Yield Curve Inversions — Any Benefit from Market Timing?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Yield Curve Inversions — Any Benefit from Market Timing?

Post by SimpleGift » Thu Jul 12, 2018 12:26 pm

Two weeks ago, a Forum member asked for some historical research to test his intriguing question: In the past, has an investor with a 70/30 portfolio as their standard asset allocation, who for an 18 month duration shifts their portfolio to 30/70 whenever the yield curve inverts, ended up with a better return than one who stuck with the 70/30 portfolio?

Since June 1976 (when the FRED inversion data starts) to December 2016, we have 40+ years that include five distinct yield curve inversion/recession events (including the double-dip inversions of the early 1980s, chart below):
Our test below uses S&P 500 and 10-year Treasury monthly returns, inflation-adjusted with dividends reinvested. Portfolio switches were made in the month following each yield curve inversion, and then held for 18 months. The full period results:
  • 70/30 Constant Portfolio............6.73% CAGR
    70/30 - 30/70 Switching............6.80% CAGR

    Image
    Note: Chart shows 2-year rolling monthly returns.
    Data sources: S&P stock returns from Shiller; 10-year Treasury returns from Medium
Discussion: Honestly, I thought the benefits would be greater than just 0.07% CAGR. Looking over the chart, it appears switching helped a few times, it hurt a few times, and otherwise didn't matter. Overall, it hardly seemed worth the effort.

Your thoughts?
Last edited by SimpleGift on Thu Jul 12, 2018 2:55 pm, edited 2 times in total.
Cordially, Todd

jminv
Posts: 576
Joined: Tue Jan 02, 2018 10:58 pm

Re: Yield Curve Inversions — Any Benefit to Market Timing?

Post by jminv » Thu Jul 12, 2018 12:39 pm

One factor you could look at is the duration of the bonds you switch to such as intermediate term treasuries.

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

Re: Yield Curve Inversions — Any Benefit to Market Timing?

Post by SimpleGift » Thu Jul 12, 2018 12:53 pm

jminv wrote:
Thu Jul 12, 2018 12:39 pm
One factor you could look at is the duration of the bonds you switch to such as intermediate term treasuries.
Possibly. Unfortunately, I don't have monthly total return data (price change plus yield) for, say, 5-year Treasuries. Also, I'm not entirely convinced it would make a great deal of difference in the results over the full 40+ year time period.
Cordially, Todd

User avatar
alpine_boglehead
Posts: 236
Joined: Fri Feb 17, 2017 9:51 am
Location: Austria

Re: Yield Curve Inversions — Any Benefit to Market Timing?

Post by alpine_boglehead » Thu Jul 12, 2018 2:56 pm

SimpleGift wrote:
Thu Jul 12, 2018 12:26 pm

Overall, it hardly seemed worth the effort.

Your thoughts?
In an ideal world, the result is a tie. With friction costs, it probably is not. If anyone would have performed the switching in a taxable account or at least some relevant trading costs (spreads, commissions, or, god forbid, front-loaded funds), the result would have been worse than just holding 70/30.

Thanks for the very interesting analyses you keep posting.

User avatar
Phineas J. Whoopee
Posts: 7566
Joined: Sun Dec 18, 2011 6:18 pm

Re: Yield Curve Inversions — Any Benefit from Market Timing?

Post by Phineas J. Whoopee » Thu Jul 12, 2018 3:38 pm

Reuters published an article about other possible indicators earlier today: Goodbye inverted yield curve? Fed looks for alternative signals to guide policy.

Please don't fail to take into account the intentionally-humorous Betteridge's law of headlines.

PJW

Pelerus
Posts: 31
Joined: Wed Mar 14, 2018 11:21 am

Re: Yield Curve Inversions — Any Benefit from Market Timing?

Post by Pelerus » Thu Jul 12, 2018 4:17 pm

SimpleGift wrote:
Thu Jul 12, 2018 12:26 pm
In the past, has an investor with a 70/30 portfolio as their standard asset allocation, who for an 18 month duration shifts their portfolio to 30/70 whenever the yield curve inverts, ended up with a better return than one who stuck with the 70/30 portfolio?
Why is this hypothetical investor switching to 30/70 as soon as the yield curve inverts? He is jumping the gun: yield curve inversion has always preceded recession by at least 6 months, and sometimes as long as 24 months. In the 24 month case, he would have already moved back into equities before the recession even began. I would suggest re-running the analysis with a lag of at least 6 months between inversion and allocation adjustment.

In practice, relying on an economic indicator to make investment decisions without pairing it with a momentum indicator is like flying on one wing, and falls victim to the old “being early is the same as being wrong” axiom.

As a future project, I would be interested in seeing the yield curve inversion indicator paired with a moving average like the 200 day SMA, in the vein of Philosophical Economics: http://www.philosophicaleconomics.com/2016/02/uetrend/ .

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

Re: Yield Curve Inversions — Any Benefit from Market Timing?

Post by SimpleGift » Thu Jul 12, 2018 4:51 pm

Pelerus wrote:
Thu Jul 12, 2018 4:17 pm
Why is this hypothetical investor switching to 30/70 as soon as the yield curve inverts? He is jumping the gun: yield curve inversion has always preceded recession by at least 6 months, and sometimes as long as 24 months.
These were the parameters specified by the Forum member in his research request. In this case, portfolio switching was done in the month following the yield curve inversion. Just eyeballing the chart in the OP, it doesn't appear that further extending the time interval between the inversions and the portfolio switching would materially affect the full period results.

And to be truthful, I've not enough interest in this topic to run a bunch more what-if scenarios. I've yet to see a market timing scheme that outperformed consistently over long periods (though others may be believers). Personally, I've been more than satisfied with the rewards of buy-hold-rebalance investing.
Last edited by SimpleGift on Thu Jul 12, 2018 5:29 pm, edited 1 time in total.
Cordially, Todd

balbrec2
Posts: 108
Joined: Mon Nov 13, 2017 3:03 pm

Re: Yield Curve Inversions — Any Benefit from Market Timing?

Post by balbrec2 » Thu Jul 12, 2018 5:24 pm

SimpleGift wrote:
Thu Jul 12, 2018 12:26 pm
Two weeks ago, a Forum member asked for some historical research to test his intriguing question: In the past, has an investor with a 70/30 portfolio as their standard asset allocation, who for an 18 month duration shifts their portfolio to 30/70 whenever the yield curve inverts, ended up with a better return than one who stuck with the 70/30 portfolio?

Since June 1976 (when the FRED inversion data starts) to December 2016, we have 40+ years that include five distinct yield curve inversion/recession events (including the double-dip inversions of the early 1980s, chart below):
Our test below uses S&P 500 and 10-year Treasury monthly returns, inflation-adjusted with dividends reinvested. Portfolio switches were made in the month following each yield curve inversion, and then held for 18 months. The full period results:
  • 70/30 Constant Portfolio............6.73% CAGR
    70/30 - 30/70 Switching............6.80% CAGR

    Image
    Note: Chart shows 2-year rolling monthly returns.
    Data sources: S&P stock returns from Shiller; 10-year Treasury returns from Medium
Discussion: Honestly, I thought the benefits would be greater than just 0.07% CAGR. Looking over the chart, it appears switching helped a few times, it hurt a few times, and otherwise didn't matter. Overall, it hardly seemed worth the effort.

Your thoughts?
In short, no you cant time the market using the yield curve.
Yes it has a high success rate of predicting recessions and possibly bear mkts
but the lag time is so variable as to render this method useless.
You'd be better off just using moving averages of the markets themselves.
That's if, you really think you'll gain any significant advantage from such
an endeavor. Good luck !

AlohaJoe
Posts: 3824
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Yield Curve Inversions — Any Benefit to Market Timing?

Post by AlohaJoe » Thu Jul 12, 2018 7:18 pm

SimpleGift wrote:
Thu Jul 12, 2018 12:53 pm
jminv wrote:
Thu Jul 12, 2018 12:39 pm
One factor you could look at is the duration of the bonds you switch to such as intermediate term treasuries.
Possibly. Unfortunately, I don't have monthly total return data (price change plus yield) for, say, 5-year Treasuries. Also, I'm not entirely convinced it would make a great deal of difference in the results over the full 40+ year time period.
I'm also not convinced, especially over just 18 months -- people often think changing bond duration has magic prophylactic powers -- but if you really wanted to try again with monthly Treasury data you can look at https://docs.google.com/spreadsheets/d/ ... 2025919577 Under the "Monthly raw data" tab and use the "Barclays 1 5" column. That's the total return data from the Bloomberg Barclays Government 1-5 Year index, which goes back to 1976.

KlangFool
Posts: 10394
Joined: Sat Oct 11, 2008 12:35 pm

Re: Yield Curve Inversions — Any Benefit from Market Timing?

Post by KlangFool » Thu Jul 12, 2018 8:02 pm

SimpleGift wrote:
Thu Jul 12, 2018 4:51 pm

I've yet to see a market timing scheme that outperformed consistently over long periods.
SimpleGift,

You had seen it. It is right in front of you. It is a fixed AA with buy, holds, and rebalancing. With a fixed AA and rebalancing, you always "buy low and sell high". It is the best market timing tool by not timing the market.

KlangFool

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

Re: Yield Curve Inversions — Any Benefit to Market Timing?

Post by SimpleGift » Thu Jul 12, 2018 8:03 pm

AlohaJoe wrote:
Thu Jul 12, 2018 7:18 pm
...but if you really wanted to try again with monthly Treasury data you can look at https://docs.google.com/spreadsheets/d/ ... 2025919577 Under the "Monthly raw data" tab and use the "Barclays 1 5" column. That's the total return data from the Bloomberg Barclays Government 1-5 Year index, which goes back to 1976.
Appreciate your sharing the data, AlohaJoe. I'm not motivated to do more what-if scenarios about market timing the yield curve — but your monthly Barclays 1-5 total returns could well come in handy for future projects. Thank you.
Cordially, Todd

AlohaJoe
Posts: 3824
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Yield Curve Inversions — Any Benefit from Market Timing?

Post by AlohaJoe » Thu Jul 12, 2018 8:11 pm

KlangFool wrote:
Thu Jul 12, 2018 8:02 pm
SimpleGift wrote:
Thu Jul 12, 2018 4:51 pm

I've yet to see a market timing scheme that outperformed consistently over long periods.
You had seen it. It is right in front of you. It is a fixed AA with buy, holds, and rebalancing. With a fixed AA and rebalancing, you always "buy low and sell high". It is the best market timing tool by not timing the market.
Rebalancing doesn't "consistently outperform". On the contrary, it usually underperforms by definition.

AlohaJoe
Posts: 3824
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Yield Curve Inversions — Any Benefit to Market Timing?

Post by AlohaJoe » Thu Jul 12, 2018 8:14 pm

SimpleGift wrote:
Thu Jul 12, 2018 8:03 pm
AlohaJoe wrote:
Thu Jul 12, 2018 7:18 pm
...but if you really wanted to try again with monthly Treasury data you can look at https://docs.google.com/spreadsheets/d/ ... 2025919577 Under the "Monthly raw data" tab and use the "Barclays 1 5" column. That's the total return data from the Bloomberg Barclays Government 1-5 Year index, which goes back to 1976.
Appreciate your sharing the data, AlohaJoe. I'm not motivated to do more what-if scenarios about market timing the yield curve — but your monthly Barclays 1-5 total returns could well come in handy for future projects. Thank you.
I don't blame you one bit :D

User avatar
Phineas J. Whoopee
Posts: 7566
Joined: Sun Dec 18, 2011 6:18 pm

Re: Yield Curve Inversions — Any Benefit from Market Timing?

Post by Phineas J. Whoopee » Thu Jul 12, 2018 8:15 pm

AlohaJoe wrote:
Thu Jul 12, 2018 8:11 pm
KlangFool wrote:
Thu Jul 12, 2018 8:02 pm
SimpleGift wrote:
Thu Jul 12, 2018 4:51 pm

I've yet to see a market timing scheme that outperformed consistently over long periods.
You had seen it. It is right in front of you. It is a fixed AA with buy, holds, and rebalancing. With a fixed AA and rebalancing, you always "buy low and sell high". It is the best market timing tool by not timing the market.
Rebalancing doesn't "consistently outperform". On the contrary, it usually underperforms by definition.
Absolutely correct. Rebalancing consistently, not each and every time but most often, moves resources from higher-performing asset classes to lower-performing ones.

Rebalancing is a risk-management tool, not a return-maximizing one.

PJW

KlangFool
Posts: 10394
Joined: Sat Oct 11, 2008 12:35 pm

Re: Yield Curve Inversions — Any Benefit from Market Timing?

Post by KlangFool » Thu Jul 12, 2018 8:23 pm

AlohaJoe wrote:
Thu Jul 12, 2018 8:11 pm
KlangFool wrote:
Thu Jul 12, 2018 8:02 pm
SimpleGift wrote:
Thu Jul 12, 2018 4:51 pm

I've yet to see a market timing scheme that outperformed consistently over long periods.
You had seen it. It is right in front of you. It is a fixed AA with buy, holds, and rebalancing. With a fixed AA and rebalancing, you always "buy low and sell high". It is the best market timing tool by not timing the market.
Rebalancing doesn't "consistently outperform". On the contrary, it usually underperforms by definition.
AlohaJoe,

Underperform what?

KlangFool

AlohaJoe
Posts: 3824
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Yield Curve Inversions — Any Benefit from Market Timing?

Post by AlohaJoe » Thu Jul 12, 2018 8:26 pm

KlangFool wrote:
Thu Jul 12, 2018 8:23 pm
AlohaJoe wrote:
Thu Jul 12, 2018 8:11 pm
KlangFool wrote:
Thu Jul 12, 2018 8:02 pm
SimpleGift wrote:
Thu Jul 12, 2018 4:51 pm

I've yet to see a market timing scheme that outperformed consistently over long periods.
You had seen it. It is right in front of you. It is a fixed AA with buy, holds, and rebalancing. With a fixed AA and rebalancing, you always "buy low and sell high". It is the best market timing tool by not timing the market.
Rebalancing doesn't "consistently outperform". On the contrary, it usually underperforms by definition.
AlohaJoe,

Underperform what?
Underperform not rebalancing ever.

KlangFool
Posts: 10394
Joined: Sat Oct 11, 2008 12:35 pm

Re: Yield Curve Inversions — Any Benefit from Market Timing?

Post by KlangFool » Fri Jul 13, 2018 7:40 am

AlohaJoe wrote:
Thu Jul 12, 2018 8:26 pm
KlangFool wrote:
Thu Jul 12, 2018 8:23 pm
AlohaJoe wrote:
Thu Jul 12, 2018 8:11 pm
KlangFool wrote:
Thu Jul 12, 2018 8:02 pm
SimpleGift wrote:
Thu Jul 12, 2018 4:51 pm

I've yet to see a market timing scheme that outperformed consistently over long periods.
You had seen it. It is right in front of you. It is a fixed AA with buy, holds, and rebalancing. With a fixed AA and rebalancing, you always "buy low and sell high". It is the best market timing tool by not timing the market.
Rebalancing doesn't "consistently outperform". On the contrary, it usually underperforms by definition.
AlohaJoe,

Underperform what?
Underperform not rebalancing ever.
Okay. My mistake. Thanks.

KlangFool

Post Reply