Point 4 is the most infuriating. Like you said, it doesn't assume an accurate counterfactual, but it also doesn't consider the lagged supply chain impacts. These could take months to fully materialize, and with the size of these tariff increases, they will be "huge."
Hey. You are a real economist with a PhD that’s at least a half century more recent than mine (‘72, though my only graduate macro course was ‘67). I have an intuition that in a 1-1-1 model the Fed never has to inflate beyond “catchup” (get the price level - the price - back on target) to maintain/restore full employment after a shock. In contrast, in a multi good model — maybe even just 2-2-5(?) — inflation may need to be more than catchup to restore full employment in the 4 (?) markets.
Do you know of a model framework that a) can derive an optimal inflation rate in a 1-1-1 model from just random, “Brownian Movement” micro shocks and a wage stickiness parameter and b) that can then be generalized to n>1?
Uh... hm. So first thing: my research focus was more on labor markets/demographics/automation/firm heterogeneity/fiscal policy (if that's not obvious :)). Penn is very fiscal policy/labor market focused. I want to expand my knowledge to monetary policy now however. You and NN may be the reason I go sit in the corner and read 20 papers on monetary policy now, so I can feel more confident calling myself an economist. Learning is part of the reason I am blogging as well. I will do my best to answer this though.
My comment was based more on propagation of idiosyncratic shocks through networks. I've read most of Acemoglu's work, so I think I subconsciously took inspiration from there. And actually, there is a recent paper that considers sector-specific propagation through an NK multi-sector model with an input-output network and finds that "final goods producers" (or sectors close to the end of the network) are likely to have more immediate inflation impacts, while more upstream sectors have persistent inflation: https://www.econstor.eu/bitstream/10419/320138/1/cesifo1_wp11917.pdf It doesn't seem as strong as I thought though. But I also don't think it accounts for frictional responses (i.e. anticipation) aside from wage stickiness.
As for your actual questions: Unsure, but I think a good start here is the following: https://www.nber.org/system/files/working_papers/w27464/w27464.pdf and is built on the works of Mankiw and Benigno (2004)? It kind of mentions the issues in the last paper I referenced. I'm much more familiar with the classical NK 1-1-1 model you're mentioning of course, but not so familiar with anything beyond that and the Taylor Rule. But different sectoral stickiness could be an issue.
You have written another excellent post, but I have to ask you a serious, practical, 2-part question: do you actually think the average American (or voter) (A) is economically literate enough to understand your valid points and/or (B) cares about how economic forces work (as long as they “feel” better about their own life).
Most Americans, as the saying goes, are rationally ignorant about economics (or about any matters of policies) or worse still they dogmatically believe their favorite demagogue.
I certainly don’t mean to curb your passion for bringing economic misunderstandings and sound economic analysis to your readers attention. But I fear that if Henry Hazlitt (Economics in One Lesson) and Milton Friedman (Free to Choose) can’t educate the average man on the street, your efforts will only be consumed by guys like me who read economic tomes (Wealth of Nations; Principles of Political Economy and Taxation; Human Action; Man, Economy and State; Basic Economics; The Theory of Price; and many other treatises) for pleasure.
I was just thinking about the sectoral shocks point:
"[The 1-1-1 model (one good, one input, one relative price) that NN points to cannot consistently analyze a sectoral shock"
I'm not sure if that would ultimately change my conclusion much (that the main margin of adjustment will be wages). If the tariffed imports can partially replaced by other sector goods, I would expect there to be more demand for labor and capital in these other sector goods. This would increase real marginal cost in that other sector, pushing prices up. Since market structure does not change, I don't think profits (mark-ups) would change in the other sectors. In order to bring back inflation to target, wages would need to fall (they might fall more in different sectors though).
And labor is not homogeneous, either. I will admit, however that this is an intuition. I cannot develop model that has that feature. [I’d love to recruit some hotshot youngster that could.]
Point 4 is the most infuriating. Like you said, it doesn't assume an accurate counterfactual, but it also doesn't consider the lagged supply chain impacts. These could take months to fully materialize, and with the size of these tariff increases, they will be "huge."
Also great follow-up article to NN!
Hey. You are a real economist with a PhD that’s at least a half century more recent than mine (‘72, though my only graduate macro course was ‘67). I have an intuition that in a 1-1-1 model the Fed never has to inflate beyond “catchup” (get the price level - the price - back on target) to maintain/restore full employment after a shock. In contrast, in a multi good model — maybe even just 2-2-5(?) — inflation may need to be more than catchup to restore full employment in the 4 (?) markets.
Do you know of a model framework that a) can derive an optimal inflation rate in a 1-1-1 model from just random, “Brownian Movement” micro shocks and a wage stickiness parameter and b) that can then be generalized to n>1?
Uh... hm. So first thing: my research focus was more on labor markets/demographics/automation/firm heterogeneity/fiscal policy (if that's not obvious :)). Penn is very fiscal policy/labor market focused. I want to expand my knowledge to monetary policy now however. You and NN may be the reason I go sit in the corner and read 20 papers on monetary policy now, so I can feel more confident calling myself an economist. Learning is part of the reason I am blogging as well. I will do my best to answer this though.
My comment was based more on propagation of idiosyncratic shocks through networks. I've read most of Acemoglu's work, so I think I subconsciously took inspiration from there. And actually, there is a recent paper that considers sector-specific propagation through an NK multi-sector model with an input-output network and finds that "final goods producers" (or sectors close to the end of the network) are likely to have more immediate inflation impacts, while more upstream sectors have persistent inflation: https://www.econstor.eu/bitstream/10419/320138/1/cesifo1_wp11917.pdf It doesn't seem as strong as I thought though. But I also don't think it accounts for frictional responses (i.e. anticipation) aside from wage stickiness.
As for your actual questions: Unsure, but I think a good start here is the following: https://www.nber.org/system/files/working_papers/w27464/w27464.pdf and is built on the works of Mankiw and Benigno (2004)? It kind of mentions the issues in the last paper I referenced. I'm much more familiar with the classical NK 1-1-1 model you're mentioning of course, but not so familiar with anything beyond that and the Taylor Rule. But different sectoral stickiness could be an issue.
Just a feeler, not seriously trying to recruit. :) Of course how shocks propagate through systems would be the plumbing of an n-n-n! model.
Thomas,
You have written another excellent post, but I have to ask you a serious, practical, 2-part question: do you actually think the average American (or voter) (A) is economically literate enough to understand your valid points and/or (B) cares about how economic forces work (as long as they “feel” better about their own life).
Most Americans, as the saying goes, are rationally ignorant about economics (or about any matters of policies) or worse still they dogmatically believe their favorite demagogue.
I certainly don’t mean to curb your passion for bringing economic misunderstandings and sound economic analysis to your readers attention. But I fear that if Henry Hazlitt (Economics in One Lesson) and Milton Friedman (Free to Choose) can’t educate the average man on the street, your efforts will only be consumed by guys like me who read economic tomes (Wealth of Nations; Principles of Political Economy and Taxation; Human Action; Man, Economy and State; Basic Economics; The Theory of Price; and many other treatises) for pleasure.
Thank you for the response!
I was just thinking about the sectoral shocks point:
"[The 1-1-1 model (one good, one input, one relative price) that NN points to cannot consistently analyze a sectoral shock"
I'm not sure if that would ultimately change my conclusion much (that the main margin of adjustment will be wages). If the tariffed imports can partially replaced by other sector goods, I would expect there to be more demand for labor and capital in these other sector goods. This would increase real marginal cost in that other sector, pushing prices up. Since market structure does not change, I don't think profits (mark-ups) would change in the other sectors. In order to bring back inflation to target, wages would need to fall (they might fall more in different sectors though).
And labor is not homogeneous, either. I will admit, however that this is an intuition. I cannot develop model that has that feature. [I’d love to recruit some hotshot youngster that could.]
If you lower policy rates lower than NIMs then bank credit will accelerate. And the FED is not currently "tight'.