FRED recession probability model

Discussion of the Stock portion of the Permanent Portfolio

Moderator: Global Moderator

Post Reply
User avatar
ochotona
Executive Member
Executive Member
Posts: 2757
Joined: Fri Jan 30, 2015 5:54 am

FRED recession probability model

Post by ochotona » Sun Jun 23, 2019 10:16 am

The St. Louis Fed's page: https://fred.stlouisfed.org/series/RECPROUSM156N

Professor Piger's page: https://pages.uoregon.edu/jpiger/us_rec ... probs.htm/ he might get the model updates out before FRED does

Twitter @JPiger

What strikes me is by the time the model gets to low single digits, there is quite a lot of danger... if it gets to 20, the economy's been turned into guacamole (the highest false alarm was in 1979, the model came in at 18.6). It's really like a step function.

Right now we're seeing April data, reported on June 1. Takes a while for data to come in, obviously.

The value is 2.68 it's really a significant warning. You get a better sense of it if you project the Y-axis as a natural log scale, by twiddling the controls at FRED. Only got to 1.45 in late 2015, and folks are drawing analogies between now and late 2015... that the chop since Feb 2018 is going to resolve, we're going to keep making new highs, etc.

I'll believe that narrative when I see this model go down in future months. If it keeps going up, that would be a real good argument for avoiding FOMO chasing the stock market. You can download the data in CSV format to understand how fast it moves from low single digits into the heart of recession / bear market. You will also realize by looking at the data that the stock market has a Beagle's nose when it comes to sniffing out recession!

Recent data -

NOV 18: 0.02
DEC 18: 0.18
JAN 19: 0.68
FEB 19: 1.18
MAR 19: 1.28
APR 19: 2.68
MAY 19 data will report by JUN 1 on Prof. Pigel's website or Twitter.
pmward
Executive Member
Executive Member
Posts: 458
Joined: Thu Jan 24, 2019 4:39 pm

Re: FRED recession probability model

Post by pmward » Sun Jun 23, 2019 10:22 am

The problem these days is that the stock market now rally's on bad news because they anticipate it will end up with easing from the Fed. People put way too much trust in the Fed these days. I could see a scenario where the market keeps climbing even as data continues to weaken. I could even see a scenario where we have a small recession that is over before it's official and the market just keeps chugging along sideways (and we could potentially even be in this recession right now and not know it yet). We have been moving sideways for 18 months now, so we might essentially "correct" by moving sideways instead of plunging down. I'm not sure what is going to happen, fundamentally the market makes no sense right now, and I think it all comes down to people overly trusting the Fed to provide a Powell PUT.
User avatar
ochotona
Executive Member
Executive Member
Posts: 2757
Joined: Fri Jan 30, 2015 5:54 am

Re: FRED recession probability model

Post by ochotona » Mon Jun 24, 2019 8:01 am

Just what does this 2.68% recession probability mean? About the same levels as:

Aug-Sep 1969
Nov-Dec 1973
Nov-Dec 1979
Jun-Jul 1981
Jun-Jul 1990
Sep-Oct 2000
Nov-Dec 2007

False alarms
Jan 1978 - 1 month spike
Apr 1979 - 1 month spike
Aug 1979 - 1 month spike
Sep 2005 - 1 month spike
Jan 2013 - 1 month spike

The current value is not a spike, it has been ramping since Dec 2018.

And here we are! "But nobody knows anything - it's all fake news".
User avatar
ochotona
Executive Member
Executive Member
Posts: 2757
Joined: Fri Jan 30, 2015 5:54 am

Re: FRED recession probability model

Post by ochotona » Mon Jun 24, 2019 1:06 pm

pmward wrote:
Sun Jun 23, 2019 10:22 am
The problem these days is that the stock market now rally's on bad news because they anticipate it will end up with easing from the Fed. People put way too much trust in the Fed these days. I could see a scenario where the market keeps climbing even as data continues to weaken. I could even see a scenario where we have a small recession that is over before it's official and the market just keeps chugging along sideways (and we could potentially even be in this recession right now and not know it yet). We have been moving sideways for 18 months now, so we might essentially "correct" by moving sideways instead of plunging down. I'm not sure what is going to happen, fundamentally the market makes no sense right now, and I think it all comes down to people overly trusting the Fed to provide a Powell PUT.
I think a very likely scenario is that the Fed does correctly "manage" the equity market, but that it chops sideways for a long, long time... like gold has chopped sideways for years, maybe only coming to life again now? If they keep it from a bear market correction, they will also end up kneecapping future upside. I think what you'll see is other beaten-down markets, segments, and assets will pull ahead of the S&P500. In a decade, the next bubble will be in (FILL IN THE BLANK) not the S&P500.
pmward
Executive Member
Executive Member
Posts: 458
Joined: Thu Jan 24, 2019 4:39 pm

Re: FRED recession probability model

Post by pmward » Mon Jun 24, 2019 2:55 pm

ochotona wrote:
Mon Jun 24, 2019 1:06 pm
pmward wrote:
Sun Jun 23, 2019 10:22 am
The problem these days is that the stock market now rally's on bad news because they anticipate it will end up with easing from the Fed. People put way too much trust in the Fed these days. I could see a scenario where the market keeps climbing even as data continues to weaken. I could even see a scenario where we have a small recession that is over before it's official and the market just keeps chugging along sideways (and we could potentially even be in this recession right now and not know it yet). We have been moving sideways for 18 months now, so we might essentially "correct" by moving sideways instead of plunging down. I'm not sure what is going to happen, fundamentally the market makes no sense right now, and I think it all comes down to people overly trusting the Fed to provide a Powell PUT.
I think a very likely scenario is that the Fed does correctly "manage" the equity market, but that it chops sideways for a long, long time... like gold has chopped sideways for years, maybe only coming to life again now? If they keep it from a bear market correction, they will also end up kneecapping future upside. I think what you'll see is other beaten-down markets, segments, and assets will pull ahead of the S&P500. In a decade, the next bubble will be in (FILL IN THE BLANK) not the S&P500.
This is inline with my thinking as well. Although, if we do get a China trade deal there is potential for a blowoff top followed by a true cyclical bear market. Then again, I could also see a trade deal being hit with a sell the news correction.
User avatar
ochotona
Executive Member
Executive Member
Posts: 2757
Joined: Fri Jan 30, 2015 5:54 am

Re: FRED recession probability model

Post by ochotona » Mon Jun 24, 2019 3:03 pm

G20, July FOMC expected cuts, debt ceiling, government shutdown... all risks. Plus we are at year end target already! Book your profits, take the rest of the year off!
pmward
Executive Member
Executive Member
Posts: 458
Joined: Thu Jan 24, 2019 4:39 pm

Re: FRED recession probability model

Post by pmward » Mon Jun 24, 2019 3:46 pm

ochotona wrote:
Mon Jun 24, 2019 3:03 pm
G20, July FOMC expected cuts, debt ceiling, government shutdown... all risks. Plus we are at year end target already! Book your profits, take the rest of the year off!
Haha, if you look back historically the market is usually up when the government is shut down, hahaha.
User avatar
Cortopassi
Executive Member
Executive Member
Posts: 1471
Joined: Mon Feb 24, 2014 2:28 pm
Location: Illinois

Re: FRED recession probability model

Post by Cortopassi » Mon Jun 24, 2019 4:15 pm

pmward wrote:
Sun Jun 23, 2019 10:22 am
I could see a scenario where the market keeps climbing even as data continues to weaken.
That is what, in the past, screwed me so many times. Bad jobs report, stocks surge! What!? Company x lays off 3,000 people. Stock surges! What?!

My brain could never wrap the logic behind those things for a long time.
Post Reply