Larry Swedroe and others: MM instead of ST bond fund?

Have a question about your personal investments? No matter how simple or complex, you can ask it here.
Post Reply
Topic Author
jefmafnl
Posts: 501
Joined: Sun Mar 25, 2007 7:11 am

Larry Swedroe and others: MM instead of ST bond fund?

Post by jefmafnl »

Larry's book suggests not lengthening maturities unless one is "rewarded" with an extra .20% per year.

With the current odd yield curve (inverted between 6 months and 5 years), should one skip short term bond funds and use a money market fund for the short term part of one's bond allocation? As of today, Vanguard Treasury Money Market yields 4.83% (7-day yield), and Vanguard Short Term Treasury yields 4.59% (30-day yield).

Joel
User avatar
Karl
Posts: 1074
Joined: Sun May 13, 2007 5:52 pm
Location: Milwaukee, WI

Re: Larry Swedroe and others: MM instead of ST bond fund?

Post by Karl »

An inverted yield curve means that investors expect ST rates to fall, so a MM doesn't provide a free lunch of more money & less risk unless the market turns out to be in error regarding the direction of future rates.
User avatar
hollowcave2
Posts: 1790
Joined: Thu Mar 01, 2007 3:22 pm
Location: Sacramento, CA

interest rate volatility

Post by hollowcave2 »

Even though MM's have stability of principal, they have most volatility (uncertainty) of interest rates. If ST rates fall, you may find yourself very unhappy with your MM and discover that the ST bond fund is more expensive to buy after the rates adjust.

Historically, ST bond funds have outperformed MM's, and that's what I would base my decision upon for long term investment.

Steve
SmallHi
Posts: 1718
Joined: Wed Feb 21, 2007 6:11 pm

Post by SmallHi »

An inverted yield curve means that investors expect ST rates to fall, so a MM doesn't provide a free lunch of more money & less risk unless the market turns out to be in error regarding the direction of future rates.
Actually, not everyone buys into this theory. Jim Davis of Dimensional Fund Advisors went back and updated Fama's research on information in the term structure, and had this to say about inverted yield curves and future interest rates:
Finally, the information contained in the term structure is independent of whether the yield curve is upward sloping or inverted. Regression tests show that inverted yield curves do not contain any more information about future interest rates than do normal yield curves
HI
strafe
Posts: 1028
Joined: Sat Mar 03, 2007 12:49 pm

Post by strafe »

That's a really interesting question, Joel. Why own bonds at all instead of cash? I've been wondering the same myself, although not with respect to the shape of the yield curve.

A paper on EMH that suggested the most efficient way to maximize return at a given level of risk is to blend risk-free cash (e.g. money market) with risky assets (presumably cap-weighted equities). This is consistent with Larry's usual advice that equity-like risk is more efficiently taken with equities rather than high-yield bonds. Regular bonds would fall somewhere in between.

Here's a totally made up, exaggerated example. Assuming that all the portfolio combos below have the same risk/volatility, this would mean something like:
80% stocks / 20% t-bills (money market)
would return more than
75% stocks / 25% short term t-note
would return more than
70% stocks/ 30% intermediate term t-note
would return more than
65% stocks / 35% long term t-bond
Last edited by strafe on Mon May 14, 2007 5:03 pm, edited 1 time in total.
User avatar
dm200
Posts: 23148
Joined: Mon Feb 26, 2007 2:21 pm
Location: Washington DC area

No expert here, but

Post by dm200 »

it seems a possibility that the shortest tem rates could drop a lot, while the short/mid term bonds could stay where they are, or drop a bit, but less than the shortest term (overnight).

In that scenario, holding MM would be a net loser.

dan
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Post by larryswedroe »

Personally I stick with the DFA 2 year global fund--and it stays very short in this type environment
But if dont have access to that I don't see any problem with staying that short--though as others point out you take on more reinvestment risk. but I certainly would not be extending long here, unless deflation was the much greater risk to me--and IMO that is true of very few people
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Post by larryswedroe »

btw--just checked the current maturity of the DFA 2 YEAR GLOBAL fund is about 8 months
SmallHi
Posts: 1718
Joined: Wed Feb 21, 2007 6:11 pm

Post by SmallHi »

Here's a totally made up, exaggerated example. Assuming that all the portfolio combos below have the same risk/volatility, this would mean something like:
80% stocks / 20% t-bills (money market)
would return more than
75% stocks / 25% short term t-note
would return more than
70% stocks/ 30% intermediate term t-note
would return more than
65% stocks / 35% long term t-bond
The best series of stock/bond data was to use 2YR Treasury Notes in combo with stocks. Based on my #s, they provided the highest possible return for a given level of risk from 1964-2006. 2YR has the enviable characteristics of being a very good inflation hedge (+ annual correlation with inflation), but also capturing much of the longer term maturity premium.

This could be duplicated using Vanguard ST Federal or Corporate depending on your taste for credit risk, or in equal allocations to DFA 2YR Global and 5YR Global, as DFA does with their Global Strategies.

HI
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Post by larryswedroe »

as I am sure you are aware but others may not be, so to clarify,

The DFA 2 year global fund is not a two year fund. Max maturity is 2 years for any single security. Result average maturity always less than 2 years. And it shifts maturities as yield curve shifts. Today is about 8 months average.
User avatar
stratton
Posts: 11083
Joined: Sun Mar 04, 2007 5:05 pm
Location: Puget Sound

Post by stratton »

Larry Swedroe wrote:
The DFA 2 year global fund is not a two year fund. Max maturity is 2 years for any single security. Result average maturity always less than 2 years. And it shifts maturities as yield curve shifts. Today is about 8 months average.
The closest thing Vanguard has is the Vanguard Short-Term Tax-Exempt Fund (VWSTX) with a 1.1 year duration. I've seen a few other ultra-short bond funds out there too. The only foreign one I've seen thats close is Payden Global Short Bond Fund Class R (PYGSX) with an average durration of 1.9 years.

Paul
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Post by larryswedroe »

One to two years should not make that much difference. In fact highest Sharpe Ratio in study I saw was about one year (of course cannot spend sharpe ratios, just returns, and 2 year about best for returns)/
User avatar
CountryBoy
Posts: 1777
Joined: Wed Feb 28, 2007 10:21 am
Location: NY

So Larry is it possible

Post by CountryBoy »

to say that by committing to a ST bond fund at say Vanguard that you are not worried about:

1) the yearly expense, even though low, of same or

2)the fact that if the interest rates change then the NAV or price (and therefore one's initial investment in the fund) could plumet? I mean in 20 yrs time at the time that one decides during retirement that it is necessary to cash in the shares then the initial investment could be radically depreciated could it not?

I am sure you have taken this scenario into account and so would appreciate your thoughts on it.

Thanks, as always.

Country Boy
SmallHi
Posts: 1718
Joined: Wed Feb 21, 2007 6:11 pm

Post by SmallHi »

Larry --

My thought on 2YR and 5YR is that the max duration assuming each fund is fully extended is 3.5yrs (2+5 / 2). That means to me in normal conditions, the funds weighted average duration will be between 1 and 3 years, which seems ideal for taxable, nominal fixed income.

As you say, right now, between DFGBX and DFGFX, the weighted average duration is about 1 year, so we are obviously at the lowest end of this range (whereas in 2002 we were probably near the highest end).

This combo seems to match up nicely with either DFA ST Muni Fund or Vanguard Limited Term Tax Exempt Fund which I really like for muni exposure. I know you prefer to go a bit longer, and like individual securities, but I think these funds do a fine job of "sticking to the plan" regarding the role of fixed income.

Also, I think the details behind Short vs. Short/Intermediate is far down the list of importance regarding portfolio building process. Much more important is the decision to diversify equities globally and determing appropriate factor exposure. Unless you are >50% fixed income, the difference between 2YR Global, 5YR Global, ST Bond Index, Int'd Treasury is gonna be pretty small relatively speaking

Incidentially, I cannot help but notice ST bonds seem to be about the most unloved investment available right now. Just a few years ago, it seemed most were on the same page regarding the importance of using ST bonds in portfolios...recently however, (as IT and LT returns have been better), I see quite a few mentions of Long Term Treasuries, Long Term Zeros, Intermediate Term Treasuries or Corporates....

Almost akin to saying LG now should play a prominent role in portfolios!

I think this is just another instance of "passive performance chasing" and highlights how hard it can be for many to stick with their portfolio policy.

:D just my 2 cents.
HI
SmallHi
Posts: 1718
Joined: Wed Feb 21, 2007 6:11 pm

Post by SmallHi »

Incidentially, the average weighted TERM coefficient for a 50/50 allocation to 2YR and 5YR global since 1996 has been 1.5 years...about the same as a 1-3 year Gov't/Credit Index, albiet with a higher after fee risk adjusted return.

HI
rokid
Posts: 509
Joined: Tue Feb 27, 2007 7:35 am

Post by rokid »

strafe wrote:A paper on EMH that suggested the most efficient way to maximize return at a given level of risk is to blend risk-free cash (e.g. money market) with risky assets (presumably cap-weighted equities).
It's my understanding that EMH/CAPM assumes bonds are one of the risky assets. Therefore, bonds are part of the tangency portfolio.

-----Jim
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Post by larryswedroe »

small hi
we agree
User avatar
alec
Posts: 3101
Joined: Fri Mar 02, 2007 2:15 pm

Re: So Larry is it possible

Post by alec »

CountryBoy wrote:to say that by committing to a ST bond fund at say Vanguard that you are not worried about:

1) the yearly expense, even though low, of same or

2)the fact that if the interest rates change then the NAV or price (and therefore one's initial investment in the fund) could plumet? I mean in 20 yrs time at the time that one decides during retirement that it is necessary to cash in the shares then the initial investment could be radically depreciated could it not?

I am sure you have taken this scenario into account and so would appreciate your thoughts on it.

Thanks, as always.

Country Boy
CB,

First, check out the NAV of VFSTX since inception. It went from $10.00 in 1982 to $10.59 today. Not that much change in 24-25 years.

Also, if you're reinvesting dividends you should be aware of the concept of duration. A few years ago Vanguard put out a bond basics thingee [which I can't seem to find now], but I did manage to save this pic:

Image

[Note, see the assumptions - initial YTM of 4%, fund duration of 5.8 years, dividends reinvested, etc.]

Basically, if you're reinvesting dividends in the bond fund, if interest rates rise, and you hold the ST fund for 20 years, you'd actually be better off, not worse off. Funny that bond investors are actually worse off since 1982 because interest rates on bonds fell, not better off.

- Alec
User avatar
CountryBoy
Posts: 1777
Joined: Wed Feb 28, 2007 10:21 am
Location: NY

Alec, thanks

Post by CountryBoy »

for taking the time to provide your answers.

CB
User avatar
Robert T
Posts: 2747
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Post by Robert T »

.
SmallHi,

I agree that the DFA two year and five global bond funds provide useful combinations for those with access to them. The shifting maturity and short duration seem consistent with Fama’s research:
  • - Using data from 1953-82 with five-year sub-periods, he found that bond funds with maturities greater than four years never had the highest average returns. [Term premiums in bond returns – Journal of Financial Economics, 1984]. Hence the focus on short duration. Although longer maturities have done better since 1982 (as indicated by Fama in his update on the DFA site)

    - He suggested that rewards and risks (in term exposure) vary with business cycle conditions (i.e. the ordering of risks and rewards across maturities change with the business cycle and its not always monotonic). [Term premiums and default premiums in money markets – Journal of Financial Economics, 1986]. Hence the focus of shifting maturities (as I understand it)
. Interestingly DFA’s first bond fund introduced in 1983 was a ‘One-Year’ fund with a primary focus on inflation protection following the high inflation in the late 1970s and early 1980s and Fama research on the subject [Asset returns and inflation – Journal of Financial Economics, 1977]. In fact, as I understand it the fund was initially called a ‘hedge fund’ (an inflation hedge fund) – TIPS bought and held to maturity now provide this inflation hedge with certainty:).

2 year Notes seems to provide a reasonably close reflection of the above and as indicated they perform well in the back-tests. I still stand by my first comment on this subject in the earlier M* thread – “A combination of TIPS [bought and held to maturity to be more explicit] and Intermediate Treasuries provide a greater TERM benefit…than 2 yr Notes while preserving some of the inflation protection of 2 yr Notes.”

The analysis below uses two time periods 1972-2006, the period I have international equity return data to allow better reflection of a globally diversified equity portfolio, and 1953-71 where I use only US equity data. The portfolio has a small and value tilt. The real yield used for TIPS was 2.5%. The assumption is TIPS are bought and held to maturity. 5yr T-Notes were used for intermediate treasuries.

Here are the useful attributes of the 50:50 TIPS:Intermediate treasury combination versus 2 yr Notes for fixed income (at least my take).

1. Greater downside risk protection (term benefit)

A 50:50 TIPS:Intermediate Treasuries combination performed better than 2 yr Notes when my equity benchmark had negative returns.

Code: Select all

Fixed income returns when equity benchmark waso negative

                         1972-2006         1953-1971
50:50 TIPS:IT combo         9.3               5.3
2 yr Notes                  8.2               4.8
The recent discussion on longer-term treasuries was to illustrate the term effect on downside risk protection (not to change ‘sticking to the plan’?: ).

2. Preserving some inflation protection

Code: Select all

Correlation with inflation (correlation coefficients)

                         1972-2006         1953-1971
50:50 TIPS:IT               0.24             0.58
2 yr Notes                  0.29             0.74
The equity value tilt also provides some inflation protection.

3. Similar mean-variance efficiency [before costs]

Here is the backtested portfolio performance when the following fixed income allocations are added to the same equity allocation. A 75:25 equity:fixed income allocation is used. The results show similar before cost results - 2 yr Notes slightly ahead.

Code: Select all

                   Annualized return     SD       Sharpe Ratio
1972-2006
50:50 TIPS:IT          14.17            13.46         0.661
2 yr Notes             14.20            13.49         0.663

1953-1971
50:50 TIPS:IT          11.54            16.52         0.578
2 yr Notes             11.55            16.55         0.577
4. Lower cost

A rough estimate suggests that a 50:50 TIPS:IT has about half the expense ratio of 2 yr Notes fund (for the former the expense ratio of the Vanguard IT fund was used and for TIPS bought and held to maturity the expense ratio used is zero, for the latter the average expense ratio of the DFA 2 and 5yr global funds was used). How did this impact the portfolio back-test results? (all portfolios assume a 0.4 percent expense ratio for the equity allocation). The results show similar before cost results but now with the 50:50 TIPS:IT slightly ahead.

The results

Code: Select all

                   Annualized return     SD       Sharpe Ratio
1972-2006
50:50	TIPS:IT         13.79           13.42        0.635
2 yr Notes              13.78           13.44        0.634

1953-1971
50:50	TIPS:IT         11.17           16.46        0.557
2 yr Notes              11.14           16.49        0.554
5. Locks in real returns

Reduces the risk - at least for the TIPS part of the fixed income portfolio.

So considering the above and the options I have available a 50:50 TIPS:IT combination seems to make sense to me.

Just my take.

Robert
.
User avatar
CountryBoy
Posts: 1777
Joined: Wed Feb 28, 2007 10:21 am
Location: NY

Bond Fund vs. Direct Treasury Paper Investment

Post by CountryBoy »

Is it ok to just once every 1-2 yrs invest in treasury paper versus putting the money in a ST or IT fund? That way I don't have to pay expenses and worry about loss of price? Yes I read what Alec said and understand his point.

Or put another way are there cost efficiencies of putting the $ in a fund and that the return would be higher that way even though I am having to pay an expense on a yearly basis?

Thanks.
CB
SmallHi
Posts: 1718
Joined: Wed Feb 21, 2007 6:11 pm

Post by SmallHi »

Robert,

Thanks for that analysis...presented a slightly different way than on the last thread.

I don't have the time to grind through it right now, but, could you remind me how you constructed your TIPS data set? Thanks!

hi
User avatar
alec
Posts: 3101
Joined: Fri Mar 02, 2007 2:15 pm

Re: Bond Fund vs. Direct Treasury Paper Investment

Post by alec »

CountryBoy wrote:Is it ok to just once every 1-2 yrs invest in treasury paper versus putting the money in a ST or IT fund? That way I don't have to pay expenses and worry about loss of price? Yes I read what Alec said and understand his point.

Or put another way are there cost efficiencies of putting the $ in a fund and that the return would be higher that way even though I am having to pay an expense on a yearly basis?

Thanks.
CB
Gee, no one wants to tackle the easy questions today. :D

CB,

If you were just going to put the bond money into a Treasury bond fund, buying the Treasury notes/bills at auction [hopefully for free] will most definitely be better than investing in a fund b/c there are less expenses. If this is an IRA, you'll have to open a brokerage acct. Note that you're essentially building your own ST or IT Treasury bond fund with less expenses. As long as you don't have to liquidate the Treasuries [in the secondary market], Woo hoo, free lunch!!

A mutual fund is just easier [for me at least] to reinvest dividends and sell whenever I want.

- Alec
SmallHi
Posts: 1718
Joined: Wed Feb 21, 2007 6:11 pm

Post by SmallHi »

Is it ok to just once every 1-2 yrs invest in treasury paper versus putting the money in a ST or IT fund? That way I don't have to pay expenses and worry about loss of price? Yes I read what Alec said and understand his point.

Or put another way are there cost efficiencies of putting the $ in a fund and that the return would be higher that way even though I am having to pay an expense on a yearly basis?
You don't get better pricing through mutual funds buying treasuries, but you are able to rebalance much easier, more efficiently target a static maturity (or variable maturity), and reinvest interest back into the fund. I assume individual bond investors invest interest back into the fund anyway...so the expense savings of doing it this way vs. using a fund seems so marginal its not worth it.

I would prefer to spend the time it takes to maintain an individual T-note portfolio every year (all of 30 minutes probably) reading the latest studies reinforcing why I want to maintain my small/value portfolio tilt -- which will be infinitely more profitable than the savings from avoiding a treasury fund and going direct. :D

Thats just me, however.
User avatar
Robert T
Posts: 2747
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

TIPS

Post by Robert T »

SmallHi,

I probably learn as much from the analyses I occasionally post as anyone else, including on how to better present my thoughts (often not easy:)). Anyway I'm still learning on every subject including fixed income.

On the TIPS series, following the definition...

"The principal of a TIPS increases with inflation and decreases with deflation, as measured by the Consumer Price Index. When a TIPS matures, you are paid the adjusted principal or original principal, whichever is greater. TIPS pay interest twice a year, at a fixed rate. The rate is applied to the adjusted principal; so, like the principal, interest payments rise with inflation and fall with deflation."

Right or wrong - I simply used inflation+2.5%. i.e. a real return of 2.5% which seems to be reflective of a buy at issue and hold to maturity approach to TIPS. Unlike the synthetic TIPS series often quoted which seems more relevant for those using a TIPS fund.

Robert
User avatar
CountryBoy
Posts: 1777
Joined: Wed Feb 28, 2007 10:21 am
Location: NY

Alec

Post by CountryBoy »

Thanks for the explanations and guidance. I know it looks simple to everyone else but I figured I had better ask just in case.

CB
SmallHi
Posts: 1718
Joined: Wed Feb 21, 2007 6:11 pm

Post by SmallHi »

Robert --

Just a few quick questions (I guess I didn't give this enough thougt the first time you presented it)....

If you just take CPI+2.5% annually, doesn't that considerably understate the amount of annual volatility, even if you plan to hold individual securities till maturity?

Doesn't it also considerably increase correlations with inflation on an annual basis? CPI +2.5% is naturally going to be very highly correlated with inflation, but its the short term price movements of actual TIPS securities that often swamp this "inflation credit" -- only washing out over periods = to maturity.

I guess I am a bit puzzeled how an actual TIPS + T-Note portfolio can have as much correlation with inflation as a much shorter term T-Note portfolio...as historically 5YR T-Notes have had 0 correlation with inflation on an annual basis. (thats asking a lot of TIPS over short intervals)

I do agree if you are looking at the asset class in isolation volatility/inflation correlations are a non-issue...but, in your examples, you are including TIPS in the context of a balanced portfolio and observing portfolio level SD and behavior. Also, this scenerio doesn't address the fact that rebalancing into/out of an individual TIPS security is pretty cumbersome, yet the portfolio is assumed to be rebalanced annually.

Finally, I guess I feel like the amount of long term inflation protection offered by 2YR T-Notes (say longer than 5 year periods) is so solid as to be almost as reliable as TIPS themselves. At the 10 yr time interval (randomly chosen as a way to compare contrast with 10YR individual TIPS), I can find only 1 time period historically where 2YR T-Notes had a negative real annualized return (and even then it was less than 0.4% annually). And, not that this should always be expected, but, during this period (1971-1980), as you know, the value and size premiums more than picked up the slack (+3.5% to +4.5% annually).

Have I totally missed the target here?

smallHI
User avatar
Robert T
Posts: 2747
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Post by Robert T »

.
SmallHi,

If you just take CPI+2.5% annually, doesn't that considerably understate the amount of annual volatility, even if you plan to hold individual securities till maturity?

I don’t think so. IMO principal is not at risk when buying a TIPS issue if held to maturity. This is not the case for a TIPS mutual fund as sales prices maybe lower than purchase prices leading to principal loss. By purchasing a TIPS issue an investor is lending money to the US government in exchange for a real return and return of an inflation adjusted principal when the terms of the loan expire [at maturity] (if indeed there is inflation over the holding period). So if held to maturity I don’t think and inflation + number understates volatility. But its not totally ‘risk-free’ IMO in that the investment is tied up until maturity (whether 5, 10 etc years) so the lower volatility of a directly held TIPS maybe some sort of illiquidity benefit. If an investor has to cash out before maturity then obviously principal is at risk. But if the investor also has more ‘liquid’ bond holdings or an emergency fund then this is not a high risk IMO.

Doesn't it also considerably increase correlations with inflation on an annual basis?

This is precisely one on the reasons to hold TIPS – to lock in a real return. This seems to be achieved more precisely through a direct holding to maturity than through a mutual fund due to the price noise of the latter. The US government will pay interest + inflation (real return) biannually and principal + inflation at maturity (if there is inflation). By definition TIPS are highly correlated with inflation as the inflation rate is a component of return.

The example in the earlier post does assume annual rebalancing and I agree that it is cumbersome to rebalance out of TIPS due to risk of principal loss as discussed above. However if only the 5-yr Notes (which is assumed to be held as a fund) is used for rebalancing I don’t think the results will be vastly different but will try a simulation of this case.

I agree that 2yr T-notes provides inflation protection – more so than a combination of 5yr-T notes and direct TIPS (as indicated in the previous post), but the latter also seems to provide more downside risk protection for a small and value tilted portfolio.

Just my take FWIW.

Robert
.
Post Reply