Taking his 2 articles together he appears to be saying something like:
“the rational expectation guys were right about the 70’s to the extent that they understood the role of expectations in inflation – but once that was factored in to the Keynesian model then it still come out ahead. This can be shown by the fact that when one compares how the different schools of thought analyzed Volker era policy, then it is rational-expectations-augmented Keynsianism that did a better job than the other schools of explaining what happened to inflation during that time”
This view seems to be consistent with both of his posts.
I think you misunderstood what Bob was trying to say. Best is to read his conclusion again:
“Now as with all things Krugman, I haven’t here shown an outright contradiction. But when Krugman in 2009 wrote “One argument you often hear from anti-Keynesians — it pops up in comments here — is that the experience of stagflation in the 1970s proved Keynesian[ism] wrong,” did he give the impression that he heard this argument from his Keynesian classmates at MIT in real-time during the stagflation itself?”
I think the impression he gives is that is was bitter supply siders who would have said “the experience of stagflation in the 1970s proved Keynesian[ism] wrong,”. His Keynesian classmates at MIT would have said something like ” the experience of stagflation in the 1970s proved Keynesian[ism] needed some adjustment to better factor in expectations.”.
His Keynesian classmates at MIT would have said something like ” the experience of stagflation in the 1970s proved Keynesian[ism] needed some adjustment to better factor in expectations.”.
Adjustment to what? The ” standard, indeed textbook, model of what should happen”?
Good question. I assume that when he said “standard, indeed textbook, model of what should happen” meant “standard after the adjustment to expectations had been added to the model” rather “standard even before RE came along”. The text book he prints the chart from has a date of 1978 so I think that is possible but don’t know enough about Keynsianism and expectations to be sure.
I think the impression he gives is that is was bitter supply siders who would have said “the experience of stagflation in the 1970s proved Keynesian[ism] wrong,”. His Keynesian classmates at MIT would have said something like ” the experience of stagflation in the 1970s proved Keynesian[ism] needed some adjustment to better factor in expectations.”.
But Transformer, we don’t need to speculate as to what Krugman’s Keynesian classmates at MIT might have said. He *told* us what they said. To wit: “The rational expectations guys were right about stagflation, so might they be right about the rest?”
Notice that that’s a heck of a lot different from the sentence you created.
The problem here is that Krugman is unfair to supply siders. You are reading Krugman most generously, but Krugman does not give that courtesy supply siders.
When supply siders say that Keynesiansim (as it was) was proven wrong during the stagflation, then obviously they were right. And they’d also claim that factoring in (after the fact!) rational expectations doesn’t rescue it, which is why they aren’t Keynesians but supply siders.
Krugman obviously admits Keynesianism was wrong without that additional theory, so he should give them this and just admit that, period. Then he, since he obviously is a Keynesian, can go on to defend Keynesianism including rational expectations as right. But in no way can he say they were wrong to say this with respect to Stagflation in the 70s. This is just making cheap shots by trying not to give your intellectual opponent even an inch…
*sigh* Transformer, suppose in 20 years I have two sets of blog posts like this:
(A) “Sometimes you see anti-Austrians argue–like LK in the comments here–that the low CPI rise of 2009-2015 was somehow a refutation of Austrian economics. But no it wasn’t. All it ‘refuted’ was the naive assumption that money velocity would remain constant after a financial crisis. Indeed, it was the spectacular failure of various rounds of QE that confirmed Austrian economics for me at the time.”
(B) “The Austrians are much less dogmatic than the Keynesians. For example, I remember back in 2009-2015, when the Keynesians were running victory laps while my crew and I in Auburn were incorporating IS/LM into Mises University lectures. Tom Woods used to ask me, ‘Bob, Krugman was right about inflation…might he be right about stimulus too?’ But no, by 2015 I realized Krugman hadn’t been right about stimulus.”
You think the above two fit together like peanut butter and jelly, Transformer?
Well maybe I’m just not reading it right but I don’t see any obvious contradiction (or even Kontradiction) between A and B. They are both consistent with a view that Austrian economics both has a better model of the economy and a better methodology for developing that model (which is what Krugman is claiming for Kerynsians).
I just can’t help it then ask you directly since you are just not getting what Bob is saying. Can you please just simply answer this question, only with a Yes or No:
“But when Krugman in 2009 wrote “One argument you often hear from anti-Keynesians — it pops up in comments here — is that the experience of stagflation in the 1970s proved Keynesian[ism] wrong,” did he give the impression that he heard this argument from his Keynesian classmates at MIT in real-time during the stagflation itself?””
It has nothing to do if those statements show an actual contradiction (what a Kontradiction is maybe be debatable). It has all to do how he debates, misleading, misrepresenting, generalizing etc…
To rehash from FB yesterday Bob I think Transformer is basically right.
The late seventies were a big win for Friedman and Phelps and a Phillips Curve that has expectations in it. I of course was not privy to discussions in the MIT econ department in the 70s (or any other time for that matter!), but understandably that would make people question whether the RE project more broadly had the same analytical strength.
What’s nuts is that a Phillips Curve with expectations in it is hardly anti-Keynesian, these insights came out many years before stagflation, particularly what was going on in the late 70s, so people were talking about it and thinking in these terms before stagflation. Even if you want to make the case that in the *late* 70s Friedman-Phelps was somehow a surprise in the MIT economics department despite being around for about a decade by the time Krugman got his PhD (much less by the time Volcker finally took action), it certainly didn’t prove Keynesianism wrong.
“What’s nuts is that a Phillips Curve with expectations in it is hardly anti-Keynesian… ”
A Phillips Curve that incorporates expectations is of no use at all, because it means you can use the trade-off inflation/unemployment only by taking the public unawares, which obviously can’t create a lasting positive effect on the unemployment situation. (Not even yet talking about the usefulness of creating economic activity that is solely driven by people who are intentionally mislead about the price structure by the central bank).
“… only by taking the public unawares, which obviously can’t create a lasting positive effect on the unemployment situation. (Not even yet talking about the usefulness of creating economic activity that is solely driven by people who are intentionally mislead about the price structure by the central bank).”
Use for people who want to steer the economy, the central bank, for economists who argue they could reduce unemployment by increasing inflation and therefore reducing wages by the back door.
Friedman/Phelps was 1967/1968 and it wasn’t hugely out of step with macro-theorizing. It was an important advance on a Keynesian concept everyone thought (and still thinks) is important: the Phillips Curve. This was most definitely well digested in Krugman’s PhD years.
DURING Krugman’s time at MIT you have the Lucas critique in 1976 and Kydland and Prescott in 1977. This is much fresher stuff, it is very much anti-Keynesian (unlike Friedman/Phelps). This is quite different stuff from Friedman/Phelps but it’s obviously related. This is what Krugman says there was soul searching over.
People seem to jumble all of this together despite the ideas being very different and separated by a decade.
“It was an important advance on a Keynesian concept everyone thought (and still thinks) is important: the Phillips Curve.”
I don’t know that the Phillips Curve (any version of it) could properly be called a Keynesian concept. Can you point me to where in Keynes’ own writings, he ever mentions anything like it?
It’s always important whether we’re using a misnomer when we say “Keynesian.” My understanding had been that the Phillips Curve had been tacked on Hicks-Hansen intepretation of Keynesianism, rather than being something Keynes himself suggested.
It’s important because during that period, generally they had the causation running from unemployment to wages to prices. It’s doubtful the author of “A Tract on Monetary Reform” would have endorsed such theories.
I think it’s best to just refer directly to Keynes if what you’re specifically interested in is what Keynes said. Keynesian refers to a set of scientific ideas.
Keynesianism in the 50s was different from Keynesianism in the 80s to be sure. There are still wrinkles. But I don’t think many people at all use the word to mean “the economics of Keynes” to borrow from Leijonhufvud. Not surprisingly, “the economics of Keynes” is quite Keynesian! But questioning the use of the term to describe later work because it’s not found in the work of Keynes isn’t helpful I don’t think.
The Phillips Curve as we know it today came even later than Hicks-Hansen (who of course Keynes was around for, and had a few objections to). But the fundamental ideas of price-level/output gap tradeoffs is in the GT.
I didn’t mean to suggest the Phillips Curve came from Hicks-Hansen, though, I meant that Hicks-Hansen was the dominant interpretation of Keynes and then the Phillips curve was added to that interpretation.
I’ll have to look at what you’ve specifically pointed to me, but I thought it wasn’t a smooth trade off in GT, but rather you could reduce the output gap with no inflation until you reduced it to zero, at which point you’d get inflation. But I guess I could be wrong about that.
Right – if I recall correctly it’s in there but not smooth. How much gets translated into wage and price increases is (like a lot of things in the GT) ultimately determined by the dominant psychology of the time.
Of course that sort of vague “dominant psychology” interpretation of the trade-off doesn’t have to be all that at odds with how we think of things. We smooth out a lot of things that probably aren’t really smooth in real life so we can do calculus on it all.
That tendency bothers some people more than others – it doesn’t bother me all that much unless it’s clear why smoothing it out does violence to the underlying idea.
I’m glad to see that you’re taking Book V of The General Theory of Employment, Interest, and Money much more seriously. Chapter XXI of J.M. Keynes’s 1936 magnum opus does expound on a theory of prices with what would be now termed “microeconomic foundations”. This would help explain upward wage/price spirals and downward wage/price spirals. However, I would like to point out that Book V of The General Theory is actually a response to the most mathematically-developed exposition of the classical theory of the labour market at the time: more specifically, it is directed at Chapter VIII to X of Part II of The Theory of Unemployment (1933) by Arthur Cecil Pigou.
Bob, concerning the main point of your post, I agree with Transformer’s view of things; I don’t see any incongruity at all between your hypothetical blog posts (A) and (B) above. But I wanted to address a side point you made in your post:
“[UPDATE: I didn’t spell it out in the original version of this post, but on second thought I probably should: Notice also that in his 2009 post, Krugman is trying to argue that the ’79-’82 situation was an embarrassment for “Lucas-type rational expectations,” and confirmed that Keynesianism was the better approach. Yet in his post from February 2015, Krugman admits that the rational expectations guys had been so right about stagflation that everybody was doing rational expectations, even Krugman and Olivier Blanchard, and “everyone learned how to emulate Lucas disciples.” Again, not an outright contradiction, but a very different spin on what actually happened in the 1970s and early 1980s.]”
When Krugman says that Lucas-type rational expectations was refuted, he means Lucas’ theory that rational expectations explained recessions and the apparent stickiness of prices, not Lucas’ belief in rational expectations. It was the latter and not the former that Keynesians incorporated into their theories.
I’m just throwing stuff to see if it sticks, here (full disclosure) …
Do the Austrians who’ve studied Keynesianism get the sense that Keynesians’ defenses of past predictive errors rely on less “aggregate-y” data than was used in the error?
In other words, would you say that Keynesians’ defense data has to borrow in the direction of Methodological Individualism (yet still be aggregate) in order to say that the errored theory wasn’t really Keynesianism?
And that so long as there exists less aggregate data, they can always have a reason to say that their predictive failures weren’t consistent with Keynesianism, so long as they don’t “de-aggregate” to the level of the individual?
Just wanted to see if I was sensing something correctly.
I score this one to Krugman.
Taking his 2 articles together he appears to be saying something like:
“the rational expectation guys were right about the 70’s to the extent that they understood the role of expectations in inflation – but once that was factored in to the Keynesian model then it still come out ahead. This can be shown by the fact that when one compares how the different schools of thought analyzed Volker era policy, then it is rational-expectations-augmented Keynsianism that did a better job than the other schools of explaining what happened to inflation during that time”
This view seems to be consistent with both of his posts.
I think you misunderstood what Bob was trying to say. Best is to read his conclusion again:
“Now as with all things Krugman, I haven’t here shown an outright contradiction. But when Krugman in 2009 wrote “One argument you often hear from anti-Keynesians — it pops up in comments here — is that the experience of stagflation in the 1970s proved Keynesian[ism] wrong,” did he give the impression that he heard this argument from his Keynesian classmates at MIT in real-time during the stagflation itself?”
I think the impression he gives is that is was bitter supply siders who would have said “the experience of stagflation in the 1970s proved Keynesian[ism] wrong,”. His Keynesian classmates at MIT would have said something like ” the experience of stagflation in the 1970s proved Keynesian[ism] needed some adjustment to better factor in expectations.”.
Transformer wrote;
His Keynesian classmates at MIT would have said something like ” the experience of stagflation in the 1970s proved Keynesian[ism] needed some adjustment to better factor in expectations.”.
Adjustment to what? The ” standard, indeed textbook, model of what should happen”?
“Adjustment to what?”
Good question. I assume that when he said “standard, indeed textbook, model of what should happen” meant “standard after the adjustment to expectations had been added to the model” rather “standard even before RE came along”. The text book he prints the chart from has a date of 1978 so I think that is possible but don’t know enough about Keynsianism and expectations to be sure.
Transformer wrote:
I think the impression he gives is that is was bitter supply siders who would have said “the experience of stagflation in the 1970s proved Keynesian[ism] wrong,”. His Keynesian classmates at MIT would have said something like ” the experience of stagflation in the 1970s proved Keynesian[ism] needed some adjustment to better factor in expectations.”.
But Transformer, we don’t need to speculate as to what Krugman’s Keynesian classmates at MIT might have said. He *told* us what they said. To wit: “The rational expectations guys were right about stagflation, so might they be right about the rest?”
Notice that that’s a heck of a lot different from the sentence you created.
I do not see those 2 quotes (from Krugman and me) as incompatible.
Question: “The rational expectations guys were right about stagflation, so might they be right about the rest?”
Answer: “No, the experience of stagflation in the 1970s just proved that Keynesian[ism] needed some adjustment to better factor in expectations”.
Transformer,
The problem here is that Krugman is unfair to supply siders. You are reading Krugman most generously, but Krugman does not give that courtesy supply siders.
When supply siders say that Keynesiansim (as it was) was proven wrong during the stagflation, then obviously they were right. And they’d also claim that factoring in (after the fact!) rational expectations doesn’t rescue it, which is why they aren’t Keynesians but supply siders.
Krugman obviously admits Keynesianism was wrong without that additional theory, so he should give them this and just admit that, period. Then he, since he obviously is a Keynesian, can go on to defend Keynesianism including rational expectations as right. But in no way can he say they were wrong to say this with respect to Stagflation in the 70s. This is just making cheap shots by trying not to give your intellectual opponent even an inch…
*sigh* Transformer, suppose in 20 years I have two sets of blog posts like this:
(A) “Sometimes you see anti-Austrians argue–like LK in the comments here–that the low CPI rise of 2009-2015 was somehow a refutation of Austrian economics. But no it wasn’t. All it ‘refuted’ was the naive assumption that money velocity would remain constant after a financial crisis. Indeed, it was the spectacular failure of various rounds of QE that confirmed Austrian economics for me at the time.”
(B) “The Austrians are much less dogmatic than the Keynesians. For example, I remember back in 2009-2015, when the Keynesians were running victory laps while my crew and I in Auburn were incorporating IS/LM into Mises University lectures. Tom Woods used to ask me, ‘Bob, Krugman was right about inflation…might he be right about stimulus too?’ But no, by 2015 I realized Krugman hadn’t been right about stimulus.”
You think the above two fit together like peanut butter and jelly, Transformer?
Well maybe I’m just not reading it right but I don’t see any obvious contradiction (or even Kontradiction) between A and B. They are both consistent with a view that Austrian economics both has a better model of the economy and a better methodology for developing that model (which is what Krugman is claiming for Kerynsians).
Transformer,
I just can’t help it then ask you directly since you are just not getting what Bob is saying. Can you please just simply answer this question, only with a Yes or No:
“But when Krugman in 2009 wrote “One argument you often hear from anti-Keynesians — it pops up in comments here — is that the experience of stagflation in the 1970s proved Keynesian[ism] wrong,” did he give the impression that he heard this argument from his Keynesian classmates at MIT in real-time during the stagflation itself?””
It has nothing to do if those statements show an actual contradiction (what a Kontradiction is maybe be debatable). It has all to do how he debates, misleading, misrepresenting, generalizing etc…
No
To rehash from FB yesterday Bob I think Transformer is basically right.
The late seventies were a big win for Friedman and Phelps and a Phillips Curve that has expectations in it. I of course was not privy to discussions in the MIT econ department in the 70s (or any other time for that matter!), but understandably that would make people question whether the RE project more broadly had the same analytical strength.
What’s nuts is that a Phillips Curve with expectations in it is hardly anti-Keynesian, these insights came out many years before stagflation, particularly what was going on in the late 70s, so people were talking about it and thinking in these terms before stagflation. Even if you want to make the case that in the *late* 70s Friedman-Phelps was somehow a surprise in the MIT economics department despite being around for about a decade by the time Krugman got his PhD (much less by the time Volcker finally took action), it certainly didn’t prove Keynesianism wrong.
“What’s nuts is that a Phillips Curve with expectations in it is hardly anti-Keynesian… ”
A Phillips Curve that incorporates expectations is of no use at all, because it means you can use the trade-off inflation/unemployment only by taking the public unawares, which obviously can’t create a lasting positive effect on the unemployment situation. (Not even yet talking about the usefulness of creating economic activity that is solely driven by people who are intentionally mislead about the price structure by the central bank).
“… only by taking the public unawares, which obviously can’t create a lasting positive effect on the unemployment situation. (Not even yet talking about the usefulness of creating economic activity that is solely driven by people who are intentionally mislead about the price structure by the central bank).”
Hold this thought for another topic, please.
Of use at all to whom?
It seems very useful if you’re trying to understand how the macroeconomy works.
There’s also other work on how the LRPC looks at low inflation levels – that’s a different topic though.
“Of use at all to whom?”
Use for people who want to steer the economy, the central bank, for economists who argue they could reduce unemployment by increasing inflation and therefore reducing wages by the back door.
I think the time ordering here is very important.
Friedman/Phelps was 1967/1968 and it wasn’t hugely out of step with macro-theorizing. It was an important advance on a Keynesian concept everyone thought (and still thinks) is important: the Phillips Curve. This was most definitely well digested in Krugman’s PhD years.
DURING Krugman’s time at MIT you have the Lucas critique in 1976 and Kydland and Prescott in 1977. This is much fresher stuff, it is very much anti-Keynesian (unlike Friedman/Phelps). This is quite different stuff from Friedman/Phelps but it’s obviously related. This is what Krugman says there was soul searching over.
People seem to jumble all of this together despite the ideas being very different and separated by a decade.
“It was an important advance on a Keynesian concept everyone thought (and still thinks) is important: the Phillips Curve.”
I don’t know that the Phillips Curve (any version of it) could properly be called a Keynesian concept. Can you point me to where in Keynes’ own writings, he ever mentions anything like it?
GT Ch. 21. Don’t have the book on me for anything more specific and I’m not going to sift through online versions.
Obviously most of the work really got started after he died, but that doesn’t seem like a sensible standard for ruling it out of being “Keynesian”.
It’s always important whether we’re using a misnomer when we say “Keynesian.” My understanding had been that the Phillips Curve had been tacked on Hicks-Hansen intepretation of Keynesianism, rather than being something Keynes himself suggested.
It’s important because during that period, generally they had the causation running from unemployment to wages to prices. It’s doubtful the author of “A Tract on Monetary Reform” would have endorsed such theories.
I think it’s best to just refer directly to Keynes if what you’re specifically interested in is what Keynes said. Keynesian refers to a set of scientific ideas.
Keynesianism in the 50s was different from Keynesianism in the 80s to be sure. There are still wrinkles. But I don’t think many people at all use the word to mean “the economics of Keynes” to borrow from Leijonhufvud. Not surprisingly, “the economics of Keynes” is quite Keynesian! But questioning the use of the term to describe later work because it’s not found in the work of Keynes isn’t helpful I don’t think.
The Phillips Curve as we know it today came even later than Hicks-Hansen (who of course Keynes was around for, and had a few objections to). But the fundamental ideas of price-level/output gap tradeoffs is in the GT.
Fair enough.
I didn’t mean to suggest the Phillips Curve came from Hicks-Hansen, though, I meant that Hicks-Hansen was the dominant interpretation of Keynes and then the Phillips curve was added to that interpretation.
I’ll have to look at what you’ve specifically pointed to me, but I thought it wasn’t a smooth trade off in GT, but rather you could reduce the output gap with no inflation until you reduced it to zero, at which point you’d get inflation. But I guess I could be wrong about that.
Right – if I recall correctly it’s in there but not smooth. How much gets translated into wage and price increases is (like a lot of things in the GT) ultimately determined by the dominant psychology of the time.
Of course that sort of vague “dominant psychology” interpretation of the trade-off doesn’t have to be all that at odds with how we think of things. We smooth out a lot of things that probably aren’t really smooth in real life so we can do calculus on it all.
That tendency bothers some people more than others – it doesn’t bother me all that much unless it’s clear why smoothing it out does violence to the underlying idea.
I’m glad to see that you’re taking Book V of The General Theory of Employment, Interest, and Money much more seriously. Chapter XXI of J.M. Keynes’s 1936 magnum opus does expound on a theory of prices with what would be now termed “microeconomic foundations”. This would help explain upward wage/price spirals and downward wage/price spirals. However, I would like to point out that Book V of The General Theory is actually a response to the most mathematically-developed exposition of the classical theory of the labour market at the time: more specifically, it is directed at Chapter VIII to X of Part II of The Theory of Unemployment (1933) by Arthur Cecil Pigou.
*Lucas (1976) was very anti-Keynesian btw. Obviously you can embrace the Lucas critique and microfoundations methodologically as a Keynesian.
Bob, concerning the main point of your post, I agree with Transformer’s view of things; I don’t see any incongruity at all between your hypothetical blog posts (A) and (B) above. But I wanted to address a side point you made in your post:
“[UPDATE: I didn’t spell it out in the original version of this post, but on second thought I probably should: Notice also that in his 2009 post, Krugman is trying to argue that the ’79-’82 situation was an embarrassment for “Lucas-type rational expectations,” and confirmed that Keynesianism was the better approach. Yet in his post from February 2015, Krugman admits that the rational expectations guys had been so right about stagflation that everybody was doing rational expectations, even Krugman and Olivier Blanchard, and “everyone learned how to emulate Lucas disciples.” Again, not an outright contradiction, but a very different spin on what actually happened in the 1970s and early 1980s.]”
When Krugman says that Lucas-type rational expectations was refuted, he means Lucas’ theory that rational expectations explained recessions and the apparent stickiness of prices, not Lucas’ belief in rational expectations. It was the latter and not the former that Keynesians incorporated into their theories.
I’m just throwing stuff to see if it sticks, here (full disclosure) …
Do the Austrians who’ve studied Keynesianism get the sense that Keynesians’ defenses of past predictive errors rely on less “aggregate-y” data than was used in the error?
In other words, would you say that Keynesians’ defense data has to borrow in the direction of Methodological Individualism (yet still be aggregate) in order to say that the errored theory wasn’t really Keynesianism?
And that so long as there exists less aggregate data, they can always have a reason to say that their predictive failures weren’t consistent with Keynesianism, so long as they don’t “de-aggregate” to the level of the individual?
Just wanted to see if I was sensing something correctly.