The final count-down: estimating the total value of all the world's natural resources and an estimate of how long they may last

1 Introduction

Seventy five percent of the world's wealth has its origin in natural resources, and it appears that about 60% of the wealth has its origin in finite, fossil natural resources, fossil energy, metals, and materials. Thus, a major part of our value creation and accumulated wealth in money and investments is fossil and finite. In this study we have attempted to count up the value of all the world's natural resources. The prospect of future prosperity depends on how well these resources are managed.

The largest sources of value for any civilization are the value of its natural resources when they are put to use in society. The economic foundation of civilization is about converting physical and intellectual resources into subsistence, protection, shelter and sophisticated aspects of value for the citizens of the state and society. This appears through production, enhancement of utility properties and their consumption. Gold, silver, and copper were very early seen as carriers of value (Currency), and these metals in standardized units also became the physical basis for the world's first coins (Kock-Johansen, 2001, 2006). A natural resource is money, and the stock existing at the beginning of modern civilization has been and still is an important economic basis for developing the modern world. The world's economy still depends heavily on these stocks. It is of importance to book keep these resources and keep accurate track of how we use them, recover them after use or how we lose them.

2 Objectives and scope

The objective of this study is to estimate the net sales value and the net profit value of the world's natural resources, and to set this in context to the total generation of income and profit in the world. Thus, the following will be addressed:

• How much of the wealth in human society has its origin in the extraction and use of natural resources, in particular the finite resources like metals, materials and fuels for energy?
• What are the natural resource basis for the estimate in physical amounts?
• How large a fraction of the total Earth's natural resources available to human society is finite, and non-renewable?

These questions will in addition be related to the earlier estimates of value of the World's ecosystem services (Costanza et al., 1997; Boumans et al., 2002). Finally, this will be related to the sustainability of human civilization, and if any advice to policy can be made (Europe, 1986). Predictions are presented at the nexus of what should be done, and evaluation in context of what would be possible to actually do in term of real change from a point of view of policy change dynamics and behavioral psychology. The materials and metals listed in Table 1 are addressed.

Table 1. Stock and flow values.

An adjustment factor of 4 from 1997 to 2026 was used to lift in the values published by Costanza et al. (1997). The stock value is the stored value in the resource deposit integrated over the time of the extraction, using the simulated price in each timestep. The annual revenue is the sales value of the extracted resource, but before added value manufacture using the resource. WORLD7 was used to simulate the future extraction rates and resource market price.

The stock types have been categorized as follows: P, Potential; R, renewable; D, Degradable; F, Physical entity; I, Invaluable. Invaluable is existential and if it fails, civilization can be disrupted.

^*Assuming production continues for 5,000 years. It would be an alternative to assume 10,000 years.

3 Earlier work

This study rest much on earlier work of others. Important for the work has been the work Howard T. Odum (examples of his writings are Odum, 1960, 1973, 1977, 1996, but he published many more articles on the subject). The take home message from Odums work during a life-time (Odum, 1996), is that nature and thus obeys the basic thermodynamic rules at all times regardless of any human opinion about it. Odum laid the foundations for ecosystem theory and made clear how the ecosystems are feedback-systems linked with the rest of the physical world as a system, parallelling the work of Meadows et al. (1972) and Forrester (1961, 1969, 1971). Mass and energy balance always rules sovereign, and any prosperity gained from harvesting above the sustainability limits will with certainty decline when the resource stock gets depleted.

Many studies have shown that the world's resources are finite, and the pioneer was Donella and Dennis Meadows and their team at the MIT in 1970–1972 (Meadows et al., 1972), developing the World3 model. Later, others followed the same logic, much based on the logic of studies, such as Heinberg (2011), Randers (2012), Bardi (2013), and Sverdrup and Ragnarsdottir (2014). The world operates at present mainly driven by finite and fossil natural resources (Heinberg, 2011; Cohen, 2007; Bardi, 2013; Boumans et al., 2015). From 2010, the work on the next versions of the World3 model started, going to WORLD4 (Sverdrup et al., 2012), WORLD5 (Sverdrup et al., 2014), to a fully detailed Integrated Assessment model (WORLD6 in 2014, Sverdrup et al., 2014, 2015; Sverdrup and Ragnarsdottir, 2016; Sverdrup et al., 2017a,b, 2019a,b, 2024a,b; and WORLD7 in 2018 (Sverdrup, 2017, 2019; 2020a; 2022; Sverdrup and Olafsdottir, 2023; Sverdrup et al., 2018; 2019a; 2019b; 2019c, 2020a; 2020b; 2021; 2022).

It is necessary to restructure the world to become material circular within sustainability limits, before 2060 (Sverdrup and Ragnarsdottir, 2014; Sverdrup, 2019; Sverdrup et al., 2020a,b). Contrary to what has been claimed, very little resource decoupling has taking place on a global scale (Fletcher and Rammelt, 2017; Sverdrup and Ragnarsdottir, 2014; Ragnarsdottir et al., 2017; Mauerhofer, 2013). The journalist Monbiot (2015) spells it straight out in an article in the British newspaper The Guardian; “Consume more, conserve more: sorry, but we just can't do both.”

On a larger scale, there are substantial problems of realism in traditional economic theory (Keen, 2011; Alexandrova and Northkott, 2013; Mauerhofer, 2013; Ward et al., 2016). The underlying assumption was that there could be increased physical consumption of mass and energy with a simultaneous decrease in negative impact on sustainability. Another underlying assumption was that negative sustainability impact would decrease faster than consumption. On closer inspection, it becomes visible that decoupling does not do decrease negative sustainability impact when externalities are included (Fletcher and Rammelt, 2017). The flaw in the decoupling concept has been discovered by many. What appeared to be observed as de-coupling, was just a manifestation of the classical archetype of “shifting the burden” (Fletcher and Rammelt, 2017; Wiedmann et al., 2015; Wolstenholme, 2004; York and McGee, 2015; Zhang et al., 2016). The apparent lack of impact of local physical actions, appeared to work because the impact was moved beyond the surveillance horizon. Globalization allowed the impacts to be moved to overseas or foreign locations, out of the system boundary applied by the decoupling studies (Roberts, 2011; Mauerhofer, 2013; Ward et al., 2016).

Costanza et al. (1997) estimated the value of the all the natural ecosystems on the Earth. He synthesized earlier studies on the subject, and added a few estimations on his own when needed to get a fuller picture. The published paper from 1997 details the methodology they used. Costanza's work rest on that of Herman Daily (Examples are Daly, 1977; Daly, 1996; Daly and Cobb, 1989). The key findings of Costanza et al., study have been worked into our estimates to attempt to generate a complete overview. They estimated then that the annual value of all production and services amounted to 33 billion $/year, but a range from 16,000-54,000 billion $/year. Their estimate is based on a supply-demand model, an approach similar to ours. Costanza et al. (1997) stress that the estimate is a minimum. They revised the values significantly upward in a study from 2014 (Costanza et al., 2014a,b). Updated valuations considering changes in land-use and present ecosystem use increase this to 125–145 trillion $ per year. The values are for different types of ecosystems and ecosystem services. Many of these are not traded as commodities like metals or energy, and thus rest on non-market methods for estimation of their values. Some of the ecosystem services do have a physical manifestation that partly appear as commodities in some markets (Forestry products, carbon sequestration in ecosystems, food derived from ecosystems, land use for food production).

Charles Hall, Professor emeritus at SUNY has worked with the concept of biophysical economics and the concept of Energy Return on Investment (EROI) and how it is an indicator of whether energy extraction is meaningful of not (Hall and Klintgaard, 2011; Melgar-Melgar and Hall, 2020). Hall studied with Odum and there is a continuity between the thinking of Odum and Hall. The concept is generalizable to a Resource Utility Return on Investment (RUROI) to be applied to the extraction of utility from any type of natural resource, natural or geological, finite or renewable. This also leads to a similar conclusion made long time ago by Hall et al., that value creation and building of wealth is linked with physical use and physical conversion of natural resources and the main source of wealth from the beginning of civilization until now.

The biophysical economy is based on the fact that any economy must at all times satisfy the basic principles of thermodynamics. Any political trick or trick of national economics that appear to divert from this will experience thermodynamic correction as a physical fact (Statement made by the physicist Eddington in 1928). The Roman philosopher Seneca said “The road to wealth is long and cumbersome, the fall into ruin is rapid Seneca (1998).” This is what Meadows et al. (1972); Odum (1996, 2011); Costanza et al. (1997), and Hall and Klintgaard (2011) all said in different words in different shapes and contexts, but with the same basic message, and it is a system, it is finite and if you overuse, you will find limitations.

4 Methods

4.1 System analysis, system dynamics, and Integrated Assessment Modeling

The Integrated Assessment Model WORLD7 is our main tool for looking into the future. We use that for exploring what will happen if we continue with Business-as-usual (BAU). We apply mass balance and energy balances to the global resource flows. System dynamics, system analysis, resource estimation and spreadsheet calculations are the main methods used in this study. The metals, energy and materials prices come from the actual prices in 2020, as well as the prices as simulated with the WORLD7 model for 1850–2250. Following is a short description of these methods and the methods included in the Integrated Assessment Model WORLD7. Our scenarios applied in the background assessment work with simulation models and manipulating data bases are:

• Business-as-usual
• Paradigm shifts

Systems analysis was used to understand the global supply system and to assess the scenarios listed above (Sverdrup and Ragnarsdottir, 2014 is an earlier study starting on this work, More is reported in Sverdrup et al., 2021). The main method includes systems thinking, systems analysis with causal loop diagrams (CLDs) and flow charts to map pathways and system behavior. In all calculations, the fundamental principles of mass and energy conservation apply, and specifically for economic calculations, which must be mass and energy balance compliant, in every time step (Eddington, 1928; Hall et al., 2001; Boumans et al., 2002; Sverdrup and Ragnarsdottir, 2014). The WORLD7 model was used for the simulations and this work rests heavily on our earlier work (Sverdrup and Olafsdottir, 2020; Sverdrup, 2019; Sverdrup and Olafsdottir, 2020a,b,c; Sverdrup et al., 2021). See Appendix Table A1 for a longer list of publication for every simulation of individual resource dynamics. Here a lot of detail is described, and this will not be repeated here.

4.2 Resource estimates and data handling

In order to set the model parameters correctly, an effort was made to find the best available estimates of the global stocks and flows of extracted natural resources in terms of energy, materials, as a basis for the study was made. Many of the estimates used here have been published earlier in peer reviewed articles, and they will not be repeated here. Some estimates have been adjusted because of new data. The information for precious metals comes from unpublished industrial data available to Sverdrup.

Included is the value taken from natural bio-geosystems, like land, soil and natural or created ecosystems. A database was created for this study for the global resources, partly compiled from earlier studies by the authors and parameterization of the WORLD7 model. The database was explored using spreadsheet methods, and used to make preliminary conclusions. It is coupled to the estimates of the annual production of these resources and the sales and profit values of that production was created. Data was harvested from a number or official databases and from earlier compilations by the authors in dedicated publications referenced in the Table A1 in the Appendix. Database handling was done using spreadsheet methods for simple calculations.

This study uses natural resource recoverable resources data collected for a number of earlier studies by the authors that are based on large literature surveys, collected databases, and evaluate these for ultimately recoverable resources (Sverdrup et al., 2017a,b, 2019a,b, 2020a,b; Sverdrup, 2017, 2019; Sverdrup and Ragnarsdottir, 2014; Sverdrup and Koca, 2018; Sverdrup and Olafsdottir, 2019a,b; Olafsdottir and Sverdrup, 2019, 2020,a,b,c; Olafsdottir et al., 2019). Details on the resource data extraction can be found in these publications. The rate of resource production and resource prices were determined from data harvesting and by simulations using the WORLD7 model. The variation in resource quality is taken care of in the ore grade classification and modeled as it changes over time with extraction of the best qualities first. Some additional tables regarding data and data sources can be found in Appendix 1. Many references that exemplify the type of literature we have used in the resource estimations are given in the reference list. The methodology was:

• Estimate the available resource and reserves size, and distribute this over cost of extraction categories. Tons in reserves (known how much and where) and resources (approximately known how much, but not exactly where).
• Estimate realistic cost of extraction for each resource quality grade, $/ton.
• Use the WORLD7 model to simulate resource prices over time, to be matched with historical data when available, and for getting future resource prices.
• Use the WORLD7 model to estimate the part of the resources present in the deposit as extractable.
• Estimate the total revenue of the extracted resource, the running extraction costs and the market profit flowing to society from the activity.

The obtained results were used to write the tables and the text. The tables were used for simple. The gross value creation was the sum of the revenues from the resource extraction. The net value creation was the sum of all resource profits made (Income revenue minus extraction costs).

5 Theory

5.1 What is money and what is wealth?

Value creation is defined in a conventional way as done in the definition of GDP. Value is expressed in terms of money, which is a proxy. Money creation used to be based on the value originating from production of energy resources in society, food for the population and metals and materials for making commodities and infrastructures. Money is value carriers, for being able to transport and store value. In older times metals gold, silver, and copper were the main value carriers. They were standardized into coinage for making transactions easier. The first paper money used paper receipts for the three metals as the primary fiat basis. Later, this system of paper receipts became further abstracted to include proxies for any type of commodity or service rendered. This was further complicated to include future production and promises made by government.

The strength of the economy was linked to how well-society could acquire resources like gold, silver, copper and iron, and wood and coal for energy because of its activity. As banking systems, paper money and financing approaches got more developed, the direct connection to physical resources got lost from vision through intermediate proxies and abstractions. This became fiat money, where we trust the proxies and abstractions to be promises that will be upheld and honored.

Today, money creation is connected to the creation of real estate and infrastructure development, production systems, political promises and virtual wealth based on trust in future anticipated incomes. If we sum up the value of all physical resources extracted globally, the value of these amount to about 50% of the global GDP, before manufacture to commodities or use to produce services. The implication is that more that 65% of the Global GDP is at present directly hard-linked to physical resources, with probably another 15%−20% linked less visibly to resources like ecosystems and use of no-monetized natural resources. Thus, ecosystem services are wealth suppliers. The implication is that natural resources are what gives fiat money its real value. Vice versa, if there are no natural resources, there is no real value in money. Thus, we may state the following Thermodynamic Laws of Real Wealth:

• Finite natural resources are finite wealth and money. Finite natural resource exhaustion is finite wealth exhaustion.
• Renewable natural resources are renewable wealth and money. Renewable natural resources when they are well-managed, will be renewable wealth.
• Finite resources will when extracted and used, only last for a limited time. The time limit depends on intensity of extraction and degree of circularity. The limit is set by resource size divided by net irreversible loss. This is a result of the second law of thermodynamics.

5.2 Defining wealth, value, and GDP

GDP has been used as an indicator for estimating a nations' wealth, income or economic turnover. It is often taken to suuvve. We use several term that for clarity will be defined here:

• Wealth is exemplified as the total value of the extractable part of a resource, thus the extractable stock in the deposit multiplied by the price at the time when it is taken out. This gives an estimate of the wealth stored in the resource deposit, be it oil, coal, iron, or gold.
• Gross value production is the value of the extracted resource multiplied with its price in the market at the time of extraction.
• Net value production is the profit made on the extracted resource multiplied with its price in the market at the time of extraction.

Per definition, “sustainable use,” implies the use can go on for eternity. Renewable resources will support regenerative wealth and money. Finite resources will support finite wealth and money for a set time.

In order to work with GDP as a proxy for gross value production, we need to define it. The GDP used is observed GDP, which has several sources according to the literature (Credit Suisse Research Institute, 2014, 2018, 2019; Investopedia, 2020; OECD, 2009):

• OECD defines GDP as “an aggregate measure of production equal to the sum of the gross values added of all resident and institutional units engaged in production and services, plus taxes, and minus subsidies, on products not included in the value of their outputs” (OECD, 2009).
• IMF defines GDP as: “GDP measures the monetary value of final goods and services, that are bought by the final user, produced in a country in a given period of time (say a quarter or a year)” (Callen, 2020).
• Investopedia (2020) defines GDP as: “The Gross Domestic Product (GDP) is the total monetary or market value of all the finished goods and services produced within a country's borders in a specific time period.”

GDP can be calculated by three methods, using expenditures, production, or income. The expenditure approach, calculates the spending by the different groups that participate in the economy. This can be calculated on National scale using:

GDP=PC+a*GC+BC+I+(E-IM)(1)

where PC, Private consumption; GC, Government spending; BC, Business consumption; I, Investment; E, export; IM, Import; a is the fraction of governmental consumption that is not a result of taxation on the other terms. The resource part of the governmental consumption not derived from taxation would be the natural resources owned in the national collective or commons. Consumer spending is the biggest component of GDP. Governments spend money on equipment, infrastructure, and payroll. Businesses spend money to invest in their business activities (buying raw materials, but also investing in machinery, for instance). Business investment increases productive capacity and boosts employment. All expenditures by companies located in the country, even if they are foreign companies, are included in the calculation. The other side of the spending aspect is income, and since expense is somebody else's income, another way to calculate GDP is the income approach. Income earned by all factors of production in an economy including wages, rent earned by land, return on capital as interest, and corporate profits are included. The production approach estimates the total value of economic output and deducts costs of intermediate goods that are consumed in the process, such as materials and services. This approach was adapted from Investopedia (2020). In the WORLD7 model, the GDP contribution was estimated from the following components:

Resource GDP contribution=VR+VM(2)

where the resource sales value (V_R) is the sales revenue of the natural resources produced in a year, composed of the annual production and the price in that year, plus the resource value (V_M) caused by the manufacture to consumer goods. The Gross Sustainable Product (GSP) will be a function of the sustainable part of: private consumption, no-tax-based government spending, business consumption, investment, export and import. For a GSP calculation, these terms must be converted to the sustainable part of the natural resource flows.

5.3 On stocks and flows

The value flow of natural resources from hidden resources to stocks in use in society and how they are lost is shown in Figure 1. The figure presents value flow of natural resources from hidden resources to stocks in use in society and how they are lost. The WORLD7 model was used to estimate the flows of “discovered,” “extracted,” “scrapped,” “recycled,” and “irreversibly lost,” and the stocks “Value in hidden stocks,” “Value in known stocks,” “Value in primary stocks-in-use,” and “Value in recycled stocks.” The runs are based on earlier work by some of the authors on estimating Ultimately Recoverable Resources (Sverdrup and Ragnarsdottir, 2014).

The estimation required that the value is establish of the stock and the costs associated with bringing the resource out from the geological resource and convert it to a useful form. Value (V) of a resource is equal to the sales value (S) minus the total extraction costs (C). The mass balance in the model follows the logic shown in Figure 1. We have for value expressed as profits:

V=S-C(3)

The extraction costs include all costs (C), thus, cost for materials (C_materials), cost for energy (C_energy), cost for work (C_work), and cost for investment (C_CapEx):

C=Cenergy+Cmaterials+Cwork+CCapEx(4)

The value of the annual sales (s) is obtained as the price (P) times the volume sold (f). The costs are different for different resource substrate qualities i, thus the profit value (v) becomes the difference between sales revenue (s) and all costs (c) associated:

v = sifi -∑i=1i=allfici(5)

s is the sale revenue and c is the cost, then v is the profit value. For the whole stock, it amounts to the whole accumulated profit over the whole time it takes to extract everything. The evaluation starts from the beginning of the European industrialization in 1850 when the WORLD7 model is used (though outputs are presented from 1950 for clarity in this paper). The end of the simulation is in 2250, 400 years later. Most natural resources have been exhausted by then under Business-as-usual scenario. The expression for the total value for each resource from the beginning of industrialization to the simulation end in 2250 is:

v = ∫t=startt=end (∑i-1i=resourcesPifi- ∑i=1i=resourcesfici) dt(6)

where v is the value of all profits and v is the value of the individual resource, and this is counted up for all resources i. Doing this by hand in a spreadsheet has some challenges. One is that the resource price (P) change over time, and is affected by the extraction costs (c) and the resource scarcity in the market.

The price is affected by the extraction cost in such a way that it must stay above the extraction cost, else no extraction will take place. The scarcity to of the resource in the markets affect the price by making it increase. This is a dynamic process and not possible to be captured in a spreadsheet model.

5.4 Using the WORLD7 model for value dynamics over the time from 1850 to 2600

Figure 2 shows an overview of the modules in the WORLD7 Integrated Assessment Model. Note that behind every little box in the figure, there are one or several sub-modules and every red line represents one or more links. In the model, a number of interdependencies are made between different modules and sectors. WORLD7 was operated in the STELLA^® Architect version 3.5 modeling environment and was used to estimate alternative energy capacities and resource use for alternative energies. The results from the WORLD7 simulations is shown using diagrams.

In the assessment input data, the content of critical metals and materials (Ge, Ga, Te, In, Ag, Cu, REE, and Co) was used to estimate the future demand in different types of photovoltaics, employing the WORLD7 model. Wind power use the metals copper, Rare Earth Elements in significant amounts for the generators and the grid that connects them. A similar assessment was done for electric vehicles, new generation of electric motors (Metals used Cu, Ag, Al, Rare Earth Elements, and Co), batteries (Li, Ni, Co, V, and graphite), magnets in general for all kinds of appliances (REE, Ni, Co, and Pt), high performance turbine blades for jet engines as well as power plants (Rh, Pd, Ta, Re, Zr, Hf, Nb, Mo, Co, and Ni). Finally, computer electronics use a range of rare metals (Ga, Ge, Ag, Cu, Co, Au, Ru, Ta, REE, Ge, Ga, Te, and Se).

The WORLD7 model operates on a global scale. Thus, there are no countries and no regions, everything is a global average. Inputs and outputs variables like commodity and resource prices are generated endogenously inside WORLD7 from first principles of supply and demand within the market arbitration mechanism over price. The WORLD7 model is focused toward biophysical modeling of the economy, resource use, industrial production, human population dynamics and a sustainable society.

For process-oriented models constrained by thermodynamics, with mass balance consistent feedbacks and linkages, and forced to be completely energy balance compliant at all times, piling up of uncertainties or biases can be avoided, even over long time periods. This is the case for the WORLD7 model where it has been shown that uncertainties remain limited and that systematic biases do not build up (Sverdrup et al., 2017a,b, 2019a, 2021). Statistically based models do not have this property and will be quickly derailed by exponentially increasing uncertainties. Figure 3 shows a causal loop diagram explaining some of the most important feedback loops in the market system supplying natural resources [modified and adapted after a diagram shown earlier by a study on copper, zinc, and lead by Sverdrup et al. (2019a,b)]. See also Sverdrup and Sverdrup and Olafsdottir (2019a,b) for a detailed explanation of the price and market model.

Econometric models of the IOT-Leontief type are statistical linear regression models for flows, but without stocks and they build up errors exponentially, limiting the forward time-span they are useful for (about 5 years). Mass- and energy-consistent models do not have this property, and in the overall mass- and energy-balances, many uncertainties cancel mass- and energy-consistent models out. In the WORLD7 model, there is an experience of no accumulation of bias or uncertainty over time, allowing long term mass- and energy-consistency with good robustness.

Figure 4 shows a causal loop for the economic drives in resource extraction in the WORLD7 model (Adapted from the copper, zinc and lead study by Sverdrup and Olafsdottir, 2019a,b, and the market model study by Sverdrup et al., 2019a,b). The red lines depict the economic feedbacks in the system, the black shows the physical parts of the system. Take note how reinforcing loops involving profits (R1), the market price (R8), the improvements in technology (R6), the supplying capacity (R2) and investments (R6, R4), backed by loan financing (R5) keeps the system going.

5.5 Economy module in WORLD7

The economy module in WORLD7 is built up in a way similar to business economics. In the model, the mass flows and logistics chains are done first. Then material flows, transformations, combinations, and transaction and manufacturing losses. Natural resources are capital stocks, containing large amounts of value that is the fundamental basis of most money. Figure 5 shows a depiction of the mass flow system for iron in the WORLD model. A similar system is made for energy of the iron system, and based on those two, a flow chart for money. Similar cycles are used for all the major metals in the model.

Figure 6 shows the flowchart and the issues of yield and losses for the circularity illustrated. All of these flow systems and value chains for physical and energy resources communicate over markets, where they get sold and bought and change ownership. The material flow represent value flows, according to the price they have. In parallel, the energy flows caused by the flows and transformations are done. Energy is produced and traded over markets, at a price. From the material flows and the energy flows, the money flows are created, much like it is done in a manufacturing company accounting books.

The material and energy stocks in the system have a value and represent stores of value. Note how the geological resource stocks are treated equivalently as restricted and conditioned bank accounts, holding currency. The WORLD7 model is used from exchanging the currency “tons of metal X” to million dollars. In the WORLD7 model, investments are made and these are financed from the money supply or loans. Loans are obtained from bank financing or government bonds. Figure 7 shows the capital stock and the main value flows in the economic module WORLD7. Note the source role of natural resources and human resources. The model is consistent with mass balance and energy.

5.6 Sustainability and time

A sustainable activity is defined as something that can go on for ever, without ruining the environment in a very comprehensive meaning. All geological deposits are finite, and most of them have a regeneration rate close to or at zero. Thus, no mining is sustainable, when the extraction rate is above the regeneration rate. For a few resources, this implies a small number, and for most elements, it is near to zero. Another approach is to set an end time for the resource use, assuming that civilization ends after that. In this study, two different alternatives were considered 5,000 years and 10,000 years.

If the Greeks of antiquity had started with using oil and coal at large scale, we would have been at the resources end now, facing civilization closure, assuming they would have behaved as we are doing now. A time of 10,000 years represents the time to the next glacial cycle, and that seems as the most reasonable (Brandon Carter, 1983; Kennedy, 1987; Gott, 1993, 1994; Leslie, 1998; Beder, 2000; Sverdrup and Ragnarsdottir, 2014) in the perspective of not destroying the opportunities of coming generations. When the next glacial cycle starts, all rules will change and the world will be a very different place.

For the simulations, the WORLD7 model outputs is presented from 1950 but it is initiated in 1850 at the beginning of the industrialization and runs until 2600, that is 750 years. This may seem long, but the run-time is determined by the need to span over three times the longest delays in the system. The system delays are of the order of 60–120 years; the lifetime for a human is 80 years, the lifetime for a major building is 100–200 years, the turnover of carbon in the soil is from 40 to 100 years, as examples. The turnover of carbon in ecosystem may be in the order of 200 years. Thus, a minimum simulation time of at least 400 years will be required. In this contexts, normal political cycles of 2 and 4 years, or budget plans of 5 or 10 years are irrelevant and short term.

6 Results

The results are presented in two ways. First simulation outputs based on the BAU case are presented. Any model predictions made from the simulation outputs are therefore based on the assumption that the future holds that same basic dynamics as the past has from the modeling perspective. The model simulations were a necessary part to estimate the resources that actually will become extracted. Extraction is dependent of being economically viable, and only the WORD7 model is capable of doing that (Sverdrup and Ragnarsdottir, 2014; Sverdrup et al., 2021). Secondly, the input data and results from the spreadsheet calculations have been presented in a series of tables (due to size limitations some tables are in Appendix). These are somewhat inaccessible, but a necessary part of a spreadsheet exercise. The raw data as it was assembled, collected or created using WORLD7 modeling is assembled in an Appendix. The results are presented below.

6.1 Reserves and resource estimates

The authors have extrapolated the available information to the whole resource quality range to be able in the model to have the economical capacity combined with the demand decide what will be extracted. It is based on data from the literature and synthesized by the authors. Examples are studies like: Dahmus and Gutowski (2007); Demarthe et al. (1990); Frenzel et al. (2017); Johnson et al. (2011); Lasky (1950); Mudd (2007a,b); Norgate and Rankin (2000); Singer and Menzie (2005, 2010); USGS Minerals Summaries (1999, 2002, 2005, 2006–2008, 2011–2017); Werner et al. (2017a,b, 2015); Yellishetty et al. (2017, 2012, 2011, 2010); Mudd (2009a,b); Mudd (2010a,b,c,d, 2012), Mudd and Weng (2012), Mudd et al. (2013a,b); Mudd et al. (2014, 2017a,b); Mudd et al. (2018a,b), Mudd and Murr (1978), Mudd and Jowitt (2014, 2018), Gutowski et al. (2008, 2016) and Hammarstrom et al. (2021), Sverdrup and Ragnarsdottir (2014) and Sverdrup and Olafsdottir (2019a). In cases where no peer reviewed information was available, number estimates were made based on available information from the internet and whatever other sources we could find.

6.2 Simulation outputs from the Integrated Assessment Model WORLD7

Simulation outputs based on the BAU case are presented in Figures 8–11 for the time period 1950–2250. Figure 8 shows examples of the simulations done with the WORLD7 model compared to data 1900–2020 for the ore quality for lead (Pb), copper (Cu), zinc (Zn), manganese (Mn), chromium (Cr), nickel (Ni), and iron (Fe). Very many runs were made with the model to investigate the robustness of the conclusions and to see the general trend in the system concerning resource supply and value generation in society.

The quality of all the resources in the model (Fossil fuels, metals, and materials) are all going down and have reached a quality between ¹/₃ in the best cases to more frequently to ¹/₃₀ of what the quality was 120 years ago. This is a case where the model has been verified on field data, and that it is an undeniable fact that we have exhausted all high quality natural resources, and that only the poorest grades remain for us and future generations (Figure 9, Mudd et al., 2013a,b, 2014, 2017a,b, 2018a,b). A stronger warning sign than this we will not get, and if we fail to internalize the fact, future generations will curse us severely for our inability to act responsibly. The simulation runs after that point in time (2020–2250) show predictions for the future based on that same BAU run. Any predictions made from the simulation outputs are therefore based on the assumption that the future holds that same basic dynamics as the past has from the modeling perspective.

Figure 9 show the stock values in different sectors in billion $ according to the model simulations. The curves are sensitive to assumptions about future legislation and quality of government, but show the peak behavior caused by increased stocks, increased and peaking resource extraction and a peak in population. Qualitatively the output resembles the outputs from the Limits to Growth study of 1972, but with some delay in the peak. In order for a society to be able to perform for the needs of the population, it needs effective institutions. That means that government has to invest in these societies to create them and to keep them updated and effective. Thus, the infrastructure needs to be in place, the technology level must be at the required and necessary level, the infrastructure must be efficiently operated and the work procedures must have the required efficacy. Maintenance is consists of physical erosion of hardware and the slow progress of equipment becoming technologically redundant. The capital stock size is determined by two elements; new investments and depreciation of the investments made. The outputs from the WORLD7 business-as-usual simulation shown are consistent with observations (Credit Suisse Research Institute, 2014).

Figure 9a shows the resource value flow from the big metals. Total value flow from all resources, energy, metals and materials. Figure 9b shows the resource revenue the WORLD7 model for the standard scenario. Figure 9c shows the energy production of the WORLD7 model for the business as usual scenario. The energy production can be maintained at 2030 level only if thorium power is fully developed by 2060. Figure 9d shows the value flow from different types of energy sources for the business as usual scenario. These energy flows were used in the value estimations in the dynamic model. The predictions can be verified quite well on observed historical data for 1900–2025. It is evident that energy flows can be said to be equivalent to huge amounts of work into society that can be converted to money flows into the same society. Energy is the enabler for converting the energy floes to monetary and social value. Without the metals, the energy flows alone lack any value to society beyond basics of heating homes and cooking. This is not fully reflected in their values in the markets, and the real value is possibly underestimated.

Figure 10 shows the global GDP originating from natural resource extraction, not including ecosystems services. The GDP per year and GDP per capita and year from 1850 to 2250 as estimated using the WORLD7 model. When simulated from the year 500 BC to 3,500 AD, the present economic boom as driven by resource use appears as a “blip” in a 4,000 year perspective. It is apparent that the world is in an overshoot and contraction mode unless we go from Business-As-Usual (BAU) to some variant of “Business-as-unusual.”

The utilization of finite reserves and resources have a clear start and a final date when the stocks run empty is suggested by the model runs. GDP will continue to grow until 2060, when it turns down in a period of ups and down until 2180, when a final decline and contraction sets in. After 2300, the final decline is fast down to a level corresponding to the GDP in about 2005. From then the global energy is mainly derived from thorium. Without any thorium energy, the plunge comes earlier, in 2130. That may seem far away, and in most political contexts of the day, it very much is. But in the perspective of normal history books, it is not such a long time.

That must be based on sustainable resource use and utilization, for a modern society to prevail. The graph to the right shows the GDP simulation for the period. In 2020, the fraction contribution was for the direct sales value 16%, if production and supply was included, it was about 30%. As the GDP is defined, about 30% of the GDP is caused by provision of services. If the manufacturing to consumer good would be included, the total would amount to about 65%. If the annual value of the ecosystem services were included for 2020, their fraction would be 61%. The total would be 156%, implying that GDP does not properly measure the value generation for society, it misses very substantial contributions especially in the informal sector as well as the environmental services as income and environmental impacts as costs. Since these terms are as large as industrial terms, the mistake is significant enough to make GDP irrelevant for many aspects of society and social sustainability.

The contribution of natural resources extraction, production, supply to GDP is large, making the concept of decoupling flawed at a fundamental level. Under the BAU scenario, renewables like wind and photovoltaic electricity will not last beyond 2300 because of lack of key technical metals and materials.

The recycling rates (0%−20%) are very small, which is making these technologies unsustainable. A fundamental system change in this arena is required. The present system under business as usual s in no way long term sustainable. No adjustment of the parameter settings will change that. The golden age of the Global Industrial Age goes from 1950 to 2250, or about 300 years. Then comes a period of about 100 years of decline and turbulence, and then the hard landing. The landing will have a parallel in what happened at the end of the Roman Empire (Tainter, 1987): the society simplified, fewer administrative levels, cancellation of large bureaucracies, collapse or complex organizations, reduction in communication efficiency. Large political units broke up into smaller and simpler states. Level of prosperity adjusted down to what the actual income could carry. Without large scale spending of present resources by wealth tax and of future resources through loans, this was a brutal landing implying a significant reduction in disposable income. Collapse of large scale organization often resulted in debt and fortune deletions and sometimes bondage (Tainter, 1987).

According to the simulations, the GDP decline in 2350 will reduce global average affluence to about the same level as 2000 in terms of GDP. This is dependent on the populations simulation shown in Figure 10 being correct, with a global population of about 3.8–4 billion people. If the global population were to stay higher (>8 billion people), the decline will be earlier (2200–2250) and reaching an average global prosperity level corresponding to 1940–1950. Many more scenarios are being explored with the model to investigate possible system changes that may make the system more long term sustainable. From Figures 8, 9 we see that the golden age of the Global Industrial Age goes from 1950 to 2250, or about 300 years. Then comes a period of about 100 years of decline and turbulence.

6.3 Accuracy and uncertainty

A valid question is how accurate are the predictions, and do we have an estimate of the uncertainty. This study rests on assessment made for 48 metals, minerals and energy commodities. These assessment have all been published and have been listed in Appendix 2, in Table A1. They were published in the time period 2011–2026. For each of the metals and minerals and commodities, the model was tested to see how well it can reproduce the observed data for extraction rate, often ore grade when such data was available and market price. Market price is modeled indigenously in the model, and perform well in the available tests (Sverdrup and Olafsdottir, 2019a,b). Data on extraction rate and price was normally was available for the time 1900 to the year before publication. Sometimes more parameters were available for testing (Stock estimates). The model is not subject to much calibration as it is based on physical properties and mass and energy balance constrained. All model parameters have physical significance and can in theory be measured. To take note of is that when the model is run, all the metals are run simultaneously and the model should recreate the observed extraction rates and price trajectories in all 48 metals, minerals and commodities simultaneously. The WORLD7 model actually does that.

In general, the historical patterns observed are simultaneously well captured. The model assessment were often also tested by plotting cumulative extraction vs. observed extraction. Those comparisons show as a general pattern that a bias do not build up with time. This implies the uncertainties do not build up with time in the model. The most uncertain parameter to the model is future demand. In many of the assessments, but not all, this was explored by running a simple scenario analysis, running demand through a number of future levels. Demand is partially estimated in the WORLD7 model, but part of that has some assumptions that needs to be set. For example, the use of different critical metal for renewable energy end electrification technologies has a higher initial value per unit and then declines as the technology matures. This was used to evaluate the validity of the conclusions of the assessments as well as suggest approximate level of uncertainty.

To be remembered is that the model outputs are forecasts, based on a model that has been tested on data series from the past. There are no future data to test on, else than the fact that the model is run on mass and energy balances. Thus, the outputs must be understood in the same sense as weather forecasts, they represent what we think is plausible, but there is no certainty that it will be like that.

A full uncertainty analysis on a large model like WORLD7 is a very large undertaking in terms of work and has not yet been done. That will remain for a future publication once the work has been funded and done. A full uncertainty analysis of the WORLD7 simulations is outside of the scope of the present essay and remains a limitation of the current study.

6.4 Monetary value creation and wealth stock estimates

Different aspects of value of production and value of stocks have been made and these have been displayed in the Tables 2, 3 and Tables A4, A5 in the Appendix. Table 2 compares the value of all-natural resources and ecosystem services. An adjustment factor of 4 from 1997 to 2026 was used to lift in the values published by Costanza et al. (1997) to the present time (2026), compensating for inflation and changed utilization of ecosystems for human society. One of the assumptions is that coal, oil, and gas production has been reduced to 10% of the 2020 level by 2060 (A global “Energiewende,” see Lorenz et al., 2017; Hirschnitz-Garbers et al., 2018a,b; Sverdrup, 2019 for further explanations). Both this study and the study of Costanza et al. (1997) indicate that the natural resource values dwarfs the annual GDP. Even if the values between Costanza et al.'s study and ours may differ some, the main overall systemic conclusions reached are the same.

Table 2. Profit value in 2020 and sales value in 2020 for the total stocks of different natural resources.

The stock profit difference between thorium and uranium come from the high expenses of final waste disposal and nuclear arms proliferation. The stock value was calculated integrating the revenue value over the full resource extraction time. Take note that thorium and uranium dwarfs everything else in stock value, and in net profit, thorium dwarfs everything else.

Table 3. The annual primary energy production (1), the supply energy production in 2020 (2), the market price in 2024 (3), and the value of the supply in 2020 (4), and in 2060 (5).

We have estimated a significantly reduced fossil fuel production by 2060.

“Invaluable” implies that this resource is existential and if it fails, civilization comes to an end. The estimate value of the solar energy power potential, assuming a 300 year lifespan of each solar power plant. An assumption was made to fill the energy gap in deficit in Table 3 between the predicted energy supply 2060 and the current energy supply in 2020 when fossil hydrocarbon fuels production have declined as a result of exhausted resources and limiting policies. The gap is the difference between the total energy demand and the actual total production achieved. When fossil fuels production declines faster than what renewables can increase, an energy gap opens up. This must be covered by other production or by savings and reduced use. To reduce total energy use, is very hard to do when the population is still growing. However, the system will be self-regulating at some point, as it is not possible to use energy that we do not have (Ehrlich, 1968; Ehrlich et al., 1992; Eisler, 1987).

The simulated decrease in fossil fuels supply can be filled with one of two potential sources; (1) Nuclear power based on fast uranium breeder technology, or (2) thorium-based breeder reactors. Conventional nuclear power based on uranium is not a long term future alternative, it is only a short term measure to bridge a gap of maximum half a century. The uranium breeding alternative (1) is suggested to be rejected, as it produces large amount of weaponizable material and must be classified as far too dangerous. The risk for nuclear arms proliferation is already too high and cannot be further increased when a decrease is more needed.

The thorium-based technology is acceptable as there are very limited nuclear arms proliferation issues and a much smaller nuclear waste issue, combined with a much larger fuel efficiency. The technology is capable of filling the gap for many millennia. The cost for the uranium and thorium is calculated in cost per oil equivalent if the uranium or thorium is used in a nuclear reactor. Thus, we express the extraction cost per oil equivalent. For the term “invaluable,” the estimate was made, monetizing the unmonetizable, using a cumulative of the income (Costanza et al., 1997) for 5,000 years. The metals and materials are 7% of the GDP, the fossil fuels represent 9% of the GDP, the ecosystem services are 62% of the GDP, in total 78%, leaving 22% to human ingenuity and enhancement. Not counted in are the interdependencies between these resource types and the fact that if one fails, then the whole system fails.

Table 2 shows value of the profits in 2020 and cumulative sales value in 2020 for the total stock of resources, taking the price from WORLD7. If a dynamic model is used, this will be different, as the price will be dynamic and the price at the time of extraction. The recycling amounts come from WORLD7 simulations. We consider the model estimates to be more accurate than the global recycling older values available in the literature. From what the authors have established, UNEP (Reuter et al., 2013a,b; Raabe, 2023) recycling estimates are educated guesses, and only to a limited extent go back to data from actual measurements. For gold, silver, copper, platinum, the authors have unpublished industrial measurements available to verify the estimates.

The cost of extraction of the available reserves for metal extraction have been expressed in in $/ton metal extracted when all costs are counted in: labor, materials, energy, and capital costs. The estimates are based on published literature (see reference list and Sverdrup and Ragnarsdottir, 2014), branch organization overviews and ephemeral trading platform information. Much of the cost for uranium comes from the extra handling and processing of radioactive material and reprocessing and storage of the final waste. For the cost of ore extraction, beneficiation, metal refining and final alloys composure, very little information is available in the published literature.

Much information is available on corporate websites, annual reports, trading blogs, as well as in proprietary reports from professional consultants. These sources are less accessible to those not familiar to the industry. The information is not less reliable, corporate profitability is often dependent on the quality of the information, creating very strong incentives to keep a high quality. Note that this may be something else than what the companies verbally report to news outlets and at political hearings. From a scientific point of view, this may seem undesirable, but such are the facts on the ground and we have to accommodate that. Gravel, sand, and stone show a large value, because humans use a huge lot of it.

Every one of the 8–9 billion people on Earth use it in many ways. Intuitively, it may appear that resources like sand or gravel should never run out. However, this is the case, we may run out of suitable sand and gravel (Sverdrup et al., 2018). Both are a scarce resource in the quality demanded. Further demand would have to be covered by using poorer qualities and processing them, increasing the cost and thus the price. Large amounts of sand cannot be mined without significant environmental and landscape stability impacts. The best qualities are quickly being consumed, and soon only more expensive artificial sand will be available.

6.5 Energy

Table 3 shows the annual primary energy production, the secondary energy production, the 2060 market price and the value of the supply. The assumption is that coal, oil and gas production has been reduced to 10% of the 2020 level by 2060. Another assumption for the spreadsheet estimates behind the tables was that the price was the same in 2020 and 2060. This is probably an underestimate for 2060, and this can only be handled in the dynamic model. Note that there is a difference between the energy sources dominating in 2020 and 2060. Table A5 in Appendix 1, shows the annual primary- and secondary production, the 2020 market price and the estimated value of the supply. The WORLD7 model was used to draw up alternative energy source like photovoltaics and windmills to a maximum as allowed by key metals and minerals and driven by subsidies and sales income.

The supply in 2060 represents the maximum amount available to get out of such energy production technologies. There is a residual of 6.945 billion ton oil equivalents per year that cannot be filled by remaining conventional methods. The energy deficit in 2060 could potentially be covered by nuclear energy based on thorium. Table A2 in the Appendix shows the estimation of available reserves and resources (URR) for fossil fuels, materials and metal extraction at the beginning of the industrialization in 1840 and Table A3, Appendix 1, show the estimated amounts in different resource grade classes. The amounts are in million ton of material type listed. This is the natural resource amount input file for the WORLD7 model as of 2020.

7 Discussion

7.1 On nuclear power and the thorium technology

The reader may wonder why we suggest to use thorium-based technologies. There are two reasons for this. One reason is that China fully embraces their own thorium technology and already have operational reactors (Michaux, 2024; Michaux et al., 2025). India is busy with an advanced thorium power program and are running experimental reactors (Michaux, 2024; Michaux et al., 2025). Canada has significant resources of thorium and research in thorium power is ongoing for a long time (Graves et al., 2015; Michaux et al., 2025). Both countries realize that oil and gas is soon ending, that continuing with coal is not long term sustainable, and that a new large scale energy source that will last a long time is needed.

The second reason is that modular thorium-based molten salt reactors are being developed in Copenhagen, Denmark, Europe, by a commercial company, Copenhagen Atomics. The two first commercial types of thorium-based molten salt reactors are standing ready in Copenhagen, Denmark, and the cost data and the energy data are based on the hardware measurements and physical tests on these units during 2023–2026 with molten salts with and without nuclear component. During late 2027 or in 2028, one MSR unit will go through technical verification, filled with liquid thorium fuel, made critical and be subjected to live performance tests and a certification run at The Paul Scherrer Institute in Villingen, ETH in Switzerland (Copenhagen Atomic Webpage May 2026). If the test go as well as the staff at Copenhagen Atomics as well as at the Paul Scherrer Institute expresses, then this is a viable technology. And it appears that provided the assembly line will stay on track, these units will be commercially available earlier than commonly anticipated. There is more information available on this at the Copenhagen Atomics webpage. IAEA on their website sound very positive about the feasibility of the Copenhagen Atomics approach.

The WORLD7 model has a nuclear energy submodel, containing options for simulating uranium-based and thorium-based reactors of different kinds. The uranium and thorium availability is simulated as a part of the resource extraction module in WORLD7. These two modules are linked in the energy module in WORLD7 and driven by demand from there.

Nuclear energy based on enriched uranium is seen as a finite resource and often classified as one too dangerous to keep, and substantial political efforts are made to phase it out. This has historical reasons and is coupled to the fear of nuclear arms proliferation and the fear of accidents. The duration of the uranium resource for conventional use is only marginally longer than fossil hydrocarbons (100 years, IAEA, 2019).

With the Molten Salt Reactor technology applying thorium breeding, the situation is different. Molten Salt Reactor technologies can be run with natural uranium, uranium 235, uranium 238, plutonium or with only thorium 232 or a mixture of them all. In short, the thorium fuel energy yield can be made to approach 95% and there is no weaponizable material created. The energy contained in the thorium resources is very large and will last far into the future (Martin, 2012; Sverdrup and Ragnarsdottir, 2014; Michaux, 2024; Michaux et al., 2025; IAEA, 2005, 2019; Francois et al., 2004).

Conventional uranium powerplants means digging up 60,000 ton/year of uranium and using 0.5%−0.8% for fission conversion to energy, thus consuming 500 ton of uranium/year. To cover for the future global deficit of 6,945 million ton oil equivalents per year in 2060, it would be necessary to extract 430,000 ton uranium per year for conventional reactors. At a total remaining global stock of 29 million ton uranium, would imply full uranium exhaustion within 58 years (Michaux, 2024; Michaux et al., 2025). This would produce 430,000 ton of radioactive waste every year (30,000 m³ per year) with significant trans-uranide elements contents, that need to be isolated for at least 100,000 years in nuclear waste repositories. Converting to uranium breeder technologies would be able to increase the yield and energy amounts significantly and the available uranium would last for the order of magnitude of 5,000 years or more. A major drawback is that during the burnup of the uranium, significant amounts of plutonium will be made (Hoisington, 1959; IAEA, 2016; Etherington, 1958; Kasten, 1998).

Using thorium for heat and electric power has the same radiation safety issues as a uranium-based power plant based on Molten Salt Technology while it is operating. The MSR plant itself will be radiation contaminated by the fission processes, but the fuel waste amounts will be 1/50 of conventional nuclear power and with less severe radioactivity. Burn-out of the thorium-related waste will be completed in 250–300 years (Dolan, 2017). To produce the energy deficit of 6,945 million ton oil equivalents per year before nuclear energy is counted in, in 2060 as compared to 2020, the thorium needed by a MSR's has been estimated at 3,100 ton/year. By estimating 90% efficiency, and considering yields and losses, 3,400 ton thorium/year would have to be extracted from the ground. In a perspective of a total global resource of minimum 25 million ton of thorium (Sverdrup and Ragnarsdottir, 2014; UNEP, 2015; Sverdrup et al., 2017a,b; IAEA, 2019; Michaux et al., 2025), that would be 25,000,000 ton divided by 3,400 ton/year is 7,300 years. The total resource is more likely to be 100 million ton if the energy content is considered, resulting in about 30,000 years.

Amounts of waste from thorium-power is in the order of magnitude of about 100 ton of waste per year, but without the trans-uranite elements, requiring storage for about 300 years (Martin, 2012; Dolan, 2017). The rest, of the total 3,300 ton waste per year (300 m³/year) would be low activity, requiring storage for about 50 years (Martin, 2012; Dolan, 2017). If the value of the energy content would be used for the breaking point for thorium, then the global resource would be at least 55 million ton thorium. The global thorium resource would then last about 16,000 years to supply the whole global energy deficit as given above. Factoring in that the Earth's population could decline during the period to about 3 billion people, then thorium could provide nearly all energy for about 50,000 years. This is consistent with press release statements released by the Chinese Energy Ministry in 2025.

To supply a total energy production of 15,000 million ton oil equivalents per year (It is 11,000 million ton oil equivalents per year now), the available thorium would last about 11,000 years. If the population declines to 3 billion people from 9, it will last 33,000 years (Michaux et al., 2025). The conclusion is that thorium is an option for replacing fossil fuels for energy production for a substantial length of time.

7.2 Policies and ramifications

The different energy technologies have been assessed for whether they are sustainable or not (Koca et al., 2017), this assessment can be found in Table A5 in Appendix 1. All technologies have sustainability issues, and only a few of them are truly sustainable.

The energy source should be sustainable, the technology should be sustainable, and within all aspects physically, economically and socially. Nearly all natural resources are finite, and there is no escape from this fundamental fact (Eddington, 1928). This is a fact known to every mining company, every oil company in the world. Not accounting carefully the natural resources is a wasteful practice and reduces the possibilities of future generations.

Some occasionally claim that substitution will solve the scarcity problem. Substitution on industrial scale only works when the element being substituted is significantly smaller in consumption volume than the resource used to substitute (Ragnarsdottir et al., 2017). Substituting copper with silver is not possible in terms of volume. A significant issue is that nearly all metal supplied is booked. Using for example aluminum to substitute for copper implies that the capacity used for substitution will have to be made available by canceling an existing use. Substituting molybdenum with niobium, means that some other use of that niobium will have to be canceled and that particular service would not be provided (Waring, 1998). Our findings imply that decoupling and substitution are dead ends (Ragnarsdottir et al., 2017; Bringezu et al., 2004). Renewable energies biofuels, carbon ecosystem sequestration, and hydropower are to a large degree dependent on significant ecosystem functionality, and can only extract material or energy within those constraints.

7.3 Creating real change

Creating change needs much more than technical innovations and political slogans. In order to make change, minds and wills must be changed. That is a difficult process if it is to be done within the bounds of democracy, without violating human rights or without disrupting society. Will to change in society must be created through understanding of the psychology of change and the psychology of motivation in order for a process of change to succeed. There are several types of change that must occur simultaneously: (1) Technical and technological change in component, (2) Change in infrastructure structure and organization, (3) Change in will for change in the public, and participating in the process of change, (4) Change in economic paradigm and business models, (5) Social adaptation, training and learning, as is helped by systems thinking (Senge, 1990) for a critical mass of the society, and finally (6) Clear visions and overarching goals.

To change society toward a path of sustainable future, many concepts will be challenged and changes will be needed (Fletcher and Rammelt, 2017; Kegan and Laskow-Lahey, 2014; Kotter and Schlesinger, 2014; Meadows et al., 1972; Heinberg, 2011; Sverdrup, 2019, 2020a, many articles by Costanza et al., 1997, 2014a,b):

• The power structures that exist in society are set in certain paradigms with a certain set of people from which they are unlikely to move. When new competences and understandings are required, new persons will move into positions of power, and old positions and structures may become redundant. This will be seen as a threat to those who are in power at present. Changes will be resisted.
• The way decisions are made must change, and operational sustainability criteria will change the way decisions are made. Making decisions about investments and giving credits will be different than today. Profitable, but unsustainable projects done in the past, can no longer be permitted or granted credit in the future.
• Consumption patterns will have to change, contingent on a behavioral change. For habits worked in over long time, this can be very challenging. Total global consumption volume is too large and must in some way be contracted significantly.
• That long term social patterns are unsustainable will challenge political paradigms, religions and beliefs. This will make many upset and could be the start of “resistance movements.”

Unless there is a situation that is recognized as a real existential crisis, it will be unlikely that anything substantial policy change will occur (Keynes, 1936; Keynes et al., 1932; Kegan and Laskow-Lahey, 2014; Eisler, 2007; Tilly, 2005, 2007). While behavioral science often uses individual perspective on “quality of life,” social science often addresses “social quality.” Stiglitz et al. (2010); Schutz (1958, 1992, 1994) were some of the pioneers in this field, working with motivational aspects for the American Army nearly 60 years ago. Both might constitute a way to address the shortcomings and go beyond GDP. Social quality is way of understanding society that has relevance for social and public policy. It has some main indicators:

• Economic security—the extent people have resources over time.
• Social cohesion—the extent to which social relations based on identities, social trust, moral values and norms are shared.
• Social inclusion—the extent to which people have access to and are integrated into institutions and everyday life social relations.
• Social empowerment—the extent to which the personal capabilities of individuals and their ability to act are enhanced by social relations.
• Social efficiency—Efficient state institutions are necessary for securing social trust in society and efficiently mitigate corruption in society.

Quality of life assessments have indicators based on individual needs like income and jobs, health, housing, education, economics, and social quality. Social quality adds indicators such as trust, norms and values, openness, identity, and personal relations etc. The indicators can be individual or society based, but are also connected in many ways. Psychology on motivation divides motivation into intrinsic motivation, or inner motivation; an activity is undertaken for its own sake, for fun, for learning or just for feeling accomplished, extrinsic motivation is driven by rewards or to avoid sanctions (Ryan and Deci, 2000; Keller, 1987; Malone, 1981; Kegan and Laskow-Lahey, 2014). This is done by carefully coordinating top-down leadership with bottoms-up motivation-driven activity, managed with adaptive management methods (Senge, 1990; Senge et al., 2008).

Figure 11 shows the concept of Gap Closure: the simplified version of a proposed methodology of systems analysis and dynamics modeling to generate input to policy proposals and policy development (Sverdrup et al., 2022). Red lines are political pathways and feedbacks, black lines are at the interface of civil executive administration and applied research. Doing policy is a complicated matter, and there is simply no quality government without systems thinking. Red lines are political pathways and feedbacks, black lines are at the interface of civil executive administration and applied research and blue lines are found at the interface between public agencies for state management and research. What to do in terms of policy and governance is far too complicated for being developed with traditional discussions and linear policy processes. This can only be developed as an adaptive learning process involving the stakeholders; policymakers, politicians, experts, scientists and representatives of the population, using quantitative set goals and integrated assessment models developed fand adapted or the purpose. Many hard choices must probably be done, assessing if enough goals can be met, if the sum of benefits outweigh the drawbacks, and that system stability within the frames of democracy can be maintained.

The diagram in Figure 11 says that for sustainability indicators to be politically effective, they must be measurable. Quantitative goals must be set. When policies are suggested, dynamic models must be used to test if the proposed policies really can achieve the goals. If not, the policy is not good enough and must be reworked. When policies are implemented, regional quantitative measurements of the indicators must be done to assess if the policy works at the sustainability targets.

8 Conclusions

The creation of wealth and money is dominated by the use and conversion of natural resources. It means that wealth in society is to a large degree a fossil wealth (70%) and it is finite. The use of advanced technologies and services depend on this primary ecosystem level and accounts for the remaining 30%.

The main overarching conclusion of this study is:

• Energy is the main world currency of the world economic system (Oil, gas, coal, thorium, uranium, biofuels, and solar energy)¹ and a main source of value creation.
• Physical materials is the second world multi-currency of the world. Each metal and materials is a currency in itself (Fe, Al, Cu, Ni, Cr, Mn, Au, and Ag++).
• Ecosystems are the third type of world currency.

This is supported by the natural resource stock and flow tables and the calculations done to interpret them. This conclusion was also reached with different wording by Odum (1996) in his key work on Environmental accounting. The implication is that any economy that does not keep track of these in proper bookkeeping, do not have the data required to do serious decisions about both short and long term sustainability. The model simulations and our simple paper desktop estimates gave the following suggested outcomes for the scenarios selected for this study:

• Business-as-usual scenario outcomes
• Paradigm shifts scenario outcomes

All use and available stocks of fossil fuels, fossil metals and fossil materials has an end date. Solving the climate change challenge with an energy transition, creates a challenges of natural resources limitation, a demand for a resource transition, which both demand a rethinking of behavioral paradigms in society, calling for a social transition (Hirschnitz-Garbers et al., 2017, 2018a,b; Sverdrup et al., 2021). What to do in terms of policy and governance can only be developed as an adaptive learning process, using quantitative set goals and integrated assessment models developed fand adapted or the purpose. The existing simulations models only address this aspect partially and with vague output formulations. This aspect is under development, but more funding and time is required for substantial progress.

Generative AI statement

The author(s) declared that Generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fsrma.2026.1841357/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Bijan Rahimi, Enaxiom Pty Ltd, Australia

Reviewed by: Elliott Thomas Campbell, Maryland Department of Natural Resources, United States

Ranaprathap Katakojwala, Greeneple Inc., Republic of Korea

Footnotes

Solar influx of energy to the earth is the source of energy for hydropower, wind power, photovoltaic power, heat harvest and growth of biofuels by photosynthesis.