Yield Curve Inversion = Here's The Truth
Read this article this morning. It's enlightening, however, let's look at the facts:
medium.com/@timothy.chang/this-graph-has-predicted-every-recession-since-1976-what-its-showing-now-56b1c7ffd004
Premises of the article is that when the 10y interest rate < 2y interest rate (yield inversion) that a recession is immanent.
1. I used FRED data T10Y2Y and Wikipedia for recession dates. I know the definition of recession is controversial, but I chose this source.
en.wikipedia.org/wiki/List_of_recessions_in_the_United_States
2. The DJIA (orange) is overlaying the white T10Y2Y curve as a reference for the market
Initial Observations:
1. My first observation is that the curve has recently crossed 0 (inverted) so recession must be "immanent". Hmm... Wikipedia
shows us being in a recession since Feb 2020 (again defined by negative GDP growth). So, the author's definition of a recession does not
match Wiki's.
2. I'm going to give the author the benefit of the doubt and use their definition of recession (though not stated). There is a very narrow line shown
in the article just after the yield inversion on 8/26/2019 that represents a very short recession and explains #1.
Assuming the author's start date of the recession is around Feb '20, there are some key concerns.
a. The inversion happened ~ 160 days before the start of the recession. That's a pretty substantial lag.
b. And, it went positive 3 days later.
This means that I'm supposed to believe that an event 160 days ago, that lasted a whole 3 days before completely turning around, predicted a recession?
My Study:
I don't know if you'll be able to scroll all the way back through the data where I lay out all of the Wiki recessions versus when the initial yield curve inversion
occurred. If you can, the just use the chart to form your own opinion. Otherwise here is some tabulated data.
1. Recession >> Jan '80 - Jul '80 (6 months total)
Date of inversion - 17 Aug '78 >> 508 days before the recession (almost 2 years)
DJIA - from 17 Aug '78 to end of recession DJIA ranged ~840 +-4%
*note: The yield curve crossed back above 0 close to the end of the recession so maybe it predicts the end of recessions as well
2. Recession >> July '81 - Nov '82 ( 1 yr and 4 months)
Date of inversion - 10 Sep '80 >> 299 days before recession
DJIA - from 10 Sep '80 to the end of the recession DJIA ranged 920 +- 8%
*note: The yield curve crossed above 0 multiple times during this recession. The second time did mark the end of the recession.
3. Recession >> Jul '90 - Mar '91 (8 months total)
Date of inversion - 13 Dec '88 >> 562 days before recession
DJIA - from 13 Dec '88 to the end of the recession DJIA ranged 2100 +900 -0 (never dropped from the start of the inversion)
*note: The yield curve actually went positive 85d before the start of the recession.
4. Recession >> Mar '01 to Nov '01 (8 months total)
Date of inversion - 27 Jan '00 >> 398 days before recession
DJIA - from 27 Jan '00 to end of recession DJIA ranged 11,000 +300 - 1800
*note: The yield curve actually went positive 66d before the start of the recession.
5. Recession >> Dec '07 to Jun '09 (1 yr and 6 months total)
Date of inversion - 1 Jan '06 >> 671 days before recession
DJIA - from 1 Jan '06 to the end of the recession DJIA ranged 10,800 +3300 -4200 (volatile)
*note: The yield curve actually went positive 185d before the start of the recession.
5. Recession >> Feb '20 to Present (2 yr and 2 months total)
Date of inversion - 7 Aug '19 >> 160 days before recession
DJIA - from 1 Jan '06 to the end of the recession DJIA ranged 10,800 +3300 -4200 (volatile)
*note: The yield curve actually went positive 3d after it went negative.
Summary:
I don't see any validity between the yield curve inverting and an accurate prediction of an immanent recession. The curve shown in the
article lacks depth of analysis. It kind of reminds me why I limit what I read because these broad statements cloud my mind
as a trader. There are macroeconomic drivers causing the yield inversion, however, those macroeconomic drivers are not neatly packaged
into this one magical inversion chart that tells us a recession is immanent. I think its much more likely that you randomly poll people
at your local Walmart and they'll get you just as close. Sorry, my data shows this article is click bait.
On a side note, I was surprised to see that recessions tend to last no more than about 1.5 years - the Great Depression being the only one
greater than 1.5 yrs at 3 yrs 7 months. I was also interested to see, as a trader, that recessions don't mean the market is always down. Each
recession was unique, however, every one seemed to offer long-side trading opportunities.
medium.com/@timothy.chang/this-graph-has-predicted-every-recession-since-1976-what-its-showing-now-56b1c7ffd004
Premises of the article is that when the 10y interest rate < 2y interest rate (yield inversion) that a recession is immanent.
1. I used FRED data T10Y2Y and Wikipedia for recession dates. I know the definition of recession is controversial, but I chose this source.
en.wikipedia.org/wiki/List_of_recessions_in_the_United_States
2. The DJIA (orange) is overlaying the white T10Y2Y curve as a reference for the market
Initial Observations:
1. My first observation is that the curve has recently crossed 0 (inverted) so recession must be "immanent". Hmm... Wikipedia
shows us being in a recession since Feb 2020 (again defined by negative GDP growth). So, the author's definition of a recession does not
match Wiki's.
2. I'm going to give the author the benefit of the doubt and use their definition of recession (though not stated). There is a very narrow line shown
in the article just after the yield inversion on 8/26/2019 that represents a very short recession and explains #1.
Assuming the author's start date of the recession is around Feb '20, there are some key concerns.
a. The inversion happened ~ 160 days before the start of the recession. That's a pretty substantial lag.
b. And, it went positive 3 days later.
This means that I'm supposed to believe that an event 160 days ago, that lasted a whole 3 days before completely turning around, predicted a recession?
My Study:
I don't know if you'll be able to scroll all the way back through the data where I lay out all of the Wiki recessions versus when the initial yield curve inversion
occurred. If you can, the just use the chart to form your own opinion. Otherwise here is some tabulated data.
1. Recession >> Jan '80 - Jul '80 (6 months total)
Date of inversion - 17 Aug '78 >> 508 days before the recession (almost 2 years)
DJIA - from 17 Aug '78 to end of recession DJIA ranged ~840 +-4%
*note: The yield curve crossed back above 0 close to the end of the recession so maybe it predicts the end of recessions as well
2. Recession >> July '81 - Nov '82 ( 1 yr and 4 months)
Date of inversion - 10 Sep '80 >> 299 days before recession
DJIA - from 10 Sep '80 to the end of the recession DJIA ranged 920 +- 8%
*note: The yield curve crossed above 0 multiple times during this recession. The second time did mark the end of the recession.
3. Recession >> Jul '90 - Mar '91 (8 months total)
Date of inversion - 13 Dec '88 >> 562 days before recession
DJIA - from 13 Dec '88 to the end of the recession DJIA ranged 2100 +900 -0 (never dropped from the start of the inversion)
*note: The yield curve actually went positive 85d before the start of the recession.
4. Recession >> Mar '01 to Nov '01 (8 months total)
Date of inversion - 27 Jan '00 >> 398 days before recession
DJIA - from 27 Jan '00 to end of recession DJIA ranged 11,000 +300 - 1800
*note: The yield curve actually went positive 66d before the start of the recession.
5. Recession >> Dec '07 to Jun '09 (1 yr and 6 months total)
Date of inversion - 1 Jan '06 >> 671 days before recession
DJIA - from 1 Jan '06 to the end of the recession DJIA ranged 10,800 +3300 -4200 (volatile)
*note: The yield curve actually went positive 185d before the start of the recession.
5. Recession >> Feb '20 to Present (2 yr and 2 months total)
Date of inversion - 7 Aug '19 >> 160 days before recession
DJIA - from 1 Jan '06 to the end of the recession DJIA ranged 10,800 +3300 -4200 (volatile)
*note: The yield curve actually went positive 3d after it went negative.
Summary:
I don't see any validity between the yield curve inverting and an accurate prediction of an immanent recession. The curve shown in the
article lacks depth of analysis. It kind of reminds me why I limit what I read because these broad statements cloud my mind
as a trader. There are macroeconomic drivers causing the yield inversion, however, those macroeconomic drivers are not neatly packaged
into this one magical inversion chart that tells us a recession is immanent. I think its much more likely that you randomly poll people
at your local Walmart and they'll get you just as close. Sorry, my data shows this article is click bait.
On a side note, I was surprised to see that recessions tend to last no more than about 1.5 years - the Great Depression being the only one
greater than 1.5 yrs at 3 yrs 7 months. I was also interested to see, as a trader, that recessions don't mean the market is always down. Each
recession was unique, however, every one seemed to offer long-side trading opportunities.
Disclaimer
The information and publications are not meant to be, and do not constitute, financial, investment, trading, or other types of advice or recommendations supplied or endorsed by TradingView. Read more in the Terms of Use.