Off topic but maybe a question other newer readers have too (and might make for a useful blog post): How should a new reader with some basic familiarity with economics and (optionally) your work start to really get into your work?
Is it by reading The Money Illusion (book) and then this blog? Maybe with an additional detour into some of the early Money Illusion (blog) posts you talked about recently? Or are there additional posts or a different orders you'd recommend?
If you are willing to invest the time, The Money Illusion book is the best place to start. Commenter Don Geddis (above) links to my 2009 Cato article, which is a short summary. I think my argument in the book is more persuasive than that article, but if limited for time it's a place to begin.
Sumner probably has hundreds (thousands?) of blog posts on the topic of the monetary origins of the 2008 recession. But perhaps this article ("The Real Problem was Nominal") is a nice summary: https://www.cato-unbound.org/2009/09/14/scott-sumner/real-problem-was-nominal/ (At the very least, it has a fantastic title!)
I don't disagree with any of what you've said. But I don't think we can rule out that the Fed has also been guided by the Sahm Rule on this occasion. It could become an example of Goodhart's Law.
Yes, and if I recall correctly that was Sahm's hope. We are so obsessed with correct predictions (big egos) that we forget the whole point of the exercise is not to predict recessions, it's to prevent them. Her achievement would be even greater if it leads the Fed to avoid an outright recession.
Over the years, I've defined "soft landing" as a situation where the economy continues to expand for three years after unemployment reaches its cyclical lows, with inflation remaining under control. Unemployment fell to cyclical lows in early 2022. If 2025 has a strong labor market and roughly 2% inflation, I'd regard that as a soft landing. America has never had a soft landing by my definition---other countries have.
Most media discussions of earlier "soft landings" (such as the 1990s) refer to interest rates, not the labor market.
It's an interesting question. Most of the "long tail" (or "fractal" or "power law") phenomena you describe have a characteristic minimum size. Like you don't often see really tiny ocean waves because surface tension just flattens them out again. So maybe the generally good economic health of the USA is such that tiny rises in unemployment get annihilated by broadly prevailing growth?
But we could look at a related phenomenon to mini-recessions, like layoffs--how many people lose their job on any given week? The minimum size is zero (if we don't count new jobs started as -1 layoff). But I bet that is something that does get really high (like a magnitude-8 earthquake) in bad weeks, and when it does, it makes a powerful ingredient to a (non-mini) recession.
Hi Scott, long time reader without formal economics training. From what I can tell, the last time you said "Money is still too loose...wake me up when the tight money starts" was April 10, 5 months ago. And I get the impression that today you might say that looser money is needed. I wonder if you could elaborate on your thinking on this point over the last 5 months, maybe in an upcoming post - did money finally become tight, and maybe become too tight, sometime over the last 5 months?
Good question. In April, we'd received three months of high inflation figures, and five year TIPS rose to a bit over 2.5%. NGDP growth had been too rapid. Things have improved since that time, and policy now seems to be roughly on course. I don't think looser money is needed; current policy seems fine. A gradual fall in interest rates may be needed, but I don't regard that as looser money.
I would say the big mistakes in monetary policy were in 2021-22, in 2023 and early 2024 policy was a bit too easy.
My knowledge of US rate moves, expected rate moves, market forecasts and market reactions is poor. Can you point to cases where the market reactions in the US to Fed rate cuts has been the opposite of now? Where they were deemed inappropriate and inflation and mid to longer term rate expectations rose in response.
There were cases where long rates rose on news of a rate cut (January 2001 and September 2007, for instance), but I don't recall cases where the higher inflation expectations were necessarily inappropriate. Maybe in the 1970s?
My question. Why does the fed keep trying to use interest rates to fight inflation? Does it actually work? FFR directly effects a narrow part of our economy. One of the places where we would expect it to have an effect would be the housing market right? Yet housing inflation has stayed relatively constant...
I once ran the regression on lagged FFR and inflation and was largely unconvinced of it being a major driver in changes to inflation. I think taking a lagging oil price inflation in particular would do a better job of modeling inflation 1970s onward, even core inflation...
My reason would obviously have to be that we undervalue the knock on effects that oil prices have within our economy. Obviously they directly effect consumers say at the gas pump, are still fairly easy to understand in cost of freight, but also are less easy to recognize but incredibly important when it comes to plastics and other particular goods. Plastics are particularly reliant on crude oil as the majority of virgin-plastics are derivates from crude oil refining (ex: polyethylene). Just think how prevalent plastics are in our wider economy, from new cars to pop-tarts, to the forks and spoons you get at restaurants to the PVC piping on large infrastructure projects to the vinyl sidings on new housing builds.... sorry went on a bit of a tangent....
I think the Economist chart relies on each country’s definition of a recession? Or is there an NBER methodology applied to the others? I know I give this lecture about how the Business Cycle Dating Committee memos go through a series of coincident indicators. Do the other countries? My impression was they do not. I feel like I should know this but I don’t.
I agree with your observation of the US not having mini-recessions and doubt it would change the direction of the result of that graph if other countries used NBER dating methods.
We (myself included) are so quick to assume the worst in people. That they're inherently evil.
Regular people in China. Regular people in Russia. Landlords. Developers. People working on Wall Street. Trump supporters in Appalachia. Illegal aliens.
What if they're not inherently evil? What if they're all semi-rational, but in ways I don't fully understand? What if they're all.......people?
Shocks should not cause recessions. Shocks (suddenly) change relative prices at which all markets clear. If the changes are small enough to be accommodated within the target rate of inflation, no problem. larger shocks like the 2008 financial crisis or COVID need temporarily over-target inflation. The Fed failed to provide that in 2008-2020 so we had the Great Recession and weak recovery. The Fed performed much better in 2020-21 but was too slow to begin returning inflation to target leading it to then be too slow to recognizing that it had achieved its objective.
It seems to me that that doesn't describe current monetary policy, it describes whether the market has faith in where monetary policy will be in 5 years. But I agree that oil price swings are a problem, and need to be corrected for in the 5-year TIPS.
Economists don't know a debit from a credit. Banks don't lend deposits. Secular stagnation was the impoundment of savings in the banks. C-19 reversed this trend. Case closed.
Off topic but maybe a question other newer readers have too (and might make for a useful blog post): How should a new reader with some basic familiarity with economics and (optionally) your work start to really get into your work?
Is it by reading The Money Illusion (book) and then this blog? Maybe with an additional detour into some of the early Money Illusion (blog) posts you talked about recently? Or are there additional posts or a different orders you'd recommend?
If you are willing to invest the time, The Money Illusion book is the best place to start. Commenter Don Geddis (above) links to my 2009 Cato article, which is a short summary. I think my argument in the book is more persuasive than that article, but if limited for time it's a place to begin.
Interesting read here. I too did not expect the Federal Reserve to cut rates by 50 bps for it would appear to be panicking. Yet, that’s what happened.
It does suggest that perhaps the Federal Reserve has learned some lessons from the past. Time will tell.
I should very much like to read your piece on how the Great Recession was not caused by a housing shock. Could you link to it?
Sumner probably has hundreds (thousands?) of blog posts on the topic of the monetary origins of the 2008 recession. But perhaps this article ("The Real Problem was Nominal") is a nice summary: https://www.cato-unbound.org/2009/09/14/scott-sumner/real-problem-was-nominal/ (At the very least, it has a fantastic title!)
Thank you. I will check it out.
I also have an entire book on the subject, entitled "The Money Illusion".
https://www.amazon.com/Money-Illusion-Monetarism-Recession-Monetary/dp/022677368X
Will check this out. Thank you.
I don't disagree with any of what you've said. But I don't think we can rule out that the Fed has also been guided by the Sahm Rule on this occasion. It could become an example of Goodhart's Law.
Yes, and if I recall correctly that was Sahm's hope. We are so obsessed with correct predictions (big egos) that we forget the whole point of the exercise is not to predict recessions, it's to prevent them. Her achievement would be even greater if it leads the Fed to avoid an outright recession.
Can you expand on how "Fed seems close to achieving its first soft landing"?
Over the years, I've defined "soft landing" as a situation where the economy continues to expand for three years after unemployment reaches its cyclical lows, with inflation remaining under control. Unemployment fell to cyclical lows in early 2022. If 2025 has a strong labor market and roughly 2% inflation, I'd regard that as a soft landing. America has never had a soft landing by my definition---other countries have.
Most media discussions of earlier "soft landings" (such as the 1990s) refer to interest rates, not the labor market.
It's an interesting question. Most of the "long tail" (or "fractal" or "power law") phenomena you describe have a characteristic minimum size. Like you don't often see really tiny ocean waves because surface tension just flattens them out again. So maybe the generally good economic health of the USA is such that tiny rises in unemployment get annihilated by broadly prevailing growth?
But we could look at a related phenomenon to mini-recessions, like layoffs--how many people lose their job on any given week? The minimum size is zero (if we don't count new jobs started as -1 layoff). But I bet that is something that does get really high (like a magnitude-8 earthquake) in bad weeks, and when it does, it makes a powerful ingredient to a (non-mini) recession.
Hi Scott, long time reader without formal economics training. From what I can tell, the last time you said "Money is still too loose...wake me up when the tight money starts" was April 10, 5 months ago. And I get the impression that today you might say that looser money is needed. I wonder if you could elaborate on your thinking on this point over the last 5 months, maybe in an upcoming post - did money finally become tight, and maybe become too tight, sometime over the last 5 months?
Good question. In April, we'd received three months of high inflation figures, and five year TIPS rose to a bit over 2.5%. NGDP growth had been too rapid. Things have improved since that time, and policy now seems to be roughly on course. I don't think looser money is needed; current policy seems fine. A gradual fall in interest rates may be needed, but I don't regard that as looser money.
I would say the big mistakes in monetary policy were in 2021-22, in 2023 and early 2024 policy was a bit too easy.
My knowledge of US rate moves, expected rate moves, market forecasts and market reactions is poor. Can you point to cases where the market reactions in the US to Fed rate cuts has been the opposite of now? Where they were deemed inappropriate and inflation and mid to longer term rate expectations rose in response.
There were cases where long rates rose on news of a rate cut (January 2001 and September 2007, for instance), but I don't recall cases where the higher inflation expectations were necessarily inappropriate. Maybe in the 1970s?
My question. Why does the fed keep trying to use interest rates to fight inflation? Does it actually work? FFR directly effects a narrow part of our economy. One of the places where we would expect it to have an effect would be the housing market right? Yet housing inflation has stayed relatively constant...
I once ran the regression on lagged FFR and inflation and was largely unconvinced of it being a major driver in changes to inflation. I think taking a lagging oil price inflation in particular would do a better job of modeling inflation 1970s onward, even core inflation...
My reason would obviously have to be that we undervalue the knock on effects that oil prices have within our economy. Obviously they directly effect consumers say at the gas pump, are still fairly easy to understand in cost of freight, but also are less easy to recognize but incredibly important when it comes to plastics and other particular goods. Plastics are particularly reliant on crude oil as the majority of virgin-plastics are derivates from crude oil refining (ex: polyethylene). Just think how prevalent plastics are in our wider economy, from new cars to pop-tarts, to the forks and spoons you get at restaurants to the PVC piping on large infrastructure projects to the vinyl sidings on new housing builds.... sorry went on a bit of a tangent....
Latest Bank of Japan minutes...not talked about much: The BoJ will radically reduce QE. They are not sure on rates.
Why reduce QE?
I think the Economist chart relies on each country’s definition of a recession? Or is there an NBER methodology applied to the others? I know I give this lecture about how the Business Cycle Dating Committee memos go through a series of coincident indicators. Do the other countries? My impression was they do not. I feel like I should know this but I don’t.
I think you are right. Some may just rely on the 2 quarters of falling GDP criterion, which is not very useful.
Nonetheless, there really is a difference in the business cycle pattern in other economies.
I agree with your observation of the US not having mini-recessions and doubt it would change the direction of the result of that graph if other countries used NBER dating methods.
Off-topic below. Apologies in advance.
What an amazing thread by King Yglesias:
https://x.com/mattyglesias/status/1838690682803720691
We (myself included) are so quick to assume the worst in people. That they're inherently evil.
Regular people in China. Regular people in Russia. Landlords. Developers. People working on Wall Street. Trump supporters in Appalachia. Illegal aliens.
What if they're not inherently evil? What if they're all semi-rational, but in ways I don't fully understand? What if they're all.......people?
But isn't it so much more comforting to assume that people who disagree with you are evil?
TIPS have been signaling inflation under-target. How does that happen w/o very slow or negative growth.
Well, stocks are hitting record highs--how does that happen with slow or negative growth? All indicators are imperfect.
Shocks should not cause recessions. Shocks (suddenly) change relative prices at which all markets clear. If the changes are small enough to be accommodated within the target rate of inflation, no problem. larger shocks like the 2008 financial crisis or COVID need temporarily over-target inflation. The Fed failed to provide that in 2008-2020 so we had the Great Recession and weak recovery. The Fed performed much better in 2020-21 but was too slow to begin returning inflation to target leading it to then be too slow to recognizing that it had achieved its objective.
https://thomaslhutcheson.substack.com/p/why-target-ngdp
Is there any better medium-term indicator of optimized monetary policy than the Russell 2000? Russell 2000 excluding tech maybe?
I think you need more than just stocks as a policy indicator.
But what if you had to pick one and only one?
Maybe 5-year TIPS spreads? But again, I would not pick just one.
Why not the 5-Year, 5-Year Forward Inflation Expectation Rate at FRED instead?
That data should be less influenced by wild swings in the oil market.
So why would 5-year TIPS spreads be a better indicator?
It seems to me that that doesn't describe current monetary policy, it describes whether the market has faith in where monetary policy will be in 5 years. But I agree that oil price swings are a problem, and need to be corrected for in the 5-year TIPS.
You'd have to count O/N RRPs high of 2022-12-30 $2553.716, and its drawdown to 2024-09-23 $380.372 as a mechanism to extend economic growth.
Economists don't know a debit from a credit. Banks don't lend deposits. Secular stagnation was the impoundment of savings in the banks. C-19 reversed this trend. Case closed.