In defense of short-term treasuries

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

In defense of short-term treasuries

Post by danielc » Sat Jul 14, 2018 4:28 pm

Hello,

Common wisdom is that Intermediate Term (IT) treasuries are better for diversifying stocks than Short Term (ST) treasuries. The resoning provided is that, when markets crash, the Fed lowers itnerest rates and ITs go up in value. However, I ran Monte Carlo simulations and it looks like (1) there isn't much of a difference, and (2) ST actually look better in most cases.

Model 1: No contribution or withdrawal, 20 yr simulation, historical returns, 20yr blocks (link vs link)

Code: Select all

                            60% stock + 40% ST   |   60% stock + 40% IT
                            -------------------------------------------
Return (10th percentile)      6.67% *            |     6.57%
Return (50th percentile)      9.59% *            |     9.50%
Return (90th percentile)     12.22%              |    12.81% *
-----------------------------------------------------------------------
Maximum Drawdown (10th)     -29.75%              |   -27.98% *
Maximum Drawdown (50th)     -22.04% *            |   -27.28%
Maximum Drawdown (90th)     -18.73% *            |   -18.97%
In all cases the results are very similar, but I put a "*" next to the preferable result (highest return, lowest drawdown). Notice that on the 50th percentile, the Short Term treasuries provide both higher return and lower drawdown. This means that more than half the time you would have wished that you had chosen ST.

Next I will test other scenarios.

Model 2: No contribution or withdrawal, 20 yr simulation, statistical returns (link vs link)

Code: Select all

                            60% stock + 40% ST   |   60% stock + 40% IT
                            -------------------------------------------
Return (10th percentile)      6.52% *            |     6.39%
Return (50th percentile)      9.39% *            |     9.36%
Return (90th percentile)     12.30%              |    12.38% *
-----------------------------------------------------------------------
Maximum Drawdown (10th)     -20.86% *            |   -22.80%
Maximum Drawdown (50th)     -14.70% *            |   -15.95%
Maximum Drawdown (90th)     -10.54% *            |   -11.48%
Model 3: Fixed inflation-adjusted contributions, 20 yr simulation, statistical returns (link vs link)

Code: Select all

                            60% stock + 40% ST   |   60% stock + 40% IT
                            -------------------------------------------
Return (10th percentile)      6.50% *            |     6.36%
Return (50th percentile)      9.35% *            |     9.34%
Return (90th percentile)     12.32%              |    12.53% *
-----------------------------------------------------------------------
Maximum Drawdown (10th)     -15.98% *            |   -17.46%
Maximum Drawdown (50th)     -11.96% *            |   -12.92%
Maximum Drawdown (90th)      -9.13% *            |    -9.79%
Model 4: Fixed 4% yearly withdrawal, 20 yr simulation, statistical returns (link vs link)

Code: Select all

                            60% stock + 40% ST   |   60% stock + 40% IT
                            -------------------------------------------
Return (10th percentile)      6.48% *            |     6.32%
Return (50th percentile)      9.34%              |     9.39% *
Return (90th percentile)     12.39%              |    12.45% *
-----------------------------------------------------------------------
Maximum Drawdown (10th)     -30.81% *            |   -32.94%
Maximum Drawdown (50th)     -20.93% *            |   -22.18%
Maximum Drawdown (90th)     -14.79% *            |   -15.63%
Model 5: No contribution or withdrawal, 20 yr simulation, statistical returns, worst 10yrs first (link vs link)

Code: Select all

                            60% stock + 40% ST   |   60% stock + 40% IT
                            -------------------------------------------
Return (10th percentile)      6.48% *            |     6.42%
Return (50th percentile)      9.33% *            |     9.32%
Return (90th percentile)     12.35%              |    12.43% *
-----------------------------------------------------------------------
Maximum Drawdown (10th)     -31.41% *            |   -34.43%
Maximum Drawdown (50th)     -19.89% *            |   -22.10%
Maximum Drawdown (90th)     -12.17% *            |   -13.17%
In summary, two results seem to be consistently true:
  1. The difference between using short-term and intermediate-term treasuries is very small.
  2. Most of the time you would have been happier with short-term treasuries. Especially in bad times.
Last edited by danielc on Sat Jul 14, 2018 5:15 pm, edited 2 times in total.

User avatar
patrick013
Posts: 2297
Joined: Mon Jul 13, 2015 7:49 pm

Re: In defense of short-term treasuries

Post by patrick013 » Sat Jul 14, 2018 4:53 pm

Image
Most people hold TRSY's for market crashes. In a 50-50
portfolio when stocks really crash the TRSY's normally
increase in value as a flight to quality starts to occur. For
other time periods, other than when crashes are expected,
a blend of TRSY and Corp bond would give a higher current
yield. So the TRSY's can provide liquidity when stocks are
stalled.

I'm semi-active regarding bonds. In other words I don't hold
short all the time or whatever, but try to balance between short
and long responding to whatever secular trend is occurring. So
it's potentially better without added expense. I don't think your
study accounts for that but just a static AA which is fine.

If you linked the MC's on PV you could post the details via the PV link
there.
age in bonds, buy-and-hold, 10 year business cycle

megabad
Posts: 362
Joined: Fri Jun 01, 2018 4:00 pm

Re: In defense of short-term treasuries

Post by megabad » Sat Jul 14, 2018 5:46 pm

Interesting analysis. Thanks for posting. I assume this is using data starting in the 1970s? I might argue the validity of data that is almost exclusive reliant on interest rates over what I would call a relatively short period of time. You could, of course, use the same argument against any backtesting though. Still interesting.

aristotelian
Posts: 4372
Joined: Wed Jan 11, 2017 8:05 pm

Re: In defense of short-term treasuries

Post by aristotelian » Sat Jul 14, 2018 6:21 pm

With the yield curve flat and ST rates rising, they make a lot of sense right now. I have moved into ST and will move back to intermediate when the 10 year gets back above 3%.

jalbert
Posts: 3426
Joined: Fri Apr 10, 2015 12:29 am

Re: In defense of short-term treasuries

Post by jalbert » Sat Jul 14, 2018 6:41 pm

What probability distributions for future returns are being used to generate simulated data in the Monte Carlo simulations?
Index fund investor since 1987.

User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Re: In defense of short-term treasuries

Post by danielc » Sat Jul 14, 2018 6:58 pm

jalbert wrote:
Sat Jul 14, 2018 6:41 pm
What probability distributions for future returns are being used to generate simulated data in the Monte Carlo simulations?
Historical values. PV's dataset seems to begin in 1972.

User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Re: In defense of short-term treasuries

Post by danielc » Sat Jul 14, 2018 6:59 pm

megabad wrote:
Sat Jul 14, 2018 5:46 pm
Interesting analysis. Thanks for posting. I assume this is using data starting in the 1970s? I might argue the validity of data that is almost exclusive reliant on interest rates over what I would call a relatively short period of time. You could, of course, use the same argument against any backtesting though. Still interesting.
PV's dataset seems to being in 1972.

User avatar
triceratop
Moderator
Posts: 5469
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: In defense of short-term treasuries

Post by triceratop » Sat Jul 14, 2018 7:02 pm

The max drawdowns are determined using monthly portfolio returns but rebalancing is only allowed to occur yearly. That is not consistent with the reasoning provided by those who use IT instead of ST. What is so special about the end of the calendar year? If the reasoning you are attempting to disprove is correct the end of the year only matters if stock crashes / yield dropping is more likely to occur at the end of the year. So, I don't think this analysis really tells us much.

(I also see patrick013 made much the same point above)
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."

jalbert
Posts: 3426
Joined: Fri Apr 10, 2015 12:29 am

Re: In defense of short-term treasuries

Post by jalbert » Sat Jul 14, 2018 7:16 pm

danielc wrote:
Sat Jul 14, 2018 6:58 pm
jalbert wrote:
Sat Jul 14, 2018 6:41 pm
What probability distributions for future returns are being used to generate simulated data in the Monte Carlo simulations?
Historical values. PV's dataset seems to begin in 1972.
That is not a Monte Carlo simulation.
Index fund investor since 1987.

jalbert
Posts: 3426
Joined: Fri Apr 10, 2015 12:29 am

Re: In defense of short-term treasuries

Post by jalbert » Sat Jul 14, 2018 7:26 pm

You can examine the last 3 bear markets. They started in Oct 1987, March 2000, and October 2007 if you take the prior market peak as starting point of the downturn. In all three, intermediate treasuries helped you more than short-term treasuries.

Short-term treasuries are not a bad choice for a bond allocation in a portfolio. If our next source of market turmoil is inflationary or stagflationary, short-term treasuries might be better than intermediate treasuries, but not as effective as TIPS.
Index fund investor since 1987.

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: In defense of short-term treasuries

Post by SimpleGift » Sat Jul 14, 2018 9:09 pm

danielc wrote:
Sat Jul 14, 2018 4:28 pm
Common wisdom is that Intermediate Term (IT) treasuries are better for diversifying stocks than Short Term (ST) treasuries.
And the long-term history of portfolio risk/returns has confirmed the common wisdom. Years ago, William Bernstein looked at this question in a book and blog article (summarized here). A Forum member updated his analysis, using data for the 1926-2015 period to compare various mixes of 10-year Treasuries and T-bills with S&P stocks (chart below):
Historically, for any portfolio with more than about 30% stocks, 10-year Treasuries were the better choice than T-bills, with higher returns per unit of portfolio volatility over the years (the orange line above the blue line above.)
Cordially, Todd

User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Re: In defense of short-term treasuries

Post by danielc » Sat Jul 14, 2018 10:57 pm

triceratop wrote:
Sat Jul 14, 2018 7:02 pm
The max drawdowns are determined using monthly portfolio returns but rebalancing is only allowed to occur yearly. That is not consistent with the reasoning provided by those who use IT instead of ST. What is so special about the end of the calendar year?
Nobody said that the end of the calendar year is special. If we agree that the end of the calendar year is not special, then there's nothing wrong with rebalancing then. To test that, I used the backetst asset allocation feature to test the effect of rebalancing. Going from 1977 to the present day, here are the results:

Code: Select all

                            60% stock + 40% ST   |   60% stock + 40% IT
                            -------------------------------------------
Rebalance annually:                              |
            Return             9.42%             |      9.90% *
            Max.Drawdown     -29.75%             |    -27.98% *
-----------------------------------------------------------------------
Rebalance semi-annually:                         |
            Return             9.36%             |      9.83% *
            Max.Drawdown     -30.61%             |    -29.00% *
-----------------------------------------------------------------------
Rebalance quarterly:                             |
            Return             9.41%             |      9.90% *
            Max.Drawdown     -31.15%             |    -29.66% *

So more frequent rebalancing seems slightly harmful, and it does not seem to fundamentally change the results. Neither type of treasury seems to benefit from more frequent rebalancing.

triceratop wrote:
Sat Jul 14, 2018 7:02 pm
If the reasoning you are attempting to disprove is correct the end of the year only matters if stock crashes / yield dropping is more likely to occur at the end of the year.
I never said that rebalancing at the end of the year is somehow better than at some other date. I don't understand why you think that I said that.
triceratop wrote:
Sat Jul 14, 2018 7:02 pm
So, I don't think this analysis really tells us much.
Well, apparently you think that the analysis was talking about calendars. I seriously can't imagine why you thought that I was talking about calendars. The goal of the analysis was to see whether there was big difference between using ST or IT treasuries as your bond allocation (answer: difference looks very small), and if so which one seems superior (answer: ST seems superior more than half the time).

User avatar
triceratop
Moderator
Posts: 5469
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: In defense of short-term treasuries

Post by triceratop » Sat Jul 14, 2018 11:04 pm

danielc wrote:Well, apparently you think that the analysis was talking about calendars. I seriously can't imagine why you thought that I was talking about calendars.
Portfolio Visualizer -- Notes on Results wrote: * Maximum drawdown statistics are calculated from simulated monthly balances.

* The results assume annual rebalancing of portfolio assets at the end of each year.
That's why I think that. Even if you don't think you were talking about calendars, it is an assumption of PV.

I would recommend you look at the relative performance of ST/IT/LT Treasuries during the 2008 (and tech blowup for that matter) crisis to see how the various assets performed.
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."

User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Re: In defense of short-term treasuries

Post by danielc » Sat Jul 14, 2018 11:11 pm

jalbert wrote:
Sat Jul 14, 2018 7:26 pm
You can examine the last 3 bear markets. They started in Oct 1987, March 2000, and October 2007 if you take the prior market peak as starting point of the downturn. In all three, intermediate treasuries helped you more than short-term treasuries.
Absolutely. The results that I posted do show that a percentage of the time intermediate treasuries do better. Neither one was better all the time at everything. In addition, my first conclusion is that the choice is not important. The returns and drawdowns never seem to be very far between IT and ST.
jalbert wrote:
Sat Jul 14, 2018 7:26 pm
Short-term treasuries are not a bad choice for a bond allocation in a portfolio. If our next source of market turmoil is inflationary or stagflationary, short-term treasuries might be better than intermediate treasuries, but not as effective as TIPS.
+1

User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Re: In defense of short-term treasuries

Post by danielc » Sat Jul 14, 2018 11:24 pm

SimpleGift wrote:
Sat Jul 14, 2018 9:09 pm
And the long-term history of portfolio risk/returns has confirmed the common wisdom. Years ago, William Bernstein looked at this question in a book and blog article (summarized here). A Forum member updated his analysis, using data for the 1926-2015 period to compare various mixes of 10-year Treasuries and T-bills with S&P stocks (chart below):
Historically, for any portfolio with more than about 30% stocks, 10-year Treasuries were the better choice than T-bills, with higher returns per unit of portfolio volatility over the years (the orange line above the blue line above.)
As you probably know, the shape of that graph will look different depending on which period you are looking at. As jalbert pointed out, if the period includes a market crash, like the big one in 1929 or the recent ones, intermediates tend to do better, but during an inflationary period, like the one in the 70s, shorts tend to do better. This is the reason for doing a Monte Carlo simulation. There are many possible futures, and you don't know which future is yours. So what you want to do is simulate a distribution of possible futures and make choices based on that distribution.

User avatar
triceratop
Moderator
Posts: 5469
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: In defense of short-term treasuries

Post by triceratop » Sat Jul 14, 2018 11:29 pm

Historical returns, from beginning of 2008 bear market (Oct 9 2007):

Image

VFISX: ST Trs
VFITX: IT Trs
VUSTX: LT Trs
VTSAX: US TSM

The return spread demonstrated above is such that the max drawdown figures provided above are called into question.
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."

User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Re: In defense of short-term treasuries

Post by danielc » Sat Jul 14, 2018 11:36 pm

triceratop wrote:
Sat Jul 14, 2018 11:29 pm
Historical returns, from beginning of 2008 bear market (Oct 9 2007):

Image

VFISX: ST Trs
VFITX: IT Trs
VUSTX: LT Trs
VTSAX: US TSM

The return spread demonstrated above is such that the max drawdown figures provided above are called into question.
Good graphic. Clearly shows the value of longer-term during a market crash. That said, most people do not advocate LT treasuries as their bond allocation, despite their superior performance during a crash. Do you agree with that advice?

User avatar
triceratop
Moderator
Posts: 5469
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: In defense of short-term treasuries

Post by triceratop » Sat Jul 14, 2018 11:42 pm

I have seen good reasons to not use LT treasuries in general, as market prices for those assets might not be determined in a way that yields a well-compensated return for the risks one takes on. Such structural demand for LT nominals which depresses yields is well-discussed.

Anyway, my point was about IT treasuries, as yours originally was.
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: In defense of short-term treasuries

Post by SimpleGift » Sat Jul 14, 2018 11:50 pm

danielc wrote:
Sat Jul 14, 2018 11:24 pm
As you probably know, the shape of that graph will look different depending on which period you are looking at.
Right — but since we don't know the future, wouldn't it be preferable to make portfolio allocation decisions based on what has worked the best on average over ALL the past periods for which we have data (back to 1926)?

Your OP analysis is well done and impressive, but its conclusions rely on data for just the 1977-2017 period, I believe. Perhaps the next 40 years will look like the last 40 years — but personally I'd be more persuaded by the long-term average data we have for 1926-2017 (which favors intermediate bonds over short for portfolios with >30% stocks).
Cordially, Todd

User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Re: In defense of short-term treasuries

Post by danielc » Sat Jul 14, 2018 11:52 pm

triceratop wrote:
Sat Jul 14, 2018 11:42 pm
I have seen good reasons to not use LT treasuries in general, as market prices for those assets might not be determined in a way that yields a well-compensated return for the risks one takes on. Such structural demand for LT nominals which depresses yields is well-discussed.

Anyway, my point was about IT treasuries, as yours originally was.
I understand that. The underlying point of my comment is that the fact that IT and LT behave better during a crash clearly doesn't tell the full story. Like most people, you think that LT is not the best choice for most people despite their better behaviour during a crash. In a similar way, I suggest that perhaps IT is not necessarily better than ST for most people despite their better behaviour during a crash. That was also the original point of the Monte Carlo simulations. Clearly there were many simulated histories in which IT came out ahead. But ST came out ahead slightly more often, and in any case the choice didn't seem to matter very much anyway.

User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Re: In defense of short-term treasuries

Post by danielc » Sun Jul 15, 2018 12:06 am

SimpleGift wrote:
Sat Jul 14, 2018 11:50 pm
Right — but since we don't know the future, wouldn't it be preferable to make portfolio allocation decisions based on what has worked the best on average over ALL the past periods for which we have data (back to 1926)?
The problem with the plot is that it is an average. It does not show a distribution of results. What is the probability that IT / ST will be the wrong choice? What happens when you choose wrong? Those are the real questions I want to answer and simple cumulative return does not help with that.

What I would ideally like to do is repeat my original analysis with the longer dataset.
SimpleGift wrote:
Sat Jul 14, 2018 11:50 pm
Your OP analysis is well done and impressive, but its conclusions rely on data for just the 1977-2017 period, I believe.
Thank you. Yes, my understanding is that it starts at 1977.
SimpleGift wrote:
Sat Jul 14, 2018 11:50 pm
Perhaps the next 40 years will look like the last 40 years — but personally I'd be more persuaded by the long-term average data we have for 1926-2017 (which favors intermediate bonds over short for portfolios with >30% stocks).
The longer dataset certainly has valuable information. What my analysis contributes is a distribution of results. For example, it shows that we don't actually know whether short or intermediate will be better. The next 20 years are not going to be an average of the last 90.

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: In defense of short-term treasuries

Post by SimpleGift » Sun Jul 15, 2018 12:16 am

danielc wrote:
Sun Jul 15, 2018 12:06 am
What I would ideally like to do is repeat my original analysis with the longer dataset.
(snip)
The longer dataset certainly has valuable information. What my analysis contributes is a distribution of results.
:thumbsup I would think such an analysis would be a valuable contribution to better understanding the allocation question.
Cordially, Todd

jalbert
Posts: 3426
Joined: Fri Apr 10, 2015 12:29 am

Re: In defense of short-term treasuries

Post by jalbert » Sun Jul 15, 2018 1:20 am

Good graphic. Clearly shows the value of longer-term during a market crash. That said, most people do not advocate LT treasuries as their bond allocation, despite their superior performance during a crash. Do you agree with that advice?
Some people advocate for it for portfolios with high equity allocations. LT treasuries are volatile enough that when combined with stock they can become a significant source of portfolio return and variance. Thus, they may protect more during disinflationary or deflationary crashes but they expose the portfolio to substantial inflation risk.

My issue with LT treasuries is that the market is driven by insurance companies and pension funds. These institutions are matching LT treasuries to nominal liabilities so they don't need a yield that compensates them for inflation and term risk. I believe the yield of LT treasuries would be much higher if it discounted in a premium to cover these risks.
Index fund investor since 1987.

Park
Posts: 524
Joined: Sat Nov 06, 2010 4:56 pm

Re: In defense of short-term treasuries

Post by Park » Sun Jul 15, 2018 1:53 am

As for using data from 1926 onwards, there are problems with that too. Prices went down about 25% during the Depression. However, since a fiat currency system was started, deflation has become much less of an issue. But iflation has become more of an issue. Inflation will make ST treasuries look better relative to IT treasuries.

jalbert
Posts: 3426
Joined: Fri Apr 10, 2015 12:29 am

Re: In defense of short-term treasuries

Post by jalbert » Sun Jul 15, 2018 2:46 am

Paul Volcker demonstrated the tools the Fed has to fight inflation. It is unlikely an increase in inflation would be managed by the Fed like it was in the 1970’s.
Index fund investor since 1987.

User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Re: In defense of short-term treasuries

Post by danielc » Sun Jul 15, 2018 2:52 am

jalbert wrote:
Sun Jul 15, 2018 1:20 am
My issue with LT treasuries is that the market is driven by insurance companies and pension funds. These institutions are matching LT treasuries to nominal liabilities so they don't need a yield that compensates them for inflation and term risk. I believe the yield of LT treasuries would be much higher if it discounted in a premium to cover these risks.
As an aside... that's fascinating.

User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Re: In defense of short-term treasuries

Post by danielc » Sun Jul 15, 2018 2:58 am

jalbert wrote:
Sun Jul 15, 2018 2:46 am
Paul Volcker demonstrated the tools the Fed has to fight inflation. It is unlikely an increase in inflation would be managed by the Fed like it was in the 1970’s.
I agree. I actually expect central banks to do a good job at managing inflation. But with that said, the 2% target might be a bit optimistic. There is an article at the WaPo today that says that inflation is at a 6-year high of 2.9% and the new trade wars are going to be an upward pressure on prices.

User avatar
triceratop
Moderator
Posts: 5469
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: In defense of short-term treasuries

Post by triceratop » Sun Jul 15, 2018 10:22 am

danielc wrote:
Sun Jul 15, 2018 2:52 am
jalbert wrote:
Sun Jul 15, 2018 1:20 am
My issue with LT treasuries is that the market is driven by insurance companies and pension funds. These institutions are matching LT treasuries to nominal liabilities so they don't need a yield that compensates them for inflation and term risk. I believe the yield of LT treasuries would be much higher if it discounted in a premium to cover these risks.
As an aside... that's fascinating.
That is what I was referring to above as well. It's not true that the same reason applies to avoiding IT in favor of ST, so I didn't quite agree with your analogy.
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."

User avatar
Doc
Posts: 8377
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Re: In defense of short-term treasuries

Post by Doc » Sun Jul 15, 2018 10:47 am

triceratop wrote:
Sat Jul 14, 2018 11:29 pm
Historical returns, from beginning of 2008 bear market (Oct 9 2007):

Image

VFISX: ST Trs
VFITX: IT Trs
VUSTX: LT Trs
VTSAX: US TSM

The return spread demonstrated above is such that the max drawdown figures provided above are called into question.
triceratop's data as a 6 month rolling return chart instead of growth chart. Look at what happens when the stock market crashes.

Image

Go for the green! :mrgreen:
A scientist looks for THE answer to a problem, an engineer looks for AN answer and lawyers ONLY have opinions. Investing is not a science.

User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Re: In defense of short-term treasuries

Post by danielc » Sun Jul 15, 2018 1:43 pm

triceratop wrote:
Sun Jul 15, 2018 10:22 am
danielc wrote:
Sun Jul 15, 2018 2:52 am
jalbert wrote:
Sun Jul 15, 2018 1:20 am
My issue with LT treasuries is that the market is driven by insurance companies and pension funds. These institutions are matching LT treasuries to nominal liabilities so they don't need a yield that compensates them for inflation and term risk. I believe the yield of LT treasuries would be much higher if it discounted in a premium to cover these risks.
As an aside... that's fascinating.
That is what I was referring to above as well. It's not true that the same reason applies to avoiding IT in favor of ST, so I didn't quite agree with your analogy.
I have a question: How do we know that the LT treasury market is driven by insurance companies and pension funds? ... I mean... it makes intuitive sense that pensions and annuities would be especially interested in LT treasuries, but do we have any data about who is buying them?

I accept the idea that LT vs IT is a different comparison from IT vs ST. I didn't mean to suggest that they were the same. I was just giving a word of caution that superior performance during market crashes does not tell you the full story. I'm sure we actually agree on that.

jalbert
Posts: 3426
Joined: Fri Apr 10, 2015 12:29 am

Re: In defense of short-term treasuries

Post by jalbert » Sun Jul 15, 2018 2:19 pm

If your concern with IT treasuries is inflation, diversify them with TIPS and you can earn the term premium of intermediate bonds while taking less inflation risk than if just holding ST treasuries.

The main benefit of ST treasuries is liquidity.
Index fund investor since 1987.

User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Re: In defense of short-term treasuries

Post by danielc » Sun Jul 15, 2018 2:40 pm

danielc wrote:
Sun Jul 15, 2018 1:43 pm
I have a question: How do we know that the LT treasury market is driven by insurance companies and pension funds? ... I mean... it makes intuitive sense that pensions and annuities would be especially interested in LT treasuries, but do we have any data about who is buying them?
I found an article about this:

https://www.reuters.com/article/us-usa- ... SKBN1HW2NN

User avatar
Kevin M
Posts: 9887
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: In defense of short-term treasuries

Post by Kevin M » Sun Jul 15, 2018 4:24 pm

jalbert wrote:
Sun Jul 15, 2018 2:19 pm
If your concern with IT treasuries is inflation, diversify them with TIPS and you can earn the term premium of intermediate bonds while taking less inflation risk than if just holding ST treasuries.
What term premium? The current yield curve is very flat for intermediate-term maturities, both nominal and real. It could be argued that the term premium actually is negative:

What We Do and Don’t Know about the Term Premium.

Fed's Crushing of Term Premium Is Key to Powell's Yield Puzzle

You only get an additional 7 basis points for extending from 5-year to 7-year nominal maturity, and only 3 basis points for extending from 7-year to 10-year. You only get 4 basis points for extending from 5-year TIPS to 7-year TIPS, and only 2 basis points for extending from 7-year to 10-year.
The main benefit of ST treasuries is liquidity.
With the current yield curve, I'd say that the main benefit of ST Treasuries is that the yield curve is very steep out to 1-year maturity (50 basis points more than 1-month), and moderately steep to 2-year year maturity (22 basis points more than 1-year), so you get sufficiently rewarded for taking the additional term risk. You only get an additional 7 bps for extending from 2-year to 3-year maturity, and it gets worse after that.

Image

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

gmaynardkrebs
Posts: 810
Joined: Sun Feb 10, 2008 11:48 am

Re: In defense of short-term treasuries

Post by gmaynardkrebs » Sun Jul 15, 2018 4:33 pm

danielc wrote:
Sun Jul 15, 2018 2:40 pm
danielc wrote:
Sun Jul 15, 2018 1:43 pm
I have a question: How do we know that the LT treasury market is driven by insurance companies and pension funds? ... I mean... it makes intuitive sense that pensions and annuities would be especially interested in LT treasuries, but do we have any data about who is buying them?
I found an article about this:

https://www.reuters.com/article/us-usa- ... SKBN1HW2NN
I think it's more leveraged carry trade. The Fed has largely taken duration risk out of the equation. Why not play the spread?

User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Re: In defense of short-term treasuries

Post by danielc » Sun Jul 15, 2018 4:37 pm

gmaynardkrebs wrote:
Sun Jul 15, 2018 4:33 pm
I think it's more leveraged carry trade. The Fed has largely taken duration risk out of the equation. Why not play the spread?
In which way has the Fed taken out duration risk? If interest rates rise (as they have been recently), long-duration bonds will lose value.

User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Re: In defense of short-term treasuries

Post by danielc » Sun Jul 15, 2018 4:41 pm

jalbert wrote:
Sun Jul 15, 2018 2:19 pm
If your concern with IT treasuries is inflation, diversify them with TIPS and you can earn the term premium of intermediate bonds while taking less inflation risk than if just holding ST treasuries.

The main benefit of ST treasuries is liquidity.
It looks like my response got lost in the aether... Anyway. I'm not worried about short term inflation. For long-term inflation I plan to rely on equities. As Kevin noted, there doesn't seem to be much of a term premium for IT treasuries. I think that my analysis showed that the benefits of ST treasuries likely go far beyond liquidity. Based on the last 40 years of data, they have done better than IT at improving the risk-adjusted return of a portfolio more than half the time. It looks like the odds are slightly in favour of ST treasuries.

gmaynardkrebs
Posts: 810
Joined: Sun Feb 10, 2008 11:48 am

Re: In defense of short-term treasuries

Post by gmaynardkrebs » Sun Jul 15, 2018 4:56 pm

danielc wrote:
Sun Jul 15, 2018 4:37 pm
gmaynardkrebs wrote:
Sun Jul 15, 2018 4:33 pm
I think it's more leveraged carry trade. The Fed has largely taken duration risk out of the equation. Why not play the spread?
In which way has the Fed taken out duration risk? If interest rates rise (as they have been recently), long-duration bonds will lose value.
They've drastically reduced duration risk out by promising to telegraph future rate increases well in advance. The great risk to the carry trade isn't increased rates you know about, but "surprise, increased rates!"

User avatar
triceratop
Moderator
Posts: 5469
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: In defense of short-term treasuries

Post by triceratop » Sun Jul 15, 2018 4:57 pm

danielc wrote:
Sun Jul 15, 2018 4:41 pm
jalbert wrote:
Sun Jul 15, 2018 2:19 pm
If your concern with IT treasuries is inflation, diversify them with TIPS and you can earn the term premium of intermediate bonds while taking less inflation risk than if just holding ST treasuries.

The main benefit of ST treasuries is liquidity.
It looks like my response got lost in the aether... Anyway. I'm not worried about short term inflation. For long-term inflation I plan to rely on equities. As Kevin noted, there doesn't seem to be much of a term premium for IT treasuries. I think that my analysis showed that the benefits of ST treasuries likely go far beyond liquidity. Based on the last 40 years of data, they have done better than IT at improving the risk-adjusted return of a portfolio more than half the time. It looks like the odds are slightly in favour of ST treasuries.
There also wasn't a positive term yield premium in Summer 2007, right at the beginning of the bear market and before there was an immense payoff for LT/IT over ST treasuries.

Obviously we don't want to focus too much on fighting the last war and it is possible next time will be different, but I think you might have the sequence of events a bit reversed here.
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."

User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Re: In defense of short-term treasuries

Post by danielc » Sun Jul 15, 2018 5:44 pm

gmaynardkrebs wrote:
Sun Jul 15, 2018 4:56 pm
danielc wrote:
Sun Jul 15, 2018 4:37 pm
In which way has the Fed taken out duration risk? If interest rates rise (as they have been recently), long-duration bonds will lose value.
They've drastically reduced duration risk out by promising to telegraph future rate increases well in advance. The great risk to the carry trade isn't increased rates you know about, but "surprise, increased rates!"
They are not going to telegraph future increases 5 years in advance, so I see limited benefit for intermediate treasuries.

stlutz
Posts: 4608
Joined: Fri Jan 02, 2009 1:08 am

Re: In defense of short-term treasuries

Post by stlutz » Sun Jul 15, 2018 6:01 pm

It looks like my response got lost in the aether... Anyway. I'm not worried about short term inflation. For long-term inflation I plan to rely on equities. As Kevin noted, there doesn't seem to be much of a term premium for IT treasuries. I think that my analysis showed that the benefits of ST treasuries likely go far beyond liquidity. Based on the last 40 years of data, they have done better than IT at improving the risk-adjusted return of a portfolio more than half the time. It looks like the odds are slightly in favour of ST treasuries.
We've had a significant period where short-term bonds were the best. More recently we've had a significant period of time where long-term bonds were the best. One can try to guess which might work best in the future or one can try to split the difference in some way by, say, owning "intermediate" term bonds or bonds with a variety of maturities.

I don't know that your determination that short-term bonds have been better ~60% of the time logically leads to the conclusion that you should use 100% short-term bonds.

That said, there is nothing fundamentally wrong with just using short-term bonds either, so I'm just posting my comment as a suggestion for further consideration as opposed to a matter of high principle.

User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Re: In defense of short-term treasuries

Post by danielc » Sun Jul 15, 2018 6:11 pm

stlutz wrote:
Sun Jul 15, 2018 6:01 pm
I don't know that your determination that short-term bonds have been better ~60% of the time logically leads to the conclusion that you should use 100% short-term bonds.
I'm sure it doesn't lead to that conclusion. But another point of my post is that the difference seems very small in practice, so it probably won't matter much.
stlutz wrote:
Sun Jul 15, 2018 6:01 pm
That said, there is nothing fundamentally wrong with just using short-term bonds either, so I'm just posting my comment as a suggestion for further consideration as opposed to a matter of high principle.
:beer

User avatar
nisiprius
Advisory Board
Posts: 36199
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: In defense of short-term treasuries

Post by nisiprius » Sun Jul 15, 2018 6:14 pm

To me, bonds are just a a way to adjust risk and return of a stock portfolio to lower values that match my personal risk tolerance. I feel that:
  • Any "low correlation," diversification, MPT benefit of stocks and bonds playing off against each other, is only significant for long-term bonds.
  • But even for long-term bonds, it doesn't amount to much. Sure, in 2008-2009 long-term bonds went up, which seems exciting when stocks were going down. But bonds only went up a little and stocks went down a lot. You wouldn't have seen bonds offsetting the loss in stocks unless your portfolio was something like 15% bonds, 85% stocks.
  • And long-term bonds aren't suitable for an ordinary retail investor's portfolio because of high interest rate sensitivity and high inflation risk.
  • For intermediate-term bonds, any "low correlation" effect is negligible and shouldn't factor into the decision.
  • The extra return of intermediate-term bonds, compared to short-term bonds (or cash!) is large enough to care about.
  • The extra risk of intermediate-term bonds, compared to short-term bonds (or cash) is lost in the noise of stock risk if your portfolio has anything like a normal amount of stocks in ti.
Thus, for example, let's consider a 60/40 portfolio using 60% Vanguard 500 Index (which I'm using because it has a longer history than Total Stock).
Portfolio 1, blue uses 40% Vanguard Short-Term Treasury Fund, VFISX.
Portfolio 2, red, uses 40% Vanguard Intermediate-Term Treasury Fund, VFITX.

Source

Image

Going from short-term bonds to intermediate-term bonds would have increased CAGR from 7.75% to 8.47%, raising the final balance from $73,160 to $87,375. That seems worthwhile enough to do, given that the standard deviation only would have increased from 8.24% to 8.28%, which to me seems truly negligible.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
Kevin M
Posts: 9887
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: In defense of short-term treasuries

Post by Kevin M » Sun Jul 15, 2018 8:37 pm

nisiprius wrote:
Sun Jul 15, 2018 6:14 pm
[*]The extra return of intermediate-term bonds, compared to short-term bonds (or cash!) is large enough to care about.
Really? Don't you mean "has been" instead of "is"? And doesn't it depend on the period we look at? How do you know what the returns will be going forward over any particular time period?
Thus, for example, let's consider a 60/40 portfolio using 60% Vanguard 500 Index (which I'm using because it has a longer history than Total Stock).
Portfolio 1, blue uses 40% Vanguard Short-Term Treasury Fund, VFISX.
Portfolio 2, red, uses 40% Vanguard Intermediate-Term Treasury Fund, VFITX.
<snip>
Going from short-term bonds to intermediate-term bonds would have increased CAGR from 7.75% to 8.47%, raising the final balance from $73,160 to $87,375. That seems worthwhile enough to do, given that the standard deviation only would have increased from 8.24% to 8.28%, which to me seems truly negligible.
Sure--that's because you picked a period when bond yields generally declined, so intermediate-term did better than short term.

Image

How about we use the same tool, but use the asset class functionality instead, so we can go back to 1977, and look at 1977-1991 instead?

Portfolio 1 is 60% large cap, 40% short-term Treasuries, and Portfolio 2 is same but with intermediate-term Treasuries.

Image
Source

Negligible difference in total return, standard deviation difference is larger, the portfolio with short-term Treasuries was ahead much of the time, and if we cut a few months off of the end, the portfolio with the short-term Treasuries did slightly better.

We could look at earlier periods, when rates generally increased, where the portfolio with short-term Treasuries would have done even better. Even with the PV asset class tool, 1978-1981, when rates generally increased, is one such period we can look at.

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

jalbert
Posts: 3426
Joined: Fri Apr 10, 2015 12:29 am

Re: In defense of short-term treasuries

Post by jalbert » Sun Jul 15, 2018 10:48 pm

What term premium? The current yield curve is very flat for intermediate-term maturities, both nominal and real. It could be argued that the term premium actually is negative...
I would gladly hold a bond allocation that is 75% IT treasuries and 25% ST TIPS today before I would hold one that is 100% ST treasuries.
Last edited by jalbert on Sun Jul 15, 2018 10:56 pm, edited 1 time in total.
Index fund investor since 1987.

jalbert
Posts: 3426
Joined: Fri Apr 10, 2015 12:29 am

Re: In defense of short-term treasuries

Post by jalbert » Sun Jul 15, 2018 10:55 pm

gmaynardkrebs wrote:
Sun Jul 15, 2018 4:33 pm
danielc wrote:
Sun Jul 15, 2018 2:40 pm
danielc wrote:
Sun Jul 15, 2018 1:43 pm
I have a question: How do we know that the LT treasury market is driven by insurance companies and pension funds? ... I mean... it makes intuitive sense that pensions and annuities would be especially interested in LT treasuries, but do we have any data about who is buying them?
I found an article about this:

https://www.reuters.com/article/us-usa- ... SKBN1HW2NN
I think it's more leveraged carry trade. The Fed has largely taken duration risk out of the equation. Why not play the spread?
It is not recent or transient issue that large life insurance companies match the duration of their assets to the duration of their liabilities. It is how they do business.
Index fund investor since 1987.

gmaynardkrebs
Posts: 810
Joined: Sun Feb 10, 2008 11:48 am

Re: In defense of short-term treasuries

Post by gmaynardkrebs » Sun Jul 15, 2018 11:37 pm

jalbert wrote:
Sun Jul 15, 2018 10:55 pm
gmaynardkrebs wrote:
Sun Jul 15, 2018 4:33 pm
danielc wrote:
Sun Jul 15, 2018 2:40 pm
danielc wrote:
Sun Jul 15, 2018 1:43 pm
I have a question: How do we know that the LT treasury market is driven by insurance companies and pension funds? ... I mean... it makes intuitive sense that pensions and annuities would be especially interested in LT treasuries, but do we have any data about who is buying them?
I found an article about this:

https://www.reuters.com/article/us-usa- ... SKBN1HW2NN
I think it's more leveraged carry trade. The Fed has largely taken duration risk out of the equation. Why not play the spread?
It is not recent or transient issue that large life insurance companies match the duration of their assets to the duration of their liabilities. It is how they do business.
Which is why they are probably not major suppliers of marginal demand -- they have no choice but to buy. However, my impression has been that they are more involved in corporates than treasuries.

User avatar
Kevin M
Posts: 9887
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: In defense of short-term treasuries

Post by Kevin M » Mon Jul 16, 2018 2:16 pm

jalbert wrote:
Sun Jul 15, 2018 10:48 pm
What term premium? The current yield curve is very flat for intermediate-term maturities, both nominal and real. It could be argued that the term premium actually is negative...
I would gladly hold a bond allocation that is 75% IT treasuries and 25% ST TIPS today before I would hold one that is 100% ST treasuries.
That doesn't answer the question. Again, where is the term premium you said one could earn with intermediate term Treasuries, whether nominal or TIPS? According to the NY Fed estimates, term premia are negative at all maturities between one and ten years, and they decrease monotonically with increasing maturity: Treasury Term Premia

Kevin
Wiki ||.......|| Suggested format for Asking Portfolio Questions (edit original post)

jalbert
Posts: 3426
Joined: Fri Apr 10, 2015 12:29 am

Re: In defense of short-term treasuries

Post by jalbert » Mon Jul 16, 2018 2:36 pm

Well the yield curve has flattened in recent weeks. I don’t jump from IT to ST or ST to IT with every short-term move in rates on the yield curve. If you are a long-term investor, intermediate treasuries have delivered a premium over short-term treasuries.
Index fund investor since 1987.

User avatar
Doc
Posts: 8377
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Re: In defense of short-term treasuries

Post by Doc » Mon Jul 16, 2018 2:41 pm

Kevin M wrote:
Mon Jul 16, 2018 2:16 pm
That doesn't answer the question. Again, where is the term premium you said one could earn with intermediate term Treasuries, whether nominal or TIPS? According to the NY Fed estimates, term premia are negative at all maturities between one and ten years, and they decrease monotonically with increasing maturity: Treasury Term Premia
I don't understand exactly what "Term Premia" means. But I think the general gist is the premium you get from rolling a one to buying a ten and holding to maturity.
Treasury term premia estimates for maturities from one to ten years from 1961 to the present.
If that's correct I don't see how that is very relevant to comparing a note short term (~3 years) to an intermediate term (~7.5 year) note.

Maybe this is all above my paygrade. :?
What is the term premium?

Briefly stated, the term premium is the excess yield that investors require to commit to holding a long-term bond instead of a series of shorter-term bonds. For example, suppose that the interest rate on the 10-year U.S. Treasury note is about 5.5%, and suppose that the interest rate on the 1-year U.S. Treasury bill is expected to average about 5% over the next 10 years (“note” and “bill” are the customary names for U.S. Treasury securities of these maturities). Then the term premium on the 10-year U.S. Treasury note would be about 0.5%, or 50 basis points.

Thus, a key component of the term premium is investor expectations about the future course of short-term interest rates over the lifetime of the long-term bond. In the example above, the term premium on the 10-year Treasury note depends crucially on financial market expectations about the course of shorter-term U.S. interest rates over the next ten years, a very long horizon. This foreshadows some of the difficulties of measuring the term premium that we will encounter below.
https://www.frbsf.org/economic-research ... m-premium/
A scientist looks for THE answer to a problem, an engineer looks for AN answer and lawyers ONLY have opinions. Investing is not a science.

User avatar
danielc
Posts: 377
Joined: Sun Dec 10, 2017 4:48 am

Re: In defense of short-term treasuries

Post by danielc » Mon Jul 16, 2018 3:25 pm

jalbert wrote:
Mon Jul 16, 2018 2:36 pm
Well the yield curve has flattened in recent weeks. I don’t jump from IT to ST or ST to IT with every short-term move in rates on the yield curve. If you are a long-term investor, intermediate treasuries have delivered a premium over short-term treasuries.
The goal of the original post was to show that that has not been the case more than 50% of the time in the last 40 years.

Post Reply