I don’t think these assumptions will ever let you create a theory that genuinely “ties physics and economics together at the fundamental level”, with anywhere close to the precision and reliability of true physical laws, or in a way that allows us to use physics equations together with economics equations to create better models of the world.
Maybe it’s possible to use something like these assumptions to “tie physics and economics together at the usual, economics-y level” of rough approximations and general guidelines to help us understand cause and effect. But I think even if this project succeeded, you’d just create basically another economic theory/perspective, which is sometimes helpful in situations where it applies well (but it doesn’t apply everywhere, and doesn’t model things perfectly). Compare for instance to various “green” or “human-centric” perspectives on economics, such as the perspective of the “circular economy”. The “circular economy” idea is based on real physical facts (matter is conserved, the earth has finite resources, etc), so you might think it would be easier to come up with precise mathematical laws, but in practice it’s been HARDER for fans of the “circular economy” to formalize their ideas, compared to traditional economic models that are based more in “illusory” human factors like skills, money, demand, etc.
That said, some of the ideas of the “circular economy” have probably been very helpful in some situations even if they haven’t led to a revolutionary improvement in economics. So, another unique perspective on value could probably be similarly helpful in certain situations, like heavy industry as you mentioned, or perhaps agriculture or construction.
“I don’t think these assumptions will ever let you create a theory that genuinely “ties physics and economics together at the fundamental level”, with anywhere close to the precision and reliability of true physical laws...”
Do you know of any true physical laws that connect physics and economics?
I don’t; I was referring to the precision with which physics connects to itself and makes great predictions (like how the rules of newtownian mechanics work extremely well together to predict the motion of objects). As opposed to the IMO lesser (but still useful) precision with which economics connects to itself. (The leading models of economics cannot do stuff like predict the exact future path of recessions or inflation or the price of commodities or etc; they are more like general principles and illustrations rather than a single giant perfectly coherent system that you could use to simulate events.)
It feels to me like the assumptions are not the clear issue. Usefulness is.
Why spend so much effort eliminating “dollar value distortions” from social cost estimates when those estimates already have margins of error due to uncertainties about manufacturing processes, modes of transportation, etc.?
Why publish economic figures in a slightly more accurate format when that format is going to be necessarily quite complex? It seems like TEMS values would be hard to use for practical purposes.
There’s definitely great work to be done around figuring out how to account for externalities. But I just don’t see the benefits of an entirely new framework.
1.-As you are proposing this, it looks you are trying to rebuild some type of “physical theory of value”, but value is subjetive. Any theory on prices, values, etc, shall be based on value subjetivity, and consequently, if properly done, you will reinvent classic micro economics:
2.- The representation of production in economics is surprisingly underdeveloped , but the problem is precisely the lack of details. Economists hate the representation of production, and they are very happy with the Cobb Douglas, while it is obviously crude, and hard to refine. Input output analysis is also crude, but at least the production is represented with some degree of sectoral detail.
So if you have any question, I can be of some help, but in my view the fruitful work is moving to a more detailed production representation, not to general physical economic theories:
There are few ideas I think I would really like to explore:
One is the detailed production representation becoming available to the public with a statutory requirement for all companies of a certain size (public and private) to quantify their TEMS inputs and outputs, similar to the way public companies are required to publish financial reports to the SEC. It would include info such as raw materials purchased, labor hours, and outputs including externalities. That would make the analysis part, much more doable, right? Because you wouldn’t have single economists trying to figure this stuff out by researching companies one at a time—it would just be available—and to everyone.
Two is the representation of TEMS EOL made more readily available in an organized format (ie per product at a single source) that doesn’t require so much time and effort to extract.
Three is studying the utility of objects in terms of their TEMS costs. So, how much TE does this new invention save the customer per unit output? My theory is that there would be a definite correlation between dollar returns, or some other measure of demand, and the provision of products with high TE utility, which would show that the assumptions hold—something I would like to work on in the near future, and something I could definitely use help with if you are available.
Regarding subjectivity, yes, I agree that a portion of value is subjective, but I also think that it is objective. It seems that many people try to fit things into neat categories of one or the other, but it seems in many cases that doesn’t hold up. There is a portion of value that is embedded in physical reality. My thinking is that it is important to understand that part and quantify it for easy analysis, otherwise the consequences of the physical reality are ignored and you have environmental disasters.
“One is the detailed production representation becoming available to the public with a statutory requirement for all companies of a certain size (public and private) to quantify their TEMS inputs and outputs, similar to the way public companies are required to publish financial reports to the SEC”
Currently the Bureu of Economic Analysis (BEA) in US summarizes inter sector relations in input output tables.
All this information is produced in monetary terms, because there is not other way to aggregate; One square kilometer of Manhattan can be hundreds or thousands times more valuable than the same “space” in rural Wyoming. Different energy sources, measured in “caloric equivalents” are extremely different in terms of cost of production and value (imagine the same Jules of coal vs. electricity!).
“Regarding subjectivity, yes, I agree that a portion of value is subjective, but I also think that it is objective.”
In economics, prices are a sufficient statistic that summarizes the subjective valuation of individuals AND the objective scarcity of each commodity. Micro economics (Competitive General Equilibrium is the canonical model) is the theory on how subjective demand and objective “production possibilities” produce price and quantities of each commodity.
Regarding economic theory, before you can change it you need to climb to the shoulders of the giants on which we stand. Only after you have assimilated microeconomics (either the reasonable advanced Micro book of Varian or the monumental Mas Collel) you can make relevant contributions. I want more centrality for natural resources and production economics, but there is an enormous body of knowledge on this fields, that in my view is the most solid (while not very central!) part of Economics.
Yes, Pigovian taxation is the natural way to deal with externalities. I summarize my position now: our economic theory is sound, and new foundations are not necessary. There is enormous room for improvement in “social engeneering”, and representation of production.
A last link. Probably what you are trying to do is EIO-LCA:
The original tool from Carnegie Mellon looks discontinued, but there are papers from 2019 (second lik above) and some other commercial tools (https://vitalmetrics.com/ghg-accounting).
I doubt that it will be possible to connect economic outcomes directly to physics fundamentals without going through the intermediate step of dealing with things like money and human values/desires. For example:
Maybe I am trying to figure out why cars are usually built out of steel, even though they could be lighter, stronger, smaller, and more energy-efficient if they were built out of aluminum. I don’t think there are any physics reasons why this is going on—aluminum seems better than steel in almost every way. The problem is simply that aluminum is more expensive than steel, precisely because it’s better in almost every way so it’s in high demand. The idea of building cars out of aluminum sounds great until you realize that other people are even more eager to build airplanes out of aluminum, and they’re already using most of the world’s supply. So it seems like you’ll still have to reckon with a world of complex human desires, at least to understand the “demand” side of the supply/demand balance that determines the availability and prices of everything. (Is the high price of gold justified by its fundamental physical properties? Or do people just want gold, because they know other people want gold, because they know other people want gold, because...)
Maybe I am wondering why a company bothered to create a large, slow, inefficient factory in a developing country, when they could have gotten the same benefit from building a small, fast, mostly-automated factory in the rich world? The answer might be hard to figure out based on physical fundamentals—it will have to involve things like “well, it’s cheaper to hire workers in the developing world, because the demand for their labor is lower, because they don’t have as many advanced skills, because there are fewer universities there”. Maybe there is some way in which you could describe “human capital” and education in terms of time, energy, matter, etc, but it seems like it would be pretty complicated!
Regarding steel, I think there is also some additional utility that you are leaving out. One would be that it is more flexible, whereas aluminum is prone to breaking or snapping when under a lot of force. I imagine that would be important as automobiles are much more prone to accidents and the flexibility adds an additional layer of safety for the driver. In general, I think that heavy industries are much less prone to subjective utilities than individuals.
As you stated, calculating TEMS values for things like human capital and subjective utilities does seem like it would involve a lot of complexity, but I still think it would be worthwhile to figure out eventually. But that isn’t necessarily what I’d focus on. It would be what investments create the best TEMS outcomes. When you compare the TEMS costs, how does that compare with the utilities?
Sure, I’d say the main difference is that traditional economic theory has no official connection to physics. It is based primarily on monetary prices and those may or may not reflect underlying physical conditions. Monetary prices are subject to distortions such as various biases of perception, power, etc whereas TEMS values are hard-coded into the underlying framework of reality.
Please let me know if that doesn’t answer your question.
I think I understand your thinking now. Wouldn’t many/most of those biases still apply to TEMS? As just one example, hyperbolic discounting is observed in relation to matter (marshmallows) in the “Marshmallow Experiment.”
I think most economists would say that many of their theories apply to anything of worth, not just money.
I’m not sure I’m catching your meaning, but if you’re talking about people making decisions regarding personal TEMS expenditure, then yes, I think many of the biases would still be present.
But here’s an example of what I’m thinking:
We want a reliable quantification of the underlying TEMS costs of production and processing, that way economic decisions, such as investing in plastics can be compared to their TEMS utility at the consumer level.
For example, how much TEMS is this plastic bottle saving the consumer and how many people are using it would be a measurement of TEMS utility. How does that compare to the TEMS cost to produce it? What are the TEMS costs of its end-of-life (EOL) cycle? I’d bet that the TEMS utility is lower than the aggregate TEMS costs, especially if you account for TEMS EOL costs in novel ways.
The theory is that the consumer demand is high for these products because of the TEMS utility at that level and that most consumers ignore the TEMS costs of production and processing. The EOL information in particular is not available in the price that they pay—it is abstracted away into taxes. I think the lack of information causes irrationality in decision-making. The idea is to reduce irrationality by improving the quality of the information—in part by making per unit TEMS values available to the consumer.
That sounds like you would be trying to do economic analysis/decision making using something like bills of material (+labor, externalities etc.). I can see how that would eliminate certain irrationalities, as you say. However, I am having a hard time imagining how one would use a the TEMS cost of a product in practice.
For example, why would it be more useful (or differently useful) to estimate a the TEMS cost of a water bottle instead of calculating the social cost (sum of private costs and external costs like EOL)?
When thinking about the plastic bottle, we not only care about TEMS we care also how many of those plastic bottles will end up in the ocean and what effects they have in the ocean.
We care about the health effects of the substances in plastic. Both those that are already scientifically known as well as health impacts we haven’t yet researched.
In Ohio, a train derailment that might have very well in the supply chain of water bottles that were produced a lot of problems.
If you focus on TEMS you are going to ignore such effects. Ideally, there are taxes that price all the externalities into the price of the plastic bottle. It’s unclear to me why it would be good for actors to focus more on TEMS values.
One of the core ideas is to quantify these externalities for ease of analysis. Many of these things would be quantified as TEMS values.
For example, bottles in the ocean is an M value in the EOL cycle.
Health effects would be T utility of the body in QALYs.
Many of things are already quantified as TEMS values in the existing economic environment, its just that the relationship between physics and economics, or physics and human behavior, is not officially recognized in the pedagogy. That is what I’m trying to address.
The great thing about monetary prices is that there are market mechanisms that keep the numbers honest.
If you want to measure your TEMS value you don’t have information about a lot of the involved factors that matter.
By forcing people to collect those values, you force people to spend a lot of work to account for those values and try to get the accounting to look the way they want it to.
To raise $4.1 trillion in total taxes, the bureaucratic work was around $313 Billion. If you force people to report those TEMS all of those terms are likely similarly complex. The questions of how the numbers will be determined are also very complex so you will need a lot of lobbyists who fight over the values. You need lawyers to litigate cases where people cheated their numbers.
If you hire a plumber then the plumber has to know the TEMS values for all his equipment and know the rules for how much of that will apply to the job for which you hire him.
@Erin @Jackson Wagner @ChristianKleineidam @Arturo Macias
All of this being said, do you think that more often than not that the assumptions would hold?
I don’t think these assumptions will ever let you create a theory that genuinely “ties physics and economics together at the fundamental level”, with anywhere close to the precision and reliability of true physical laws, or in a way that allows us to use physics equations together with economics equations to create better models of the world.
Maybe it’s possible to use something like these assumptions to “tie physics and economics together at the usual, economics-y level” of rough approximations and general guidelines to help us understand cause and effect. But I think even if this project succeeded, you’d just create basically another economic theory/perspective, which is sometimes helpful in situations where it applies well (but it doesn’t apply everywhere, and doesn’t model things perfectly). Compare for instance to various “green” or “human-centric” perspectives on economics, such as the perspective of the “circular economy”. The “circular economy” idea is based on real physical facts (matter is conserved, the earth has finite resources, etc), so you might think it would be easier to come up with precise mathematical laws, but in practice it’s been HARDER for fans of the “circular economy” to formalize their ideas, compared to traditional economic models that are based more in “illusory” human factors like skills, money, demand, etc.
That said, some of the ideas of the “circular economy” have probably been very helpful in some situations even if they haven’t led to a revolutionary improvement in economics. So, another unique perspective on value could probably be similarly helpful in certain situations, like heavy industry as you mentioned, or perhaps agriculture or construction.
“I don’t think these assumptions will ever let you create a theory that genuinely “ties physics and economics together at the fundamental level”, with anywhere close to the precision and reliability of true physical laws...”
Do you know of any true physical laws that connect physics and economics?
I don’t; I was referring to the precision with which physics connects to itself and makes great predictions (like how the rules of newtownian mechanics work extremely well together to predict the motion of objects). As opposed to the IMO lesser (but still useful) precision with which economics connects to itself. (The leading models of economics cannot do stuff like predict the exact future path of recessions or inflation or the price of commodities or etc; they are more like general principles and illustrations rather than a single giant perfectly coherent system that you could use to simulate events.)
It feels to me like the assumptions are not the clear issue. Usefulness is.
Why spend so much effort eliminating “dollar value distortions” from social cost estimates when those estimates already have margins of error due to uncertainties about manufacturing processes, modes of transportation, etc.?
Why publish economic figures in a slightly more accurate format when that format is going to be necessarily quite complex? It seems like TEMS values would be hard to use for practical purposes.
There’s definitely great work to be done around figuring out how to account for externalities. But I just don’t see the benefits of an entirely new framework.
Duly noted, thank you
Hello,
1.-As you are proposing this, it looks you are trying to rebuild some type of “physical theory of value”, but value is subjetive. Any theory on prices, values, etc, shall be based on value subjetivity, and consequently, if properly done, you will reinvent classic micro economics:
https://www.amazon.com/-/es/Hal-R-Varian/dp/0393957357
2.- The representation of production in economics is surprisingly underdeveloped , but the problem is precisely the lack of details. Economists hate the representation of production, and they are very happy with the Cobb Douglas, while it is obviously crude, and hard to refine. Input output analysis is also crude, but at least the production is represented with some degree of sectoral detail.
I used to work on this field: https://www.sciencedirect.com/science/article/pii/S0301421515300392
So if you have any question, I can be of some help, but in my view the fruitful work is moving to a more detailed production representation, not to general physical economic theories:
https://link.springer.com/book/10.1007/978-3-540-75751-1
There are few ideas I think I would really like to explore:
One is the detailed production representation becoming available to the public with a statutory requirement for all companies of a certain size (public and private) to quantify their TEMS inputs and outputs, similar to the way public companies are required to publish financial reports to the SEC. It would include info such as raw materials purchased, labor hours, and outputs including externalities. That would make the analysis part, much more doable, right? Because you wouldn’t have single economists trying to figure this stuff out by researching companies one at a time—it would just be available—and to everyone.
Two is the representation of TEMS EOL made more readily available in an organized format (ie per product at a single source) that doesn’t require so much time and effort to extract.
Three is studying the utility of objects in terms of their TEMS costs. So, how much TE does this new invention save the customer per unit output? My theory is that there would be a definite correlation between dollar returns, or some other measure of demand, and the provision of products with high TE utility, which would show that the assumptions hold—something I would like to work on in the near future, and something I could definitely use help with if you are available.
Regarding subjectivity, yes, I agree that a portion of value is subjective, but I also think that it is objective. It seems that many people try to fit things into neat categories of one or the other, but it seems in many cases that doesn’t hold up. There is a portion of value that is embedded in physical reality. My thinking is that it is important to understand that part and quantify it for easy analysis, otherwise the consequences of the physical reality are ignored and you have environmental disasters.
“One is the detailed production representation becoming available to the public with a statutory requirement for all companies of a certain size (public and private) to quantify their TEMS inputs and outputs, similar to the way public companies are required to publish financial reports to the SEC”
Currently the Bureu of Economic Analysis (BEA) in US summarizes inter sector relations in input output tables.
https://www.bea.gov/resources/methodologies/concepts-methods-io-accounts
In Europe we have the KLEMS project for additional information on capital, labout, etc:
https://economy-finance.ec.europa.eu/economic-research-and-databases/economic-databases/eu-klems-capital-labour-energy-materials-and-service_en
All this information is produced in monetary terms, because there is not other way to aggregate; One square kilometer of Manhattan can be hundreds or thousands times more valuable than the same “space” in rural Wyoming. Different energy sources, measured in “caloric equivalents” are extremely different in terms of cost of production and value (imagine the same Jules of coal vs. electricity!).
http://www.sterndavidi.com/Publications/Aggregation.pdf
Finally the most important:
“Regarding subjectivity, yes, I agree that a portion of value is subjective, but I also think that it is objective.”
In economics, prices are a sufficient statistic that summarizes the subjective valuation of individuals AND the objective scarcity of each commodity. Micro economics (Competitive General Equilibrium is the canonical model) is the theory on how subjective demand and objective “production possibilities” produce price and quantities of each commodity.
Regarding economic theory, before you can change it you need to climb to the shoulders of the giants on which we stand. Only after you have assimilated microeconomics (either the reasonable advanced Micro book of Varian or the monumental Mas Collel) you can make relevant contributions. I want more centrality for natural resources and production economics, but there is an enormous body of knowledge on this fields, that in my view is the most solid (while not very central!) part of Economics.
This is a very helpful post, thank you. I especially appreciate the names and the resources.
I think where I would disagree is that price is a sufficient statistic, especially for externalities.
Your assessment would be that those need to be taxed into price?
Yes, Pigovian taxation is the natural way to deal with externalities. I summarize my position now: our economic theory is sound, and new foundations are not necessary. There is enormous room for improvement in “social engeneering”, and representation of production.
A last link. Probably what you are trying to do is EIO-LCA:
https://en.wikipedia.org/wiki/EIO-LCA
https://www.sciencedirect.com/topics/engineering/economic-input-output-life-cycle-assessment
The original tool from Carnegie Mellon looks discontinued, but there are papers from 2019 (second lik above) and some other commercial tools (https://vitalmetrics.com/ghg-accounting).
I doubt that it will be possible to connect economic outcomes directly to physics fundamentals without going through the intermediate step of dealing with things like money and human values/desires. For example:
Maybe I am trying to figure out why cars are usually built out of steel, even though they could be lighter, stronger, smaller, and more energy-efficient if they were built out of aluminum. I don’t think there are any physics reasons why this is going on—aluminum seems better than steel in almost every way. The problem is simply that aluminum is more expensive than steel, precisely because it’s better in almost every way so it’s in high demand. The idea of building cars out of aluminum sounds great until you realize that other people are even more eager to build airplanes out of aluminum, and they’re already using most of the world’s supply. So it seems like you’ll still have to reckon with a world of complex human desires, at least to understand the “demand” side of the supply/demand balance that determines the availability and prices of everything. (Is the high price of gold justified by its fundamental physical properties? Or do people just want gold, because they know other people want gold, because they know other people want gold, because...)
Maybe I am wondering why a company bothered to create a large, slow, inefficient factory in a developing country, when they could have gotten the same benefit from building a small, fast, mostly-automated factory in the rich world? The answer might be hard to figure out based on physical fundamentals—it will have to involve things like “well, it’s cheaper to hire workers in the developing world, because the demand for their labor is lower, because they don’t have as many advanced skills, because there are fewer universities there”. Maybe there is some way in which you could describe “human capital” and education in terms of time, energy, matter, etc, but it seems like it would be pretty complicated!
Regarding steel, I think there is also some additional utility that you are leaving out. One would be that it is more flexible, whereas aluminum is prone to breaking or snapping when under a lot of force. I imagine that would be important as automobiles are much more prone to accidents and the flexibility adds an additional layer of safety for the driver. In general, I think that heavy industries are much less prone to subjective utilities than individuals.
As you stated, calculating TEMS values for things like human capital and subjective utilities does seem like it would involve a lot of complexity, but I still think it would be worthwhile to figure out eventually. But that isn’t necessarily what I’d focus on. It would be what investments create the best TEMS outcomes. When you compare the TEMS costs, how does that compare with the utilities?
Could you please clarify how this differs from traditional economic theory?
Sure, I’d say the main difference is that traditional economic theory has no official connection to physics. It is based primarily on monetary prices and those may or may not reflect underlying physical conditions. Monetary prices are subject to distortions such as various biases of perception, power, etc whereas TEMS values are hard-coded into the underlying framework of reality.
Please let me know if that doesn’t answer your question.
I think I understand your thinking now. Wouldn’t many/most of those biases still apply to TEMS? As just one example, hyperbolic discounting is observed in relation to matter (marshmallows) in the “Marshmallow Experiment.”
I think most economists would say that many of their theories apply to anything of worth, not just money.
I’m not sure I’m catching your meaning, but if you’re talking about people making decisions regarding personal TEMS expenditure, then yes, I think many of the biases would still be present.
But here’s an example of what I’m thinking:
We want a reliable quantification of the underlying TEMS costs of production and processing, that way economic decisions, such as investing in plastics can be compared to their TEMS utility at the consumer level.
For example, how much TEMS is this plastic bottle saving the consumer and how many people are using it would be a measurement of TEMS utility. How does that compare to the TEMS cost to produce it? What are the TEMS costs of its end-of-life (EOL) cycle? I’d bet that the TEMS utility is lower than the aggregate TEMS costs, especially if you account for TEMS EOL costs in novel ways.
The theory is that the consumer demand is high for these products because of the TEMS utility at that level and that most consumers ignore the TEMS costs of production and processing. The EOL information in particular is not available in the price that they pay—it is abstracted away into taxes. I think the lack of information causes irrationality in decision-making. The idea is to reduce irrationality by improving the quality of the information—in part by making per unit TEMS values available to the consumer.
That sounds like you would be trying to do economic analysis/decision making using something like bills of material (+labor, externalities etc.). I can see how that would eliminate certain irrationalities, as you say. However, I am having a hard time imagining how one would use a the TEMS cost of a product in practice.
For example, why would it be more useful (or differently useful) to estimate a the TEMS cost of a water bottle instead of calculating the social cost (sum of private costs and external costs like EOL)?
Regarding use in practice—in a database accessible from one source linked to the product via the label, or on the product page.
Something like:
Product A
Production Cycle:
T34, E45, M23, S15
Externality Watchlist:
CO2 = 300kg
P = 150kg
Utility
T3, E22, M4, S4
EOL Cycle:
T15, E25, M35, S10
Alerts:
Company B of product production line implicated in Ohio train wreck with (TEMS) externality outputs.
///
Then maybe you could click on the individual letter-number pairs to get more info on the underlying values. Something along those lines.
Regarding the second question, precisely because the social cost is prone to dollar value distortions.
When thinking about the plastic bottle, we not only care about TEMS we care also how many of those plastic bottles will end up in the ocean and what effects they have in the ocean.
We care about the health effects of the substances in plastic. Both those that are already scientifically known as well as health impacts we haven’t yet researched.
In Ohio, a train derailment that might have very well in the supply chain of water bottles that were produced a lot of problems.
If you focus on TEMS you are going to ignore such effects. Ideally, there are taxes that price all the externalities into the price of the plastic bottle. It’s unclear to me why it would be good for actors to focus more on TEMS values.
One of the core ideas is to quantify these externalities for ease of analysis. Many of these things would be quantified as TEMS values.
For example, bottles in the ocean is an M value in the EOL cycle.
Health effects would be T utility of the body in QALYs.
Many of things are already quantified as TEMS values in the existing economic environment, its just that the relationship between physics and economics, or physics and human behavior, is not officially recognized in the pedagogy. That is what I’m trying to address.
The great thing about monetary prices is that there are market mechanisms that keep the numbers honest.
If you want to measure your TEMS value you don’t have information about a lot of the involved factors that matter.
By forcing people to collect those values, you force people to spend a lot of work to account for those values and try to get the accounting to look the way they want it to.
To raise $4.1 trillion in total taxes, the bureaucratic work was around $313 Billion. If you force people to report those TEMS all of those terms are likely similarly complex. The questions of how the numbers will be determined are also very complex so you will need a lot of lobbyists who fight over the values. You need lawyers to litigate cases where people cheated their numbers.
If you hire a plumber then the plumber has to know the TEMS values for all his equipment and know the rules for how much of that will apply to the job for which you hire him.