You missed the Judea Pearl monomaniacal critique, namely: economists spend a lot of time doing causal inference but by and large refuse to use DAGs and structural causal models to represent their assumptions and determine their identification strategies, even though this would make things much clearer with no downside.
It's interesting that I think Macro has some of the most obvious examples how it's improved the world. Like, comparing the response to 08/09 to the Great depression is pretty amazing. And the role of the world bank in China in 1983 was also pretty amazing and seems to me at least played a pretty signifigant part in China's liberalisation (you guys should read Vogel's biography of Deng, it's pretty epic.) I know much less about this case, but also feels pretty signfigant for India's liberalisation in the 1990s. If macro get's get a few percent of the credit for the declines of poverty in India and China that's pretty epic.
It's accurate to point out that academic economics is doing a decent job of moving past the once-standard assumptions of perfect markets, perfect rationality, etc. *However*, communications with non-economists, especially with regard to policy and politics, are still overwhelmingly dominated by "free market" economics. How much do you hear on media or in legislative action on the overwhelming market power we see in field after field from eyeglasses to paper products to communications? Not much, and the little we do hear is mostly from when some info sector megacorp steps on some glibertarian toes.
Granted, this is mostly because the people who pay to have certain economic voices amplified, via think tanks and media, are overwhelmingly wealthy (duh) and have an intense vested interest in amplifying low-tax and let-the-rich-run-wild messages. As a for example, monetarist talk is still ubiquitous decades after it failed in actual practice and more than a decade after the non-inflation from QE made it a joke. But still, anybody who doesn't deliberately hunt out the academic literature hears little else from economists, and has no good way to learn how limited the economic evidence for these views is.
This reminds me of a complaint my Mediaeval History Professor had about the rapid decline within the History department at my school of the COMMERCE historians. His basic complaint was that nobody is really taught about what commerce and economics was like BEFORE mercantalism/capitalism grew up, displaced "just price theory" religious laws etc. and "rational/invisible hand" markets became the primary way individuals bought and sold goods.
"Just price theory" still survives when applied to pricing labor. I was just reading an article on it today. Most workers are offered jobs at a price considered "just" by their employer on a take it or leave it basis. Negotiation is out of the question, and we've seen the current labor shortage in which employers have been doing everything they can to avoid paying "unjust" wages.
Sadly I do not. The professor wasn't a specialist in commerce so we didn't read anything explicitly on it (other than some primary sources, including one where some 11th century Brits got arrested for cornering the market on clams)
Having done macro professionally in the financial world, it's interesting to see how far apart the academic concept of macro is from financial macro. People need to reasonably forecast growth rates, value currencies and public debt instruments, predict how the macro climate will feed into other asset prices. They have learned ways of doing it. Indeed many trillions of dollars worth of asset markets are influenced by macro analysis on a daily basis. The causes and effects are not quite so unpredictable as some academics seem to imagine.
I'm on the buy side and, err, I've never seen sell side macro research being overly correct. It makes for good conversation fodder and help determine what's today's/this week's narrative (that's useful) but, just like the Fed, their projections at 1+ yr aren't exactly reliable.
I was also, and my view is the quality sell side e.g. Goldman is reasonably good up to a year ahead. When monetary loosens or tightens, or fiscal loosens or tightens, macro analysts who know the country they're writing on can generally say pretty well what the effects to growth and inflation will be. There's no real conservative/liberal freshwater/saltwater divide, at least not over the short-term effects. I'm speaking at least of real professional analysts who are writing for the mainstream buy side, not the various odd hucksters and shills writing for bear fundraising or yammering on the tube or issuing policy advice.
Fair enough. I like Goldman stuff in general (JPM EOTM and SocGen Quant stuff as well) and below 1 yr, you actually can make informed bets/projections, though I'm not really sure you need much macro training/PhD to do so.
I agree - I learned on the job, covering a small number of EMs for a country-specific fund. Wanting accurate projections more than a year out isn't realistic, too many active factors at play, but it's not nearly as turbulent as meteorology. One has to limit oneself to the largest, most predictable causes and effects and accept some uncertainty. And if there's trouble brewing below the surface sell side's not going to be out front investigating
A way one can confirm what I'm saying is to compare measured macro indicators with market indicators of expectations. There's always some gap but it's pretty clear that the average person with money at stake is not helpless at making predictions.
I do see one thing in common: skepticism in perfect rationality and information. I wasn't around that far back but my understanding it that the financial world totally ignored that idea from day one.
I think you miss the 'not even wrong' critique - the 'string theory critique' that like modern physicists criticizing the giant wrong turn of strong theory; modern 'dgse dogmatist' are modelling a world that doesnt exist because they are using simply bad math. I.e. lack of understanding of basic modern mathematical modeling techniques used in every other science and social science to model the real world and using equilibrium assumptions, using disproved concepts from 17th probability theory etc. Using an outdated conception of Utility and for a discipline about money pretending money is an illusion. Steve Keen and Ole Peters leaders of this group.
You carefully avoided some of the most poignant criticisms of neoclassical economics. You're fighting straw men (and women). And like many who choose the opponents you fight you win! It ain't macro that's played out. It's microeconomics. One of the CORE authors you cite, Sam Bowles starts out his graduate micro text, Microeconomics: Behavior, Institutions and Evolution (Princeton U. Press,) with a startling repudiation of some basic axioms: Take away self-regarding utility preferences, admit that increasing returns (and network effects) are ubiquitous and the sense that preferences are not changing / adaptive. Look very closely at general equilibrium theory. How realistic are assumptions behind the First/Second Welfare Theorems. How complete are contracts, etc.? Use game theory, look to evolutionary biology (and behavior) and the discipline is no longer your grandfather's economics. I graduated with a degree in Economics 40 years ago (Samuelson's world) Last year I read Bowles' text (for no good reason), then tried to read Mankiw's text, believing that he would put the house in order and restore things to their natural order, but Mankiw's text just seemed silly, concocted, artificial, which I think was the motivation of the 70 or so students (I don't know the exact number) that walked out of Mankiw's Harvard Ec10 a decade ago. It wasn't entirely a UK phenomenon as you imply, not at all.
Part of the problem from the Left is historians, sociologists, anthropologists etc. assuming the guy they're criticizing on Fox News, who probobly has at best an MBA, represents 'economists'. Witness Erik Loomis, left-wing historian, who frequently charactrizes such figures as representing the current state of the field.
Without getting into deeper debates about philosophy of econ/soc science, here's my twopence as a grad student:
1. In the same vein as Rodrik's take on value of reading classics, I think some more qualitative and ethnographic studies of how specific markets function have a similar value. They often have a richer account of how agents behave in markets - how they collect information about prices, the interplay of strategic behavior while being in seemingly long-term economic relations etc. I think this is especially valuable for understanding developing country markets, where social ties in market processes and weak enforcement of formal norms is very common. Basically, write models and generate evidence, but look for richer mechanisms in classics and other disciplines studying at markets!
2. While credibility revolution is a very important step forward, it has made economists feel very much a "jack of all social science questions". This is quiet true for graduate students in training now, where they wantonly take on subjects like domestic violence and classroom teaching. This is not a problem in its own right. However, I wager that analysing these non-market "action spaces" using only individual utility maximisation models is going to be grossly underpowered as an intellectual framework. I think the norms of the profession need to evolve such that researchers interested in such traditionally non-economic questions start to appreciate (publically) the value of non-economic theory in understanding these domains. They would also value greatly by collaborating across disciplines more freely, as well as publishing and presenting outside economics circles. Basically, there needs to be more differentiation of disciplines along lines of subject matter, and less along methods - methods ought to be more catholic!
Speaking of historically informed science-learning, the snarky comment that "Nobody learns calculus by reading De analysi per aequationes numero terminorum infinitas" is misleading, because in fact some of the best mathematics education out there *does* go back to the roots of the big concepts and trace why and how they were developed. This can give a better sense of why those concepts matter than the traditional presentation of fully-realized sprung-from-the-head-of-Athena theorems, and for math majors especially, it gives a much better sense of what working mathematicians do and how they do it, i.e. what kind of mindset allows you to prove new theorems by reasoning from first principles.
You don't actually read the original "classic" math texts to do this, because they're effectively written in a different language: the terminology and style we use to do math today is a product of the last ~250 years. Instead you read modern glosses written by professors who have gone back and slogged through the classics. I was very fortunate to be taught by one of the finest such professors, David Bressoud, whose text _A Radical Approach to Real Analysis_ I recommend to anyone who has taken a standard calculus class and wants to understand how we can put calculus concepts on a more rigorous logical foundation, and what we learn when we do.
You sure are right that the language has changed. Reading a 17th century mathematical text is like trying to read Chaucer. On the other hand, the concepts addressed are often reintroduced in modern mathematics classes if only to provide insight. Mathematics is a long term project. Some theorems have taken centuries to prove.
I share your suspicion of DSGE modeling and my conclusion in grad school was that the entire framework should be trashed in favor of something new.
What I don't get is why there isn't more of an emphasis on predictive power. I don't expect any macro model to be especially good at prediction, but certainly some can be marginally better than others. Since theory doesn't seem to help with prediction at all, why not dump it altogether and just make the best predictive model possible using ML/bayes structural time series/etc? Macro should lean into its one great advantage: a constant stream of new data. It has a great method for detecting overfitting, why not make use of it?
"Since theory doesn't seem to help with prediction at all, why not dump it altogether and just make the best predictive model possible using ML/bayes structural time series/etc?"
DSGE models *are* structural time-series models. Certainly they have loftier theoretical assumptions that ground that structure, but that is still literally what they are. Maybe the following serves only to justify my own decision to specialize in DSGE, but I would give two reasons that DSGE is still a worthwhile use of economists' time:
1) DSGE models at least in principle can survive the Lucas critique by stating explicitly how expectations are formed and how they affect the economy. The same cannot be said of reduced-form VARs. This may or may not be helpful to a banker or anyone whose job is only to forecast the economy, but it is of critical importance to central bankers and other monetary and fiscal authorities.
2) DSGE models allow for welfare analysis by stating, explicitly, people's welfare functions.
Here are some answers to question about MMT posed by Noah back in 2019 in an article I came across today, namely:
".... does MMT..... allow us to determine whether it describes the economy better than, say, Old Keynesian IS-LM models, or New Keynesian models with financial frictions?
Also, how does MMT model productivity in the economy? It seems like hiring a bunch of firefighters in the absence of fires would negatively impact productivity and reduce living standards. Does MMT assume that productivity is exogenous, does it assume that productivity reductions will always be of secondary concern relative to the importance of full employment, or does it have some other way of dealing with potential hits to productivity?"
Answer: If people are unemployed and living in poverty, they are certainly "unproductive". Nevertheless MMT recognizes a nation's productivity is 'exogenous' (?) re its potential aggregate output with full employment, and yet endogenous (?) re the impact of money-incentivized, self-interested individuals exhibiting various degrees of morality and real creative talent in the 'earning' of that money.
So why will state guaranteed employment of presently unemployed workers, via a central bank-funded JG scheme, reduce productivity? Extra care for the environment (public parks etc) won't reduce a nation's productivity; and if the private sector needs more workers and can pay more than the JG wage, the private sector can entice workers from the JG pool.
Btw, I was pleased to see a banner in a New York climate protest on TV today, proclaiming "Down with Powell"... many people see his economic orthodoxy as part of the problem and responsible for Manchin's un-knowing, unyielding pro-fossil stance.; the green transition can indeed be 'paid for' by Powell...to bypass the profit-seeking private sector market which can never transition quickly enough without the imposition of crippling carbon taxes which most negatively affect those least able to pay. That's why Glasgow will fail. Meantime the SCOTUS is being urged to rule against the legality of mandated emissions reductions...not a good look at Glasgow...
Back to MMT and productivity: full employment and maximum productivity are compatible - as reason itself would suggest...and there need be no hits to productivity, via state employment of workers not wanted by the private sector, in fact the opposite is likely to be the case, as crime, depression and demoralization resulting from unemployment are reduced, and demand in the economy is increased in the sectors of the economy that can easily satisfy the extra demand in a full employment economy, thereby increasing the size of the economy, not reducing it.
But who is really responsible for a nations' "productivity"? The guy who claims he is "doing God's work" manipulating clever derivatives, at Goldman Sachs? Or the person - with his God given talent - designing and engineering a new computer chip? Or someone supporting residents in a nursing home?
I know who is the least in terms of contributing to a productive AND well-ordered society, in my estimation.....and oddly that guy is also the highest paid of the three examples of "productivity".
Unfortunately economics' mathematical models don't take into account a 'well-ordered', 'just' society....
Noah, when Doctorow begins by saying "neoclassical" economics, he´s probably thinking that Economics is divided in "schools of thought", and he uses that to criticize a subset of past economists whose thought (he probably imagines) still dominate political and international institutions like the World Bank, IMF, etc...
This idea of "schools of economic thought" is taught on some non-US Universities and seems to be specially beloved by both Marxists and Austrians, perhaps because it gives them a way of justifying their pseudo-science.
I believe that "schools of thought" are for cults. It´s much better to think in terms of the "mainstream" of economics and the "heterodox" rest, because it gives for much richer discussions and explanations, like the ones that you do.
Full disclosure: I have no formal training in economics, and no interest in one due to ingrained skepticism of the discipline. Bear with whatever ignorance and misuse of terminology I display while I try to explain why I'm not finding this post especially convincing or reassuring.
1. Essentially the only way of using the macro/micro opposition I've ever encountered is "micro deals with individual agents interacting with the system, macro with the system as a totality", and the essential criticism of macro as currently (stereotypically?) practiced is that is confuses the two and tries to apply assumptions from the former to the latter. In light of this, I can only see the redefinition of the terms as "things with evidence vs. things without" as meaningless muddling of the waters, and a sales pitch for a macroeconomic work that praises it for "using evidence from microeconomics" as... well, the exact opposite of reassuring.
2. The thing about classics, often about the only thing they have in common, and pretty much THE thing that invariably made them insightful compared to the field, is that they tried to introduce, again and again, the idea that economics is not describing objective laws of reality, but of a particular, historically-specific setting, on one hand embedded in society and determined by social relations, on the other shaped by changes in laws and policy, often prescribed by economists. This is hardly a subtle insight, it's a crucial observation that is bound to get lost in "distilling to equations", again and again. It's exactly what makes economics different from formal and hard science like math and physics, and misunderstanding this and pretending to be math or physics is exactly what economists get accused of, over and over. It makes utterances like Ederer's seem almost willfully oblivious, and those like Rodrik's - technically true, but utterly missing the point.
3. I feel that dismissal of economics as a field of science is often a dismissal of its expertise. To illustrate what I mean: I have no beef with thousands (millions?) of researchers in economics departments, I'm sure that many of them are doing good work, that the discipline as a whole is evolving and learning from past mistakes. Even with all pitfalls of social science research in mind, I wouldn't shy away from citing individual economics papers to substantiate or corroborate some of my arguments and claims. However, I think it'd be extremely beneficial if economic credentials weren't treated as a marker of an arcane knowledge or a deeper understanding of reality, if economists were unable to influence policy by dictating prescriptions from positions of authority. If what it takes to achieve that is sacrificing the credibility of the entire field, then well, I won't expect you to forgive me, but that's a price that I, along with other people who aren't economists invested in their discipline, am willing to pay. Especially since I'm sure that valuable research, the kind that can stand on its own, would survive this.
The macro/micro economics thing always bothers me. I'm always seeing macro-economic arguments that make no sense at the micro level. People can't spend money that don't have and can't borrow. Physics and other fields have done a lot of work on bridging the gap with statistical and system approaches. There's the math for it, and economists don't have to deal with hard to observe things like subatomic particles or protein interactions. Even at the micro level, things are visible.
About 1, the reason that Noah defines the macro-micro distinction in this way is to accommodate the large change in economics in the last 70 years. The field has become extremely empirical and most of that empirical work is categorized as micro. However, that doesn't really fit the individual-aggregate definition so well, so Noah is changing that definition a bit. Since empirical econ is now probably the largest part of econ, his definition is more useful.
About 2, I wonder what you consider classics. Sort of by definition, most classics aren't railing against the field, because there wasn't much of a field before the classics. The idea that econ at any time was not interested in policy or social relations is wrong and I am not aware of any classic that argues (wrongly) along those lines. Criticism of economics as too abstract are reasonable, but since the turn to abstraction is associated with "modern" econ (i.e. the last 75 years), such criticism does not come from classics. When the classics were written, econ wasn't so formalistic. So I wonder what you are referring to with the classics.
I don't understand what you mean by 3, but I will make a related observation. Academic economists have no power. Suppose that business wants to implement policy A. In academic economics, there is a discussion about policy A, with some economists in favour and some against. Business lobbyists will then put the economist in favour on tv and claim economics supports policy A. Clearly economists have no power in this story, even if tv viewers might associate the policy with the economist. I think it's a pretty reasonable description of the role economists play in policy (except for their role in the Fed, which is more dominated by actual academic economists).
You missed the Judea Pearl monomaniacal critique, namely: economists spend a lot of time doing causal inference but by and large refuse to use DAGs and structural causal models to represent their assumptions and determine their identification strategies, even though this would make things much clearer with no downside.
I'll get that one next year!
It's interesting that I think Macro has some of the most obvious examples how it's improved the world. Like, comparing the response to 08/09 to the Great depression is pretty amazing. And the role of the world bank in China in 1983 was also pretty amazing and seems to me at least played a pretty signifigant part in China's liberalisation (you guys should read Vogel's biography of Deng, it's pretty epic.) I know much less about this case, but also feels pretty signfigant for India's liberalisation in the 1990s. If macro get's get a few percent of the credit for the declines of poverty in India and China that's pretty epic.
It's accurate to point out that academic economics is doing a decent job of moving past the once-standard assumptions of perfect markets, perfect rationality, etc. *However*, communications with non-economists, especially with regard to policy and politics, are still overwhelmingly dominated by "free market" economics. How much do you hear on media or in legislative action on the overwhelming market power we see in field after field from eyeglasses to paper products to communications? Not much, and the little we do hear is mostly from when some info sector megacorp steps on some glibertarian toes.
Granted, this is mostly because the people who pay to have certain economic voices amplified, via think tanks and media, are overwhelmingly wealthy (duh) and have an intense vested interest in amplifying low-tax and let-the-rich-run-wild messages. As a for example, monetarist talk is still ubiquitous decades after it failed in actual practice and more than a decade after the non-inflation from QE made it a joke. But still, anybody who doesn't deliberately hunt out the academic literature hears little else from economists, and has no good way to learn how limited the economic evidence for these views is.
This reminds me of a complaint my Mediaeval History Professor had about the rapid decline within the History department at my school of the COMMERCE historians. His basic complaint was that nobody is really taught about what commerce and economics was like BEFORE mercantalism/capitalism grew up, displaced "just price theory" religious laws etc. and "rational/invisible hand" markets became the primary way individuals bought and sold goods.
"Just price theory" still survives when applied to pricing labor. I was just reading an article on it today. Most workers are offered jobs at a price considered "just" by their employer on a take it or leave it basis. Negotiation is out of the question, and we've seen the current labor shortage in which employers have been doing everything they can to avoid paying "unjust" wages.
Do you know of any good books or papers that do talk about pre-mercantilist markets?
Sadly I do not. The professor wasn't a specialist in commerce so we didn't read anything explicitly on it (other than some primary sources, including one where some 11th century Brits got arrested for cornering the market on clams)
Having done macro professionally in the financial world, it's interesting to see how far apart the academic concept of macro is from financial macro. People need to reasonably forecast growth rates, value currencies and public debt instruments, predict how the macro climate will feed into other asset prices. They have learned ways of doing it. Indeed many trillions of dollars worth of asset markets are influenced by macro analysis on a daily basis. The causes and effects are not quite so unpredictable as some academics seem to imagine.
I'm on the buy side and, err, I've never seen sell side macro research being overly correct. It makes for good conversation fodder and help determine what's today's/this week's narrative (that's useful) but, just like the Fed, their projections at 1+ yr aren't exactly reliable.
I was also, and my view is the quality sell side e.g. Goldman is reasonably good up to a year ahead. When monetary loosens or tightens, or fiscal loosens or tightens, macro analysts who know the country they're writing on can generally say pretty well what the effects to growth and inflation will be. There's no real conservative/liberal freshwater/saltwater divide, at least not over the short-term effects. I'm speaking at least of real professional analysts who are writing for the mainstream buy side, not the various odd hucksters and shills writing for bear fundraising or yammering on the tube or issuing policy advice.
Fair enough. I like Goldman stuff in general (JPM EOTM and SocGen Quant stuff as well) and below 1 yr, you actually can make informed bets/projections, though I'm not really sure you need much macro training/PhD to do so.
I agree - I learned on the job, covering a small number of EMs for a country-specific fund. Wanting accurate projections more than a year out isn't realistic, too many active factors at play, but it's not nearly as turbulent as meteorology. One has to limit oneself to the largest, most predictable causes and effects and accept some uncertainty. And if there's trouble brewing below the surface sell side's not going to be out front investigating
A way one can confirm what I'm saying is to compare measured macro indicators with market indicators of expectations. There's always some gap but it's pretty clear that the average person with money at stake is not helpless at making predictions.
I do see one thing in common: skepticism in perfect rationality and information. I wasn't around that far back but my understanding it that the financial world totally ignored that idea from day one.
I think you miss the 'not even wrong' critique - the 'string theory critique' that like modern physicists criticizing the giant wrong turn of strong theory; modern 'dgse dogmatist' are modelling a world that doesnt exist because they are using simply bad math. I.e. lack of understanding of basic modern mathematical modeling techniques used in every other science and social science to model the real world and using equilibrium assumptions, using disproved concepts from 17th probability theory etc. Using an outdated conception of Utility and for a discipline about money pretending money is an illusion. Steve Keen and Ole Peters leaders of this group.
I didn't see anyone making that this year! Perhaps it shall go on next year's list.
You carefully avoided some of the most poignant criticisms of neoclassical economics. You're fighting straw men (and women). And like many who choose the opponents you fight you win! It ain't macro that's played out. It's microeconomics. One of the CORE authors you cite, Sam Bowles starts out his graduate micro text, Microeconomics: Behavior, Institutions and Evolution (Princeton U. Press,) with a startling repudiation of some basic axioms: Take away self-regarding utility preferences, admit that increasing returns (and network effects) are ubiquitous and the sense that preferences are not changing / adaptive. Look very closely at general equilibrium theory. How realistic are assumptions behind the First/Second Welfare Theorems. How complete are contracts, etc.? Use game theory, look to evolutionary biology (and behavior) and the discipline is no longer your grandfather's economics. I graduated with a degree in Economics 40 years ago (Samuelson's world) Last year I read Bowles' text (for no good reason), then tried to read Mankiw's text, believing that he would put the house in order and restore things to their natural order, but Mankiw's text just seemed silly, concocted, artificial, which I think was the motivation of the 70 or so students (I don't know the exact number) that walked out of Mankiw's Harvard Ec10 a decade ago. It wasn't entirely a UK phenomenon as you imply, not at all.
Part of the problem from the Left is historians, sociologists, anthropologists etc. assuming the guy they're criticizing on Fox News, who probobly has at best an MBA, represents 'economists'. Witness Erik Loomis, left-wing historian, who frequently charactrizes such figures as representing the current state of the field.
Without getting into deeper debates about philosophy of econ/soc science, here's my twopence as a grad student:
1. In the same vein as Rodrik's take on value of reading classics, I think some more qualitative and ethnographic studies of how specific markets function have a similar value. They often have a richer account of how agents behave in markets - how they collect information about prices, the interplay of strategic behavior while being in seemingly long-term economic relations etc. I think this is especially valuable for understanding developing country markets, where social ties in market processes and weak enforcement of formal norms is very common. Basically, write models and generate evidence, but look for richer mechanisms in classics and other disciplines studying at markets!
2. While credibility revolution is a very important step forward, it has made economists feel very much a "jack of all social science questions". This is quiet true for graduate students in training now, where they wantonly take on subjects like domestic violence and classroom teaching. This is not a problem in its own right. However, I wager that analysing these non-market "action spaces" using only individual utility maximisation models is going to be grossly underpowered as an intellectual framework. I think the norms of the profession need to evolve such that researchers interested in such traditionally non-economic questions start to appreciate (publically) the value of non-economic theory in understanding these domains. They would also value greatly by collaborating across disciplines more freely, as well as publishing and presenting outside economics circles. Basically, there needs to be more differentiation of disciplines along lines of subject matter, and less along methods - methods ought to be more catholic!
Speaking of historically informed science-learning, the snarky comment that "Nobody learns calculus by reading De analysi per aequationes numero terminorum infinitas" is misleading, because in fact some of the best mathematics education out there *does* go back to the roots of the big concepts and trace why and how they were developed. This can give a better sense of why those concepts matter than the traditional presentation of fully-realized sprung-from-the-head-of-Athena theorems, and for math majors especially, it gives a much better sense of what working mathematicians do and how they do it, i.e. what kind of mindset allows you to prove new theorems by reasoning from first principles.
You don't actually read the original "classic" math texts to do this, because they're effectively written in a different language: the terminology and style we use to do math today is a product of the last ~250 years. Instead you read modern glosses written by professors who have gone back and slogged through the classics. I was very fortunate to be taught by one of the finest such professors, David Bressoud, whose text _A Radical Approach to Real Analysis_ I recommend to anyone who has taken a standard calculus class and wants to understand how we can put calculus concepts on a more rigorous logical foundation, and what we learn when we do.
You sure are right that the language has changed. Reading a 17th century mathematical text is like trying to read Chaucer. On the other hand, the concepts addressed are often reintroduced in modern mathematics classes if only to provide insight. Mathematics is a long term project. Some theorems have taken centuries to prove.
I share your suspicion of DSGE modeling and my conclusion in grad school was that the entire framework should be trashed in favor of something new.
What I don't get is why there isn't more of an emphasis on predictive power. I don't expect any macro model to be especially good at prediction, but certainly some can be marginally better than others. Since theory doesn't seem to help with prediction at all, why not dump it altogether and just make the best predictive model possible using ML/bayes structural time series/etc? Macro should lean into its one great advantage: a constant stream of new data. It has a great method for detecting overfitting, why not make use of it?
"Since theory doesn't seem to help with prediction at all, why not dump it altogether and just make the best predictive model possible using ML/bayes structural time series/etc?"
DSGE models *are* structural time-series models. Certainly they have loftier theoretical assumptions that ground that structure, but that is still literally what they are. Maybe the following serves only to justify my own decision to specialize in DSGE, but I would give two reasons that DSGE is still a worthwhile use of economists' time:
1) DSGE models at least in principle can survive the Lucas critique by stating explicitly how expectations are formed and how they affect the economy. The same cannot be said of reduced-form VARs. This may or may not be helpful to a banker or anyone whose job is only to forecast the economy, but it is of critical importance to central bankers and other monetary and fiscal authorities.
2) DSGE models allow for welfare analysis by stating, explicitly, people's welfare functions.
Here are some answers to question about MMT posed by Noah back in 2019 in an article I came across today, namely:
".... does MMT..... allow us to determine whether it describes the economy better than, say, Old Keynesian IS-LM models, or New Keynesian models with financial frictions?
Also, how does MMT model productivity in the economy? It seems like hiring a bunch of firefighters in the absence of fires would negatively impact productivity and reduce living standards. Does MMT assume that productivity is exogenous, does it assume that productivity reductions will always be of secondary concern relative to the importance of full employment, or does it have some other way of dealing with potential hits to productivity?"
Answer: If people are unemployed and living in poverty, they are certainly "unproductive". Nevertheless MMT recognizes a nation's productivity is 'exogenous' (?) re its potential aggregate output with full employment, and yet endogenous (?) re the impact of money-incentivized, self-interested individuals exhibiting various degrees of morality and real creative talent in the 'earning' of that money.
So why will state guaranteed employment of presently unemployed workers, via a central bank-funded JG scheme, reduce productivity? Extra care for the environment (public parks etc) won't reduce a nation's productivity; and if the private sector needs more workers and can pay more than the JG wage, the private sector can entice workers from the JG pool.
Btw, I was pleased to see a banner in a New York climate protest on TV today, proclaiming "Down with Powell"... many people see his economic orthodoxy as part of the problem and responsible for Manchin's un-knowing, unyielding pro-fossil stance.; the green transition can indeed be 'paid for' by Powell...to bypass the profit-seeking private sector market which can never transition quickly enough without the imposition of crippling carbon taxes which most negatively affect those least able to pay. That's why Glasgow will fail. Meantime the SCOTUS is being urged to rule against the legality of mandated emissions reductions...not a good look at Glasgow...
Back to MMT and productivity: full employment and maximum productivity are compatible - as reason itself would suggest...and there need be no hits to productivity, via state employment of workers not wanted by the private sector, in fact the opposite is likely to be the case, as crime, depression and demoralization resulting from unemployment are reduced, and demand in the economy is increased in the sectors of the economy that can easily satisfy the extra demand in a full employment economy, thereby increasing the size of the economy, not reducing it.
But who is really responsible for a nations' "productivity"? The guy who claims he is "doing God's work" manipulating clever derivatives, at Goldman Sachs? Or the person - with his God given talent - designing and engineering a new computer chip? Or someone supporting residents in a nursing home?
I know who is the least in terms of contributing to a productive AND well-ordered society, in my estimation.....and oddly that guy is also the highest paid of the three examples of "productivity".
Unfortunately economics' mathematical models don't take into account a 'well-ordered', 'just' society....
http://bilbo.economicoutlook.net/blog/?p=48571
Headlined:
"Federal Reserve research paper kills another core New Keynesian idea about "inflation expectations".
http://bilbo.economicoutlook.net/blog/?p=48571
Headlined:
"Federal Reserve research paper kills another core New Keynesian idea about "inflation expectations".
Noah, when Doctorow begins by saying "neoclassical" economics, he´s probably thinking that Economics is divided in "schools of thought", and he uses that to criticize a subset of past economists whose thought (he probably imagines) still dominate political and international institutions like the World Bank, IMF, etc...
This idea of "schools of economic thought" is taught on some non-US Universities and seems to be specially beloved by both Marxists and Austrians, perhaps because it gives them a way of justifying their pseudo-science.
I believe that "schools of thought" are for cults. It´s much better to think in terms of the "mainstream" of economics and the "heterodox" rest, because it gives for much richer discussions and explanations, like the ones that you do.
Full disclosure: I have no formal training in economics, and no interest in one due to ingrained skepticism of the discipline. Bear with whatever ignorance and misuse of terminology I display while I try to explain why I'm not finding this post especially convincing or reassuring.
1. Essentially the only way of using the macro/micro opposition I've ever encountered is "micro deals with individual agents interacting with the system, macro with the system as a totality", and the essential criticism of macro as currently (stereotypically?) practiced is that is confuses the two and tries to apply assumptions from the former to the latter. In light of this, I can only see the redefinition of the terms as "things with evidence vs. things without" as meaningless muddling of the waters, and a sales pitch for a macroeconomic work that praises it for "using evidence from microeconomics" as... well, the exact opposite of reassuring.
2. The thing about classics, often about the only thing they have in common, and pretty much THE thing that invariably made them insightful compared to the field, is that they tried to introduce, again and again, the idea that economics is not describing objective laws of reality, but of a particular, historically-specific setting, on one hand embedded in society and determined by social relations, on the other shaped by changes in laws and policy, often prescribed by economists. This is hardly a subtle insight, it's a crucial observation that is bound to get lost in "distilling to equations", again and again. It's exactly what makes economics different from formal and hard science like math and physics, and misunderstanding this and pretending to be math or physics is exactly what economists get accused of, over and over. It makes utterances like Ederer's seem almost willfully oblivious, and those like Rodrik's - technically true, but utterly missing the point.
3. I feel that dismissal of economics as a field of science is often a dismissal of its expertise. To illustrate what I mean: I have no beef with thousands (millions?) of researchers in economics departments, I'm sure that many of them are doing good work, that the discipline as a whole is evolving and learning from past mistakes. Even with all pitfalls of social science research in mind, I wouldn't shy away from citing individual economics papers to substantiate or corroborate some of my arguments and claims. However, I think it'd be extremely beneficial if economic credentials weren't treated as a marker of an arcane knowledge or a deeper understanding of reality, if economists were unable to influence policy by dictating prescriptions from positions of authority. If what it takes to achieve that is sacrificing the credibility of the entire field, then well, I won't expect you to forgive me, but that's a price that I, along with other people who aren't economists invested in their discipline, am willing to pay. Especially since I'm sure that valuable research, the kind that can stand on its own, would survive this.
The macro/micro economics thing always bothers me. I'm always seeing macro-economic arguments that make no sense at the micro level. People can't spend money that don't have and can't borrow. Physics and other fields have done a lot of work on bridging the gap with statistical and system approaches. There's the math for it, and economists don't have to deal with hard to observe things like subatomic particles or protein interactions. Even at the micro level, things are visible.
About 1, the reason that Noah defines the macro-micro distinction in this way is to accommodate the large change in economics in the last 70 years. The field has become extremely empirical and most of that empirical work is categorized as micro. However, that doesn't really fit the individual-aggregate definition so well, so Noah is changing that definition a bit. Since empirical econ is now probably the largest part of econ, his definition is more useful.
About 2, I wonder what you consider classics. Sort of by definition, most classics aren't railing against the field, because there wasn't much of a field before the classics. The idea that econ at any time was not interested in policy or social relations is wrong and I am not aware of any classic that argues (wrongly) along those lines. Criticism of economics as too abstract are reasonable, but since the turn to abstraction is associated with "modern" econ (i.e. the last 75 years), such criticism does not come from classics. When the classics were written, econ wasn't so formalistic. So I wonder what you are referring to with the classics.
I don't understand what you mean by 3, but I will make a related observation. Academic economists have no power. Suppose that business wants to implement policy A. In academic economics, there is a discussion about policy A, with some economists in favour and some against. Business lobbyists will then put the economist in favour on tv and claim economics supports policy A. Clearly economists have no power in this story, even if tv viewers might associate the policy with the economist. I think it's a pretty reasonable description of the role economists play in policy (except for their role in the Fed, which is more dominated by actual academic economists).