Danielle DiMartino Booth has been beating the drums for weeks that the US is in recession and has been since October. No so fast says Jim Bianco. The lead chart shows Booth’s view.
Warning: This is a lengthy post. I review the positions of Jim Bianco and Danielle DiMartino Booth.
I also explain how I went wrong on my recession forecasts and why I believe Booth is early in her position that a recession started in October 2023.
Finally, I discuss when I think recession is likely.
The Pending Recession
Jim Bianco
Expansions do not die of old age; they are murdered.
This has been true for the last 50 years:
2020 = COVID
2008 = Financial Crisis, $145 crude oil
2001 = Tech crash/9/11
1991 = Iraq invaded Kuwait (400% rise in crude)
1982 = 200-year high in interest rates
1980 100-year high in inflation
1974 = Arab Oil EmbargoToo many think the economy will either roll over or “pop.” It does not work that way. It does not die of old age.
So, all the talk about deficits and inflation will not cause a recession. However, these issues can, and probably do, make the economy vulnerable to “murder.” But it will still take that murder to cause a recession. Think pandemic, spiking crude oil, political crisis, all of the above simultaneously. This unexpected event(s) could be later this year, in 10 years, or anywhere between. It really cannot be predicted, like COVID, especially by economists. This is not a knock on economists. No amount of analysis of payroll numbers or inflation reports can tell you a pandemic, a spike in oil prices, or a political crisis is coming.
How do we know when the murder has happened? Investors lose a lot of money. A recession means that even investors’ good ideas act like bad ones.
Expansions Don’t Die of Old Age
Bianco has had a hot hand. Since I have been following him, no one has called the economy better.
When most, including me, were discussing or predicting recession, and nearly everyone else predicting a soft landing, Bianco proposed the “No Landing” scenario.
By no landing, Bianco meant continued inflation but no recession, thus higher-for-longer Fed interest rates than most believed, coupled with sticky inflation.
That has been the right call. Congrats to Bianco.
I disagree with one point above, Bianco’s very last paragraph “How do we know when the murder has happened? Investors lose a lot of money. A recession means that even investors’ good ideas act like bad ones.“
Stock markets are terrible predictors. The market was at a then all-time high in November of 2007. Recession hit two months later with hardly any overall losses, and the Fed in total denial. I think a recession was obvious then, but Bernanke denied one in March of 2008 when the recession was already strongly underway.
But this is nitpicking. Bianco makes a very good case.
With that, let’s turn our attention to Danielle DiMartino Booth who has stated for weeks a recession has started.
Danielle DiMartino Booth
That was on May 2, repeated May 9.
Epic Bankruptcies
The Recession Indicator
Liquidations, Not Just Bankruptcies
Sahm “Real-Time” Recession Indicator
You can download a chart on the St. Louis Fed. It’s called the Real-time Sahm Rule Recession Indicator (SAHMREALTIME)
Please consider The Sahm Rule: Step by Step written December 7, 2023.
Smooth out the monthly ‘bumps and wiggles.’
Being data-driven is good, but being data-ridden is not. So, the Sahm rule uses the 3-month average of the monthly unemployment rate, not the monthly rate. Calculate that average in each month and create a new series, “3-Month Average of the Monthly Rate.”
Look back over the prior year.
The logic of the Sahm rule is that when the unemployment rate starts rising, it often picks up steam, and we end up in a recession. A key input to the rule is the lowest value of the 3-month average in the prior 12 months. Note the 12-month look-back does not include the current month.
Put it all together.
The most important part of the Sahm rule is the change. Take the current value of the 3-month average and subtract the 12-month low, and if the difference is 0.50 percentage point or more, then we have historically been in a recession.
In October 2023, the Sahm rule was 0.33 percentage point. That’s below the trigger but also up notably from its values earlier this year.
In closing.
I created the Sahm rule, and it’s on me to communicate it well. I try. If you have any questions, please add them to the comments.
Key Points
- Sahm computes her indicator to two decimal points. However, the input data is one decimal place. Mathematically, you cannot have two decimal points of accuracy when the input is only to one place. (Not that anyone places a lot of faith in these numbers in the first place)
- Instead of using Sahm’s numbers directly, I used labor force and unemployed numbers to calculate the unemployment rate and 3-month moving average unemployment rate to two decimal places.
- By calculating the numbers myself, I can go back further, and did. My chart starts in 1948.
- The McKelvey recession indicator is at 0.3 percent, whereas the Sahm rule is at a much higher 0.5 percent, reducing the number of false positives.
Here is my chart again for ease in reading.
I count five false positive using 0.3 percent as the rule but only one false positive if the trigger is 0.5 percent.
Did Sahm Invent the Methodology?
EconBrowser Calculation
Econbrowser looks at the data in his post In Recession? Real Time vs. Final Revised Data
Notice there are no false negatives. However, this graph (apparently) uses revised data. In real time, one would get a different set of results.
My chart only shows one false positive in the timeframe the Econbrowser chart shows three.
Sahm Real-Time Lags
- 1953-08: Late 3 months
1957-09: Late 2 months
1960-05: Late 5 months
1970-01: Late 2 months
1973-12: Late 7 months
1980-02: On Time
1981-08: Late 3 months
1990-08: Late 2 months
2008-01: Late 1 month
2020-03: Late 1 month
That is with revised data. Since revisions are negative, real-time data would generally be worse.
Sahm also had a false positive in 1959-11 at 0.59 repeating in 1959-12 at 0.51. Does this explain why Sahm starts tracking in 1960?
McKelvey Five False Signals
- 1951-11: 0.35
1957-07: 0.31
1959:10: 0.39 with 1959-11 at 0.59 and 1959-12 at 0.51
1963-03: 0.34 with 1963-04 at 0.31
2003-05: 0.31 with 2003-06 at 0.45 (6 consecutive months above 0.30)
If we conveniently start at 1968 and add a new rule that the indicator does not reset until it first drops to zero (the 2003 low was 0.17) then we can erase the 2003 false signal.
Whereas Sahm is late, McKelvey is early.
We can fix that with another new rule stating that the indicator needs to be above 0.30 for two consecutive months. But that still does not fix 1959 when the indicator fell all the way to -0.25 before crossing the 0.3 percent trigger.
Perfect Track Record?
Sorry, that is wrong. Neither has a perfect track record.
But I am very familiar with being too early. I have a history of being early and in 2022, flat out wrong.
“Sure Looks Like Recession Started in 2022 Q4“
The Philadelphia Fed has a measure called GDPplus that’s a blend of GDP and GDI, not an average. It appears to lean more heavily on GDI.
I discussed the setup August 30, 2023 in Philadelphia Fed GDPplus Measure Sure Looks Like Recession Started in 2022 Q4
In 100 percent of the cases, with no false signals, no misses, and no lead times more than two quarters, every time GDPplus had two consecutive quarters of negative growth, the economy was in recession.
I was positive we were in recession. Confirmed for four months, with what I thought was an indicator that gave no false signals.
What Happened?
Positive Revision Shock!
The Philadelphia Fed revised GDPplus higher reversed string of negative GDPplus numbers.
As revised, there was not a second consecutive quarter of negative GDPplus.
Q:Why the revision?
A: Don’t blame GDPplus or the Philadelphia Fed. GDPplus is subject to huge revisions when the BEA revises GDP and GDI.
GDI lags GDP by at least one month. And the GDP report lags the economy greatly. For example, The BEA released the advance (preliminary) estimate GDP for 2024 Q1 on April 25, 2024.
The second estimate of GDP for 2024 Q1 will be on May 30. That is also when the preliminary estimate of GDI is released.
GDPplus Since 1960
GDPplus Recession Track Record
That’s a very impressive track record, with no misses and no false positives as long as one waits for the second revision to GDI.
Where Are We Now?
In 2022, we had two consecutive quarters of GDP as noted in the square box above, but that is not the definition of a recession.
Real final sales were +1.5 percent. A huge inventory adjustment took GDP to -0.6 percent but the NBER discounts that. So do I, and I said so in real time.
Now we have pretty strong numbers.
Shocks Run in Two Directions
Q: What happened to stave off recession in 2022Q4-2023Q1?
A: A huge tax cut coupled with big jumps in minimum wages in 27 states!
Please recall my February 28, 2023 post Explaining a Huge Inflationary Jump in Disposable Personal Income
Disposable personal income is income after taxes. Since the tax rate went down, income went up.
Personal Income Percent Change From Previous Month
Anything Else? Yes!
Minimum Wage in States with Increases in 2023
In every 2023 paycheck, low-to-medium wage earners got more money in their paychecks relative to 2022 due to tax cuts.
In addition, there were minimum wage hikes in 27 states.
Recession Murdered
Q: How did the tax cuts happen?
A: The IRS Changed its Tax Brackets for 2023.
Americans could save on taxes this year because of historically large inflation adjustments set by the IRS.
The agency adjusted many of its 2023 tax rules to help taxpayers avoid “bracket creep.” That’s when workers get pushed into higher tax brackets due to the impact of cost-of-living adjustments to offset inflation, despite their standard of living not having changed. On average, the IRS pushed up each provision by about 7% for 2023.
I am not aware of anyone who saw this coming. I didn’t. The above link is from February of 2023.
Did Biden influence this or is this standard IRS procedure to cut taxes?
Bianco discussed the economy being murdered. In this case, a pending recession was murdered on sight, ironically, by the IRS.
Tax Cuts Explain Surge in Consumer Spending in 2023
On January 29, 2024, I commented Tax Cuts, Not Bidenomics Explains Surge in Consumer Spending in 2023
On January 1, 2023, 38 states had noteworthy tax changes. 37 of the changes put extra money in people’s pockets. Here’s the result in pictures.
Another Positive Shock
The ridiculously named Inflation Reduction Act was another positive shock.
The IRA introduced consumer tax credits for EVs, gave inflation rebates to low income groups, and handed money to auto makers to invest in EVs.
But what’s happening now?
Negative Revisions Multiple Places
Negative revisions are running rampant in jobs, homes, and durable goods.
On May 24, I commented Another Massive Revision, This Time Durable Goods, What’s Going On
The Commerce Department revised March durable goods orders from +2.6 percent to +0.8 percent. Now it reports a 0.7 percent gain vs an expectation of -0.5 percent.
Also see New Home Sales Sink 4.7 Percent on Top of Huge Negative Revisions
Arguably, the biggest negative revision is in jobs.
Expect Big Negative Revisions to BLS Monthly Jobs
On April 24, the BLS released a little-read jobs report that shows reported jobs in 2023 may be wildly overstated. In turn, that means GDP is likely overstated as well.
On April 25, I commented Expect Big Negative Revisions to BLS Monthly Jobs in 2023, GDP Too
Business Employment Dynamics (BED) Chart Notes
- Data is from the BLS Business Employment Dynamics (BED) report and the BLS monthly jobs reports (CES).
- BED data is less timely but far more accurate than the BLS monthly jobs reports/
- For 2023 Q3, the BED reports shows gross job gains of 7.559 million and gross job losses of 7.751 million for a net loss of 192,000 jobs.
- The BLS monthly jobs reports show a gain of 640,000 jobs.
56 percent Think We Are In Recession
It seems Ed Krassenis, shall we say, mystified as to why 56% of Americans “incorrectly think we are in recession” Perhaps it’s because they’re CORRECT. Per the BLS,job losses began in Q3 2023 & CONTINUED into Q4 2023 per today’s data.
Q: How does the QCEW and BEDs reports relate to the unemployment rate?
A: The Establishment Survey (monthly jobs report), QCEW, and BEDs measure jobs with varying degrees of accuracy. BEDs and QCEW are highly accurate. Unemployment is based on the household survey, a completely different dataset.
With the unemployment rate, you are either employed or you aren’t. However, not employed does not mean unemployed. One has to be seeking employment to be counted as unemployed. In contrast, someone working multiple part-time jobs is counted multiple times in the jobs reports.
Q: Are those job losses reflected in retail sales?
A: Yes.
Real GDP, Real Final Sales, Real GDI
We do not have real GDI for 2024 Q1 yet, but it would take one hell of a set of negative revisions to wipe away Real GDI of 4.8 percent in 2023 Q4.
As for sales, RFS was 2.0 in 2024 Q1 and a whopping 3.9 in 2023 Q4.
Recession Started?
Like Booth, I expect more negative revisions but not enough to revise away a 2023-Q4 GDI print of +4.7 percent, a RFS print of 3.9 percent, and a GDPplus print of 3.4 percent.
The idea that a recession started in October of 2023 seems far fetched at the moment.
Even 2024 Q1 seems early, pending the GDI report and the related GDPplus number.
The Fed’s Big Problem
On February 20, 2024 I noted The Fed’s Big Problem, There Are Two Economies But Only One Interest Rate
Those renting and looking to buy a home are very angry at rent prices up at least 0.4 percent for 32 straight months while home prices are the least affordable in history.
No one wants to trade a 3.0 percent mortgage for a 7.5 percent mortgage so the housing market is essentially locked up.
For discussion, please see Home Prices Hit New Record High, Don’t Worry, It’s Not Inflation
There is no solution to this self-made Fed problem. The Fed has never admitted it created this mess or any mess.
New Home Sales Huge Negative Revisions
New Home Sales plunged in April. And the Census Department completely revised away the fictional 8.8 percent rise in March.
For discussion, please see New Home Sales Sink 4.7 Percent on Top of Huge Negative Revisions
Discretionary Spending Tumbles at Target
On May 22, I noted Discretionary Spending Tumbles at Target, Shares Drop 10 Percent
Target CEO Brian Cornell said the results show “continued soft trends in discretionary categories.”
The interesting word above is “continued”.
EVs Hit Brick Wall
The Inflation Reduction Act provided stimulus for a while, but it also caused the auto manufacturers to gear up for cars that few want now.
EVs have hit a brick wall. IRA stimulus has gone into reverse.
On April 2, I commented Tesla’s Deliveries Drop for First Time Since 2020, It’s Demand Not Supply
On April 15, I noted Elon Musk Fires 10 Percent of Tesla Workforce, Prepares for “Next Phase of Growth”
On April 26, I noted Ford Loses $132,000 on Each EV Produced, Good News, EV Sales Down 20 Percent
And repetitive minimum wage hikes are now pressuring restaurants into layoffs.
Factor in tariffs, Biden’s regulatory errors, and a huge slowdown in the number of student debt cancellations.
What will Biden or the Fed do for an encore?
Murder by Poison?
Q: Looking back to Bianco’s murder shock thesis, is it possible the economy was murdered and we don’t see it yet?
A: Yes.
A combination of Bidenomics, massive immigration and energy policy errors, coupled with Fed QE and housing errors, and you have murder by slow-acting poison instead of one big shock.
The positive shocks are over. And a rate cut by the Fed, whenever it comes (July or September) may mean nothing more than a recession has started.
Final Thoughts
- There is no magic real-time recession indicator.
- Recession? Not in 2023 Q4, but headed that way. GDI, GDPplus, and Real Final Sales are all too high to be revised away enough to indicate a recession.
- As an indicator, I put more faith in GDPplus than Sahm-based rules. But GDPplus is lagging. We do not have GDI for 2024 Q1 yet and that is the key leading input to GDPplus.
- Looking back, we were headed for recession in 2022 Q4. However, two positive shocks, tax cuts and the IRA, murdered the then-pending recession.
- Bianco has called the economy accurately so far. But murder by slow-acting poison is finally taking a huge toll.
Recession did not start in October 2023 and probably not in 2024 Q1 pending the next GDP/GDI report.
But I do expect we will be in recession this year. Bidenomics plus Fed policy errors will finally kill the patient. AI might even play a small role. Companies are getting rid of employees, turning instead to AI.
Label the recession murder by poison.
Tweet Comments
Finding a View You Like
Poison Has Taken Its Toll, We Just Await the Coroner