Yeah, perhaps I was being too harsh. However, the baseline scenario should be that current trends will go on for some time, and they predict at least cheap batteries and increasingly cheaper H2.
I mostly focussed on these two because the current problem of green energy sources is more related to energy storage than production, photovoltaic is currently the cheapest in most places.
I agree the baseline scenario is that current trends will go on.
In geology the resources availability trend (for both fossil energy and mining) follows the Hubbertâs curve. It doesnât follow a straight line up to the infinite. After a period of going up, it follows a period of going down, once we pass the peak. The peak doesnât mean that the resource is completely depleted, but it means that the amount we can extract this year is less than previous year.
To the date Iâm not aware of any other scientific explanation better than Hubbertâs curve, and this should be our baseline.
It is more difficult to predict exactly where we are in the Hubbertâs curve for each resource and whether the peak will happen this decade, but it is a fact that it will take place.
I think the Hubbert curve makes sense for something that is meaningfully finite, like fossil fuels. However, minerals are not meaningfully finite. For instance, aluminum, iron, and silicon are very common in the Earthâs crust, so it would not be that difficult to mine from âcommon rock.â Even for something like phosphorus, if you had to mine from common rock, it would not increase the price of food very much. Some rare minerals may very well peak this century (and platinum and palladium may have already), but I donât think that low concentration ores are a significant impediment to energy increase at least over this century.
Weâve already discussed this (of course we did ^^), but Iâd like to point out once again that the amount of minerals we can extract from the ground depends on how much energy we have: aluminum in common ground has a concentration 5 times inferior to that in ores. Had we infinite energy, this wouldnât be a problem (and we didnât have this issue so far as energy kept growing). But the issue wâre dealing with is precisely that of having less energy to mine them. This is why energy prices and metal prices are highly correlated. Less energy for mining means less metals being available (which is a problem since renewables require more metals). (for readers, all of that is detailed here).
I agree that aluminum, iron and silicon are abudant (as well as rare earths, ironically), but this study pointed out that most minerals are expected to peak before 2100, given exponential growth. This is probably not accurate for relatively ânewâ resources (like lithium), where there still is the potential for more discoveries, but this sounds reasonable for resources where most discoveries have already been done.
Iâm quite skeptical that we can run our complex society with common minerals onlyâespecially for computing.
A smart phone may be more difficult, but basic computing is not a challenge because silicon is abundant and the dopants are used in minuscule quantities. But the more important issue is that wind turbines, solar cells, and even energy storage in the form of thermal energy storage can all be done with common minerals (limestone for cement is about 10% of the Earthâs surface). So if that means we donât have an energy descent, that means we have energy in order to continue mining the rarer minerals from common rock. You link to your report, but I assume you are referring to your reproduced figure 4. Of course if you assume a Hubbert curve, itâs going to go to zero in a century or two. But if we do have sufficient energy, Iâm arguing that curve is fundamentally not appropriate for minerals.
Oh, yeah, while there is ever more energy, metals are not a problem. Although with constant energy, a Hubbert curve is appropriate, since weâll get to crummy ores with low concentration which will mean a decline at some point.
But the main problem is that keeping the current level of energy is unlikely. Which means less minerals. Even if we can build solar cells and windmills with common materials (with of course lower efficiency), we canât build an infinity of themâthere are limits of low EROI, transport, financial and energetic cost, locations⌠(thermal storage is also good for half a day, but do not work for seasonal). So we may still be able to extract a few rarer minerals, but we still have the original problem: we donât run out, but there is a large decline in availability.
Since wind power peaks in the winter and solar power peaks in the summer, we can generally handle seasonal variation by appropriate fractions of these energy sources. Backup plans include overbuilding, hydrogen, and biofuels. As you know, I donât think EROI, transport, financial and energetic cost, or locations would be limiting factors.
Well, as pointed out in the âenergy storageâ section : âA study in Europe found that even with a giant supergrid across the European Union, North Africa, and the Mediterranean, a much larger and sunnier area than the United States, there would still need to be 1 month of energy storage to keep the grid up during seasonal variations (Droste-Franke 2015). Palmer (2020) thought that up to 7 weeks of storage would be required as well as large amounts of renewable overbuild. â
Overbuilding would add to the energetic, material and financial cost of the energy system. As I already pointed out, using biofuels or hydrogen also imply a loss in energy efficiency. Transportation also requires materials and takes time to scale up.
What I still do not understand, however, is why you think that financial and energetic costs are not limitations? I mean, if more and more of societiesâ resources go to energy production (financial, material, energetic resources), for the same amount of energy as an output, this would mean that these resources cannot go toward providing food, housing, heating, and investment in capital for more economic growth.
Surely this would reduce GDP at some point, no ? I mean, even if efficiency can increase, yes, it cannot do so indefinitely.
It is true that hydrogen is inefficient and expensive. But biofuels, especially from agricultural and logging residues, can be quite land and energy efficient. I believe energy makes up about 8% of the global economy. We will need to produce more energy as the EROI falls and as minerals become less concentrated, and I think energy prices will increase. However, the economy is expected to get much larger (expansion of services, etc), so the percentage of GDP that is energy may not increase. Even if it doubles over the next 30 years, the drag on economic growth is ~0.3%/âyr, much smaller than the current 2% growth. So I donât see energy limitations as preventing the world from achieving developed country wealth and expanding into space in the next centuries.
Yeah, if biofuels can be made from agricultural and logging residues, that would be interesting. But as I pointed out here, there is almost no commercial facility that does that today, despite many companies trying to do that, billions in funding and tax incentives (2 factories are doing that today with sugarcane bagasse, but it is simpler to treat than other residues). It may be possible to do that someday, but by then I do not count on it given the short timing we have. Weâd also have to solve the logistical issue of collecting a huge amount of sparse and fluffy residues and bring them to the factory.
âthe economy is expected to get much larger (expansion of services, etc)â
Yeah but getting the energy to be larger means using more energy, as we still have relative decoupling, not absolute. There can be increases in efficiency but as I mentioned they have slowed down in the last couple decades.
(For other readers I give more details about the economy in this section).
âif the proportion of GDP in the economy doubles over the next 30 years, the drag on economic growth is ~0.3%/âyrâ
Well, empirical data rather seems to indicate that when energy reaches about 10% of GDP, a recession ensues. The current small share of energy in GDP is not representative of its importance. This is because energy has a multiplier effect in GDP: the price of oil is counted in GDP when it is bought at the gas station, but also in the price of trucking services, and in the price of any product transported by truck out of a factory (so about everything), etc.
I mean, say Russia were to cuts gas exports to Germany. Itâs about 10% of energy, so by your reasoning, one could conclude that this would affect Germanyâs GDP by only 0.5-0.8%. If you were to go to the German government and tell them this is not a big deal, what do you think they will answer?
Youâre referring to liquid biofuels, which I agree is more challenging (and required for aircraft, or hydrogen). But in that case, even if cellulosic ethanol isnât economical, we can turn solid residues into Fischer Tropsch liquids, as was done in World War II with coal.
To solve the seasonal mismatch of renewable energy and electrical demand, we can just burn agricultural residues or logging residues in repurposed coal power plants.
Since the solar resource is orders of magnitude greater than what we need, we donât need absolute decoupling (and we do have examples of absolute decoupling for some countries).
We are talking about different timescales. I agree that if energy increases from 8% to 10% of GDP in one year, then that would overwhelm the 2% economic growth. However, for it to overwhelm economic growth over decades, the price of energy would have to increase an order of magnitude or more, which I donât think is plausible. I donât think there is a multiplier effect in equilibriumâit would be great for an economist to weigh in here. For instance, people have modeled the impact of a carbon tax-sometimes it is paid directly in fuel, and other times it is paid indirectly such as in buying other products. But the overall drag on the economy I donât think is any larger than the total carbon tax.
Again, in the German case, itâs all about speed. If you have an abrupt reduction in energy supply, that is very disruptive to the economy. But resource exhaustion is not nearly that abrupt.
âHowever, for it to overwhelm economic growth over decades, the price of energy would have to increase an order of magnitude or more, which I donât think is plausible. I donât think there is a multiplier effect in equilibriumâ
Do you have any empirical data on that claim? Or is that just a guess on your part?
On my part, I personally tend to think that my personal instinctive takes about how the economy works are false, so I try to rely a lot on empirical data. For the multiplier effect, see the paper I mention here.
For a carbon tax, the main perk of a progressive carbon tax is that it is predictable and gives time to adapt, so it would be a good thing, I agree. Unfortunately, there are many things preventing a strong carbon tax, like fear of offshoring or overall very low public support (see the Yellow Vests in France).
âResource exhaustion is not nearly that abruptâ
Iâd argue it is quite abrupt, at least when it comes to impact on prices. Like oil price going from $20 in 2002 to $140 in 2008 (when overall oil supply didnât really decline, it just didnât grow fast enough). See this graph. Geopolitical factors like Russia exporting less are expected to increase in the future, not decrease.
To solve the seasonal mismatch of renewable energy and electrical demand, we can just burn agricultural residues or logging residues in repurposed coal power plants.
For biomass, using Fischer Tropsch liquids for coal has been done (and has only been viable with low coal prices and subsidies, like in Germany). But doing that with biomass is far less mature, I havenât seen a commercial plant for that (See this report page 7-8).
Limits for solar are less about the theoretical potential, but about the materials needed to harness it (electric cars and batteries), deployment speed (with the rehauling of the grid), and land use (see Halsteadâs report, page 52-54).
Iâd argue it is quite abrupt, at least when it comes to impact on prices. Like oil price going from $20 in 2002 to $140 in 2008 (when overall oil supply didnât really decline, it just didnât grow fast enough). See this graph.
Link didnât go to specific part of document. But even if it were a shortfall from business as usual demand of 2%/âyear for oil, that is ~0.8%/âyear for all energy, which is a different order of magnitude from 10%/âyear for energy.
Geopolitical factors like Russia exporting less are expected to increase in the future, not decrease.
Renewable energy is better distributed across countries than fossil fuels, so I would expect geopolitical disruptions to decrease in impact.
Well, the shortfall for oil in the 2000s was still big enough to be highly linked to the 2008 financial crisis. You can check it hereâif the link doesnât work, search for the title âThe 2008 financial crisis: the third oil shock?â. The graph I refered to that didnât work was this one:
Relationship between oil price and oil production. Jancovici, based on data from BP statistical review
Renewable energy is more distributed, yes, but when I talk about supply shocks, Iâm talking about the fossil fuels dependency that we have right nowâand that is likely to stay there well into the next 3 decades.
Renewables also rely on metals, some of which are poorly destributed (like lithium and copper in China and Australia, and platinum in South Africa) . China also directly controls approximately 80% of the raw materials value chain (mining, refining, smelting, manufacture and recycling). This does not account for Chinese-held corporate foreign investment in industrial assets worldwide. Specifically, the country has reduced its exports to attract more industry to the country. The Made in China 2025 plan is designed to secure the remaining 20% for Chinese interests in the name of long-term security (see here, page 61).
I looked at the reference and I donât see evidence for the 80% number. The majority of the mineral budget (total ~1% of GDP) is cement, iron, and aluminum. It looks like China mines little iron and aluminum, though it does refine a lot of them. Eyeballing it looks like China is ~half production and consumption minerals, which is a lot. But the idea that China would control 100% of the worldâs mining, refining, smelting, manufacture and recycling is hyperbole.
Ok, I looked again and the 80% figure is a bit a stretch compared to the initial formula, I can agree. I think itâs not just âChina is mining these mineralsâ but âChina is involved in the material chain at some point, through mining or refining or smelting or manufacturing or recyclingâ (with varying degrees of dependency). That could be where the 80% figure comes from. 100% is not realistic, but Chinaâs share is the value chain is increasing.
But even if we were to stick to 50% of control as you suggest, or 60%, this would not change much of the issue that there is a lot of dependency, indeed. A lot of potential vulnerability would arise if China were to cut down some of its exports (whether voluntarily, or by accident, or because of a pandemic).
Yeah, perhaps I was being too harsh. However, the baseline scenario should be that current trends will go on for some time, and they predict at least cheap batteries and increasingly cheaper H2.
I mostly focussed on these two because the current problem of green energy sources is more related to energy storage than production, photovoltaic is currently the cheapest in most places.
I agree the baseline scenario is that current trends will go on.
In geology the resources availability trend (for both fossil energy and mining) follows the Hubbertâs curve. It doesnât follow a straight line up to the infinite. After a period of going up, it follows a period of going down, once we pass the peak. The peak doesnât mean that the resource is completely depleted, but it means that the amount we can extract this year is less than previous year.
To the date Iâm not aware of any other scientific explanation better than Hubbertâs curve, and this should be our baseline.
It is more difficult to predict exactly where we are in the Hubbertâs curve for each resource and whether the peak will happen this decade, but it is a fact that it will take place.
I think the Hubbert curve makes sense for something that is meaningfully finite, like fossil fuels. However, minerals are not meaningfully finite. For instance, aluminum, iron, and silicon are very common in the Earthâs crust, so it would not be that difficult to mine from âcommon rock.â Even for something like phosphorus, if you had to mine from common rock, it would not increase the price of food very much. Some rare minerals may very well peak this century (and platinum and palladium may have already), but I donât think that low concentration ores are a significant impediment to energy increase at least over this century.
Weâve already discussed this (of course we did ^^), but Iâd like to point out once again that the amount of minerals we can extract from the ground depends on how much energy we have: aluminum in common ground has a concentration 5 times inferior to that in ores. Had we infinite energy, this wouldnât be a problem (and we didnât have this issue so far as energy kept growing). But the issue wâre dealing with is precisely that of having less energy to mine them. This is why energy prices and metal prices are highly correlated. Less energy for mining means less metals being available (which is a problem since renewables require more metals). (for readers, all of that is detailed here).
I agree that aluminum, iron and silicon are abudant (as well as rare earths, ironically), but this study pointed out that most minerals are expected to peak before 2100, given exponential growth. This is probably not accurate for relatively ânewâ resources (like lithium), where there still is the potential for more discoveries, but this sounds reasonable for resources where most discoveries have already been done.
Iâm quite skeptical that we can run our complex society with common minerals onlyâespecially for computing.
A smart phone may be more difficult, but basic computing is not a challenge because silicon is abundant and the dopants are used in minuscule quantities. But the more important issue is that wind turbines, solar cells, and even energy storage in the form of thermal energy storage can all be done with common minerals (limestone for cement is about 10% of the Earthâs surface). So if that means we donât have an energy descent, that means we have energy in order to continue mining the rarer minerals from common rock. You link to your report, but I assume you are referring to your reproduced figure 4. Of course if you assume a Hubbert curve, itâs going to go to zero in a century or two. But if we do have sufficient energy, Iâm arguing that curve is fundamentally not appropriate for minerals.
Oh, yeah, while there is ever more energy, metals are not a problem. Although with constant energy, a Hubbert curve is appropriate, since weâll get to crummy ores with low concentration which will mean a decline at some point.
But the main problem is that keeping the current level of energy is unlikely. Which means less minerals. Even if we can build solar cells and windmills with common materials (with of course lower efficiency), we canât build an infinity of themâthere are limits of low EROI, transport, financial and energetic cost, locations⌠(thermal storage is also good for half a day, but do not work for seasonal). So we may still be able to extract a few rarer minerals, but we still have the original problem: we donât run out, but there is a large decline in availability.
Since wind power peaks in the winter and solar power peaks in the summer, we can generally handle seasonal variation by appropriate fractions of these energy sources. Backup plans include overbuilding, hydrogen, and biofuels. As you know, I donât think EROI, transport, financial and energetic cost, or locations would be limiting factors.
Well, as pointed out in the âenergy storageâ section : âA study in Europe found that even with a giant supergrid across the European Union, North Africa, and the Mediterranean, a much larger and sunnier area than the United States, there would still need to be 1 month of energy storage to keep the grid up during seasonal variations (Droste-Franke 2015). Palmer (2020) thought that up to 7 weeks of storage would be required as well as large amounts of renewable overbuild. â
Overbuilding would add to the energetic, material and financial cost of the energy system. As I already pointed out, using biofuels or hydrogen also imply a loss in energy efficiency. Transportation also requires materials and takes time to scale up.
What I still do not understand, however, is why you think that financial and energetic costs are not limitations? I mean, if more and more of societiesâ resources go to energy production (financial, material, energetic resources), for the same amount of energy as an output, this would mean that these resources cannot go toward providing food, housing, heating, and investment in capital for more economic growth.
Surely this would reduce GDP at some point, no ? I mean, even if efficiency can increase, yes, it cannot do so indefinitely.
It is true that hydrogen is inefficient and expensive. But biofuels, especially from agricultural and logging residues, can be quite land and energy efficient. I believe energy makes up about 8% of the global economy. We will need to produce more energy as the EROI falls and as minerals become less concentrated, and I think energy prices will increase. However, the economy is expected to get much larger (expansion of services, etc), so the percentage of GDP that is energy may not increase. Even if it doubles over the next 30 years, the drag on economic growth is ~0.3%/âyr, much smaller than the current 2% growth. So I donât see energy limitations as preventing the world from achieving developed country wealth and expanding into space in the next centuries.
Yeah, if biofuels can be made from agricultural and logging residues, that would be interesting. But as I pointed out here, there is almost no commercial facility that does that today, despite many companies trying to do that, billions in funding and tax incentives (2 factories are doing that today with sugarcane bagasse, but it is simpler to treat than other residues). It may be possible to do that someday, but by then I do not count on it given the short timing we have. Weâd also have to solve the logistical issue of collecting a huge amount of sparse and fluffy residues and bring them to the factory.
Yeah but getting the energy to be larger means using more energy, as we still have relative decoupling, not absolute. There can be increases in efficiency but as I mentioned they have slowed down in the last couple decades.
(For other readers I give more details about the economy in this section).
Well, empirical data rather seems to indicate that when energy reaches about 10% of GDP, a recession ensues. The current small share of energy in GDP is not representative of its importance. This is because energy has a multiplier effect in GDP: the price of oil is counted in GDP when it is bought at the gas station, but also in the price of trucking services, and in the price of any product transported by truck out of a factory (so about everything), etc.
I mean, say Russia were to cuts gas exports to Germany. Itâs about 10% of energy, so by your reasoning, one could conclude that this would affect Germanyâs GDP by only 0.5-0.8%. If you were to go to the German government and tell them this is not a big deal, what do you think they will answer?
Youâre referring to liquid biofuels, which I agree is more challenging (and required for aircraft, or hydrogen). But in that case, even if cellulosic ethanol isnât economical, we can turn solid residues into Fischer Tropsch liquids, as was done in World War II with coal.
To solve the seasonal mismatch of renewable energy and electrical demand, we can just burn agricultural residues or logging residues in repurposed coal power plants.
Since the solar resource is orders of magnitude greater than what we need, we donât need absolute decoupling (and we do have examples of absolute decoupling for some countries).
We are talking about different timescales. I agree that if energy increases from 8% to 10% of GDP in one year, then that would overwhelm the 2% economic growth. However, for it to overwhelm economic growth over decades, the price of energy would have to increase an order of magnitude or more, which I donât think is plausible. I donât think there is a multiplier effect in equilibriumâit would be great for an economist to weigh in here. For instance, people have modeled the impact of a carbon tax-sometimes it is paid directly in fuel, and other times it is paid indirectly such as in buying other products. But the overall drag on the economy I donât think is any larger than the total carbon tax.
Again, in the German case, itâs all about speed. If you have an abrupt reduction in energy supply, that is very disruptive to the economy. But resource exhaustion is not nearly that abrupt.
Do you have any empirical data on that claim? Or is that just a guess on your part?
On my part, I personally tend to think that my personal instinctive takes about how the economy works are false, so I try to rely a lot on empirical data. For the multiplier effect, see the paper I mention here.
For a carbon tax, the main perk of a progressive carbon tax is that it is predictable and gives time to adapt, so it would be a good thing, I agree. Unfortunately, there are many things preventing a strong carbon tax, like fear of offshoring or overall very low public support (see the Yellow Vests in France).
Iâd argue it is quite abrupt, at least when it comes to impact on prices. Like oil price going from $20 in 2002 to $140 in 2008 (when overall oil supply didnât really decline, it just didnât grow fast enough). See this graph. Geopolitical factors like Russia exporting less are expected to increase in the future, not decrease.
Well, current burning of biomass for electricity in Europe already contributes to deforestation. So I donât think residues will be enough.
For biomass, using Fischer Tropsch liquids for coal has been done (and has only been viable with low coal prices and subsidies, like in Germany). But doing that with biomass is far less mature, I havenât seen a commercial plant for that (See this report page 7-8).
Limits for solar are less about the theoretical potential, but about the materials needed to harness it (electric cars and batteries), deployment speed (with the rehauling of the grid), and land use (see Halsteadâs report, page 52-54).
Link didnât go to specific part of document. But even if it were a shortfall from business as usual demand of 2%/âyear for oil, that is ~0.8%/âyear for all energy, which is a different order of magnitude from 10%/âyear for energy.
Renewable energy is better distributed across countries than fossil fuels, so I would expect geopolitical disruptions to decrease in impact.
Well, the shortfall for oil in the 2000s was still big enough to be highly linked to the 2008 financial crisis. You can check it hereâif the link doesnât work, search for the title âThe 2008 financial crisis: the third oil shock?â. The graph I refered to that didnât work was this one:
Relationship between oil price and oil production. Jancovici, based on data from BP statistical review
Renewable energy is more distributed, yes, but when I talk about supply shocks, Iâm talking about the fossil fuels dependency that we have right nowâand that is likely to stay there well into the next 3 decades.
Renewables also rely on metals, some of which are poorly destributed (like lithium and copper in China and Australia, and platinum in South Africa) . China also directly controls approximately 80% of the raw materials value chain (mining, refining, smelting, manufacture and recycling). This does not account for Chinese-held corporate foreign investment in industrial assets worldwide. Specifically, the country has reduced its exports to attract more industry to the country. The Made in China 2025 plan is designed to secure the remaining 20% for Chinese interests in the name of long-term security (see here, page 61).
I looked at the reference and I donât see evidence for the 80% number. The majority of the mineral budget (total ~1% of GDP) is cement, iron, and aluminum. It looks like China mines little iron and aluminum, though it does refine a lot of them. Eyeballing it looks like China is ~half production and consumption minerals, which is a lot. But the idea that China would control 100% of the worldâs mining, refining, smelting, manufacture and recycling is hyperbole.
Ok, I looked again and the 80% figure is a bit a stretch compared to the initial formula, I can agree. I think itâs not just âChina is mining these mineralsâ but âChina is involved in the material chain at some point, through mining or refining or smelting or manufacturing or recyclingâ (with varying degrees of dependency). That could be where the 80% figure comes from. 100% is not realistic, but Chinaâs share is the value chain is increasing.
But even if we were to stick to 50% of control as you suggest, or 60%, this would not change much of the issue that there is a lot of dependency, indeed. A lot of potential vulnerability would arise if China were to cut down some of its exports (whether voluntarily, or by accident, or because of a pandemic).