This American Enterprise Institute chart, which breaks down how price changes for different types of goods and services in the consumer price index, has by now become very widely known.
You really can't run this analysis treating energy prices as if they were just like any other commodity. Energy is one of the most significant inputs for all commodities. It's either required directly in production/extraction (e.g., operating machinery, running factories, powering smelters, etc.) or indirectly (e.g., transportation costs). It's a very major factor in agricultural prices, not only in obvious ways, but because taken together, fertilizers and pesticides all by themselves represent something like 3-5% of the world's total fossil fuel consumption annually. That's in addition to the energy costs of operating agricultural machinery and shipping agricultural products absurd distances for packaging, distribution, and retail sale.
Like the tiny jar of honey I saw in my office breakroom this week. The honey itself was produced in New Zealand, it was packaged in the UK, and sold in Pennsylvania. The total cost of the product I held in my hands was probably mostly energy. The actual honey was almost a rounding error.
So when you look at all these various commodities, particularly agricultural prices, you should really account for movements in energy prices, as they make everything more expensive, in and of themselves.
I'd be very interested to see an update to this analysis with such an adjustment.
I assume it has something to do with economies of scale, because if it's not that, I have no earthly idea. Arbitraging labor and/or energy rates is the only thing I can think of.
Or more likely: the actual cost of production (labor, capital investments, operating expenses), is much larger than a rounding error. The statement just doesn't pass the smell test.
If energy was 90% of honey production (at least an order of magnitude difference from a rounding error), we would notice all sorts of energy dependent things:
- Local honey would in general be much cheaper
- Areas with low cost of energy would produce more honey (similar to how aluminum production occurs in Iceland because of cheap energy)
- The price would fluctuate significantly with the price of energy
Direct, sure. But it's an input for all of the other inputs. That's my point. Equipment manufacturers have energy costs. Labor expenses include laborers paying for energy for themselves. When energy costs go up, everything gets more expensive, not just those budget items that are obviously related to energy.
A comment on Natural Gas. I am not sure how relevant my comment is, as this is just one among many commodities examined, but here goes:
I think US prices were used here. Nat Gas is unlike coal or oil in that it has been impossible to cheaply transport it across oceans for most of the period examined. (That is changing with LNG although that is recent, somewhat expensive, and still scaling up.) Hence prices in Europe and Asia have been much higher than in the US for decades (I think). In this sense, it is not a "commodity" - it is landlocked.
Again, many commodities have been examined here, and this is a bit of quibble. But on the other hand, natural gas is an input into many other prices listed here - particularly agriculture in the US, via the fertilizer production process. So perhaps this has some relevance?
Also worth noting that as LNG capacity scales up, at some point we should reach a "tipping point" where US prices more or less converge on global prices (less the direct cost associated with liquefaction).
The timing of the Ehrlich-Simon bet wasn’t random, it was at a time when prices had recently risen sharply. The debate was over whether this was due to exhaustion of resources (Erhlich’s argument) or just short term cyclical factors (Simon’s). Simon won the bet as of course supply of pretty much all commodities vastly increased and we didn’t all starve to death.
Importantly Simon didn’t ever say commodities would always decrease in price from any point, he was well aware of commodity cycles and trivializing the argument to this misses the point. The real lesson of this bet is the abundance one; we can rely on the power of the market to incentivize human innovation to increase supply. This contrasts with what superficially seems the more “sensible” doomster approach of government top down intervention and control but which has led to many disastrous and inhumane policies like Indian sterilization programs, or the Chinese one child policy.
>But folks have pointed out that if the bet had been over a different 10-year window, Ehrlich would have won the bet.
Ehrlich picked what to bet on. He has been embarrassingly wrong about everything forever. But True Believers will never face facts - Doom is always right!
Ehrlich's population predictions were correct. As noted in the article, he was wrong about some commodities on a ten year basis but correct over a twenty year basis. He proposed more than one scenario including one with the widespread introduction of birth control. It took longer than his book predicts, but it did slow population growth.
He wasn't embarrassingly wrong about everything. In fact, he got a lot of things wright. He just had an inconvenient message, so a lot of effort has gone into smearing him.
People often confuse price spikes with Malthusianism. Economically speaking there is no deleption, there is always supply, reflected by the usage, price, technology. The more we look the more we find and the more we use the more we look.
Well the earth is finite and at some point we are bound to have depleted most of the useful ressources.
Of course we are going to recycle what can be but the question is how many people can be supported for a given quality of life if all ressources are in circulation.
This seems somewhat consistent with your data, the modal annual increase in resource price is ~1%, a little lower than real interest rates though (1-4%):
The recent increase might be due in part to the recent increase in interest rates this decade.
Theory and data seem close enough for now that I'm going to stick to the null hypothesis that exhaustible resources grow in (real) price at real interest rates? Though there will still be exogenous shocks and manufacturing innovations that dwarf this long-term trend.
Interesting article. I did not note a mention of (governmental) subsidies that some industries receive (e.g. farming) and I am wondering how much, if any, those subsidies can impact price?
a) Looking at the food available years ago, not the premium stuff, just the quality that the average working family would expect; and considering the portion of the family budget that food cost;
b) Can you still get that quality today if you’re willing to pay?
No doubt it would be more expensive than the average food prices today, but in real terms, or in terms of the portion of the family budget, how would prices compare?
Note I asked an AI (alter.systems) that question. It reports that in 1950, 25 to 30% of the family budget went on food, while today it is 10 to 15%. However, to eat high quality whole food meals today, would take it back to a similar 25 to 35% of the budget.
Isn't tracking price more about supply and demand and producers profitability and not the cost of production? It isn't reflected in your oil graph, but for a short period of time during the covid pandemic the price of oil went negative and producers/dealers were having to pay to have their inventory taken off their hands. Surely a extreme case of cost [of production] separating from price.
Perhaps an interesting analysis would be to compare the profitability of the commodity producers compared to the price. Industry consolidation and competitiveness would be another factor.
In that AEI graph its very notable that the highest price increases are in the industries most notorious for private equity funded monopoly roll-ups.
Here in the UK, we have the impression that American food is much more intensively farmed and processed than in the ‘70s, and also is of much poorer quality. Are there still farmers rearing animals and produce to a similar quality to the average standard 50 or 60 years ago. It would be interesting to compare how these prices have changed over time. I suspect they have held pretty steady in real terms , maybe risen as it became harder to find.
If you are comparing today with the 1950s or 1960s, American food has changed. Chicken used to taste like chicken. Now it has no intrinsic flavor. They did the same thing with pork when they started pushing it as "the other white meat". You can still find chickens that taste like chicken and pork that tastes like pig, but you have to pay a premium for it. It's similar with a lot of produce.
The meat industries are highly industrial and centralized, so a handful of companies make the breeding and feeding decisions. Grain was standardized in the 19th century. Meat was standardized in the 20th. I get the impression Americans hate the taste of food, so if you want food with flavor you have to search for it and pay more for it.
I believe it is just an impression. I have looked at a few things and actually the quality has stayed the same or even increased.
The impression comes from the obesity crisis which is an oversupply/cheapness problem more than anything. Historically humans never had access to so much food for so little effort.
And the cheap stuff is what is calorically dense and very easy to make because industrial processes have become extremely efficient.
Nutrient density has lowered but that’s just because vegetables are much bigger (they are mostly water in the end). Total nutrient output isn’t different.
The other changes are related to specific crop selection for various reasons.
Unless you are vegan you probably don’t lack much of anything.
The problem is essentially that people have « forgotten » how to eat, for convenience, pleasure and because women are too busy to cook and share/transfer recipes to the next generation.
Excellent info. Would be interested in seeing the total cost of crop subsidies as well as profit to the large entities in crop science as well as farmer net income.
Commodities are raw materials predominantly. And these are processed in chemical engineering type plants. Which lend to manufacturering productivity.
Farming is also a manufacturing process.
Services are people based. People services scale by adding people so little productivity.
However the great service productivity and cousin, white collar productivity came from a quiet silent revolution called the Personal Computer and early software. It so subtle that even brilliant Jack Welch at GE was clueless. But he saw the benefit as he increased productivity by lowering the work force. Because work of secretaries, travel agents, draftsman were all absorbed by the white collar professionals.
Very interesting article. My prior assumption would have been that anything that we put any serious effort into producing would have seen large price decreases over this period.
How much do energy costs hold everything else hostage? Everything has to be transported. Most of these things also involve significant use of heavy machinery. Some of the mined/refined metal products are notorious for massive energy consumption, aluminum comes to mind and I'm sure other metals are similar.
I wonder if there is a way to analyze energy costs as a percent of total costs and see how that effects things? I've often wondered, especially with the developments in solar and battery technology, how abundant energy could completely change some of these dynamics. If transportation cost was no issue, how many new locations become profitable? If electricity was basically free how does that change how heavy equipment is used? What kids of machines would be invented that just can't make economic sense right now?
For a longer term view, check out Hackett-Fischer's The Great Wave. European historians have done a lot of research on pricing, but not a lot of it has been published in English. There appears to be a roughly 200 year cycle with perhaps a century of relative price stability alternating with roughly a century of rising prices, but those prices didn't rise uniformly. Technology was less advanced back then, so prices that were based on the value of land - fuel i.e. wood and food - would rise with rents while the price of manufactured goods - i.e. labor - would fall. There's a suggestion of a Malthusian effect with the land-population ratio driving prices, but there has to be a strong political effect to account for the long periods of stability.
You really can't run this analysis treating energy prices as if they were just like any other commodity. Energy is one of the most significant inputs for all commodities. It's either required directly in production/extraction (e.g., operating machinery, running factories, powering smelters, etc.) or indirectly (e.g., transportation costs). It's a very major factor in agricultural prices, not only in obvious ways, but because taken together, fertilizers and pesticides all by themselves represent something like 3-5% of the world's total fossil fuel consumption annually. That's in addition to the energy costs of operating agricultural machinery and shipping agricultural products absurd distances for packaging, distribution, and retail sale.
Like the tiny jar of honey I saw in my office breakroom this week. The honey itself was produced in New Zealand, it was packaged in the UK, and sold in Pennsylvania. The total cost of the product I held in my hands was probably mostly energy. The actual honey was almost a rounding error.
So when you look at all these various commodities, particularly agricultural prices, you should really account for movements in energy prices, as they make everything more expensive, in and of themselves.
I'd be very interested to see an update to this analysis with such an adjustment.
Could not agree more. Energy is the Master Resource. Humanity is the Ultimate Resource.
I'm very interested in how that's cost-effective. You can keep bees in Pennsylvania.
I assume it has something to do with economies of scale, because if it's not that, I have no earthly idea. Arbitraging labor and/or energy rates is the only thing I can think of.
Or more likely: the actual cost of production (labor, capital investments, operating expenses), is much larger than a rounding error. The statement just doesn't pass the smell test.
If energy was 90% of honey production (at least an order of magnitude difference from a rounding error), we would notice all sorts of energy dependent things:
- Local honey would in general be much cheaper
- Areas with low cost of energy would produce more honey (similar to how aluminum production occurs in Iceland because of cheap energy)
- The price would fluctuate significantly with the price of energy
I just spent more time than I care to admit looking at honey production [1](https://blythewoodbeecompany.com/blogs/news/is-beekeeping-profitable), [2](https://honestbeeltd.com/faqs/how-does-the-logistics-of-migratory-beekeeping-influence-the-overall-production-cost-of-honey), [3](https://financialmodelslab.com/blogs/operating-costs/honey-production) and my very unscientific guess is that energy is 20% of the cost of honey.
Direct, sure. But it's an input for all of the other inputs. That's my point. Equipment manufacturers have energy costs. Labor expenses include laborers paying for energy for themselves. When energy costs go up, everything gets more expensive, not just those budget items that are obviously related to energy.
A comment on Natural Gas. I am not sure how relevant my comment is, as this is just one among many commodities examined, but here goes:
I think US prices were used here. Nat Gas is unlike coal or oil in that it has been impossible to cheaply transport it across oceans for most of the period examined. (That is changing with LNG although that is recent, somewhat expensive, and still scaling up.) Hence prices in Europe and Asia have been much higher than in the US for decades (I think). In this sense, it is not a "commodity" - it is landlocked.
Again, many commodities have been examined here, and this is a bit of quibble. But on the other hand, natural gas is an input into many other prices listed here - particularly agriculture in the US, via the fertilizer production process. So perhaps this has some relevance?
Also worth noting that as LNG capacity scales up, at some point we should reach a "tipping point" where US prices more or less converge on global prices (less the direct cost associated with liquefaction).
Thank you for the article. I enjoyed it.
The timing of the Ehrlich-Simon bet wasn’t random, it was at a time when prices had recently risen sharply. The debate was over whether this was due to exhaustion of resources (Erhlich’s argument) or just short term cyclical factors (Simon’s). Simon won the bet as of course supply of pretty much all commodities vastly increased and we didn’t all starve to death.
Importantly Simon didn’t ever say commodities would always decrease in price from any point, he was well aware of commodity cycles and trivializing the argument to this misses the point. The real lesson of this bet is the abundance one; we can rely on the power of the market to incentivize human innovation to increase supply. This contrasts with what superficially seems the more “sensible” doomster approach of government top down intervention and control but which has led to many disastrous and inhumane policies like Indian sterilization programs, or the Chinese one child policy.
>But folks have pointed out that if the bet had been over a different 10-year window, Ehrlich would have won the bet.
Ehrlich picked what to bet on. He has been embarrassingly wrong about everything forever. But True Believers will never face facts - Doom is always right!
https://www.mattball.org/2024/04/weekend-reading-more-from-not-end-of.html
Ehrlich's population predictions were correct. As noted in the article, he was wrong about some commodities on a ten year basis but correct over a twenty year basis. He proposed more than one scenario including one with the widespread introduction of birth control. It took longer than his book predicts, but it did slow population growth.
He wasn't embarrassingly wrong about everything. In fact, he got a lot of things wright. He just had an inconvenient message, so a lot of effort has gone into smearing him.
People often confuse price spikes with Malthusianism. Economically speaking there is no deleption, there is always supply, reflected by the usage, price, technology. The more we look the more we find and the more we use the more we look.
Was inflation included in your graphs?
Well the earth is finite and at some point we are bound to have depleted most of the useful ressources.
Of course we are going to recycle what can be but the question is how many people can be supported for a given quality of life if all ressources are in circulation.
The universe is bigger than Earth.
Yeah but for now the dream of setting up shop somewhere else is just that.
We already have a hard time shipping stuff to the moon, so it’s not happening in a timeline that doesn’t have to deal with oil scarcity first.
The theory of exhaustible resources (e.g. ores and fuels; but not food) suggests real prices should grow at least as fast as real interest rates:
https://www.economicforces.xyz/p/how-should-we-think-about-the-strategic
This seems somewhat consistent with your data, the modal annual increase in resource price is ~1%, a little lower than real interest rates though (1-4%):
https://fred.stlouisfed.org/series/REAINTRATREARAT10Y
The recent increase might be due in part to the recent increase in interest rates this decade.
Theory and data seem close enough for now that I'm going to stick to the null hypothesis that exhaustible resources grow in (real) price at real interest rates? Though there will still be exogenous shocks and manufacturing innovations that dwarf this long-term trend.
Interesting article. I did not note a mention of (governmental) subsidies that some industries receive (e.g. farming) and I am wondering how much, if any, those subsidies can impact price?
That is really the crux of my question.
a) Looking at the food available years ago, not the premium stuff, just the quality that the average working family would expect; and considering the portion of the family budget that food cost;
b) Can you still get that quality today if you’re willing to pay?
No doubt it would be more expensive than the average food prices today, but in real terms, or in terms of the portion of the family budget, how would prices compare?
Note I asked an AI (alter.systems) that question. It reports that in 1950, 25 to 30% of the family budget went on food, while today it is 10 to 15%. However, to eat high quality whole food meals today, would take it back to a similar 25 to 35% of the budget.
Isn't tracking price more about supply and demand and producers profitability and not the cost of production? It isn't reflected in your oil graph, but for a short period of time during the covid pandemic the price of oil went negative and producers/dealers were having to pay to have their inventory taken off their hands. Surely a extreme case of cost [of production] separating from price.
Perhaps an interesting analysis would be to compare the profitability of the commodity producers compared to the price. Industry consolidation and competitiveness would be another factor.
In that AEI graph its very notable that the highest price increases are in the industries most notorious for private equity funded monopoly roll-ups.
I’m interested to note that precious metals, which are thought of as an inflation hedge, come very close to performing that function over 100+ years.
Here in the UK, we have the impression that American food is much more intensively farmed and processed than in the ‘70s, and also is of much poorer quality. Are there still farmers rearing animals and produce to a similar quality to the average standard 50 or 60 years ago. It would be interesting to compare how these prices have changed over time. I suspect they have held pretty steady in real terms , maybe risen as it became harder to find.
If you are comparing today with the 1950s or 1960s, American food has changed. Chicken used to taste like chicken. Now it has no intrinsic flavor. They did the same thing with pork when they started pushing it as "the other white meat". You can still find chickens that taste like chicken and pork that tastes like pig, but you have to pay a premium for it. It's similar with a lot of produce.
The meat industries are highly industrial and centralized, so a handful of companies make the breeding and feeding decisions. Grain was standardized in the 19th century. Meat was standardized in the 20th. I get the impression Americans hate the taste of food, so if you want food with flavor you have to search for it and pay more for it.
I believe it is just an impression. I have looked at a few things and actually the quality has stayed the same or even increased.
The impression comes from the obesity crisis which is an oversupply/cheapness problem more than anything. Historically humans never had access to so much food for so little effort.
And the cheap stuff is what is calorically dense and very easy to make because industrial processes have become extremely efficient.
Yes, bred or created to be calorie dense, but unfortunately much lower in essential nutrients.
Nutrient density has lowered but that’s just because vegetables are much bigger (they are mostly water in the end). Total nutrient output isn’t different.
The other changes are related to specific crop selection for various reasons.
Unless you are vegan you probably don’t lack much of anything.
The problem is essentially that people have « forgotten » how to eat, for convenience, pleasure and because women are too busy to cook and share/transfer recipes to the next generation.
Excellent info. Would be interested in seeing the total cost of crop subsidies as well as profit to the large entities in crop science as well as farmer net income.
Commodities are raw materials predominantly. And these are processed in chemical engineering type plants. Which lend to manufacturering productivity.
Farming is also a manufacturing process.
Services are people based. People services scale by adding people so little productivity.
However the great service productivity and cousin, white collar productivity came from a quiet silent revolution called the Personal Computer and early software. It so subtle that even brilliant Jack Welch at GE was clueless. But he saw the benefit as he increased productivity by lowering the work force. Because work of secretaries, travel agents, draftsman were all absorbed by the white collar professionals.
Can you clarify what inflation-adjusted means? How are you correcting for "inflation," whatever that might mean?
Very interesting article. My prior assumption would have been that anything that we put any serious effort into producing would have seen large price decreases over this period.
How much do energy costs hold everything else hostage? Everything has to be transported. Most of these things also involve significant use of heavy machinery. Some of the mined/refined metal products are notorious for massive energy consumption, aluminum comes to mind and I'm sure other metals are similar.
I wonder if there is a way to analyze energy costs as a percent of total costs and see how that effects things? I've often wondered, especially with the developments in solar and battery technology, how abundant energy could completely change some of these dynamics. If transportation cost was no issue, how many new locations become profitable? If electricity was basically free how does that change how heavy equipment is used? What kids of machines would be invented that just can't make economic sense right now?
For a longer term view, check out Hackett-Fischer's The Great Wave. European historians have done a lot of research on pricing, but not a lot of it has been published in English. There appears to be a roughly 200 year cycle with perhaps a century of relative price stability alternating with roughly a century of rising prices, but those prices didn't rise uniformly. Technology was less advanced back then, so prices that were based on the value of land - fuel i.e. wood and food - would rise with rents while the price of manufactured goods - i.e. labor - would fall. There's a suggestion of a Malthusian effect with the land-population ratio driving prices, but there has to be a strong political effect to account for the long periods of stability.