed
Norbert
Again my apologies on that citation, I checked it from a old early draft that had been changed in the editing and thus cited the wrong study
Handbook of Inflation Indexed Bonds, edited by John Brynjolfsson and
Frank Fabozzi (Hoboken, New Jersey: Wiley, 1999). page 13
That is the correct citation for that study
Thanks Larry for starting this thread - I've found it very useful and informative. At the risk of beating up on a possibly already dead horse, I just wanted to try and draw an actionable conclusion regarding the issue of duration risk of TIPS:
SmallHi, who believes ST bonds are preferable to TIPS going forward said:
SmallHi, who believes ST bonds are preferable to TIPS going forward said:
Larry said:In the next decade, if rates rise again to 3.5%, you could see the Vanguard TIPS Index lose 1.6% (8YR Duration * 2% change in IRs). This is currently more than the real yield of Vanguard TIPS
ST bonds are reasonable alternative to TIPS since they don't take the duration risk and don't have much inflation risk. But that means you have to stay short. You give up the yield curve premium for term risk. With TIPS you get the term premium but don't have the inflation risk. Clearly that has to be an advantage, unless the risk premium for unexpected inflation is very high.
So, bottom line - is the best course of action to stay short and phase into TIPS as the coupon (real rate) on them rises and be "all in" when the coupon approaches 2.7%?For you to believe the real yield on longer term bonds will rise and stay at 3.5% you have to think real gnp would rise and stay at 4.5% or so real growth--don't think anyone believes that is even remotely likely. No developed country has sustained that growth rate for very long. And if the longer real rates were that high nominal ST bonds likely stay 1% or so less than that (maybe 50-150 range depending on how short is short). So I don't see you much better off even if that happened.
Now I certainly would prefer to see higher TIPS rates. And in fact currently I have only about 10% in TIPS and rest in ST nominals (Not counting taxable fixed income which is all in munis). But if rates rise I will get back to the 100% TIPS I was not that long ago--when real yields went above 2.7%. I do happen to believe valuations matter..
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard |
|
"You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs
-
- Posts: 16022
- Joined: Thu Feb 22, 2007 7:28 am
- Location: St Louis MO
Lbill
few thoughts
a) IMO SH is wrong on his analysis because if one simply holds the TIPS to maturity then there is no loss. You earned EXACTLY the real rate you bought at. With a nominal bond that will not be the case.
b) another example, as I stated, using TIPS allows one to go longer, and since yield curves are normally upward sloping you gain yield/return by doing so. But with nominal bonds you pick up incremental correlation with stocks and the inflation risks. But with TIPS that is not true. You get the negative correlation and you don't have the inflation risk thus your returns should be (they are expected to be) higher.
Here is perfect example. Today the 2 year Treasury about 2.2% NOMINAL but the 20 year TIPS now at 2.1 REAL. Almost same REAL return as the 2 year is NOMINAL return for the same credit risk and similar inflation risk. So which looks better to you? Especially in light of the historical data on real rates.
All I can say as to best course is that currently because of the steep yield curve on TIPS I prefer the 20year or so (bit shorter better, find the sweet spot where curve stops going up). It is not far below the intermediate to long term real returns for nominal bonds. If the curve was flat I would buy shorter TIPS at these levels, say 5-10 year. Then if real rates rose I would extend maturity to lock in those higher real rates for longer time frame.
Right now a good strategy seems to me to be a bar bell--buy some 20 year and some 1 year CDs where can get much better rates, about 2% above Treasuries with no credit risk so that looks to me to be better than ST TIPS. Then revaluate when they mature.
Of course nothing wrong with simply building a ladder of TIPS or owning the Admiral shares of the Vanguard fund.
few thoughts
a) IMO SH is wrong on his analysis because if one simply holds the TIPS to maturity then there is no loss. You earned EXACTLY the real rate you bought at. With a nominal bond that will not be the case.
b) another example, as I stated, using TIPS allows one to go longer, and since yield curves are normally upward sloping you gain yield/return by doing so. But with nominal bonds you pick up incremental correlation with stocks and the inflation risks. But with TIPS that is not true. You get the negative correlation and you don't have the inflation risk thus your returns should be (they are expected to be) higher.
Here is perfect example. Today the 2 year Treasury about 2.2% NOMINAL but the 20 year TIPS now at 2.1 REAL. Almost same REAL return as the 2 year is NOMINAL return for the same credit risk and similar inflation risk. So which looks better to you? Especially in light of the historical data on real rates.
All I can say as to best course is that currently because of the steep yield curve on TIPS I prefer the 20year or so (bit shorter better, find the sweet spot where curve stops going up). It is not far below the intermediate to long term real returns for nominal bonds. If the curve was flat I would buy shorter TIPS at these levels, say 5-10 year. Then if real rates rose I would extend maturity to lock in those higher real rates for longer time frame.
Right now a good strategy seems to me to be a bar bell--buy some 20 year and some 1 year CDs where can get much better rates, about 2% above Treasuries with no credit risk so that looks to me to be better than ST TIPS. Then revaluate when they mature.
Of course nothing wrong with simply building a ladder of TIPS or owning the Admiral shares of the Vanguard fund.
Thanks for your comments on this Larry. When purchasing TIPS directly, do you recommend the secondary market or at issue (treasury auction)? The question I'm focussing on here is the price relative to par. Elsewhere, you said, and I agree, that it is better to purchase TIPS at or below par. Purchasing at auction avoids a brokerage fee (thru Vanguard); however, the auction results may have the purchase price at issue above par - I don't know if you can predict this in advance or whether it's a big deal or not. In the secondary market, you know the purchase price and can try to time purchases to buy at or below par. Appreciate any guidance here. Thanks.
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard |
|
"You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs
-
- Posts: 3665
- Joined: Thu Oct 18, 2007 11:42 am
Typically, the coupon is set quite close to the market rate for an issue of similar maturity, so that the price at auction is very close to par, and most often at a modest discount. It is possible to predict in advance, to the extent that the Treasury does declare the coupon rate beforehand and that you can reasonably guess the auction yield based on the current market.Lbill wrote:. . . however, the auction results may have the purchase price at issue above par - I don't know if you can predict this in advance or whether it's a big deal or not.
-
- Posts: 16022
- Joined: Thu Feb 22, 2007 7:28 am
- Location: St Louis MO
Lbill
Have to look at the sweet spot on the curve--so generally not going to be lucky enough to be the new issue. So need to buy on secondary and thus will want to do so if can get "institutional" type pricing--which I can get because we have a trade desk that buys about $2billion a year of bonds.
When I bought my recent round I thought the sweet spot was about the 18 year period. All else equal would prefer below vs above par, but I really don't worry about that much because as I have shown even above par bonds have deflation protection. It is just that the less above par the more opportunity to BENEFIT from deflation as any deflation that leads to prices below par at maturity becomes par and you have then have REAL GAINS.
Have to look at the sweet spot on the curve--so generally not going to be lucky enough to be the new issue. So need to buy on secondary and thus will want to do so if can get "institutional" type pricing--which I can get because we have a trade desk that buys about $2billion a year of bonds.
When I bought my recent round I thought the sweet spot was about the 18 year period. All else equal would prefer below vs above par, but I really don't worry about that much because as I have shown even above par bonds have deflation protection. It is just that the less above par the more opportunity to BENEFIT from deflation as any deflation that leads to prices below par at maturity becomes par and you have then have REAL GAINS.
In my case I'm thinking about TIPS inside my IRA at Vanguard, which require bond purchases thru the Vanguard Brokerage bond desk. I'm assuming that "institutional" pricing is not accessible to individuals in my situation, correct? Is that enough of a disadvantage that I'm better off looking a purchases "at auction?" Realizing I would not be able to pick or time the "sweet spot" in this way, but for purposes of slowly assembling a TIPS bond ladder over the next 5 year period.need to buy on secondary and thus will want to do so if can get "institutional" type pricing
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard |
|
"You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs
Lbill,
I am not debating the benefits of TIPS, I am just saying -- I am not confident most investors can look past the signficant volatility necesssary to capture guaranteed 2.5% (or whatever) real returns for the next 18 years. Bottom line is, we look at portfolio results. Risk and return. A TIPS bear market will be tough to swallow even if you'll be made whole in 18 years.
If we take this 18YR TIPS, for example, you are locking in a real rate of 2.5%, with the short term risk that if real rates rise by 2%, you'll see about 36% of your principal temporarily disappear. (18YR Duration * 2% interest rate change).
That is long term bond volatility, which can be tough to stomach. Just to give you some approximation, since 1964, a portfolio that was 55% Global Equities, 45% 1YR T-Notes had the same annual volatility as LT bonds. That portfolio, however, had an annualized real return of 6.5% a year through 2008. While the past is not prologue, I am not sure a long term guarantee is worth giving up the potential of 4% more real returns.
Now, maybe you can ingore TIPS volatility (in this case, look at your portfolio piece by piece, and not as a portfolio like you are supposed to). Maybe you can differentiate between some investments that have nominal volatility, and others that have real volatility.
I just think there is more than one way to do this, and personally, I find Equity/ST bond portfolios more compelling in most cases. With ST bonds, you are also able to take a bit of extra multifactor risk -- in terms of AAA/AA credit risk, and increase portfolio returns further without significantly increasing portfolio risk*. Overall, not a really big deal, however.
sh
*in the best example I can come up with, DFA has run two identical ST bond funds for almost 20 years -- DFA 5YR Gov't and DFA 5YR Global...only difference being that "Global" can buy Treasuries/Gov'ts AND AAA/AA corporate bonds <its fully hedged, so there is no impact from currencies>. The added flexibility of 5YR Global has led to 0.7% higher annual returns vs. 5YR Gov't without any increase in risk. 65/35 portfolios that use 5YR Global vs. 5YR Gov't also exibit no difference in risk, just higher returns...
First off, apologize for the delayed response, I haven't been keeping up with this thread.SmallHi, who believes ST bonds are preferable to TIPS going forward said:
I am not debating the benefits of TIPS, I am just saying -- I am not confident most investors can look past the signficant volatility necesssary to capture guaranteed 2.5% (or whatever) real returns for the next 18 years. Bottom line is, we look at portfolio results. Risk and return. A TIPS bear market will be tough to swallow even if you'll be made whole in 18 years.
If we take this 18YR TIPS, for example, you are locking in a real rate of 2.5%, with the short term risk that if real rates rise by 2%, you'll see about 36% of your principal temporarily disappear. (18YR Duration * 2% interest rate change).
That is long term bond volatility, which can be tough to stomach. Just to give you some approximation, since 1964, a portfolio that was 55% Global Equities, 45% 1YR T-Notes had the same annual volatility as LT bonds. That portfolio, however, had an annualized real return of 6.5% a year through 2008. While the past is not prologue, I am not sure a long term guarantee is worth giving up the potential of 4% more real returns.
Now, maybe you can ingore TIPS volatility (in this case, look at your portfolio piece by piece, and not as a portfolio like you are supposed to). Maybe you can differentiate between some investments that have nominal volatility, and others that have real volatility.
I just think there is more than one way to do this, and personally, I find Equity/ST bond portfolios more compelling in most cases. With ST bonds, you are also able to take a bit of extra multifactor risk -- in terms of AAA/AA credit risk, and increase portfolio returns further without significantly increasing portfolio risk*. Overall, not a really big deal, however.
sh
*in the best example I can come up with, DFA has run two identical ST bond funds for almost 20 years -- DFA 5YR Gov't and DFA 5YR Global...only difference being that "Global" can buy Treasuries/Gov'ts AND AAA/AA corporate bonds <its fully hedged, so there is no impact from currencies>. The added flexibility of 5YR Global has led to 0.7% higher annual returns vs. 5YR Gov't without any increase in risk. 65/35 portfolios that use 5YR Global vs. 5YR Gov't also exibit no difference in risk, just higher returns...
Last edited by SmallHi on Sun Sep 14, 2008 10:33 pm, edited 3 times in total.
-
- Posts: 16022
- Joined: Thu Feb 22, 2007 7:28 am
- Location: St Louis MO
Lbill
Just ask Vanguard for a bid offer spread on the bond you are asking about and don't tell them which side you are on. That will tell you the spread and how close to interbank price you are getting. The interbank market is very tight.
You should not have to give up say more than 2bp in yield I think/hope.
Just ask Vanguard for a bid offer spread on the bond you are asking about and don't tell them which side you are on. That will tell you the spread and how close to interbank price you are getting. The interbank market is very tight.
You should not have to give up say more than 2bp in yield I think/hope.
-
- Posts: 16022
- Joined: Thu Feb 22, 2007 7:28 am
- Location: St Louis MO
Sh
Note missing one important point--don't think of these things in isolation
When you have negative correlation than volatility is not necessarily a bad thing----unless you make the mistake of thinking about the asset in isolation.
Great example is using the GSCI vs DJAIG--GSCI lower returns and higher volatility but actually had better impact on portfolio (not by much but better). Looking at the SD would have been very misleading.
The reason for the outcome was the negative correlation.
Note missing one important point--don't think of these things in isolation
When you have negative correlation than volatility is not necessarily a bad thing----unless you make the mistake of thinking about the asset in isolation.
Great example is using the GSCI vs DJAIG--GSCI lower returns and higher volatility but actually had better impact on portfolio (not by much but better). Looking at the SD would have been very misleading.
The reason for the outcome was the negative correlation.
- market timer
- Posts: 6535
- Joined: Tue Aug 21, 2007 1:42 am
Even if there were a futures market for inflation, it would not predict inflation rates any different from the TIPS - nominal bond spread, or else an arbitrage opportunity would exist.larryswedroe wrote:The problem here is that we cannot directly calculate the inflation estimate or the risk premium because there is no futures market on inflation
I'd like to bet that inflation will average > 1.9% annually for the next 10 years, but have no room in my portfolio for TIPS. Many investors would probably like to make such a bet. A company like Profunds could make a lot of money creating a 5-10x leveraged TIPS ETF.
Is there any theoretical reason to expect the inflation risk premium should be positive? In other words, why shouldn't the market reward people for taking the risk that inflation will not turn out to be as high as expected? Inflation, after all, is correlated with GDP growth.
-
- Posts: 16022
- Joined: Thu Feb 22, 2007 7:28 am
- Location: St Louis MO
Market timer
I did not say anything about a futures market accuracy in predicting future inflation. Only that it would provide us with the market's estimate and then we could directly compute the risk premium for unexpected inflation. Then an investor could make a more informed decision about whether they felt the premium was worth it or not.
I did not say anything about a futures market accuracy in predicting future inflation. Only that it would provide us with the market's estimate and then we could directly compute the risk premium for unexpected inflation. Then an investor could make a more informed decision about whether they felt the premium was worth it or not.
- market timer
- Posts: 6535
- Joined: Tue Aug 21, 2007 1:42 am
I think this was asked before, but what is the leading/lagging relationship between the spread and CPI? For example, recent monthly CPI data have been showing an notable annualized increase in CPI, with lots of talk about rising prices for food, fuel, etc. Yet, the TIPS spread was under 2%, indicating much lower long-term inflation expectations. Now commodity prices seem to be tanking and there is even the possibility being discussed about "deflation", due to credit contraction and recession. Was the spread telling us that the spike in CPI wasn't going to stick?
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard |
|
"You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs
-
- Posts: 112
- Joined: Sat Aug 16, 2008 2:12 pm
- Location: Colorado
God knows you don't need an organized futures market for one to develop on its own. There are over the counter derivatives for just about every economic eventuality--which, of course, is one of the things getting us in to deeper trouble. If the market thought it needed a better predictor of unexpected inflation than the TIPS nominal spread, it would have arisen.
-
- Posts: 16022
- Joined: Thu Feb 22, 2007 7:28 am
- Location: St Louis MO
.
Sh,
There seems to be two views in this thread on the risks of buying TIPS directly. Lets assume an investor requires a real return of 2.5% from their investment over the next 18 years (to stick with the above example).
First view: Annual volatility (SD): If you buy a 2.5% yield/real rate TIPS at issue and hold it to maturity, the annual price volatility (even if held) is the risk of holding this issue. And if the standard deviation of US TIPS returns are similar to UK inflation-indexed bonds (just as US nominal bonds are to UK nominal bonds), then the real return SD of TIPS will likely be close to (slightly less than), similar maturity nominal bonds. And TIPS prices can decline significantly if real rates rise (as illustrated in the UK - table below), so the risks can be high.
Second view: Odds of success: If you buy a 2.5% yield/real rate TIPS at issue and hold it to maturity (assuming the 18YR TIPS in the example above), the risk is virtually zero (i.e. it has a US government guarantee of success).
IMO the first view will likely lead to a portfolio that includes ST nominal bonds (i.e. a portfolio with an expected real return of 2.5% with minimum expected standard deviation). The second view will likely lead to a portfolio that includes TIPS.
Which approach has the highest likelihood of achieving the investor’s objective? I guess that’s for each investor to decide (just trying to summarize my understanding of some of the differences being discussed).
Robert
.
Sh,
Don’t think the calculation is exactly right. As you know, maturity and duration measures are different, and TIPS prices respond more to changes in the real rate of interest (not sure if this is what you meant) than to the nominal rate. https://personal.vanguard.com/us/Vangua ... 04_ALL.jsp But agree prices decline when real rates rise as happened in the UK.If we take this 18YR TIPS, for example, you are locking in a real rate of 2.5%, with the short term risk that if rates rise by 2%, you'll see about 36% of your principal temporarily disappear. (18YR Duration * 2% interest rate change).
There seems to be two views in this thread on the risks of buying TIPS directly. Lets assume an investor requires a real return of 2.5% from their investment over the next 18 years (to stick with the above example).
First view: Annual volatility (SD): If you buy a 2.5% yield/real rate TIPS at issue and hold it to maturity, the annual price volatility (even if held) is the risk of holding this issue. And if the standard deviation of US TIPS returns are similar to UK inflation-indexed bonds (just as US nominal bonds are to UK nominal bonds), then the real return SD of TIPS will likely be close to (slightly less than), similar maturity nominal bonds. And TIPS prices can decline significantly if real rates rise (as illustrated in the UK - table below), so the risks can be high.
Code: Select all
UK Inflation-Indexed Bonds (20 yr average maturity)
Years of Negative Annual Return: 1981-2000
1981 -15%
1983 -4%
1985 -6%
1990 -4%
1994 -10%
Source: Triumph of the Optimists
IMO the first view will likely lead to a portfolio that includes ST nominal bonds (i.e. a portfolio with an expected real return of 2.5% with minimum expected standard deviation). The second view will likely lead to a portfolio that includes TIPS.
Which approach has the highest likelihood of achieving the investor’s objective? I guess that’s for each investor to decide (just trying to summarize my understanding of some of the differences being discussed).
Robert
.
-
- Posts: 213
- Joined: Mon Aug 25, 2008 3:01 am
Sorry to interject an elementary question here, but in looking at the historical TIPS prices for the past year, I see places where a couple of cents' change in the NAV was accompanied by a much larger increase in the yield. Typically they move inversely, but here it happened once when the NAV declined and once when it increased, each time by .02. Both times the yield increased, once by .12 and once by .09. Is there a reason for this? Is it the Fed's semiannual revaluation? It seems to have occurred twice in August:
08/07/2008 $12.64 1.02%
08/08/2008 $12.62 1.14%
08/11/2008 $12.53 1.14%
08/12/2008 $12.59 1.14%
08/13/2008 $12.58 1.14%
08/14/2008 $12.64 1.14%
08/15/2008 $12.66 1.23%
08/18/2008 $12.68 1.23%
08/07/2008 $12.64 1.02%
08/08/2008 $12.62 1.14%
08/11/2008 $12.53 1.14%
08/12/2008 $12.59 1.14%
08/13/2008 $12.58 1.14%
08/14/2008 $12.64 1.14%
08/15/2008 $12.66 1.23%
08/18/2008 $12.68 1.23%
"In theory, there is no difference between theory and practice, but in practice there is." -- Jan L.A. van de Snepscheut (1953-1994), late of CalTech
Robert,
An excellent summary. I might add, for investors torn between the two, a 50/50 split is fine. Just as investors who believe in the size/value premiums still usually don't invest 100% in SV, there is plenty of room for compromise.
I don't want to come across as thinking your relative fixed income decision makes a huge amount of difference. What % of stocks vs. bonds is far more important than if you decide to go 100% TIPS (w/i fixed income) vs. 100% ST Bonds or some combo of the two.
Later!
sh
An excellent summary. I might add, for investors torn between the two, a 50/50 split is fine. Just as investors who believe in the size/value premiums still usually don't invest 100% in SV, there is plenty of room for compromise.
I don't want to come across as thinking your relative fixed income decision makes a huge amount of difference. What % of stocks vs. bonds is far more important than if you decide to go 100% TIPS (w/i fixed income) vs. 100% ST Bonds or some combo of the two.
Later!
sh
-
- Posts: 213
- Joined: Mon Aug 25, 2008 3:01 am
Another (perhaps) elementary question. ST nominals have about 1% lower yield now than IT. Larry Swedroe mentioned he had reduced his TIPS to 10% and was going to buy again when they improved. Why not do the same thing with IT compared to ST? i.e., buy IT now and if they decline compared to ST, sell them and buy ST?
"In theory, there is no difference between theory and practice, but in practice there is." -- Jan L.A. van de Snepscheut (1953-1994), late of CalTech
SH- not sure I completely agree. At my stage of the journey I have a lot more in bonds than stocks and that makes me more concerned about how to allocate the bond portion between ST, IT, TIPS, and even Foreign. A 100% TIPS allocation concerns me if, as you state, I might be sitting on a sizeable capital loss if interest rates rise. This could be particularly disconcerting because I could probably expect my equity holdings to be tanking at the same time.don't want to come across as thinking your relative fixed income decision makes a huge amount of difference. What % of stocks vs. bonds is far more important than if you decide to go 100% TIPS (w/i fixed income) vs. 100% ST Bonds or some combo of the two.
"Life can only be understood backward; but it must be lived forward." ~ Søren Kierkegaard |
|
"You can't connect the dots looking forward; but only by looking backwards." ~ Steve Jobs