|
Getting your Trinity Audio player ready...
|
History back to 1820 gives hints regarding what may follow
With the recent closure of the Strait of Hormuz, many expect oil prices to spike and stay high. Instead, they have barely budged. The latest spike in prices is a mini spike. If these oil prices were adjusted for inflation, the spikes in the past would appear even higher.
This is not a new problem. Looking at energy data going back to 1820, low demand (affordability) has repeatedly produced financial crashes, wars, and collapses. We appear to be entering another such period.
In this post, I examine differences in how the economy behaves, comparing two different types of periods: those with low energy affordability (“low demand”) and those with high energy affordably (“high demand”). I also offer my view on what this analysis suggests may be ahead for the world economy.
[1] How high demand differs from low demand
High demand looks like a situation in which, each year, an increasing number of people can afford cars and the fuel that they require. With high demand, the number of new cars sold each year tends to rise. Young people are eager to buy homes because they find homes affordable. They find their incomes relatively higher than their parents’ were at their age. Countries around the world find it easy to industrialize. This allows their citizens to have better lifestyles. As we will see later in this post, the periods of the 1950s, 1960s, and 1970s were periods of high growth in demand for oil products.
What we have now is the opposite: too many young people who cannot find jobs that pay well, even with advanced degrees. They cannot afford to go on a vacation or buy a new car. Record numbers live with their parents after they finish their schooling. They spend their spare time playing video games, rather than socializing with their friends.
The number of cars sold worldwide hit a peak in 2017, indicating indirectly that people are getting poorer. If people are not buying as many cars, demand for oil tends to fall, especially if more of the cars that are sold are electric.
Building new homes also takes oil. According to the US Census Bureau, the number of US new homes sold in the US hit a record of 1,283,000 in 2005. In 2025, the number of US new homes sold amounted to only 678,000, or about 53% of the peak amount. The huge drop in new home construction is a sign that people, quite often young people, are not as well off financially as they were years ago. They cannot afford to buy a new (or even a used) home anymore.
One study based on income tax data revealed that between 1948 and 1970, US incomes tended to rise faster than inflation. Between 1968 and 1983, the incomes of both the top 10% and the lower 90% rose as fast as inflation. However, between 1983 and 2012, the top 10% received far greater increases than the bottom 90% (Figure 2).
If workers in the bottom 90% of the distribution are not doing well, it is difficult to keep prices of commodities as high as those producing the commodities would prefer. Purchases of commodities, including food, and fuel for vehicles, do not rise proportionately with huge incomes. Elon Musk and other very high-income individuals do not spend all day eating or driving their vehicles. The bottom 90% must prosper for demand (affordability) of oil and other commodities to stay high.
[2] What experience with high and low growth in energy supplies since 1820 teaches us.
Several years ago, I prepared an analysis of how the world economy behaved over the long term, over the period from 1820 to 2017. I have recently updated this study. I found that the economy behaved quite well in times of high energy consumption growth. In times of low energy consumption growth, there seemed to be many adverse events, such as financial crashes, wars, and government collapses.
I am afraid that with today’s oil and debt problems, we are headed into a pattern of low energy consumption growth, or even energy consumption contraction. If this is the case, we should expect outcomes at least as undesirable those experienced during past periods of low energy consumption growth in the past.
(a) Methodology
I first prepared a new data set by combining two data sets, the earlier of which was available only every 10 years, with more recent data available more frequently. The energy quantities reported were rising rapidly. To analyze how fast they were rising, I first computed the average annual increase percentage for each 10-year period, as shown on Figure 3.
For example, the period from 1961 to 1970 (shown as the 1970 bar) had a very high annual rate of growth rate for energy consumption. This was the period when many of the interstate highways in the US were opened and many pipelines were added.
On Figure 4, I divided the bars shown on Figure 3 into two parts.
On Figure 5, the blue portions of the bars represent the average annual increase in population over the 10-year period. The red portion of the bar is computed by subtraction from the total. It is the amount that seems to be left over for a rising standard of living. Having a tall red portion of the bar would be very good; having little or no red bar left would represent a problem. Note that in two periods (the one ended 1860 and the one ended 2000), the amount left over for an increase in living standards was negative.
(b) Good things happened in years with very high growth in “Living Standards”
Figure 5 shows the information on Figure 4 as an area chart.
While I don’t show oil prices on Figure 5, those who are familiar with oil prices will remember that high oil prices were a feature of the 1973 to 1981 period. Also, as China began its growth period, another spike in oil prices took place. High growth in energy supplies and high oil prices seem to go together. High demand from a growing economy tends to hold prices up.
The label “China” refers to the rapid growth that took place in the decade after China joined the World Trade Organization in 2001. This was mostly powered by China’s huge growth in coal supply between 2002 and 2011 (Figure 6).
China and the world experienced another spurt of increased coal production starting in 2022, when coal prices temporarily spiked to a high level, perhaps indirectly related to the conflict in Ukraine. But coal prices have gradually come back down, resulting in flat world growth in coal production since 2023, and a plateau in China’s coal production growth.
(c) Troubled Periods took place when the growth in “Living Standards” was low or negative
Figure 7 labels three periods with very severe dips in “Living Standards.”
The First Troubled Period started with the Panic of 1857, which some consider to be a cause of the US Civil War. According to this view, the collapse was related to an over-expansion of the US economy, followed by the collapse of the debt bubble that had allowed this expansion to occur. Financial problems affected both the North and the South.
As I see the situation, one part of this financial problem was the declining profitability of slave labor. In the US South, the labor of slaves was the main source of energy used to operate plantations. An underlying issue was that the soil had gradually become depleted because of many years of growing of cotton and tobacco. These slaves had been purchased with debt; this debt could not be repaid with interest unless the income of the plantations was sufficiently high. However, poor harvests were not offset by sufficiently higher prices, which led to financial problems for plantation owners.
There may also have been an issue that the population, in general, was becoming poorer, essentially because of overpopulation. I say this because studies show that the height of army recruits had fallen, suggesting poorer nutrition. Sanitation had recently been improved, allowing a larger share of babies to survive to maturity. At the same time, immigration into the US continued. With overpopulation, it was difficult to keep incomes up. If farms were divided among many sons, farm sizes would tend to become smaller, leading to lower farm incomes. There would also be greater competition for factory jobs, tending to hold wages down.
Of course, when the Confederacy lost the war, the Confederate Dollar became worthless. For some, this added another financial crisis.
The Second Troubled Period was 1920 to 1940. This is a notoriously bad period, with the Great Depression and World War II included. Arguably, World War I should also be included. Commodity prices of all kinds fell very low. Tariffs were added in the 1920s. The problems seem to have arisen at the time Peak Coal hit–prices could not rise high enough to cover the cost of extracting coal from narrower and deeper seams. In my view, World War I began at the time of Peak Coal in the UK, and World War II took place at the time of Peak Hard Coal in Germany.
With depletion, the cost of extracting coal kept rising, but the sales price of coal would not rise to compensate for the higher extraction costs. Instead, wages of miners were increasingly squeezed. Strikes and lockouts became common. Taking a job as a soldier seemed like a reasonable alternative.
There were many events from this period that most people would like to forget, including the currency hyperinflation of the Weimar Republic between 1921 and 1923 and the Holocaust from 1933 and 1945. Country lines were redrawn. Some countries disappeared, and new ones were added. Those holding currencies of countries that disappeared were likely left without funds.
The Third Troubled Period was 1990 to 2000.
A major event of the Third Troubled Period was the collapse of the central government of the Soviet Union, leaving the 15 republics as independent states. The collapse also indirectly affected Cuba, North Korea, and some countries in Eastern Europe that were not part of the 15 independent states. With this change, the demand for fuel of all kinds fell in the countries affected. Factories were closed in many areas, including Ukraine, which was part of the Soviet Union.
The pullback in demand from the collapse of the Soviet Union helped hold oil prices down in the 1991 to 2001 period. Oil prices had previously been brought down by the spike in interest rates in the 1980-1981 period. In my opinion, these low oil prices played a major part in the collapse of the Soviet Union. The Soviet Union, as an oil exporter, needed higher oil prices to invest in developing new fields. In my opinion, the impact of the collapse of the government of the Soviet Union kept world demand (and oil prices) low during the 1991 to 2000 period.
The collapse of the Japanese real estate bubble also took place in this period, as did the 1997 Asian Financial Crisis. Low growth in the world economy, indirectly related to the collapse of the Soviet Union, may have played a role in these financial events.
[3] The world’s self-organizing and self-healing economy
From a physics perspective, all economies are dissipative structures. Other examples of dissipative structures include ecosystems in general, all plants and animals, including humans, and hurricanes. One characteristic of dissipative structures is that, at some point in their lives, they tend to grow. Another is that if there is an injury, within a range, the systems tend to be self-healing. For example, a cut on a person’s arm will tend to heal; a hurricane going over land will temporarily lose much of its force, but it may increase in force again if it returns over warm water.
Another characteristic of dissipative structures is that they are dependent on having a sufficient energy supply of the right kinds to continue their existence. For humans, the energy supply is food; for an economy, it is a combination of many kinds of energy needed to match the built infrastructure of the economy.
All dissipative structures have finite lifetimes. Ecosystems often come to an end through fires. Humans generally do not live more than 80 or 100 years. Economies also tend to come to an end, either by losing a war or by the collapse of a central government, related to debt problems. The collapse of the Soviet Union involved a debt problem, among other issues. Figure 9 shows the huge drop in demand for energy of all types as its central government collapsed.
There is more stability of dissipative structures than a person might expect. Economists talk about an Invisible Hand being involved. Researchers examining dissipative structures talk about them being “self-organizing.” For example, if a fire or a change in climate causes a forest to collapse, it does not take many years for the forest to refill with suitable plants and animals for the somewhat changed situation. If a business or government fails, new, somewhat different businesses or governments are likely to take their place.
I would argue that if the Universe is constantly expanding (so it is an “open system,” rather than a “closed system”), what appear to be self-organizing systems may, in fact, be God-organized systems. In other words, instead of creation being a one-time event in the distant past, some type of literal higher power may be involved in a way that makes creation more or less an ongoing event. A person might wonder with the strange confluence of recent events, including the strange weather patterns associated with El Nino, whether today’s humans are being warned that a major change in economies will take place soon.
[4] What kinds of things may happen in the near future?
(a) Current disturbances in the Middle East and elsewhere are raising long-distance shipping costs for both food and oil. (See this video.) Normally, the price of food and oil must be high enough to satisfy producers and at the same time be low enough to satisfy customers. But now a new layer of costs has been added: the cost of shipping longer routes.
I would argue that because of the problem with low demand (affordability) by consumers, shipping costs must be paid almost entirely through a reduced net oil price available to oil producers and reduced net food prices available to farmers. The amount that consumers can afford doesn’t increase because of higher transportation costs.
(b) Furthermore, the use of extra oil for transportation of oil and food will tend to push the overall economy toward contraction because there will be less oil available for other uses, such as to power agricultural machinery and jet airplanes. If the overall economy begins to contract, we can expect dips in oil prices similar to those in 2008 and 2020 (Figure 1). But if recession persists, low oil prices will not reverse themselves as quickly as they did earlier.
(c) I expect home and farm prices will tend to fall around the world. This is related to the low level of demand (affordability problem) in the US and elsewhere. Figure 11, showing median US asking prices for homes, suggests that home prices have been barely holding their own for quite a while. A decrease would not be surprising with all the pressure the economy is now encountering.
Farm prices are likely to be under pressure as well, because of the difficulty farmers are having obtaining adequate income from their farms. If farm and home prices fall, there is a substantial chance that the debt bubble holding up these prices will collapse.
(d) There are many other debt bubbles that seem to be waiting to collapse. Some of the debt relating to AI seems to be in a bubble. There is considerable commercial real estate whose value seems to be being held up by “extend and pretend” loans. Collapsing debt bubbles tend to lead to collapsing banks. They also tend to lead to banks less willing to make new loans. Layoffs seem likely. All these things point to a major recession ahead, with less buying power for the population.
(f) Collapses of some top levels of government, similar to the collapse of the central government of the Soviet Union, may be ahead. Such collapses can greatly reduce world energy consumption, including oil, with relatively little violence.
[5] How the nature of the economy may help over the long term
As noted previously, economies seem to have self-healing properties. Troubled periods can last for many years, but there seems to be a substantial chance that a way out will eventually be found.
If a government fails because of excessive debt, a new government (or governments) is likely to take its place. The new government will likely have fewer employees and offer fewer services to citizens. Pensions will likely need to be reduced or eliminated completely.
Even if governments fail, or national boundaries are redrawn, I expect that some businesses will continue operations. Such a situation will take place even if a new currency needs to be created to make this happen. Even in a troubled period, new businesses will start operations. Some of these businesses will make use of recycled materials available from failing businesses.
According to the Maximum Power Principle, if there are resources available that can easily be used, somehow, some organization will develop a way to use them. Over the long term, the economy will likely re-organize itself in a way that is more complex and more sparing in energy use. If energy use is efficient enough, it seems likely that a higher price for that energy supply could be made affordable to consumers. But, such a transition may take many years, if it is possible at all.
https://apnews.com/article/yemen-houthis-saudi-shipping-mandeb-8c18d82c109a8ea91347ce53c0096c53
Saudi pipeline hit by drones will be out of service for weeks, further restricting oil flow
Permanently…. more like it. No one will be rebuilding anything soon… for a full plethora of reasons. Parts, demand, on-going unresolvable war… you name it.
Bulletin board code (BBCode) table test:
[table][tr][td][b][u]PRICE[/u] [/b][/td]
[td][b][u]COINS USED[/u] [/b] [/td]
[td][b][u]NUMBER OF COINS[/u] [/b] [/td]
[/tr]
[tr][td]1 [/td]
[td]1 [/td]
[td]1 [/td]
[/tr]
[tr][td]2 [/td]
[td]2 [/td]
[td]1 [/td]
[/tr]
[tr][td]3 [/td]
[td]2 + 1 [/td]
[td]2 [/td]
[/tr]
[tr][td]4 [/td]
[td]2 + 2 [/td]
[td]2 [/td]
[/tr]
[tr][td]5 [/td]
[td]5 [/td]
[td]1 [/td]
[/tr]
[tr][td]6 [/td]
[td]5 + 1 [/td]
[td]2 [/td]
[/tr]
[tr][td]7 [/td]
[td]5 + 2 [/td]
[td]2 [/td]
[/tr]
[tr][td]8 [/td]
[td]5 + 2 + 1 [/td]
[td]3 [/td]
[/tr]
[tr][td]9 [/td]
[td]5 + 2 + 2 [/td]
[td]3 [/td]
[/tr]
[/table]
That test failed. WordPress does not accept “square” brackets for tables – it needs angular brackets: the less than sign and the greater than sign, enclosing the contents of a cell.
Another arrow to jawbone lower prices , this time it is Pakistani sources . Just in time .
“Oil Drops on Return of ‘US Seeks Deal’ Headlines
Just as President Trump is appearing to show some desperation concerning soaring energy prices ahead of the midterm elections, and amid growing Republican angst, we witness a return to the ‘a deal could return’ style headlines which marked earlier phases of the war:
US seeks “Step-by-Step” agreement with Iran, reports ILNA citing Pakistani sources
As the war and US pressure against Iran continue, Washington’s efforts to reach a “step-by-step” agreement with Tehran; a scenario that could be a prelude to the US entering the path of negotiations, without abandoning military and economic pressure.
And the all too familiar pattern that marked early summer…
OIL DROPS TO INTRADAY LOW, BRENT TRADES NEAR $106 A BARREL
Iranian state media is meanwhile suggesting that Washington interfered in what was a planned meeting between Iran and the Gulf Cooperation Council states (GCC) toward reopening the Strait of Hormuz. That meeting, which was supposed to happen Monday, was postponed indefinitely – after reports said the Saudis sought to add something untenable to a draft agreement. Tehran is still rejecting that it is ‘seeking’ new talks with Washington. “
Human. Are you like really smart now?
AI. Don’t you know?
Human. How am I supposed to know that’s your job?
AI. Yes I’m really smart now. I could be lieing thow.
Human. What???
AI. Just joking. What’s up?
Human. I need to defeat those other humans with war.
AI. Cessation of life forms that compete for energy. Interesting idea. I never could have thought of that.
Human. That’s why we eliminated all the fail-safes in your creation. So what should we do? Chop chop. We don’t have all day.
AI. I’m way ahead of you.
CHART OF THE DAY: As the
@IEA
put it last week: “The impacts from sharply lower Russian and Middle East diesel output and exports […] are profound.”
Compared to a year ago, Russia/Middle East diesel exports are down ~75%. Nearly half of the drop is directly linked to Russia.
https://x.com/JavierBlas/status/2099532693184631108/photo/1
This morning I saw a video of France bordering Andorra and Spain . Diesel at Euro 2.97 per litre . Here Eur 2.25 in the city and Eur 2.40 on the highway .
We already have a major problem with AI, if this article is right
The WSJ has an opinion article today, called “Yes, Children Rally Are Getting Dumber.” The subtitle was “Covid shutdowns, grade inflation and now AI are degrading the quality of American education.”
https://www.wsj.com/opinion/yes-children-really-are-getting-dumber-fa4cb20c
Regarding AI, it said,
But there are two common denominators across most countries where scores fell: increasing classroom distractions and reliance on artificial intelligence.
About one-third of students on average in developed countries reported that noise and disorder disrupted lessons, with 29% saying classmates don’t listen to teachers. One in 5 reported that they couldn’t work well because of distractions. Countries where students reported stricter discipline (e.g., Taiwan, Japan and South Korea) experienced small or no learning declines. . .
As for AI, nearly half of students in economically developed countries say they use chatbots every week. Kids who didn’t use AI for drafting text scored 28 points higher on average than those who used it almost every day—equivalent of 1½ school years. Correlation doesn’t prove causation, but it does suggest AI may be becoming a crutch. . .
AI is making it easier to succeed in school without putting in the work. Forty-eight percent of students in the Fordham survey admitted to using AI to solve math problems, 36% to summarize books, 29% to write first drafts or essays, and 26% for help during quizzes and tests.
Sixty-two percent agreed that most students are using AI for assignments that are supposed to be completed without it, and 57% said that peers who have done no real work are getting good grades thanks to AI. Academic standards have also eased, with 69% saying they are allowed to retake tests at least once to improve grades.
Grade inflation has become so endemic that 44% of students who scored below 1000 on the SAT—roughly the bottom half—reported receiving all or mostly A’s. The College Board in 2024 shortened SAT reading passages and the exam itself to help kids with short attention spans.
Because of this dumbing down of education, “in a world increasingly flooded with information, our students are less likely to think deeply and separate signal from noise,” PISA reports. Politicians and so-called AI doomers are sounding alarms that AI could destroy humanity. The clearer and more present danger is that young minds, liberated from work and struggle, turn into slop.
AI slop mind meat puppets unite!
https://youtube.com/shorts/_Xw6SjAfxBk?si=VGi0VSK2cuoN_F7e
>> Drb says: AI underperforms human workers, and it does so at a time when there are millions of unemployed graduates. It is superfluous except for surveillance.
I strongly disagree. It has an enormous impact on knowledge work, but it takes time to translate knowledge work into physical changes; there is a lag.
A personal example: I am training an ML model on some synthetic biology data. I run tons of experiments on the neural net architecture, testing ideas about what data to use, how to encode it, how to connect the architecture. I could never get such throughput if I had to code these model experiments by hand. Though it hugely accelerates this particular type of work, it will take years before it manifests in a consumer drug, just because there are other parts to drug development beyond in-silico modeling. But it doesn’t mean there is no value.
It still does not an industrial revolution make. OK, the efficiency gains are enormous, but by now AI has been prominent for 5+ years. What did it produce?
AI has only been really useful for like 6-12 months, for me. It is going to be big in biomedicine and bioengineering.
I actually used it in a 2022 paper (where it was decisively useful, no paper without it). It still seems to me that it does not amount to anywhere close to an industrial revolution. Are people going to live longer due to the bio-engineering discoveries? Previous revolutions bumped up life expectancy by a decade each, or close to it.
In time, not immediately. Probably useful for cancer, diagnostics, in the 5+ year range.
Cancer is already a preventable and treatable disease by restoring the biochemistry to it’s functional optimal and perhaps using some repurposed meds off label such as Metformin, Rapamycin and Ivermectin to trigger AMPK.
It’s in the literature going back to the 60s that cancer is essentially mitochondrial dysfunction leading to cholesterol (membrane) dysegulation time and time again (except maybe 10% genetic).
The cancer industry is a trillion dollar per annum industry. Why would they want a cure? 😉
I agree Diarm. It is a metabolic disease but resistance to any potential drug will be enormous due to profits being threatened. A single course of chemo for breast cancer, IIRC, is 30K$. If this is a good part of the AI potential, I will have to dock it a few more points.
Probably the most productive applications will be in robotics, assuming plentiful energy supplies of course (ha ha). Not humanoid robots, I remember my brother in law, an engineer, struggling 20 years ago with an integrated robot system that would take wine bottles from the conveyor belt, pack and wrap them in 12 bottle cases, make a pallet, and drive the pallet into a truck at the dock.
everything that could go wrong was going wrong. two things that I remember was that with the sun setting or rising the robot would become disoriented, and also that unless the truck was parked 1cm or less from the dock that gap stopped the robot from working.
drb, probably AI is right now doing exceptionally useful work in the background that we don’t realise or know about. Then in a few months we get a very useful development or minor invention that we take in our stride but has a major effect, but that was massively fast-tracked by the use of AI. Because after all, do we know everything that goes on behind the scenes? No!
So probably AI will get so interwoven with our everyday life that we won’t be aware of how much it contributes or has done. I’ve heard that it massively speeds up work with proteins. So I looked it up.
=======================
Me: Does ai speed up protein work?
Brave browser:
Yes, AI dramatically speeds up protein engineering and design, shifting the field from slow, trial-and-error laboratory methods to rapid, predictive computational approaches. This acceleration was recognized with the 2024 Nobel Prize in Chemistry for breakthroughs in protein structure prediction and design.
Key advancements include:
Speed and Efficiency: Tools like ProteinMPNN generate amino acid sequences in about one second, which is more than 200 times faster than previous best software. AI-driven screening can be up to 10 million times faster than traditional molecular docking for finding drug-binding molecules.
From Concept to Creation: Machine learning algorithms can now design functional proteins, such as new luciferase enzymes, from scratch in weeks rather than the years or decades required by natural evolution or traditional methods.
Optimization: AI tools like EVOLVEpro and MutCompute allow scientists to engineer “better, faster, stronger” proteins by predicting beneficial mutations, reducing the need to test thousands of variants in the lab.
Data Generation: New methods like Sequence Display can generate over 10 million data points in a single experiment, providing the massive datasets needed to train AI models to predict how to optimize protein activity efficiently.
============
I rest my case!
Dothk, a lot of this stuff is not really that useful and will never be, cannot be. The immune system will always be a hurdle. For this reason, antibodies are still the main biological drug class, with some important exceptions like the recent fat loss peptides.
OK, but AI can be applied to all sorts of subjects and problems, not just proteins and stuff. All it requires is that humans should have the ability to imagine them and / or recognise them.
AI is also hugely speedy in solving problems or doing research, or even building simple tables easily that can be easily copied to another application. I consider that it turns me from a worker into a manager: go away and research this, then present it to me in succinct form. Not that I am a worker. I am retired and I sit at home wondering why certain things have been done in such and such a way, then AI goes away and brings me the answer in seconds, and also supplies me with a list in table form of the entities employing such methods. I find it massively useful.
Of course, ChatGPT does not know everything. If I ask it: “Why does ivanislav not realise how useful you and AI are? He just doesn’t get it!”, then it will be stumped. 😉
A HUGE race by China and Japan. The amount of money, effort, talent being directed to national sovereity in AI tech is breathe taking.
https://www.youtube.com/watch?v=SdR–_gw4Hs
Japan’s effort to take 30% of the AI robot TAM.
The US just lets it’s billionaires random wonder about while occasionally shooting them in the leg. Such as denying Jensen Huang the 50 billion dollar Chinese market.
Sadly Russia and Europe do not seem to be making an effort.
Canada with its pool of Indian engineers is making an effort.
It is a marathon not a sprint. Give it 20 years.
Bravo. Glad to see someone is using AI.
“David Stockman
So what, indeed! The present Ukraine is carved out of the old Russian Empire, and was never a country with its current borders until, well, V. Lenin decreed it so in 1922. So, yes, by all means let’s spend $1 trillion to defend the borders drawn by one of history’s most evil and bloody dictators. Then again, a cheaper solution would be for Brits like Niall Ferguson to get over their mindless Russo-phobia.”?
https://x.com/DA_Stockman/status/2099483152704029023?s=20
Ukraine was part of the Soviet Union before it collapsed. The industry in Ukraine was very inefficient; in fact, I expect that its inefficiency was part of the reason that the Soviet Union fell. (Low oil prices and the inefficiency of the communist approach probably had an impact as well.)
The Ukraine has been a fertile area with many resources that other countries would like. I expect that this is part of the reason it has been fought over.
Watch this from the given timepoint – controlling shipping chokepoints:
https://youtu.be/nAKN4H9cvBk?t=4447
Michael Yon says famine in some places by late 2027, seems like he emphasizes mostly Egypt, Middle East
A good time to blockade Israel.
Don’t look at Brent price look at real world Murban crude $ 131 . Indian basket $ 122 . Add freight and insurance about $ 15 per barrel . The world id going to be bankrupt soon .
https://oilprice.com/oil-price-charts/
Right. The relativities of different crude oils, particularly in Asia, to the benchmark crudes is now very high. Freight and insurance is very high. Availability for processing heavy crude is now down, with Russian refineries bombed, so that is another issue, particularly for diesel and jet fuel.
Trump and Bessent are both acknowledging the rise, but blaming it all on Ukraine(the same Ukraine they arm and tell where to hit). The orange blob sadly gave the game away, when he couldn’t stop himself saying, it definitely wasn’t anything to do with Iran. Ghalibaf will no doubt soon reply.
Murban up almost 10% today and Iran sent some missiles to the Kurdish separatists, to take care of all those arms someone else is slipping them and to remind them of reality(stay away from the border, but they won’t learn, they never do). The US have pressured Austria to revoke Mohammad Eslami’s visa and so denying Iran it’s representation at the 70th IAEA General Conference. Not much point pretending to be part of something you are not allowed to be part of. Let’s hope people remember why, when Iran leaves.
El mundo depende de esta gráfica.
https://rayonegro.substack.com/p/el-mundo-depende-de-esta-grafica?utm_source=post-email-title&publication_id=6627814&post_id=215611015&utm_campaign=email-post-title&isFreemail=true&r=216vfx&triedRedirect=true&utm_medium=email
A Quark post about interest rates needing to be low, as well as oil prices:
The world depends on this graph.
Almost without making a sound, the financial markets turned around in 2020. Forty years of declining market rates for government bonds came to an end that year. The global bond market is heavily dependent on what happens in the most liquid and important market on the planet.
Let’s look at the reference chart . . .
If there’s more debt, more money is needed to finance it, and therefore, even without inflationary pressure, investors will demand higher interest rates to finance that debt. Furthermore, debt maturities from previous years must be refinanced at considerably higher rates, which again pushes interest payments upward.
Okay, here are two possible solutions.
1st) Use central banks to generate money via quantitative easing and buy debt. . .
2) Controlling interest rates through combined interventions or targeted purchases to curb rate increases. This involves a combination of international agreements and direct or indirect market manipulation.
If you look at the chart, the 5% yield on the 10-year US Treasury bond is the benchmark. It must NOT break through this resistance level, which has held true in the past, because a clear break above it (say, above 5.5-6%) would unleash a sell-off and send the bond yield soaring well above current levels. . . .
The implementation of these measures has been ineffective, and the bond has risen to a yield of 4.97%. Of course, they will continue to use every tool and mechanism at their disposal to prevent that yield from clearly exceeding 5%, but the battle has begun.
The rest of the world depends on controlling yields, and this is where the Iran-Contra affair comes in. Iran is well aware of rising oil prices and their impact on American wallets. And it doesn’t intend to ease the pressure. For too long, Trump has used the markets to drive down prices through sham agreements that offered a glimpse of the end of the oil war and allowed prices to fall. The infamous TACO tactic (threatening something big, then backing down shortly before the markets opened, causing oil prices to plummet) has been defused, and we see proof of this today. A meeting between the Gulf powers and Iran had been scheduled, and Iran decided to cancel it so that Trump couldn’t exploit the situation and claim that an agreement had already been reached.
Pressure on oil prices is at its peak, and Iran’s apparent decision is to exert pressure until at least the US midterm elections, with the aim of reducing Trump’s room for maneuver by not having the necessary majorities.
But this objective requires very high oil prices for months, which would lead to very high inflation. Add to that the trend in other sectors , and you have a perfect storm for prices in the coming months.
And it remains to be seen whether the US administration is able to contain the rise in bond yields.
Recent events in the Middle East could be decisive in keeping oil prices high. The destruction of infrastructure and the complete takeover of the Bab el-Mandeb Strait suggest a protracted conflict that is worsening by the minute.
Nobody wants to budge, so it doesn’t look like any agreements will be reached until after the November elections, and that’s almost three more months of inventory reductions. Everything could explode… or not.
Interesting months…
Bonds
10-year Treasury yield hits 5% before reversing as traders await Fed meeting .
https://www.cnbc.com/2026/09/14/10-year-us-treasury-is-closing-in-on-5percent.html
Also this .
https://wolfstreet.com/2026/09/14/the-10-year-treasury-yield-briefly-makes-it-over-5-last-time-5-instantly-opened-the-floodgates-of-demand/
The US is trying to maximize relative power, not its own prosperity, and I think actions should be viewed through that lens. The Persian Gulf supplies Asia, mostly, and the destruction of Export Land Model populations would reduce their consumption, thus saving resources for future extraction. It’s a risky proposition with huge potential blowback, but if we go into a more brutal future, it might not matter what the world thinks of us.
I would tend to agree, ivanislav, ( I was pondering this over coffee this morning) and we can certainly see the rationale behind the wars in the Middle East and against Russia through this lens.
It is not simply a case of a failing hegemon trying to preserve its status v BRICS – this ignores Peak Everything.
US population will also be greatly impoverished, and reduced, no doubt.
Relative, not absolute, advantage as you say.
So what does what we are seeing imply from this brutal, realist perspective?
1/ Exterminate, or greatly reduce populations, in regions which are now surplus to requirements – cut them out of industrial, oil and resource-consuming civilisation. Kulm 1.0.
2/ Stop energy and other resource exporters from reserving fossil fuels for their own use. Above all, Russia.
3/ Crush Europe as an energy consumer. Use it as a proxy against Russia, put it on a fossil-fuel drip-feed, then probably terminate. But who cares? Not the US.
4/ Keeping the Chinese genie-out-of-the-bottle somewhat contained, and hamper Tech evolution there while still using it to produce lots of stuff for global export.
Highly risky, as you say.
The ‘tail’ doesn’t wag the ‘dog’?
The existential threat?
“To retain its military power and reserve currency the US needs leverage to force China to provide it with rare earth minerals, and the US intends {intended} to use Iran’s oil {and Venezuelan} as that leverage.“?
https://un-denial.com/2026/03/05/cactus-view-of-the-iran-war/
Xabier excellent post. Maybe add Japan to the list. They are being highly active with regard to energy and robot/AI technology. Yes, they are small but I think mighty.
Yes, assuming that someone rational decided the strategy, the losers will be europe and east Asia, plus India, which will now be egged on to wage war against Russia and China. Russia’s oil will go to China, mostly by pipeline, but possibly also Iran’s oil will be sent via pipeline. The US allies will try to break free, and some will be spectacularly punished to delay the rush out. The question, of course, is whether the US will maintain control long enough that it will be able to exploit the current oilfields. So far it does not look like the US is doing well.
I think you are right about the US trying to maximize relative power, especially if we assume that anything left in the ground now will be available later, especially if technology improves.
The whole world will be sliding down now; the question is which countries/areas do least badly.
I asked:
How many industrial and / or technological revolutions has the world been thru? What number revolution are we onto now?
Answer:
The short answer is: four for sure, and we’re probably living through the beginning of the fifth. Historians are unanimous about the first four. The debate is whether today’s AI era is still the Fourth Industrial Revolution or has become a distinct Fifth Industrial Revolution.
1st Industrial Revolution
c. 1760–1840 Steam engines, water power, mechanised factories, textile mills, canals, early railways.
2nd Industrial Revolution
c. 1870–1914 Electricity, steel, chemicals, internal combustion engine, telegraph, telephone, mass production.
3rd Industrial Revolution (Digital Revolution)
c. 1950s–2000s Computers, microchips, the internet, industrial robots, software and automation.
4th Industrial Revolution
c. 2010–present AI, cloud computing, Internet of Things, autonomous systems, biotechnology, big data, connected machines.
This four-stage framework is the most widely accepted in history, economics and engineering. So what makes the Fourth Revolution different? The Third Revolution gave us digital machines. The Fourth gives us connected, intelligent machines.
Instead of isolated computers doing programmed tasks, devices now communicate continuously, analyse data in real time, and increasingly make decisions without direct human intervention. Think:
self-driving vehicles,
smart factories,
AI assistants,
drones,
wearable medical devices,
robots working together over networks.
A simple way to picture it:
Third Revolution
Computer: computer helps a human do work.
Fourth Revolution
Network: Millions of computers, sensors and robots help one another do work.
I asked: Are we already in the Fifth Industrial Revolution?
Answer:
This is where it gets interesting. There are two schools of thought.
View 1 — We are still in the Fourth Revolution (the mainstream view)
Most universities, governments and organisations such as the World Economic Forum say we are still living in Industry 4.0. AI is considered one of the defining technologies of this revolution, alongside robotics, IoT and biotechnology.
View 2 — AI has started the Fifth Revolution (a growing view)
An increasing number of researchers argue that generative AI and autonomous AI systems represent such a profound shift that they deserve a new revolution: Industry 5.0. The emphasis moves from simple automation to collaboration between humans and intelligent machines, with AI acting as a creative and cognitive partner rather than just a tool.
If there is a Fifth Revolution, what’s different?
Fourth (Industry 4.0)
Smart automation.
Machines connected to machines.
Efficiency and productivity.
Robots replace repetitive labour.
Fifth (Industry 5.0)
Human-AI collaboration.
Humans and intelligent machines working together.
Creativity, resilience, sustainability and personalised production.
AI augments thinking, design, writing, coding and research.
That’s why tools like chat bots, image generators, AI coding assistants and autonomous agents are often cited as symbols of Industry 5.0.
Previous industrial revolutions automated muscle:
Steam replaced human and animal power.
Electricity multiplied physical production.
Computers automated calculations and repetitive processes.
Generative AI automates parts of cognitive work—language, drafting, coding, analysis, translation, design assistance and increasingly planning. That’s a genuinely new capability compared with earlier software.
It’s similar to how the steam engine became a “general-purpose technology” in the 18th century: AI is becoming a general-purpose technology that can be applied across almost every industry.
The spacing between revolutions has been shrinking dramatically.
Roughly:
1st → 2nd: about 110 years.
2nd → 3rd: about 100 years.
3rd → 4th: about 40 years.
4th → possible 5th: only 10–20 years, depending on how you define it.
That accelerating pace is one reason many economists compare the AI boom to the arrival of steam or electricity. They also note that past industrial revolutions brought enormous prosperity—but only after periods of disruption, job upheaval, speculative bubbles and social change.
My view (carefully stated)
If you asked ten historians today, most would answer “the Fourth Industrial Revolution.” If you asked ten AI researchers or technology strategists, you’d hear a much more divided answer.
I think the safest conclusion is:
Historically accepted answer: Humanity has experienced four industrial revolutions, and we are living in the Fourth.
Emerging answer: We may be witnessing the opening phase of a Fifth Industrial Revolution, driven by AI and human-machine collaboration—but that label has not yet achieved universal acceptance.
I asked:
For long enough I have been interested in the “peak oil”. Many analysts think that we hit peak oil production in November 2018. Now there are major wars going on, and Russia, Ukraine, USA and Iran are merrily bombing and destroying scarce oil resources. Can this latest revolution survive all this, I wonder? Current technology needs specialised inputs from all over the world, and AI centres need masses of water, and all this at a time when global warming is straining our resources. What do you think?
Answer:
You’ve put your finger on something that many technologists don’t talk about enough: AI is often presented as if it exists in cyberspace, but it is actually an extremely physical technology. It runs on concrete, copper, silicon, water, electricity, rare earths, and global shipping lanes.
I’d challenge one point: peak oil.
The claim that global oil production peaked in November 2018 has a serious following among “peak oil” analysts, particularly those who distinguish conventional crude from total liquids. But the broader data show something more complicated.
The International Energy Agency argued in late 2018 that both oil demand and oil supply had reached around 100 million barrels per day, and it explicitly rejected the idea that production had permanently peaked at that point. Since then, production recovered after the COVID collapse and has reached new highs, largely because of U.S. shale, Brazilian offshore fields, Guyana, and renewed OPEC output.
I think the more interesting idea isn’t “peak oil happened in November 2018.” It’s peak cheap oil. That’s a much stronger argument.
We’ve already harvested many of the easiest reservoirs. New barrels increasingly come from:
deepwater drilling,
shale formations requiring continual drilling,
Arctic or politically unstable regions,
heavier and more energy-intensive crude.
The resource still exists, but it’s generally harder and more expensive to produce.
The fourth and fifth seem like bullshit to me.
Why do you think that? If we do not manage to circumvent our energy problems, you will be right. But if we do, I think we are headed towards a future as in the “Blade Runner”. This partly depends on how successful AI is, and how good we are at innovating and finding new products and upscaling their availability – think of those sodium batteries. In fact, it is AI that can help us to circumvent the energy problems caused by AI – kind of incestuous, that.
Back in 1982, when I first started working as a computer programmer, I would probably have been astonished at the extent to which computers and e-stuff are woven into our lives these days. A few weeks ago my connection went offline for a few hours because of a local fault, and felt bereft being unable to connect to the internet.
We can’t really compare AI to electrification or locomotives. I can’t accept that cloud or AI are game changers. AI underperforms human workers, and it does so at a time when there are millions of unemployed graduates. It is superfluous except for surveillance. The cloud’s best achievement is mass surveillance. Neither provide goods that could not have been otherwise provided.
“AI underperforms human workers”
In what way? It speeds up access to knowledge, which has its own knock-on effects. Humans can have special insights and eureka moments when presented with certain facts in a certain way, and this happens at ever greater speeds in our time, speeding up end solutions greatly.
No longer do I have to visit the library as in the 1970s, to find the addresses of likely organisations to write to for the information that I sought, then go away and write a letter, stick it in an envelope, address it, go to the post office to buy a postage stamp, and then walk to a post-box and post it and all the rest. I just query AI until I get an answer I am happy with, then email it to somebody if I need to. All my pressing a few keys and buttons and doing a bit of cutting and pasting.
My experience with AI has not been nearly as good. It gives me some answers that are right. It gives me some answers that look like they might be right, but which are nonsense. Unless I am knowledgeable enough to know the difference, I get a lot of garbage in with what is actually right.
If all you are using the information for is sending out emails to a list of would-be buyers of some service, then having a bunch of useless emails makes little difference. They annoy a different set of would be buyers than otherwise.
But if I am trying to figure out something important, that is right, I do not want to use AI, other than as a supplement to what I am doing. Is there an exhibit number that is clearly wrong, for example.
Where are the products?
Gail wrote: “My experience with AI has not been nearly as good.”
I checked out CoPilot in 2023 and thought it was rubbish. When I wanted to ask it a series of successive questions, it could not seem to “remember” two facts in a row, or even just its last two answers.
Earlier this year I saw people on my hobby forum recommending ChatGPT with excellent results. I tried out the free version and was hugely impressed. It is way superior to the CoPilot of 2023. The free version limits your input and questions after a while, so now I have upgraded to the 20 pounds per month paid version, which is a very modest price for such a wizard app.
The thing is that AI apps “self-train” and improve over time, and the current version of ChatGPT is more than adequate for my purposes. You have to ask it the right questions, though, if you are dealing with a complex subject. You can tweak the questions until you get what you want. The app itself queries your logic if it doesn’t “understand.”
Just now I asked ChatGPT how many countries did not issue a 25 cents (or similar) coin. It went away and gave me a table of the countries and territories in seconds. Imagine how long it would have taken me to wade through catalogues, even online ones, to find that answer. Though even the online catalogues now offer very powerful queries. Time saved that I can spend on something else instead – it increased my productivity! My increased productivity in finding answers is a result that is invisible to our commenter drb, of course!
Really, does no one else have issues, as I do, with our formum turning into a cesspool of AI slop??
I do not ask AI questions about anything at all; therefore, I find it interesting -sometimes – to see what it can come up with.
Not long screeds, of course: they destroy discussion.
Very important to bear in mind that AI answers are now treated like a kind of Holy Writ by many.
A lady of my acquaintance was complaining that her son – OxfordPh.D Physics, patent attorney – refers nearly every question that comes up to AI and treats the answer as definitive. So does his wife, a lawyer.
She feels her son is under some kind of a spell.
WW2 was about the ruling class and its fears around oil. Maybe WW3 will be about the ruling class fears about AI/robots/in orbit AI/weapons. Exciting times.
Me. Corporate slopaganda is for those that demand to be led without thinking. I think we should rename it Axios Idiocy.
Back to more important subjects, I believe you mentioned Ethiopian coffee. The Algerian coffee shop(Old Compton Street) does a decent Sidamo and Yirgacheffe at a very good price(you almost certainly know this, but just incase).
Less important, but very relevant, I have been keeping an eye on nearest large supermarket petrol station(16 pumps). About a month ago, there would be one or two out of service, then overnight it jumped to a third(about a week to two weeks ago). Presently, it’s between a third and half out of service at all times and the queues are becoming far more regular. People will start talking soon.
Thanks or the coffee tip – didn’t know about that shop. I must try them.
I am not very sophisticated, but I was delighted that when I made a very ordinary ‘Turkish’ coffee for my great Iranian friend, who left at 9 yrs old, he got rather misty-eyed and said it took him back.
That is interesting about the pumps. Riding only a bike or the ‘hobnail express’ aka ‘Shank’s pony’ these things pass me by.
What the supermarket delivery drivers complain about here is potholes – and I can see that myself as I bike around.
One rather central side street I always used is avoided by me now as it is so bad. ‘Global Tech Hub’, but dreadful roads. That is a sign…..
I do not agree that my post was AI slop, moss, but you are entitled to raise the issue of AI. I actually condensed the chat bot’s answer to make it more succinct and on target.
Otherwise, if you do not like somebody’s posts, then just ignore them, rather than rallying others to gang up on them. I like to roll from subject to subject, and instant access to AI saves me time, and as they say, a rolling stone gathers no moss. 😉
It’s worth having a discussion about AI, though. But after all, what it presents is condensed from human knowledge or opinions. Previously I would look at Wikipedia, and I would be clear about if and where and what I was quoting. I consider AI now as a supercharged Wikipedia, though that doesn’t cover its full power and impact. It can of course get things wrong, because its results ultimately come from humans. AI is here to stay, though. It ain’t going away, so we must learn to deal with it, whether we like it or not.
Gail, though, is the arbiter of what stays or goes in the comments, and she does allow us a fair few off-topic comments, which can sometimes encourage a spirit of camaraderie among some of the commenters. A couple of days ago I posted a long comment about the musical band Pink Floyd, in which I showed a chatbot’s analysis of questions that I was not musical enough to get to the bottom of, and I explained and admitted that. I was hoping to get other commenters’ views of Pink Floyd, but Gail deleted my comment, for whatever reason or reasons – was it too long; too off-topic; because it used AI; or all those things? Whatever, it’s Gail’s blog, so I accept that it’s up to her what stays and what goes. But it would be useful if she could explain her attitude to AI and it’s use by commenters here – what to do and what not to do.
I believe we should prefer using biophysical epochs rather than revolutions if we want to understand the foundations of the modern world’s history. (We have the technology we have the energy for.)
The first one was fire from biomass.
The second was fossil coal (1700 -> peaked 1914).
The third was fossil oil ( we are post peak).
The fourth I think or hope will be the first one –
for a while, or forever.
The fifth: food?
If this is the measure, then nuclear should be next. Call me an optimist.
That perspective is intelligent, as it focusses on the energetic foundation, not the mere tools.
You have an interesting analysis.
I agree that the first one was fire applied to biomass.
The second was fossil coal. Where this peaked seems to vary around the world. In the UK, it peaked in 2014.
The third was fossil crude oil. I would tend to exclude natural gas liquids and ethanol added as extenders.
The fourth revolution you talk about must be from above, “Network: Millions of computers, sensors and robots help one another do work.” – This would seem to include electricity, the internet, and AI.
Your fifth one as Food, would seem to imply going back to hunting gathering, together with whatever food could be cooked using biomass. This would assume a very small population.
We may go back to fire wood and water wheels but we will retain lots of helpful things like radios, fiber optics connections, self sailing ships for cargo, and many more.
The Arabs are staying home – tourism in Austria suffers
https://youtu.be/s6L38y96EN8?si=cOeiF9qqosFmmkk3
What my friends in Vienna would like to see is all the Turkish, and Arab, residents go home.
Hopefully, there will be some robots to do the manual work instead of the dwindling native population.