Matter/scatter, part 7: Bonds during the financial crisis

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
Topic Author
nisiprius
Advisory Board
Posts: 42242
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

Matter/scatter, part 7: Bonds during the financial crisis

Post by nisiprius »

Part 1: Introduction, Part 2: Bond fund choices,, Part 3, International bonds, Part 4: Bond allocation. Part 5: International stocks Part 6: Expense ratios Brief notes: 1) In these Monte Carlo simulations, the return for a given month is varied by randomly choosing the actual historical return or the actual historic return for an adjacent month. 2) They show the final value from $100 monthly contributions made over the whole time period. 4) Green and red crosses mark actual historical values. 5) Green and red bars show the 10% percentile, median, and 90% percentile of the range of outcomes. 5) The actual data source is PortfolioVisualizer.com, Backtest Portfolio, Monthly Returns; this is used as calculation input; and no PV content or numeric values are directly reproduced.

This is an exploration in response to comment by Doc in a PM. In my earlier "part 4" post, it didn't look to me as if the choice of bond fund mattered an awful lot: a small difference riding on top of the huge variability of overall portfolio returns. Doc remarked that "The difference between the covariance of you bond types and equities at times of stock market stress are not random events that are conducive to being analyzed by scatter charts." This is true, and my scatter charts are based on simulations that can be criticized (and will be when I get around to posting more details).

However, my charts are not (directly) based on "covariances" or other long-term descriptive statistics, but on a random choice of monthly returns that are closely linked in time to the historical returns. That is, the Monte Carlo simulation, during a period of stock market stress, is based on returns drawn from that specific period of stock market stress.

In any event, let's look at the performance of intermediate-term Treasuries and intermediate-term corporate bonds during the financial crisis. I decided to look at the three-year period 2007-2009 inclusive. For Treasuries, I used VFIUX (blue). Vanguard doesn't have a pure investment-grade corporate bond fund that goes back to 2007, so I used LQD, the iShares $ iBoxx Investment-Grade Corporate Bond Fund (orange) As you can see, the corporate-heavy Vanguard Intermediate-Term Investment-Grade Bond Fund, VFIDX, was almost identical. In any case, during 2007-2009, the Treasury fund got a small upward kick, presumably the "flight to safety," while the corporate bond funds took a nasty knock. At one point, on a $10,000 investment, VFIUX is up almost $1,500 while the corporate bond funds are down almost $1,000... a 25% difference. So, certainly, if your portfolio had been 100% VFIUX or LQD, and you had been forced to sell it at the worst time, the difference mattered.
Source

Image

Now, keep that in mind. It happened, it is one aspect of reality, it might influence a sane investor's choice. What I'm about to show is a view from another angle, not a contradiction.

In this scatter chart, I am breaking my rule of using all available data to focus on a very short three-year period when I expected the difference to have mattered as much as it ever would. I also deliberately picked an endpoint of year-end 2010, so that I know in advance that corporates had not yet recovered and were still lower than Treasuries; that is, I know that the swarm of red dots will be lower than the green dots.

The comparison is between 60/40 balanced portfolios of 60% Vanguard Total Stock, and 40% of either VFIUX or LQD.

Image

What's happening here? Why is the visual difference so small in this comparison, when it was so large in the "straight" Morningstar chart? I think that three things are going on.

1) We are looking at the three-year time period 2007-2009 inclusive. Even over such a short time period, both stocks and corporate bonds had time to recover, and had recovered almost completely. The difference between Treasuries and corporates would have mattered much more to an investor who had to sell at the bottom, than they to an investor able to hang on through the end of 2009.

2) We are looking at a 60/40 portfolio, not bonds in isolation, and we are actually looking mostly at the huge variability of stocks. The difference between the bond funds rides on top of the stock market crash and is small by comparison.

3) We are not looking at the growth of $10,000. We are looking at the total final outcome of an investment of $100 per month. Stock and bond purchases were made at all points, some near the bottom as well as some near the top.

The portfolio with corporate bonds had volatility, shown both in the red swarm appearing to the right of the green swarm, and also in the wider spread of final outcomes.

If we let the comparison run for four years instead of three, over that time period the portfolio with corporate bonds actually did slightly better--and spread out the outcomes so that the best outcome was better, but the worst outcome was worse, than the portfolio with treasuries.

Image
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
Doc
Posts: 9850
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Re: Matter/scatter, part 7: Bonds during the financial crisis

Post by Doc »

nisiprius wrote: Fri Aug 03, 2018 11:35 am In any case, during 2007-2009, the Treasury fund got a small upward kick, presumably the "flight to safety," while the corporate bond funds took a nasty knock. At one point, on a $10,000 investment, VFIUX is up almost $1,500 while the corporate bond funds are down almost $1,000... a 25% difference. So, certainly, if your portfolio had been 100% VFIUX or LQD, and you had been forced to sell it at the worst time, the difference mattered.
Restated from a different point of view: So, certainly, if your <bond> portfolio had been 100% VFIUX or LQD, and you had been forced wanted to sell it at the worst best time, the difference mattered.

It's the short term negative covariance that often occurs when equities tank that is difficult to evaluate with a monte carlo or a total return type chart. There is little doubt that Treasuries perform better in that type situation but it is enough better to make up for the lower return in more normal times? And perhaps moreover for many if the S&P drops by 30 or 40 percent are you going to be mentally able to sell the high flying Treasuries and buy the tanking stock?

I'm buying equities on weakness but many people are doing the opposite or just doing nothing i.e. "staying the course".

It would be nice to quantify the choices but I don't know how myself.

In any case good work Nisiprius.

Aside: Nisi prius is sometimes thought about as "that which has already been decided" or something to that effect. Unfortunately in investing its application is less useful than it is in a court of law. :(
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
CULater
Posts: 2832
Joined: Sun Nov 13, 2016 10:59 am
Location: Hic sunt dracones

Re: Matter/scatter, part 7: Bonds during the financial crisis

Post by CULater »

Are commodities up next? I've noticed with these also that the downside negative correlation benefit with equities is often short-term. If you wait a bit, the diversification benefit disappears just as it appeared to do with bonds. Time horizon is an often-neglected aspect when considering asset correlations.
On the internet, nobody knows you're a dog.
User avatar
ram
Posts: 1698
Joined: Tue Jan 01, 2008 10:47 pm
Location: Midwest

Re: Matter/scatter, part 7: Bonds during the financial crisis

Post by ram »

So nisiprius what are your thoughts on the following strategy:
1. For those that are >5 years away from retirement: Continue investing in corporate bonds. Even if they fall they will recover in the next 5 years.

2. For retired people and those nearing retirement keep bonds worth 1 to 2 years of expenses in treasuries and the rest in corporate bonds.
Ram
RetiredAL
Posts: 1070
Joined: Tue Jun 06, 2017 12:09 am
Location: SF Bay Area

Re: Matter/scatter, part 7: Bonds during the financial crisis

Post by RetiredAL »

ram wrote: Sat Aug 04, 2018 5:19 pm So nisiprius what are your thoughts on the following strategy:
1. For those that are >5 years away from retirement: Continue investing in corporate bonds. Even if they fall they will recover in the next 5 years.

2. For retired people and those nearing retirement keep bonds worth 1 to 2 years of expenses in treasuries and the rest in corporate bonds.

+1

For me, my monthly withdrawal $ comes from my S/T Treasuries Fund, thus I could hardly care what either my longer bonds or the stock market are doing each month.
User avatar
Doc
Posts: 9850
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Re: Matter/scatter, part 7: Bonds during the financial crisis

Post by Doc »

ram wrote: Sat Aug 04, 2018 5:19 pm So nisiprius what are your thoughts on the following strategy:
1. For those that are >5 years away from retirement: Continue investing in corporate bonds. Even if they fall they will recover in the next 5 years.

2. For retired people and those nearing retirement keep bonds worth 1 to 2 years of expenses in treasuries and the rest in corporate bonds.
1. The flip side of this is that means you don't rebalance if stocks tank.

2. Use the flip side here as well. If you're saying 1 to 2 years what significance is there that your short corporate fund declines by 10 %. That's a whole months cash flow. My reserve is only 22 months instead of 24.
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
ram
Posts: 1698
Joined: Tue Jan 01, 2008 10:47 pm
Location: Midwest

Re: Matter/scatter, part 7: Bonds during the financial crisis

Post by ram »

Doc wrote: Sat Aug 04, 2018 5:59 pm
ram wrote: Sat Aug 04, 2018 5:19 pm So nisiprius what are your thoughts on the following strategy:
1. For those that are >5 years away from retirement: Continue investing in corporate bonds. Even if they fall they will recover in the next 5 years.

2. For retired people and those nearing retirement keep bonds worth 1 to 2 years of expenses in treasuries and the rest in corporate bonds.
1. The flip side of this is that means you don't rebalance if stocks tank.

2. Use the flip side here as well. If you're saying 1 to 2 years what significance is there that your short corporate fund declines by 10 %. That's a whole months cash flow. My reserve is only 22 months instead of 24.
Doc,
Look at the data shown by Nisiprius. Lets assume annual expenses are 60K. 2 yrs of annual expenses are in Treasuries = 120K. During bad times like 2008 this will become 130 to 140 K. DO NOT spend the extra 10- 20K. During the 3rd year use it to supplement the draw down from the still recovering corporates. By 4th year the recovery of corporates is complete and you are back to square one. For all I know one may be slightly ahead.

I think the ratio of stocks to bonds is a separate issue. The re balancing will be done using the corporate bonds. These may be 'only' 20% down whereas the stocks are 50% down. Alternately the 10% gains in Treasuries can be utilized to buy stocks. (I do get it that the 10% increase can be used to supplement the loss from corporates OR used for buying stocks and the choice will depend on how aggressive you feel at that time.)
Ram
User avatar
Doc
Posts: 9850
Joined: Sat Feb 24, 2007 1:10 pm
Location: Two left turns from Larry

Re: Matter/scatter, part 7: Bonds during the financial crisis

Post by Doc »

ram wrote: Sun Aug 05, 2018 11:24 am By 4th year the recovery of corporates is complete and you are back to square one.
You are missing a major point in the Treasury/Corporate trade off analysis. That is what does one do when the stock market drops by say 40%. Do you use your fixed income to buy equities to maintain your AA or do you just use your fixed income to meet ongoing expenses. If you are going to rebalance during the crisis you will be better off with Treasuries. Holding Treasuries instead of corporates is going to cost you something if that stock market crash doesn't happen or if you don't have the willingness to buy equities when they are tanking. What I was trying to point out to you was that a loss of a few months of your reserve capacity is nothing compared to this scenario.

To summarize: Don't make the Treasury/Corporate decision based on assumptions that exclude the equity portion of your portfolio.

I sold my entire Treasury ladder in '08 to rebalance. If I had to sell corporates I would at minimum have had a bigger loss and maybe even been unable to rebalance completely. As you noted by the 4th year of recovery I was able to establish my desired fixed income portfolio to the original Treasury/corporate mix. (Actually sooner but that's not the issue.)
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.
rgs92
Posts: 2812
Joined: Mon Mar 02, 2009 8:00 pm

Re: Matter/scatter, part 7: Bonds during the financial crisis

Post by rgs92 »

Awesome post Nisprius. Thanks!
Post Reply