Will You Use the New Vanguard Short-Term TIPS Index Fund?

[natureexplorer]

https://advisors.vanguard.com/VGApp/iip/site/advisor/researchcommentary/article/IWE_NewsInflationHedging

Benchmark: Barclays U.S. Treasury Inflation-Protected Securities (TIPS) 0-5 Year Index
Expense Ratio (ER): 0.10% for ETF

There are also Admiral mutual fund shares with the same ER, but they come with a 0.25% purchase fee.

[dyangu]

Nope. I would buy the full $10k/year in I Bonds before considering short term TIPS.

[Phineas J. Whoopee]

Nope. I would buy the full $10k/year in I Bonds before considering short term TIPS.

In today's interest rate environment that would be wise.

PJW

[gkaplan]

We need a "I'm not sure" option.

[jebmke]

I only buy TIPS in individual bonds.

[G-Money]

I dislike both the shorter duration and the purchase fee.

[Van]

I'm considering it, but, quite frankly, I'm failing around right now trying to figure out what to do. I don't have a clue.

Mr. Bogle recently remarked that this is the toughest investment environment that he has seen since he started investing in 1951. It's no wonder that someone like me has a tough time trying to invest wisely at this time!

[stevewolfe]

I'd buy the mutual fund after purchasing my limit in I-Bonds.

[Doc]

No. There are many less expensive options to accomplish the same thing at lower cost. Many have been mentioned already. Not mention include: Existing short TIPS ETFs already exist. A short nominal Treasury ladder or fund also is an alternative since the main benefit of TIPS is to allow you to go long when real yields are attractive without taking on the inflation risk. In the short time the baked in expected inflation component is more likely to be correct simply because we are better able to estimate the immediate future.

An intriguing aspect of the .25% purchase fee is its effect on the ETF class. By charging the other classes extra you effectively subsidize the ETF class. That's an interesting way to get the ETF volume up quickly so that the necessary NAV/price arbitrage mechanism can work.

[dbr]

I think the comments by Doc above are most helpful. It is a little hard to understand what real attraction lies in offering this fund.

I wonder if there is not enough business coming to the existing short TIPS ETF's and/or to short bond funds that Vanguard would like to capture part of that market, whether or not it makes sense for anyone in particular to invest in that asset.

PS As discussed, I think correlation per se with inflation is a red herring.

[abuss368]

No interest here for pretty much the same reasons Doc noted.

[huntertheory]

Setting aside the necessity of the fund/expenses/other ways to accomplish the same thing, what is the rationale? Is it to get inflation protection with reduced interest rate risk, to the extent they aren't exactly the same thing? I.e. if nominal rates move up for policy or other reasons, then you don't get hurt like a long-term TIPS fund, but if they move up because of an increase in CPI you are protected?

Just trying to figure out the theory behind it.

[DaveS]

Well It's nice to know you wont do worse than inflation. I suppose you could use it for something like an emergency fund, but I would prefer I bonds.

[natureexplorer]

Well It's nice to know you wont do worse than inflation.

I guess you haven't been watching real yields in a couple years.

[nisiprius]

I'm sorry you didn't set up the poll to allow changing votes. I started out voting "no, I don't like the short duration," because I was thinking of it vis-a-vis my existing holding of the intermediate-term Vanguard Inflation-Protected Securities (VIPSX), which is used to complement my individual TIPS and is roughly similar in composition to my individual TIPS.

But then I realized that I also have a small-medium holding of Vanguard Short-Term Bond Index, that I use because, well, I think I have enough "true" cash-like emergency funds that I decided, yeah, I could take a little risk with some of it. And I have to think about this, but on the whole I'm thinking that maybe the Short-Term TIPS Index Fund would be no worse, and I sort of like the idea of keeping a reasonably high percentage of my portfolio indexed to inflation, and I might just shift it to the short-term TIPS fund, for no particular reason.

In this case, the attraction lies in its "frictionlessness." I wouldn't have considered using STPZ because that would mean an ETF in the VBS account and all the minor nuisances that entails.

So, never say never.

By the way, if Vanguard is going to do TIPS indexes, why don't they switch VIPSX to being a true index fund, too?

[VennData]

Already in. Great compliment to a portfolio, the ETF is subsidized by the fee to the other funds. This is the ultimate safe, short-term investment.

[Doc]

But then I realized that I also have a small-medium holding of Vanguard Short-Term Bond Index, that I use because, well, I think I have enough "true" cash-like emergency funds that I decided, yeah, I could take a little risk with some of it. And I have to think about this, but on the whole I'm thinking that maybe the Short-Term TIPS Index Fund would be no worse, and I sort of like the idea of keeping a reasonably high percentage of my portfolio indexed to inflation, and I might just shift it to the short-term TIPS fund, for no particular reason.

Nisi, you might want to rethink what "no worse" means.


(Adjusted Price Chart)

(The Vg short Treasury price actually went up from 10.66 to 10.92 during the latter half of '08.

http://quote.morningstar.com/fund/chart ... %2C0%22%7D

It was this precipitous fall of TIPS prices during the banking crisis that caused me to replace the short rungs of my TIPS ladder with similar duration nominal Treasuries. The '08 event may not occur again in my lifetime but the cost of swapping short nominals for short TIPS before a crisis is minimal and if done only at the short end the extra inflation risk is also minimal.

(I am also currently using a short TIPS ETF but this is really only a placeholder for the longer rungs of my ladder which I am not ready to replace at the current low real rates.)

[Phineas J. Whoopee]

...
An intriguing aspect of the .25% purchase fee is its effect on the ETF class. By charging the other classes extra you effectively subsidize the ETF class. That's an interesting way to get the ETF volume up quickly so that the necessary NAV/price arbitrage mechanism can work.

Hi Doc,

Little as I relish the prospect of disagreeing with you, I don't think that can be the case.

Let's suppose the 0.25% purchase fee reasonably represents the transaction costs of buying the requisite short-term TIPS.

If that's true, then Vanguard, with its extremely efficient execution skills, isn't the only firm that would have to pay it. Authorized Participants, which exchange securities for creation units and thereby supply the market with ETF shares, would also have to pay at least that much.

With relatively high transaction costs, the arbitrage equation would change slightly for the Authorized Participants. That means there will be a drag on providing new ETF shares for which there is demand, and for destroying ETF shares for which there is insufficient demand.

I believe, although if I'm wrong I'm sure you, #Cruncher, or some other helpful poster ( ) will tell me, I believe the net effect will be increased spread, and increased tracking error, for the ETF shares.

If the purchase fee of 0.25% legitimately represents transaction costs, there would be no way ETF shareholders could escape something similar to it.

PJW

[Carpe]

I voted no - I prefer the certainty in maturity value of individual TIPS holdings.

[Jack]

I have to agree with nature explorer on the previous TIPS thread. What you are really talking about is duration and there are better ways to adjust that than shortening TIPS.

There are two risks you are trying to balance. There is real interest rate risk, which you reduce by shortening duration. And then there is unexpected inflation risk, which you reduce by buying TIPS.

Either long or short TIPS will protect against unexpected inflation risk, but long TIPS work better because they provide longer protection. If unexpected inflation shows up, then short TIPS will become more expensive for the fund to roll over while long TIPS do not need to roll over as often. If you try to reduce duration risk by shortening TIPS, you have the side effect of also reducing your long term inflation protection and that may not be an optimal strategy.

TIPS are an insurance policy for inflation protection. Just like term life insurance, you want to buy 20 or 30 years guaranteed when you are young and healthy. You don't want to buy term life insurance for 5 years at a time and risk not being able to get it when you need it or the price going up due to bad health. The same applies to a short term TIPS fund. You don't want to be rolling TIPS over after inflation takes off. It is better to have long term TIPS inflation insurance and compensate for duration by shortening your other bond holdings.

[gkaplan]

There seems to be a negative consensus here toward the Vanguard Short-Term TIPS Index Fund, so I wonder what was the reasoning for Vanguard to roll out this fund.

[BornInCA]

Would this be a good addition for savers who want to build and maintain a diversified short term fixed income portfolio in a taxable account? I'm just listing the asset class and their Vanguard funds and ETFs for reference.

Short-Term Government (traditional) bonds: Vanguard Short Term Treasury (vfisx/vfirx) OR Vanguard Short Term Bond Index (vbisx/vbirx/bsv) OR Vanguard Short-Term Government Bond ETF (VGSH)

Short-Term Municipal bonds: Vanguard Short-Term Tax-Exempt (vwstx/vwsux) OR Vanguard Limited-Term Tax-Exempt (VMLTX/VMLUX)

Short-Term Inflation Protected Bonds: Vanguard Short-Term Inflation Protected Securities Index (VTIPX/VTAPX/VTIP)

Since it's a taxable account, I excluded short-term Corporate bonds from this post. Depending on one's tax situation, I would guess that it would consist of at least one of the top two asset classes. I would only need to pick one fund for each asset class above.

Can it be said that "Short-Term Inflation Protected Bonds" is the third asset class? I know it's different from short-term municipal bonds but how does it compare/contrast to short-term government bonds? How much correlation can we expect from VTAPX against VFIRX?

For example, let's say a saver in the middle federal income tax bracket has his short term fixed income portfolio structured this way:
80% VBIRX and 20% VWSTX. VWSTX is added more for the diversification benefit than the tax benefit.

What if he were to modify his short term fixed income porfolio to include Short Term TIPS to this fund allocation:
60% VBIRX, 20% VWSTX, and 20% VTIP

Would VTIP provide an added diversification benefit? why or why not? How so or how not?

[Phineas J. Whoopee]

Would this be a good addition for savers who want to build and maintain a diversified short term fixed income portfolio in a taxable account? I'm just listing the asset class and their Vanguard funds and ETFs for reference.

Short-Term Government (traditional) bonds: Vanguard Short Term Treasury (vfisx/vfirx) OR Vanguard Short Term Bond Index (vbisx/vbirx/bsv) OR Vanguard Short-Term Government Bond ETF (VGSH)

Short-Term Municipal bonds: Vanguard Short-Term Tax-Exempt (vwstx/vwsux) OR Vanguard Limited-Term Tax-Exempt (VMLTX/VMLUX)

Short-Term Inflation Protected Bonds: Vanguard Short-Term Inflation Protected Securities Index (VTIPX/VTAPX/VTIP)

Since it's a taxable account, I excluded short-term Corporate bonds from this post. Depending on one's tax situation, I would guess that it would consist of at least one of the top two asset classes. I would only need to pick one fund for each asset class above.

Can it be said that "Short-Term Inflation Protected Bonds" is the third asset class? I know it's different from short-term municipal bonds but how does it compare/contrast to short-term government bonds? How much correlation can we expect from VTAPX against VFIRX?

For example, let's say a saver in the middle federal income tax bracket has his short term fixed income portfolio structured this way:
80% VBIRX and 20% VWSTX. VWSTX is added more for the diversification benefit than the tax benefit.

What if he were to modify his short term fixed income porfolio to include Short Term TIPS to this fund allocation:
60% VBIRX, 20% VWSTX, and 20% VTIP

Would VTIP provide an added diversification benefit? why or why not? How so or how not?

Hi BornInCA,

I'm sorry, but didn't you just ask that question word for word?
http://www.bogleheads.org/forum/viewtopic.php?f=10&t=104253&p=1513611#p1513611

And didn't someone answer?
http://www.bogleheads.org/forum/viewtopic.php?f=10&t=104253&p=1513611#p1513695

I'm not sure I understand. Is everything OK? If you want more attention paid to your question you could start a new thread and ask it there.

PJW

[BornInCA]

Hi BornInCA,

I'm sorry, but didn't you just ask that question word for word?
http://www.bogleheads.org/forum/viewtopic.php?f=10&t=104253&p=1513611#p1513611

And didn't someone answer?
http://www.bogleheads.org/forum/viewtopic.php?f=10&t=104253&p=1513611#p1513695

I'm not sure I understand. Is everything OK? If you want more attention paid to your question you could start a new thread and ask it there.

PJW

Yes, I copied and pasted here. More opinions are always better.

[ftobin]

With relatively high transaction costs, the arbitrage equation would change slightly for the Authorized Participants. That means there will be a drag on providing new ETF shares for which there is demand, and for destroying ETF shares for which there is insufficient demand.

I believe, although if I'm wrong I'm sure you, #Cruncher, or some other helpful poster ( :happy ) will tell me, I believe the net effect will be increased spread, and increased tracking error, for the ETF shares.

If the purchase fee of 0.25% legitimately represents transaction costs, there would be no way ETF shareholders could escape something similar to it.

I was watching realtime quotes in VTIP for a bit, and I was seeing a one-penny spread on a $50 stock almost all the time. While 0.25% might represent transaction costs, that doesn't mean that translates into the ETF.

Consider that VWO, an emerging markets stock index, trades with a one-penny spread as well, when the underlying markets are closed. I find it difficult to think people providing liquidity are accurately pricing the underlyings listed in closed, more-difficult-to-trades markets down to 0.02%. There is more going on than simple constituent arbitrage.

[#Cruncher]

Either long or short TIPS will protect against unexpected inflation risk, but long TIPS work better because they provide longer protection. If unexpected inflation shows up, then short TIPS will become more expensive for the fund to roll over while long TIPS do not need to roll over as often. If you try to reduce duration risk by shortening TIPS, you have the side effect of also reducing your long term inflation protection...

One does not reduce inflation protection with short term TIPS. TIPS of all maturities provide almost exactly the same inflation protection. The only significant way return can vary between TIPS of different maturities is if their real yields differ.

Jack, you state that if unexpected inflation occurs, "short TIPS will become more expensive". This may or may not be the case. But even if it is true, it has nothing to do with their inflation protection. It simply means that their real yield will go down. This is known as re-investment risk, the flip side of interest rate risk.

Consider two alternatives:

• Buy and hold a 30-year TIPS until maturity.
• Buy and roll over 5-year TIPS at maturity for 30 years.

Apart from an unfortunate 15-day gap in the issuance of 5-year TIPS, * the inflation protection will be identical. The only difference in return will be due to the difference in real yield between the 30-year TIPS on the one hand and the real yields of the six 5-year TIPS on the other. Because 30-year real yields are currently higher than those for 5-years, the yields would have to rise for the last five 5-year TIPS in order to break-even with the single 30-year.

There is certainly a risk that the yields on the 5-year TIPS won't rise enough to match the locked in yield on the 30-year. This is re-investment risk. You believe that if there is unexpected inflation, the real yield on TIPS will fall. This is possible. But it's also possible that in an environment of rising inflation, the real yield of TIPS would rise.

* For example, the last 5-year TIPS matured on 4/15/2012. The next issue wasn't until 4/30/2012. (See the "Issued" column in the row near the bottom of this table of 5-year TIPS Auctions.)

[Doc]

Regarding the .25% purchase fee:

If the purchase fee of 0.25% legitimately represents transaction costs, there would be no way ETF shareholders could escape something similar to it.

But it doesn't represent the legitimate transaction cost as ftobin indicated. For example the bid ask spread on the '17 from the WSJ is only 2/32 with the market closed. Last week I got a quote from Schwab on the same issue and the bid/asked was only 117.683/118.08 for a small lot. The .25% is just Vg's "standard" purchase fee. In my opinion it is just Vg's way to juice the initial returns on the ETF class to get the trading volume up quickly. The fee will go away when the fund gets bigger and will not be an issue. And in the meantime it helps get the ETF volume up because the arbitragers are getting subsidized.

In the long run different expenses for different share classes can only be justified if the difference truly represents the difference in costs to the fund. In this case it does not but it is also not the "long run".

There are few reasons for any investor to use a short Treasury fund instead of their own ladder. These reasons include small portfolio, short term investment and convenience. Transaction cost should not be a major consideration for moderate sized portfolios. There is however one additional reason for Vg clients to use the new fund - you don't need to deal with the bond desk at VBS.

[Phineas J. Whoopee]

Regarding the .25% purchase fee:

If the purchase fee of 0.25% legitimately represents transaction costs, there would be no way ETF shareholders could escape something similar to it.

But it doesn't represent the legitimate transaction cost as ftobin indicated. For example the bid ask spread on the '17 from the WSJ is only 2/32 with the market closed. Last week I got a quote from Schwab on the same issue and the bid/asked was only 117.683/118.08 for a small lot. The .25% is just Vg's "standard" purchase fee. In my opinion it is just Vg's way to juice the initial returns on the ETF class to get the trading volume up quickly. The fee will go away when the fund gets bigger and will not be an issue. And in the meantime it helps get the ETF volume up because the arbitragers are getting subsidized.

In the long run different expenses for different share classes can only be justified if the difference truly represents the difference in costs to the fund. In this case it does not but it is also not the "long run".

There are few reasons for any investor to use a short Treasury fund instead of their own ladder. These reasons include small portfolio, short term investment and convenience. Transaction cost should not be a major consideration for moderate sized portfolios. There is however one additional reason for Vg clients to use the new fund - you don't need to deal with the bond desk at VBS.

Thanks ftobin and Doc. I understand your empirical observations of the spread.

With respect to my other point, increased tracking error, it cannot be possible relative to the mutual fund shares because the ETF is simply a different share class with an equal claim on the underlying assets.

At present, at least, the ETF shares certainly would seem to be a better buy, if one's portfolio already included short-term fixed income and one was considering moving it.



PJW