The unemployment rate increased to 3.9% in October; job gains slowed to 150,000--a sign that the recent burst in economic activity and jobs was likely unsustainable. But that's not a recession.
The fact that interest payments go to net savers is reason to think again about the usefulness of those payments as an economic buffer. The fact that they haven’t worked for the money is a supply-side issue. The fact that they won't spend it is a demand-side issue. S=I isn't all that reliable as a business cycle predictor.
It's going to take a lot more than a so called "soft" jobs number of 150k to trigger this recession so many are clamoring for. Some people need to get over themselves, this was not a terrible jobs report yesterday. I also wouldn't be in any rush to credit Jerome Powell for the moderation in job growth or even wages which have slowed to some extent as well.
Net-transfers from the Treasury into private bank accounts is still running at about $1.7T/year rate. No way a recession happens with that kind money-flow... regardless of what happens to the Sahm rule.
I would not bank on that. But programs like the infrastructure, CHIPS, and Inflation Reduction Act have likely softened the blow of high interest rates.
The interest rate rise has ADDED to the net-transfer from the Treasury...although it is a regressive increase since it is UBI for the rich ( risk-free income for no work and in proportion to how rich one is).
This is correct. Its unfortunate too many mainstream economists fail to understand this concept but again a big part of the 9% of GDP via fiscal spending is indeed the massive interest payments to US treasury holders.
It's not the size of deficit that matters, what matters is composition. If Govt deficit is mostly accumulating in upper class (bond holders) savings, then deficit is bad. The high interest rate fiscal welfare for rentiers & much of the deficit gets captured by bond holders. Bond holders in high interest rate environment may chose to save more and consumption & investment can stagnate.
I just wanted to say it is excellent that that the Federal Reserve actually has a chart named after you. Way to go. It is especially nice that the tool has been so accurate. I'm adding this one to my toolbox.
thank you for the article and your beautiful, simple indicator. Question: why did you say if the 3m avg of unemployment rate hits 4%, it triggers the Sahm rule? since the lowest unemployment rate was 3.4% (april 23), a 3-month avg of 3.9% will trigger it already, right? in other word, even if unemployment rate stays at 3.9% for the next 2 months, the 3 month avg will automatically increase to 3.9%, which trigger the Sahm rule. thanks.
Does/should the Sahm Rule apply when the unemployment rate is below CBO's Non-cyclical Rate of Unemployment (NROU)? Currently calculated as 4.42%, which is well above the current 3.9% estimate. Given that it is possible with very low unemployment that the national unemployment rate rises by 0.50 percentage points or more relative to its low during the previous 12 months and still be below NROU. I'd be curious to hear your thoughts. Enjoy reading your work - keep it up!
A low level of unemployment has preceded recessions in the past, at times. That's why I use changes, not levels. Also, no one knows what that natural rate of unemployment is. The moving around of labor force participation and other labor supply issues is a more likely cause if it 'breaks.'
Claudia - read a note earlier today that employment has been revised down for 8 of the last 9 months. Have you looked at the preceding periods prior to the Sahm Rule triggering as to the nature of employment revisions? Were they also in one direction? Or variable (some up, some down)? I do NOT know how often revisions are so consistently in one direction. Seems odd. Either a data reporting problem or model problem on part of BLS.
The data sources for payroll employment (establishment survey) and unemployment are different (household survey), and I have not looked at revisions in one predicting revisions in another. The unemployment rate only revises once a year, and it's only the seasonals. Part but not all of the payroll revisions have been seasonals, which are updated monthly.
But what does the rule imply about the Fed policy? Could it not looking at the rule and change policy instrument settings on the basis of it so as to avoid the recession it is indicating? Or the Fed is setting policy instruments on the basis of data that is or is not correlated with the SAMH rule? Or it makes policy on the basis of totally orthogonal data? But there ought to be an assumption.
Ah, that's a whole other conversation. One I plan to write more about soon.
Big picture: the Fed does not follow policy rules, though it watches Taylor's rules and related that suggest an 'optimal' federal funds rate. A recession indicator is not (explicitly) in them.
Have you had a look at how accurate a similar rule would behave for any of the segments of unemployment? I had a go at calculating this rule for different genders, and it looks like its triggered for men (https://fred.stlouisfed.org/series/LNS14000001). However it may not "work" as well for different segments - interesting none the less!
It's important to use a 3-month moving average on the national rate; otherwise there are more false positives. Also, when you are looking at sub-aggregates by gender or race, these tend to be much noisier, so 6 months might be better than 3 months. But again, I have not looked at that.
The fact that interest payments go to net savers is reason to think again about the usefulness of those payments as an economic buffer. The fact that they haven’t worked for the money is a supply-side issue. The fact that they won't spend it is a demand-side issue. S=I isn't all that reliable as a business cycle predictor.
It's going to take a lot more than a so called "soft" jobs number of 150k to trigger this recession so many are clamoring for. Some people need to get over themselves, this was not a terrible jobs report yesterday. I also wouldn't be in any rush to credit Jerome Powell for the moderation in job growth or even wages which have slowed to some extent as well.
Net-transfers from the Treasury into private bank accounts is still running at about $1.7T/year rate. No way a recession happens with that kind money-flow... regardless of what happens to the Sahm rule.
I would not bank on that. But programs like the infrastructure, CHIPS, and Inflation Reduction Act have likely softened the blow of high interest rates.
The interest rate rise has ADDED to the net-transfer from the Treasury...although it is a regressive increase since it is UBI for the rich ( risk-free income for no work and in proportion to how rich one is).
This is correct. Its unfortunate too many mainstream economists fail to understand this concept but again a big part of the 9% of GDP via fiscal spending is indeed the massive interest payments to US treasury holders.
Sounds like someone has been listening to Mosler. Endorsed.
Also, Kelton, and Wray.
It's not the size of deficit that matters, what matters is composition. If Govt deficit is mostly accumulating in upper class (bond holders) savings, then deficit is bad. The high interest rate fiscal welfare for rentiers & much of the deficit gets captured by bond holders. Bond holders in high interest rate environment may chose to save more and consumption & investment can stagnate.
I just wanted to say it is excellent that that the Federal Reserve actually has a chart named after you. Way to go. It is especially nice that the tool has been so accurate. I'm adding this one to my toolbox.
thank you for the article and your beautiful, simple indicator. Question: why did you say if the 3m avg of unemployment rate hits 4%, it triggers the Sahm rule? since the lowest unemployment rate was 3.4% (april 23), a 3-month avg of 3.9% will trigger it already, right? in other word, even if unemployment rate stays at 3.9% for the next 2 months, the 3 month avg will automatically increase to 3.9%, which trigger the Sahm rule. thanks.
I'm wondering the same thing. Why isn't Sahm rule at .43 since 3.4% is the low in past 12 months?
Does/should the Sahm Rule apply when the unemployment rate is below CBO's Non-cyclical Rate of Unemployment (NROU)? Currently calculated as 4.42%, which is well above the current 3.9% estimate. Given that it is possible with very low unemployment that the national unemployment rate rises by 0.50 percentage points or more relative to its low during the previous 12 months and still be below NROU. I'd be curious to hear your thoughts. Enjoy reading your work - keep it up!
A low level of unemployment has preceded recessions in the past, at times. That's why I use changes, not levels. Also, no one knows what that natural rate of unemployment is. The moving around of labor force participation and other labor supply issues is a more likely cause if it 'breaks.'
And thanks for the kind words!
Claudia - read a note earlier today that employment has been revised down for 8 of the last 9 months. Have you looked at the preceding periods prior to the Sahm Rule triggering as to the nature of employment revisions? Were they also in one direction? Or variable (some up, some down)? I do NOT know how often revisions are so consistently in one direction. Seems odd. Either a data reporting problem or model problem on part of BLS.
The data sources for payroll employment (establishment survey) and unemployment are different (household survey), and I have not looked at revisions in one predicting revisions in another. The unemployment rate only revises once a year, and it's only the seasonals. Part but not all of the payroll revisions have been seasonals, which are updated monthly.
But what does the rule imply about the Fed policy? Could it not looking at the rule and change policy instrument settings on the basis of it so as to avoid the recession it is indicating? Or the Fed is setting policy instruments on the basis of data that is or is not correlated with the SAMH rule? Or it makes policy on the basis of totally orthogonal data? But there ought to be an assumption.
Ah, that's a whole other conversation. One I plan to write more about soon.
Big picture: the Fed does not follow policy rules, though it watches Taylor's rules and related that suggest an 'optimal' federal funds rate. A recession indicator is not (explicitly) in them.
As an aside - would love to hear your take on the gangbuster productivity numbers! Really great news, and hopefully a sign we can see a soft landing?
Some here on productivity: https://stayathomemacro.substack.com/p/jay-powells-greenspan-moment. And I have one coming next week (that should have been done if the Sahm rule had not moved up).
Have you had a look at how accurate a similar rule would behave for any of the segments of unemployment? I had a go at calculating this rule for different genders, and it looks like its triggered for men (https://fred.stlouisfed.org/series/LNS14000001). However it may not "work" as well for different segments - interesting none the less!
I have not. I have done state-level same rule to examine localized contractions. See here: https://stayathomemacro.substack.com/p/looking-local-for-early-signs-of
Also I may have not used the 3mma correctly - as you point out one print at 4% is not enough...
It's important to use a 3-month moving average on the national rate; otherwise there are more false positives. Also, when you are looking at sub-aggregates by gender or race, these tend to be much noisier, so 6 months might be better than 3 months. But again, I have not looked at that.
Claudia: Should 4% 3-month MA of the unemployment rate trigger the Sahm Rule, would this be the lowest trigger value of U?
Yes, before 1970.