This chapter covers the valuation of mineral and energy resources, distinguishing between non-renewable mineral and energy resources and renewable energy resources. Both types of resources are treated systematically. First, the scope, definition, and classification of the assets to be included in the national accounts is provided. Then four compilation stages are presented and explained. Finally, resource specific compilation issues are covered as well as modifications to the standard approach.
Measuring Natural Resources in the National Accounts
A Compilation Guide
Abstract
241. This chapter covers the valuation of mineral and energy resources, distinguishing between non-renewable energy and mineral resources (Section 4.2) and renewable energy resources (Section 4.3). Subsoil assets will be used in this chapter as shorthand for non-renewable mineral and energy resources (AN321 in 2025 SNA classification of assets).1
242. The 2008 SNA already required valuation of subsoil assets as well as their cost of depletion (as an economic disappearance of non-produced assets, K21). In the 2025 SNA, however, the depletion of natural resource will be recorded as a cost of production, no longer as other changes in the volume of assets.
243. Guidelines for measuring and valuing subsoil assets already exist in the form of the Eurostat guidelines (2003[1]), the SEEA Central Framework (CF) (United Nations et al., 2014[2]), SEEA Energy (United Nations, 2019[3]), the IMF Guide to Analyzing Natural Resources in National Accounts (IMF, 2017[4]) and results of the OECD Taskforce on SEEA Implementation (Pionnier and Yamaguchi, 2018[5]). In addition, there is quite some country experience: about a dozen countries regularly compile estimates for mineral and/or energy resources. This chapter structures the existing guidance and country experiences, and when necessary, updates it to take the 2025 SNA recommendations into account. It also contains a number of concrete compilation steps, as well as recommendations to enhance international comparability. In addition, two accompanying workbooks (one for subsoil assets and another for renewable energy resources) are included to facilitate and illustrate compilation.
244. The situation regarding renewable energy is different from that of subsoil assets. The 2025 SNA for the first time, explicitly recognises renewable energy resources such as hydro, wind, geothermal and solar resources as economic assets. The changes will primarily affect the balance sheet. They will have no impact on the production or income accounts.2 In the case of renewable energy, it is assumed that no depletion takes place because such resources do not (at least in the relevant time frame) decrease due to their use, nor do they regenerate.
245. The point of departure for the guidance on renewable energy resources provided in this Chapter is the 2025 SNA, complemented by Guidance Note WS.11 (SNA Update, 2023[6]) which provides a lot more detail, and the results from early implementation by countries participating in the EGNC. While Guidance Note WS.11 restricted its discussion of renewable energy resources to solar, wind, hydro (water) and geothermal sources for electricity production, the guidance provided here widens the treatment towards all possible renewable energy sources within the 2025 SNA asset boundary including assets for the production of heat and cool air.
246. A three-tier approach (basic, standard, advanced) is used in the chapter to provide different options for compilers depending on their data availability and resources. The standard approach will be described as the “default”, with a specific section dedicated to describing modifications in case a basic or advanced method is preferred by countries.
247. The outline of the chapter is as follows. Sections 4.2 and 4.3 follow the same structure. First, the scope and definition of the assets to be included in the national accounts is provided, as well as relevant classifications. Second, four compilation stages are presented and explained: identifying the types of assets to be included; collecting the physical data; building the monetary asset accounts; and integration of the results into the sequence of economic accounts. Third, we discuss specific compilation issues; and finally, modifications to the standard approach. A text box with a summary of key recommendations concludes the chapter. Two accompanying Excel workbooks have been developed: Workbook: subsoil assets, and Workbook: renewable energy resources. These facilitate and illustrate compilation according to the 2025 SNA and are referred to in Sections 4.2 and 4.3.
4.2.1. What to include in the national accounts
248. The choice of subsoil assets to include in national accounts estimates is based on the United Nations Framework Classification for Resources - Update 2019 or “UNFC-2019” (UNECE, 2020[7]), which is a standardised classification system for defining the environmental-socio-economic viability and technical feasibility of projects to develop resources (both renewable and non-renewable). The classification categorises mineral and energy resources through determining whether, and to what extent, projects for the extraction and exploration of the resources have been confirmed, developed or planned. The underlying resources are classified based on the maturity of the projects in terms of economic and social viability (E), field project status and feasibility (F) and geological knowledge (G), see Figure 4-1.
Source: UNECE (2020[7])
249. In the SEEA CF, the known deposits are categorised in three classes, while each is defined according to combinations of criteria derived from UNFC-2019:
Class A: commercially recoverable resources.
Class B: potentially commercially recoverable resources.
Class C: non-commercial and other known deposits.
250. To facilitate the delineation and international comparability of mineral and non-renewable energy resources in 2025 SNA, only Class A resources are to be included (§11.186): “The measurement of monetary estimates is typically restricted to the first class, which in practice could be approximated by those resources for which permissions to exploit have been granted, and/or those for which the existence is explicitly recognised by (past) monetary transactions. Potential mineral and energy resources where it is not foreseen that they will be exploited in the near future are thus explicitly excluded.”
251. Class A is equal to the class of viable projects defined according to the UNFC consisting of classes E1, F1, G1-3 (see Table 4‑1). The class therefore includes both resources that are currently exploited as well as resources that are expected to be exploited (with a high degree of certainty).
4.2.2. Compilation stages
252. Regarding the measurement of subsoil assets, four compilation stages are distinguished: identifying the types of assets to be included; collecting the physical data; building the monetary asset accounts; and integration of the results into the sequence of economic accounts.
Stage 1 (subsoil assets): Identifying types of assets
253. In the first step, the different types of subsoil assets that fall within Class A, and are therefore “in scope”, need to be clearly identified. The 2025 SNA asset classification distinguishes four categories: AN321S1 coal and lignite, AN321S2 oil (petroleum) and natural gas (further disaggregated into oil and natural gas), AN321S3 minerals, and AN321S9 other non-renewable mineral and energy resources (see Section 2.5). Although these are supplementary items, countries are encouraged to provide this breakdown (as a minimum) for the dissemination of asset values. However, production of the estimates for these categories following the recommendations set out in this chapter will be at the level of more specific subsoil asset categories. A more detailed break-down (see country example of Canada) such as the one proposed in the SEEA CF3 could be disseminated as part of the environmental-economic accounts (either in physical or monetary units); but this goes beyond the 2025 SNA requirements.
254. The SNA is exhaustive, implying that all assets with economic value (the Class A resources discussed above) are within the asset boundary and should be estimated. However due to the resource intensive nature of the work required to value subsoil assets, it is proposed to apply a materiality threshold. It is therefore suggested that it would be reasonable to focus on valuation of subsoil assets that contribute more than 5% of output in ISIC Section B: Mining and quarrying (see Table 4‑2) and for which the long-term average contribution of mining and quarrying to GDP is at least 0.1%. When a reasonable estimate can be made of how much of the asset value is missed, it is recommended to gross up the asset value.
255. In the absence of an internationally agreed detailed classification for mineral and energy resources suitable for statistical purposes (as noted in SEEA CF, §5.181), it is recommended to use the following list of commodities as a checklist to assess which energy and mineral resources are available in the country and are above the materiality threshold in terms of output:
Coal and lignite: Hard coal; Lignite4 (ISIC Division 5).
Crude oil (petroleum) and natural gas: Crude oil; Natural gas (ISIC Division 6).
Minerals: Iron ore; Bauxite; Copper; Tin; Zinc; Lead; Nickel; Gold; Silver; Uranium and other metal ores used as nuclear fuels (ISIC Division 7); and Phosphate (ISIC Division 8).
This list of subsoil assets has been selected based upon economic and environmental significance. However, if other commodities are significant in a particular country, they should also be considered.5
Note: There has been no change in Section B of the ISIC between Revision 4 and Revision 5.
Stage 2 (subsoil assets): Collecting the physical data
256. Once the relevant subsoil assets are identified, the following stage is to compile a physical asset account for the individual resources in scope (see Table 4‑3). The physical asset account provides information on stocks and changes in stocks which are important for the estimation of the asset life, cost of depletion and other entries in the non-financial accounts and the financial accounts and balance sheets.
257. A full physical asset account distinguishes between additions due to discoveries, upward reappraisals and reclassifications, and deductions distinguishing between extraction, downward reappraisals, reclassifications as well as catastrophic losses.6 It would be ideal to obtain or compile a full physical asset account for all the individual resources in scope of the national accounts, as this allows to make a clear distinction between revaluation and other changes in volume (see Stage 4 (subsoil assets): Integration). However, as a minimum, information is needed about opening stocks, extraction, and closing stocks.
Source: Based on SEEA CF Table 5.8
258. An issue may be that these data sources may not align fully with the UNFC classes in which case a crosswalk (mapping) needs to be undertaken between the national classification of resources and the UNFC classes, such as the one shown in Figure 4-2. A full mapping is undertaken for all resources, but in the UK case only those coloured in dark blue in the mapping, which correspond to Class A, would be included in the national accounts. Other examples of crosswalks (mappings) can be found in the correspondence tables of the OECD dataset on Mineral and Energy Resources.7
Source: Adapted from Pionnier and Yamaguchi (2018[5]).
259. In the absence of national data on reserves or extraction rates (or in case of difficulties obtaining the national data) Table 4‑4 describes several databases with global coverage that may contain relevant information on extraction / production, physical proven reserves (and how they change over time), as well as unit prices and costs of commodities.
Stage 3 (subsoil assets): Building the monetary asset accounts
260. Once the physical asset accounts have been compiled for each resource, the next stage is to compile the monetary asset accounts. These provide information on stocks and changes in stocks that are needed to populate the relevant accounts in the 2025 SNA, including estimates of the cost of depletion and various entries in the capital accounts and balance sheets as explained in Stage 4.
261. Eight steps are required to compile the monetary asset accounts, as shown in the example for the year 2023 in the Workbook: subsoil assets, which is discussed below. The eight steps are:
Calculate resource rents (past and present).
1. Project the physical asset account and physical output until end of the asset life of the resource.
2. Calculate the unit resource rent.
3. Smooth unit resource rents to address price volatility.
4. Project future resource rents.
5. Calculate NPV for the opening stocks.
6. Calculate NPV for the closing stocks.
7. Put together the monetary asset account.
262. It should be noted that when valuing subsoil assets, it is preferable to use observable market prices for transactions. As discussed in Chapter 3, such markets are often very thin or do not exist at all, and valuation based on permits may only be feasible in certain circumstances. Therefore, in most cases, the recommended valuation method is to calculate asset values as the Net Present Value (NPV) of future resource rents, which are projected from actual (past and present periods) resource rents calculated with the Residual Value Method (RVM).
Step 1: Calculate resource rents (past and present)
263. When using the RVM to calculate resource rents (see Section 3.3), a distinction can be made between a top-down approach using information included in the national accounts or a bottom-up approach applying information from other data sources such as business statistics. For valuing subsoil assets, Chapter 3 recommended applying the top-down approach.8 Table 4‑5 presents an example, which follows the steps of the RVM top-down approach set out in Figure 3-1. The calculation is shown in the Workbook: subsoil assets in rows 6-25 of the Year 1 and Year 2 worksheets.
Notes: Cells in green indicate input data; blue indicates calculated data. Specific taxes on products/specific other taxes on production should be recorded as rent payment (D45) when government is the legal owner. Depreciation includes decommissioning costs (see Section 0).
Source: Workbook: subsoil assets, Year 2 worksheet.
264. The main difficulty here is the aggregated nature of the top-down information. In some cases there may not be a one-to-one link between the physical data (which may be for an individual resource such as iron or bauxite) and the monetary data which will probably be at ISIC group or class level, as shown in Table 4‑2 and may cover multiple resources. There are two ways to proceed.
265. The first option is to aggregate the physical data so that it matches the available national accounting aggregates (e.g. aggregate hard coal and lignite reserves). This would only work for resources that are measured in the same units and have similar prices per unit, which is unlikely in most cases.
266. The second option – which is recommended in this guide – is to compile the monetary estimates to the same level of disaggregation of the physical data. A partitioning of the national accounts aggregates by individual natural resource should be made based on relevant data on output, costs and capital stocks for each resource. While this may introduce some uncertainty regarding resource rents of specific resources and eventually also asset values, the advantage is that it would not affect the total value of subsoil assets which is based on the national accounts data. As a result, one should obtain a time series of actual (past and present) resource rents for each of the individual resources in scope.
267. Countries that extract subsoil assets already compile figures for the production and generation of income account and therefore have estimates of gross operating surplus (GOS) for the relevant industries as a starting point for the top-down approach, thus it is not necessary to elaborate on compilation of these figures in this guide. However, some guidance may be helpful for estimating specific taxes/subsidies. This is discussed in detail in Section 3.4.2.
268. As explained in Section 3.3.1 the user costs consist of two elements: depreciation and a net return to produced assets, specifically fixed capital. The value of depreciation (consumption of fixed capital) is usually derived from a perpetual inventory model (PIM) and available by industry as part of the national accounts. The asset life of the fixed assets could differ from the asset life of the natural resource, but in principle it should not be longer than the life of the natural resource.
269. The return to fixed capital is estimated by multiplying the value of fixed assets (including mineral exploration and evaluation) (row 22) with the rate of return, which is specified in cell H24. The rate of return is assumed to be 6% real in the example, but countries should apply their own rate of return.
270. After deducting both elements – depreciation and the return to fixed capital – from GOS for the derivation of resource rent, we obtain the resource rent (consisting of depletion and net return to subsoil assets).
Step 2: Project the physical asset account and physical output until end of the asset life of the resource
271. In the Workbook: subsoil assets (Year 1 worksheet, rows 33-43) and Table 4‑6, we assume that in 2023 we have an opening stock of 1000 physical units (e.g. tonnes of coal), and an expected extraction of 100 units per year, which is based on the last year and is assumed to remain constant (see Chapter 3). Alternatively, it is possible to use a specific extraction profile when available, for instance when prescribed by government. When making projections, we only project extractions, as by definition we do not have information about discoveries, reappraisals or reclassifications. The assumption of a constant extraction of 100 units implies that the resource will be exhausted by the end of 2032 (see orange column). When making projections, the key restriction is that the total projected extraction over the asset life at the start of the accounting period should be equal to the total physical opening stock (Class A) as included in the physical asset account for the resource in question.
Note: Cells in green indicate actual data; yellow indicates projections; orange indicates end of asset life.
Source: Workbook: subsoil assets, Year 1 worksheet.
272. The use of a constant level of extraction is also supported by findings of the OECD Task Force on SEEA CF implementation (Pionnier and Yamaguchi, 2018[5]) which investigated this issue, looking also at the scientific literature. One of the basic models describing the extraction and valuation of subsoil assets is the so-called Hotelling (or net price) model which assumes producers can optimize when they extract (e.g. the extraction rate is endogenous) in which case the valuation of the resource simplifies to the unit resource rent times the total physical stock (the unit resource rent of optimising producers would increase at the discount rate). Empirical evidence however supports the assumption of constant output along mines’ service lives, as there oftentimes are capacity constraints on production (it is not possible to all of a sudden extract more due to restrictions in capacity of machinery; extracting less also does not happen much due to investments in fixed capital which can be seen as sunk costs). This supports the recommendation to assume extraction will continue at the same level as in the (recent) past since this is the level for which an appropriate amount of produced assets have been acquired.9
For Steps 1-2, the following data are required:
Physical information on stocks and extraction: this may already be available as part of the energy statistics or may need to be collected from agencies such as the natural resources department or technical agencies.
Monetary information from national accounts including GOS, consumption of fixed capital (depreciation) and value of fixed assets.
Monetary information from environmental-economic accounts or government finance statistics to assess specific taxes and subsidies (for different economic activities).
Step 3: Calculate the unit resource rent
273. The unit resource rent is the resource rent (from Step 1) divided by the physical amount extracted during the same accounting period. It can be considered a price measure. The unit resource rent needs to be calculated for several years, as this is required for the next step. This is done in row 49 of the workbook.
274. In the workbook it is assumed (as a convention) that the resource rent is generated in the middle of the accounting period and therefore reflects the average price level of the accounting period.10
275. In order to apply smoothing of unit resource rents in Step 4, we need to first bring the unit resource rent of the previous years to the same price level as the current accounting period (in this case 2023). This is done in the Year 1 worksheet by applying a price deflator in row 50 to the unit resource rent figure (row 49), obtaining – for each past year – the unit resource rent in mid-2023 prices (row 51). We use a fixed price deflator of 2% in the example, but countries should apply their own price index (which may differ from year to year).
Step 4: Smooth unit resource rents to address price volatility
276. We recommend assuming that the unit resource rent will remain constant in the projection period unless specific policies have been implemented which would allow us to estimate a specific path of future unit resource rents. As discussed in Chapter 3, it is recommended to project future unit resource rents based on an average of actual unit resource rents for several years. Due to volatility in commodity prices for several natural resources, if we were to use only the unit resource rent of the last year, asset values would become highly volatile and hard to cope with in the national accounts. Moreover, current unit resource rent will likely not be a good predictor of future resource rents. The number of years used for smoothing will depend on the type of resource, but typically would range from three to ten years.
277. Under certain circumstances there may be good reasons not to smooth, for instance when futures markets provide a different signal compared with the long-term price trend or if there are expected to be changes in the regulatory regime.
Step 5: Project future resource rents
278. We now multiply the smoothed unit resource rent in mid-2023 prices (cell F55) by the projected physical extraction for the year in question (Year 1 worksheet row 39). This results in projections of future resource rents in mid-2023 prices in row 60.
279. Next, we project discounted future flows of resource rents using a discount factor for each projected year (Year 1 worksheet row 62). The discount factors are calculated from a real discount rate (cell B60). The opening stock is to be calculated (in Step 6) for the start of the accounting period (1 January) and the resource rents are assumed to arise in the middle of the accounting period as these activities occur mid-year on average, so we halve the discount factor in the first period (in this case 2023).
280. As the resource rent in future periods is expressed in constant prices, the discount rate used must be “real” (excluding inflation), as noted in Section 3.3.2. The Workbook: subsoil assets example uses the real discount rate of 2% that is recommended as the common, stable rate by the EGNC. The resulting discounted projections of future resource rents is shown in the Year 1 worksheet row 63.
281. Countries may prefer to use a real discount rate that is higher or lower than the common, stable rate agreed by the EGNC. As noted in Section 3.3.2, countries are free to set their own discount rates as long as they also include a valuation using the common agreed rate as part of sensitivity analysis. This is simple to do as part of Step 5: compilers need only change the figure in cell B61 from 0.02 (2%) to the desired rate.
282. Countries may also prefer to project resource rents including future price increases (e.g. price of a barrel of oil will increase with X percent per year). If so, a nominal discount rate which includes price changes must be used. However, it is easier to assume that the price of the resource remains constant and apply a real discount rate, and this is the method recommended in this compilation guide.
Step 6: Calculate NPV for the opening stocks
283. Now we are able to estimate the opening stock value (in this case of the year 2023) by applying the NPV equation (see Section 3.3.2 Equation 3.1).
284. In the Workbook: subsoil assets (Year 1 worksheet cell F67), we sum the discounted future resource rents to give the opening stock of assets. We obtain an opening asset value as of 1 January 2023 of 204. If a country were to change the discount rate from 2% to 4%, the resulting value would be 186. In this case, the value of 186 would be used by the country in its accounts, and the value of 204 would also be reported (as part of sensitivity analysis).
Source: Workbook: subsoil assets.
Step 7: Calculate NPV for the closing stocks
285. A year goes by, after which we redo compilation steps 1-6 using information now available (Year 2 worksheet in the Workbook: subsoil assets) in order to estimate, in Step 7, the opening stock value of the year 2024 (which gives us the closing stock value of the year 2023).
Note: Cells in green indicate actual data; yellow indicates projections; orange indicates end of asset life.
Source: Workbook: subsoil assets, Year 2 worksheet.
286. Suppose during the accounting period we had some discoveries of 20 physical units, as well as reappraisals of +10 and -2, and a slightly higher extraction level than anticipated (102 units), resulting in a closing stock of 926 units in 2023 (Table 4‑7). The column for 2023 is now coloured green as the figures are all actuals. Using the same assumptions of constant extraction, we now have revised projections as shown in the table above, with extraction continuing into 2033 (albeit with only 8 units).
287. For the year 2023 we now also have measured data on output and user cost of fixed assets (included in column F of the Year 2 worksheet). We again estimate the resource rent, unit resource rent, but now expressed in mid-2024 prices. Again, we do smoothing, and we have an opening stock value for 2024 of 200, which is also the 2023 closing stock value. This figure can be found in cell G71 of the Year 2 worksheet. The opening stock for 2023 (cell F66) is not re-calculated in the Year 2 worksheet, but instead taken from the Year 1 worksheet. This is because of the forward-looking (or ex ante) nature of balance sheets: their main purpose is to describe how the value of assets change over time, for instance due to discoveries, extractions and reclassifications.
288. It would also be possible to revise the discount rate when calculating the 2024 opening stock value: if so, the resulting change should be recorded as a revaluation during the year in question, so that the balance at the end of the year is affected but not the opening balance. The value of this change can be estimated as the difference between the asset value based on the old discount rate and the new rate (all else remaining equal).
Step 8: Put together the monetary asset account
289. We now have the NPV of the subsoil asset at the start of the accounting period (204 as of 1 January 2023) and end of the accounting period (200 as of 1 January 2024, or end of 2023). It is important to realise that these NPV estimates represent the value of the assets in current prices. The compilation of the monetary asset account in constant prices in discussed in Chapter 6.
290. Table 4‑8 shows how this can be used to calculate the price of the resource11 at the beginning and end of 2023, as well as the 2023 average price and average physical stock.
Source: Workbook: subsoil assets, Year 2 worksheet.
291. Depletion in 2023 (Table 4‑9) can then be calculated as the average price from Table 4‑8 multiplied by the physical amount extracted during the year from Table 4‑7 (see also country example from Norway). As part of the sensitivity analysis we can also look at the impact of changing the discount rate on the depletion figure. In this example, if a country were to change the discount rate from 2% to 4%, the resulting value for depletion would be 20.6 instead of 21.5.
292. In the monetary asset account (Table 4‑9), depletion replaces the “extraction” row in the physical asset account. The other rows of the physical asset account retain the same labels and, like depletion, can be calculated by multiplying the average price from Table 4‑8 by the estimates in physical units from Table 4‑7. Revaluation can be calculated by multiplying the average physical stock with the change in the price of the resource in situ. Revaluation will pick up both the effect of changes in the resource rent as well as changes in the extraction path to the extent they lead to changes in asset value. In the Workbook: subsoil assets, a check is included to ensure that the sum of opening stock value + all changes results in a closing stock equal to the opening stock estimate of the next year.
Source: Workbook: subsoil assets, Year 2 worksheet.
Stage 4 (subsoil assets): Integration
293. The information from the monetary asset account for subsoil assets will be used in the sequence of economic accounts (standard SNA presentation) (see Table 4‑10).
294. In the 2025 SNA, depletion of subsoil assets is recorded as a cost of production, similar to depreciation (instead of other changes in the volume of assets and liabilities as it was in the 2008 SNA). This will be recorded in the production account, in the earned income, the transfer of income account, and the capital account.
295. Discoveries and reclassifications (upward or downward) and catastrophic losses are recorded as other changes in volume. Reappraisals (upward and downward) are also treated as other changes in volume. Revaluation is to be recorded in the revaluation account (see also Section 4.2.3 Stranded assets). Finally, opening and closing stocks are part of the national accounts balance sheets (opening and closing balance sheet).
Note: * if split asset: part of the cost of depletion allocated to the legal owner.
4.2.3. Specific issues
Mineral exploration and evaluation
296. Before the extraction of subsoil assets can take place, companies first engage in mineral12 exploration and evaluation to assess possible prospects. These expenditures can be significant and are treated as follows: 2025 SNA (§7.258): “Expenditures on mineral exploration and evaluation are not treated as intermediate consumption. Whether successful or not, they are needed to acquire new reserves and so are all classified as gross fixed capital formation.” Mineral exploration and evaluation (AN1152) are classified under intellectual property products (AN115), i.e. as a component of fixed assets AN11.
297. Consistently, the SEEA CF (§5.201 – 5.205) says that mineral exploration is to be treated as an intellectual property product (IPP). Thus, the user costs implicitly include both depreciation (consumption of fixed capital) and a return to these fixed assets (IPPs) when deriving the resource rent. Compilers may find it useful to include them as of which items when calculating resource rent for subsoil assets in Step 1, Table 4‑5. Such a presentation is shown in Table 4‑11.
Note: Value of mineral exploration and evaluation is input data (green cells); depreciation and return to mineral exploration and evaluation (blue cells) are estimated in the same way as for depreciation and return to fixed capital generally (see Step 1), although the specific assumptions used may be different.
298. The 2025 SNA describes mineral exploration and evaluation in §11.108 - 11.111, while §14.40 explains that mineral exploration and evaluation should be valued either on the basis of the amounts paid under contracts awarded to other institutional units for this purpose or on the basis of the costs incurred for exploration undertaken on own account.13 These costs should include a return to the fixed capital used in the exploration activity. That part of exploration undertaken in the past that has not yet been fully written off should be revalued at the prices and costs of the current period.
299. UNECE (2024, pp. 12-13[8]) recommends applying the same service life for mineral and energy exploration as that used for the associated subsoil assets, with 30 years as recommended average service life, while 2025 SNA 11.110 notes that “depreciation may be calculated for such assets by using average service lives similar to those used by mining or oil corporations in their own accounts”.
Decommissioning costs
300. 2025 SNA (§1.58) explains the importance of considering costs associated with the decommissioning of assets such as oil rigs and nuclear power at the end of their productive lives. The SEEA CF contains an elaborate discussion of the treatment of decommissioning costs (§4.194 – 4.209; §5.206) distinguishing between terminal costs and remedial costs. “Terminal costs are costs that can and should be anticipated during the production periods prior to closure; provision should be made for meeting them during the life of the fixed asset. Remedial costs are incurred when production has already ceased, with no provision having been made for the taking of remedial action while production was in progress. Examples are the rehabilitation of sites contaminated by past activities, for example, fuel storage sites, and former landfill and abandoned mining sites.” (SEEA CF, §4.195).
301. According to the 2008 SNA terminal costs should be included in the estimates of depreciation (leading to a negative asset). The European System of Accounts (ESA) 2010 has a different treatment of terminal costs: they are booked when they occur and directly depreciated (as a result of which there can never be a negative asset in the balance sheet) (European Commission, 2014, p. 33[9]).
302. The treatment of terminal costs changes in the 2025 SNA towards including the expected terminal costs in the value of fixed assets upfront (this is consistent with IAS 37/IPSAS 19 on recording of provisions). With the new SNA (as detailed in Chapter 17, D.6 – Terminal costs), these estimates will be part of the fixed assets estimate and hence be automatically included when estimating the user costs of produced assets using the same assumptions for service life and rate of return. CFC of terminal costs and return to terminal costs may be shown as of which items under User costs of produced assets when calculating resource rent for subsoil assets in Step 1, Table 4‑5 (similar to the treatment for mineral exploration and evaluation shown in Table 4‑11).
303. Contrary to 2025 SNA which does not explicitly deal with remedial costs, the SEEA CF has an extensive discussion on this issue, with the following recommendations:
Costs of a remedial nature are often incurred after a site has been closed and the operator has left. There are two main types of remedial costs: (a) expenditures to restore land to allow its use for some other purpose; and (b) expenditures to ensure that no harmful emissions from deposits of pollutants and other residuals from past activity are able to leach into the surrounding environment and cause environmental damage. In both cases, the relevant expenditures should be treated as gross fixed capital formation and give rise to a fixed asset: land improvement.
For remedial costs, there is no special consideration required regarding the timing of reporting nor are there questions regarding whether the costs are anticipated, since, by definition, these costs are incurred after the operations at the site have ceased and are not incurred by the operator of the site who caused the need for the remediation.
In cases where environmental protection expenditures are incurred on an ongoing basis so that environmental damage is either inhibited or reduced on a continuing basis, then these expenditures should be treated as intermediate consumption or gross fixed capital formation of the owner at the time they are incurred and not recorded as either terminal or remedial costs. (SEEA CF, §4.207-9)
Stranded assets
304. Stranded assets are mineral and non-renewable energy resources that were at one point considered assets but are now no longer likely to be extracted due to for instance changes in regulations and/or changing energy markets.
305. Stranded assets should be recorded as (downward) reappraisals in the physical asset account (see Year 2 worksheet in the Workbook: subsoil assets, Step 7). This will impact the life length of the asset and hence the asset value. The 2025 SNA (§13.28) (see also the discussion in Guidance Note WS.9 Recording of Provisions, SNA Update 2023c) recommends recording such downward reappraisals as other changes in volume (Step 8). It is helpful to clarify that there can be different types of situations that require a different recording in the national accounts:
In case of a change in relative prices, this could have both a volume effect (lower reserves considered as economically viable to exploit) and a price effect (the reserves extracted generate a lower price). It is recommended to split these two impacts when compiling the monetary asset account.
In case of a change in the extraction path: if the total amount of reserves that is extracted remains the same, this is not a volume effect but (due to the effect of discounting) a value effect, which should be recorded as a revaluation.
In case there is more (or less) of the resource accessible (e.g. due to legal changes), this should be recorded as an “other change in the volume of assets”.
Statistics Canada's Natural Resource Asset Accounts (NRAA) provide a broader dimension to our national wealth by reporting the stocks of natural resources such as energy resources (e.g. oil and gas), mineral resources (e.g. zinc, potash), and timber. These accounts form the basis of the estimates of Canada's quarterly natural resource wealth that are integrated into the National Balance Sheet. Integrating natural resource assets with other income-generating assets, such as buildings, machinery and equipment, and intellectual property products, is essential for tracking the national wealth of a resource-based economy like Canada’s.
The calculation of subsoil asset wealth (Figure 4-4) involves myriad data sources to calculate the net present value of thirteen different commodities, including crude bitumen, iron, potash and gold. Data relating to the costs and value of production are obtained from annual surveys of the resource-producing industries. The total value, or wealth, associated with the stock is calculated as the present value of all future annual resource rent that the stock is expected to yield. In each year, it is assumed that the physical depletion of energy and mineral stocks will continue at the current rate until the stock is fully depleted. Statistics Canada produces quarterly estimates by projecting forward annual data using a selection of quarterly indicators. These projections are calculated using price and production data for the respective commodities along with wages and salary estimates and establishment based operating cost survey data.
In the compilation of these estimates, one challenge that Canada has encountered is that of negative resource rent, which occurs in each period the resource extraction is not economically viable and, as a result, the resource has no value (or carries a cost). This implies that the surplus generated does not exceed the return to capital and depreciation of produced capital. There is some evidence that stranded assets may exist in the capital stock of certain resource industries in Canada, which would inflate the return to capital and depreciation estimates used in the net present value calculation.
In Canada, the current treatment for negative resource rent is to, instead of attributing negative wealth to this commodity, set the commodity’s wealth to zero. Because Canada’s estimates are calculated on a commodity-by-commodity basis, negative resource rent can arise more frequently than if all commodities were aggregated together. Additionally, negative resource rent poses a related issue when it comes to depletion. If depletion is calculated as the quantity of extraction by the average in situ price then the implied price resulting from negative resource rent would either be zero, if this is the chosen treatment, or negative.
Presently, Canada’s calculation of resource rent does not include any smoothing of unit resource rent, while international guidance recommends considering some degree of smoothing. This is something that the team at Statistics Canada plans to reconsider based on emerging guidance, which could serve to mitigate issues arising from negative resource rent. At the moment, Canada compiles quarterly estimates of wealth and resource rent that give a more volatile series that aligns more closely with operating surplus of resource extractors.
Canada was the first country to sector natural resource wealth using a split-asset approach (Figure 4-5). The stocks of natural resource wealth are allocated between the government (federal, provincial) and the extractor (corporate sector), using the royalty receipts of government as the resource rent captured by the government sector, while the remaining resource rent would be captured in the economic sector extracting the asset (mostly private non-financial corporations). This treatment is also impacted by negative resource rent in situations where royalties are received by government despite having no wealth attributed to the commodity.
Norway has historically been a country with abundant natural resources. Among these are biological resources, such as fish and timber, and non-biological renewable resources, such as hydropower. Since the discovery of offshore petroleum resources in 1969 however, the stock of crude oil and natural gas has been the paramount contributor to Norwegian national wealth. The Norwegian Government regulates extraction licenses which are only distributed to legal units registered in Norway. All corporations involved in extraction on the Norwegian Continental Shelf are therefore obliged to pay taxes to the Norwegian government.
The Norwegian Offshore Directorate14 was established in 1972. The main tasks of the Directorate are to manage extraction and exploration licenses and analysing petroleum resource deposits. As such, they publish annual data on extracted petroleum of different categories, as well as the monthly production/extraction of the different petroleum products in physical units. They also estimate and publish figures for expected future extraction from the Norwegian Continental Shelf, as well as estimates of the physical assets based on a national classification. These asset categories however can be easily mapped onto the SEEA CF classes (Liu and Midttun, 2024[10]). Only the category “commercially recoverable resources” would be regarded as assets within the 2025 SNA framework. These resource estimates are updated on an annual basis.
Statistics Norway has historically made several estimates on the value of natural capital, in particular of petroleum resources, however with different approaches. Calculations of resource rents have been used for policy analysis, such as analysing whether the taxation of the industry captures a “reasonable” part of the resource rent. Updated calculations based on the most recent international recommendations have now been carried out, financed with grants from Eurostat. To estimate the resource rent stemming from petroleum deposits on the Norwegian continental shelf, different data sources must be combined. The data on extraction and estimated deposits are available on a disaggregated level, i.e. per oil and gas field. On many fields oil, different petroleum liquids and natural gas are extracted together. Also, multiple license holders can operate simultaneously on a given field, and each license holder can operate on several fields. The main operator on any given field is responsible for reporting extraction per product, such as natural gas and crude oil. These companies report their annual income, investments and compensation of employees on a company-wide basis, not per field or per petroleum product extracted. All these factors make it difficult to estimate resource rent per field, and per petroleum product. Consequently, a top-down approach is the only feasible method, given the data sources currently available in Norway. Even though a bottom-up approach would be favoured, it can be shown that under reasonable assumptions, a top-down approach will yield the same result (Liu and Midttun, 2024[11]).
In addition to taxes applied on all companies, legal units involved in extraction are obliged to pay higher taxes on their net income. This is a way for the government to capture a share of the resource rent derived from extraction. The government also has a direct financial interest in the oil and gas fields. As this data is readily available, applying a split asset approach on the Norwegian petroleum deposits is feasible (Liu, 2023[12]).
The real rate of return to fixed assets favoured by the Norwegian Ministry of Finance is 4% per year. Additionally, a resource rent using a rate of return estimated by the net operating surplus divided by the net stock of produced assets in Norway is used.15 The reasoning behind this last choice of the rate of return is that, in equilibrium, any investor should expect the same asset return from the market-oriented sector in Mainland Norway (excluding government assets) (Liu, 2023[12]).
Even if the extraction, measured in physical assets, is relatively stable over time, there might be substantial fluctuations in monetary values from one year to another, caused by extremely volatile energy prices.
Current prices, NOK million.
Source: Liu and Midtun (2024[13]). Authors’ own calculation based on data from Statistics Norway.
Estimated nominal resource rent does not differ significantly whether one applies a 4% real rate or an estimated rate of return as a basis for the calculation (Figure 4-6). Whether the calculations include specific taxes or not, also has limited influence on the final results.
The asset value of Norwegian petroleum resources is, however, more sensitive to the choice of variables used in the NPV calculations. In Figure 4-7 below (2024[13]), one can see that for example in 2020, the asset value, measured in constant 2021 prices, will vary considerably depending on the choices made on nominal rate of return, annual real rate of return and real discount rate one applies to the calculation.
In constant 2021 prices, NOK billion.
Source: Liu and Midttun (2024[13]). Authors’ own calculation based on data from Statistics Norway.
This rather large variation in estimates is consistent whether one makes the calculation in constant 2021 prices, constant 1970 prices or current prices (Liu and Midttun, 2024[13]).
Statistics Norway has also made some experimental calculations of depletion of oil and gas resources, in line with the recording proposed in the 2025 SNA.
The value of depletion is calculated as recommended in this guide, i.e. using the physical extraction multiplied with the average price of the resource in situ. The calculations are based on Norwegian time series data for the value of the resources from Liu and Midttun (2024[10]), assuming an estimated nominal annual rate of return to produced capital for Mainland Norway, and an annual real discount rate of 4%.
Figure 4-8 below shows the impact, based on the preliminary experimental calculations, of including depletion as a production cost on the Net Domestic Product (NDP) for total economy (S1) and on the net value added for non-financial corporations (S11) over the period 1978-2021.
The impact of depletion on NDP and net value added according to these experimental calculations was most significant during the 1990s and early 2000s, with a reduction in NDP of up to 9%. In recent years, the reduction in NDP is around 4%, while net value added in S11 is reduced by 6-8%.
Even though the Norwegian authorities possess comprehensive data sets regarding extraction of crude oil and natural gas, it should be noted that some of the components in the calculations might be uncertain. Especially we will draw attention to the large fluctuations in crude oil and gas prices, which might have large impact on expectations of future resource rents. Even when using historic resource rents backward, for more than one year (smoothing), instead of the last observed value, as a basis for expectation of future rents, there are no guarantees that this will reduce the uncertainty. The calculations also depend on other assumptions, where different “sets” of assumptions might give different results. Some assumptions, like the discount rate and rate of return, might be harmonised between countries, but the compatibility between countries is still blurred when countries have different sets of data. Some countries have scarcer data than others and might apply the data differently.
In percentage.
Source: Experimental calculations based on data from Statistics Norway and Offshore Directorate.
It should also be noted that so far only annual calculations have been carried out, with a time lag of two years. For quarterly and preliminary annual figures, the available date is more scare. Timely information about the development in the prices and incomes is available, as well as quarterly information about gross fixed capital formation. However, timely information about extraction costs, taxes paid, etc., is not available. If we would compile quarterly (and preliminary annual figures), these figures would contain even more uncertainty than the annual figures.
4.2.4. Modifications to the standard approach
306. The approach described in Section 4.2.2 is to be understood as the Tier 2 (default) method. As countries differ in their data availability and resources for conducting valuation of natural resources, this section describes also a Tier 1 (basic) approach that would typically be followed in case of limited data availability and/or resources, as well as Tier 3 (advanced) methods requiring high data availability and resources.
Basic approach
307. A Tier 1 method would consist of applying a product-based application of the RVM for deriving resource rent in Step 1, instead of the top-down method that is used for the default approach. The product-based method does not use national accounts data or business statistics data for estimating resource rent for activities (e.g. oil and gas extraction). Rather, it estimates all elements required for calculating resource rent (e.g. revenues, costs, user cost of fixed capital, taxes / subsidies) directly for specific products (e.g. oil, copper) from a range of data sources including commodity market prices, industry reports on cost to revenue ratios etc.
308. A good example of this method is the World Bank methodology used for the Changing Wealth of Nations (CWON) (World Bank, 2021[14]) which estimates mineral and energy asset values for all 14 individual resources mentioned in Section 4.2.1, for a large portfolio of countries (> 140), including a time series (1995-2021). CWON uses a range of global data sources for the estimates, including from commercial providers (Table 4‑12). When applying the product-based approach, it is recommended to use national data whenever possible instead of the global data sources.
309. A disadvantage of the Tier 1 method is that the estimates may not fully align with national accounts concepts and therefore differences in results are to be expected. For instance, by using market prices for resources, it is not entirely clear to what extent taxes/subsidies are taken into account.
Advanced methods
310. Several more advanced methods exist compared with the default approach. One advanced method involves applying more advanced modelling of future prices and/or extraction rates. For example, SEEA Energy (United Nations, 2019, p. 108[3]) notes that “if extraction profiles are available from experts, energy agencies, geologic institutes, etc., those profiles should be used.”
311. For projecting future prices, instead of assuming that the smoothed unit resource rent remains constant, an interesting possibility that deserves further research is to use information from futures markets. This could especially be relevant in case of extreme events or major societal issues (e.g. wars) that are likely to cause upheaval in markets in the short to medium term but not necessarily the long term. Futures markets cover multiple commodities (but not all) and there exist multiple exchanges. A possibility could be to use futures market prices as a point on the horizon (to understand what prices will do say X years from now); an alternative could be to use their information to assess how long it will take for prices to return to their long-term average trend. However, before we can turn this into a more specific recommendation, future research would be needed to assess how accurate these predictions tend to be.
312. It is also possible to compile the value of mineral and energy deposits at a disaggregated level, ideally at the individual deposit level. Guidance Note WS.10 recommends this. The main reason is the heterogeneity of extraction costs across space, which would be problematic if deposits with low extraction costs are extracted first, as a result of which current costs would be poor predictors of future costs.
313. This approach could consist in applying a micro-level bottom-up approach (e.g. using deposit level data) allowing to estimate values for a larger number of individual resources. This approach would allow to capture heterogeneity in revenues and costs across deposits; but it is highly resource intensive.
314. The implementation of a deposit level approach may prove difficult for countries in the following circumstances: if they have a very large number of mines (e.g. a country like Canada produces 60 minerals and metals at almost 200 mines and 6 500 sand, gravel and stone quarries); when mining takes place to a large extent in the informal sector (a country like Ghana); or due to a lack of sufficient information at deposit level that would permit a resource rent calculation at that level; see Liu and Midttun (2024[11]). A further challenge arises when multiple minerals are mined in the same operation (e.g. mixed ore mines).
4.3.1. What to include in the national accounts
315. Renewable energy resources are shown separately in the 2025 SNA classification of assets as AN322.
316. As for subsoil assets, their scope is grounded in the United Nations Framework Classification for Resources (UNFC) (UNECE, 2020[7]), a standardised classification system for all mineral and energy resources (both renewable and non-renewable). The 2025 SNA states:
In the case of mineral and energy resources, SEEA Central Framework distinguishes three classes based on the United Nations Framework Classification (UNFC) for Fossil Energy and Mineral Resources: class A: commercially recoverable resources; class B: potentially commercially recoverable resources; and class C: non-commercial and other known deposits. The measurement of monetary estimates is typically restricted to the first class, which in practice could be approximated by those resources for which permissions to exploit have been granted, and/or those for which the existence is explicitly recognised by (past) monetary transactions. Potential mineral and energy resources where it is not foreseen that they will be exploited in the near future are thus explicitly excluded. (2025 SNA, §11.186)
Based on the UNFC, the 2025 SNA glossary defines renewable energy resources as “energy resources which comprise the cumulative quantities of kinetic, radiative and thermal energy recoverable from moving water (hydro and ocean energy), moving air (wind energy), hot underground and surface rock and water (geothermal resources) and incident solar radiation (solar resources).”
317. Renewable energy assets to be included in the national accounts correspond to a subset of renewable energy resources, namely those resources “viable for use in economic production under prevailing technological and economic conditions” (Smith, Ilas and Peszko, 2021[15]). This means that the treatment of renewable energy is consistent with that of non-renewable energy and mineral resources (subsoil assets), where the scope of measurement is restricted to SEEA Class A which also corresponds with the class of viable projects. Viable projects (called commercial projects in Guidance Note WS.11) are defined according to the UNFC as consisting of classes E1, F1, G1-3 (see Section 4.2), however as shown in Box 4-1 there are some differences with respect to the interpretation of the Geological or G-axis.
According to UNFC (2016[16])
It is recognized that the reference to “geological knowledge” [the interpretation applied for subsoil assets] is not generally applicable to Renewable Energy Resources. Therefore, when applied to Renewable Energy Resources, the G axis should be understood to reflect the “level of confidence in the potential recoverability of the quantities”.
Thus, the G axis categories shown in Figure 4-1 are intended to reflect all significant uncertainties impacting the estimated Renewable Energy Resources quantities that are forecast to be produced by the Project and typically would include (but not be limited to) areas such as meteorology, climatology, topography and other branches of geography, ecology and, for geothermal Projects, geology. Uncertainties include both variability in the Renewable Energy Source and the efficiency of the extraction and conversion methodology (where relevant).
The level of confidence for quantities that are classified on the G axis as G1, G2 and G3 is defined as “high”, “moderate” and “low”, respectively. In order to maintain alignment between different Renewable Energy Resources, as well as with non-renewable fossil energy and mineral reserves and resources, specifications for application of the G-axis categories to Renewable Energy Resources are:
Where the “probabilistic” approach is used, the cumulative probability levels associated with G1, G1+G2 and G1+G2+G3 shall be 90%, 50% and 10% respectively, where each probability level reflects the probability of exceeding the estimated Renewable Energy Resource quantities for that level.
Where the “scenario” approach is used, the low, best and high estimates shall reflect the same principles, and approximately the same probabilities, as would be associated with estimates derived from a probability analysis as described above for the “probabilistic” approach.
318. UNFC makes a helpful distinction between products and sources:
Products of the project may be bought, sold or used, including electricity, heat, hydrocarbons, hydrogen, minerals, and water. It is noted that with some projects, such as for renewables, the products (electricity, heat etc.) are different from the sources (wind, solar irradiation etc.). In other projects the products and sources may be similar e.g. in petroleum projects both the sources and products are oil and/or gas, although the fluid state and properties may change from reservoir to surface conditions. (UNFC, 2016[16])
In other words, we can draw a clear conceptual distinction between the source (such as the oil in the ground; or wind), and the product (the oil produced; electricity generated). It is the source that we seek to capture when measuring natural resource assets (in this case, renewable energy resources).
319. Renewable energy resources consist of the cumulative quantity of renewable energy resources harvestable by viable renewable energy projects. This means for instance that the energy of a river without hydroelectric generation (existing or planned at a point in time) does not constitute an asset, and that only the solar radiation capturable by (existing or planned) solar panels is considered an economic asset. Physical stocks of renewable energy resources may increase over time due to additional equipment being installed or due to better technology – these would be recorded as (upward) reappraisals in a physical asset account.
320. The scope for renewable energy resources includes resources that generate electricity and/or heat or cool air. As we have seen in Section 2.5.2, the 2025 SNA asset classification of renewable energy resources consist of wind, solar, water, geothermal and other renewable energy resources. Additionally, the following recommendations regarding what resources to include in the national accounts are made:
Biomass/solid biofuel is in many countries an important source of energy (either in the form of heat or electricity production or combined heat and electricity production). However, the value of biomass should in principle be captured under timber resources if it pertains to fuelwood. Other biomass used for energy production such as crop residue or felling residues can be considered as being produced within the economy (United Nations, 2019[3]) and would therefore be considered as products not sources (following the earlier UNFC categorisation). It is therefore recommended not to account for biomass under renewable energy assets.
Other biofuels (e.g. renewable municipal waste, biogas, liquid biofuels) can also be considered as produced within the economy (United Nations, 2019[3]) constituting products (or sometimes residuals in case of waste (water) or sewage), not as sources and hence would be out of scope as renewable energy assets.
Geothermal energy resources. A distinction is commonly made between deep subsoil sources (e.g. in the Dutch context > 500 m) and shallow sources (e.g. used for heat pumps) – although there does not appear to be an internationally standardised depth/classification. Heat-pumps can be used both for heating and cooling with heat (or cool air) as main product. Both deep and shallow sources are in scope, as is hot surface water (e.g. from geysers).
Aerothermal energy is captured by heat-pumps but using outside air rather than geothermal energy. These devices seem to be predominantly used by households. Aerothermal energy is also in scope of measurement.
Solar thermal may be captured either by a small installation that warms water directly or large installations like concentrated solar power plants. Solar thermal is also in scope.
Water energy resources, more generally known as hydropower, can be understood as the use of moving water (either natural or artificial, after storage behind a dam) to generate electricity. Wave and tidal energy are in scope but probably insignificant in most countries.
4.3.2. Compilation stages
321. For renewable energy resources, we will distinguish four compilation stages: identifying the types of assets to be included; collecting the physical data; building the monetary asset accounts; and integration of the results into the accounts. The four compilation stages follow the same approach as for non-renewable energy and minerals; but they are simpler than for non-renewables and there is (by definition) no cost of depletion to be recorded in the accounts at the end of the compilation process.16
Stage 1 (renewable energy): Identifying types of assets
322. In the first stage, the types of renewable energy assets for which accounts are to be compiled need to be identified. The 2025 SNA asset classification distinguishes between wind energy resources (AN322S1), solar energy resources (AN322S2), water energy (hydro) resources (AN322S3), geothermal energy resources (AN322S4) and other renewable resources (AN322S9). Countries are encouraged to provide breakdowns at least at this level of detail, because they are of considerable policy interest. A more detailed break-down could be disseminated as part of the environmental-economic accounts (either in physical or monetary units) but this goes beyond the 2025 SNA requirements.17
323. The SNA is exhaustive, implying that all assets with economic value (the in-scope Class A resources discussed above) should be estimated. Due to the resource intensive nature of the valuation of renewable energy assets, it is proposed to apply a materiality threshold and focus on the valuation of those renewable energy resources that contribute more than 5% of the national energy supply. The easiest way to assess this is by looking at the generation of energy by source (in physical units) based on energy statistics. However, data and resources permitting, countries are encouraged to compile accounts for any renewable resources that are likely to grow and become economically important in future. When a reasonable estimate can be made of how much of the asset value is missed, it is recommended to gross up the asset value.
324. The materiality threshold should be applied at a slightly more detailed level of resources than the five categories mentioned above. In the absence of an internationally agreed detailed classification for renewable energy resources suitable for statistical purposes,18 it is recommended to use the following list as a checklist to assess which renewable energy resources are available in the country and would be above the threshold in terms of energy generation:
Wind (offshore)
Wind (onshore)
Solar photovoltaic
Solar thermal
Hydro
Geothermal (deep)
Geothermal (shallow)
Aerothermal
Wave and tidal
This list has been drawn up based upon existing country practices, taking into account the exclusion of biomass as discussed in the previous section.
Stage 2 (renewable energy): Collecting the physical data
325. The next stage is to collect the physical data for the renewable energy resources that have been identified and are above the materiality threshold. In the case of renewable energy, it is necessary to obtain information on renewable energy generation in physical units. No information is required about installed capacity. This information may already be available as part of energy statistics, in some cases it would need to be collected and estimated in order to express various energy resources in the same units.19
326. The use of renewable energy resources is growing in most if not all countries. This raises the question how one should make projections of future production in physical terms. Some projects (such as the construction of a hydropower dam) may take several years. If you know when the dam would start producing (say five years from now with an expected lifetime of 50 years), it would be possible to take this into account when making future projections of electricity generation. While this may be feasible for a technology such as hydropower with a limited number of locations, for other resources such as wind or solar PV this may pose significant information demands.
327. While in theory plants or installations that are not yet producing energy (to the extent they are considered viable projects according to the UNFC) should be included, in the standard approach they are therefore excluded for pragmatic reasons.
Stage 3 (renewable energy): Building the monetary asset accounts
328. Once the physical data has been collected for each renewable energy resource, the next stage is to compile the monetary asset accounts. These provide information on stocks and changes in stocks that are needed to populate the relevant accounts in the 2025 SNA, such as various entries in the capital accounts and balance sheets as explained in Stage 4.
329. When valuing renewable resource assets, as discussed in Chapter 3, the preferred valuation method would be to use observable market prices for transactions. Under some circumstances, the valuation of renewable energy will already be captured in associated land values; for instance, in the case of onshore wind energy or in the value of dwellings equipped with solar panels. This issue is recognised in the 2025 SNA (§11.197): “The value of land may be higher due to the availability of subsoil resources, or the possibility to exploit renewable energy by having permission to put, for example, wind turbines or fields of solar panels on the land.” In such cases a hedonic pricing method could be followed, which may provide a direct estimate of the renewable energy asset value (e.g. a house with solar panels fetches a higher market price). However, care would need to be taken to obtain separate values for the land or dwelling, the equipment (e.g. the solar panel) and the renewable energy asset value itself to avoid double counting. Furthermore, the use of hedonic pricing methods is highly resource intensive, requiring extensive micro-data. For these reasons, the use of hedonic pricing is not recommended as the standard approach, but as an advanced method (see Section 4.3.4). As explained in SNA update Guidance Note WS.11, in many instances no associated land exists (e.g. in case of offshore wind) or installations are placed on land that is not recognised as economic asset in the SNA (e.g. desert areas/wastelands), which would rule out the possibility of doing hedonic pricing.
330. In some instances, permits to access the natural resource exist but as discussed in Chapter 3, valuation based on permits may only be feasible in certain circumstances. Therefore, in most cases, the recommended valuation method is to calculate asset values as the Net Present Value (NPV) of future resource rents, which are projected from actual (past and present periods) resource rents calculated with the Residual Value Method (RVM). This has the additional advantage also of consistency in treatment with the valuation of subsoil assets.
331. However, Guidance Note WS.11 indicated concerns about using the NPV method in cases where markets are immature or heavily distorted: “Though we recommend the RVM in most instances, we acknowledge that the least-cost alternative (LCA) method is worthy of consideration in cases where subsidies remain significant and markets are likely still far from long-term equilibrium (mainly for solar and wind energy assets).” The EGNC has investigated how the LCA method could be operationalised for different technologies based on Levelised Cost of Electricity Estimates (LCOEs), see Annex A. The investigation concluded that the LCOE is not suitable for use in compiling the national accounts.
332. Stage 3: Building the monetary asset accounts therefore focuses on demonstrating how to calculate asset values for renewable energy resources as the NPV of future resource rents, which are projected from actual (past and present periods) resource rents calculated with the RVM (see also country examples from Costa Rica and Indonesia). While most renewable electricity and heat is generated by companies, we will also include a discussion of the production of electricity and heat by households.
333. As in the case of subsoil assets, eight steps are required to compile the monetary asset accounts for renewable energy resources, as shown in the example for the year 2023 in the Workbook: Renewable energy, which is discussed below. The eight steps are:
1. Calculate resource rents (past and present).
2. Project the physical asset account and physical output until end of the asset life of the resource.
3. Calculate the unit resource rent.
4. Smooth unit resource rents to address price volatility.
5. Project future resource rents.
6. Calculate NPV for the opening stocks.
7. Calculate NPV for the closing stocks.
8. Put together the monetary asset account.
Step 1: Calculate resource rents (past and present)
334. The calculation of resource rents is shown in the Workbook: Renewable energy in rows 6-25 of the Year 1 and Year 2 worksheets. Table 4‑13 presents an example.
Table 4‑13 is similar to Table 4‑5 used for subsoil assets. However, it also includes gross mixed income (GMI) of the household sector to allow for the inclusion of the profits associated with renewable energy production by households (as unincorporated enterprises).
Note: Cells in green indicate input data; blue indicates calculated estimates. Specific taxes on products / specific other taxes on production should be recorded as rent payment (D45) when government is the legal owner.
Source: Workbook: renewable energy assets, Year 2 worksheet.
335. For calculating resource rents associated with renewable energy production, it is recommended to use a bottom-up approach rather than the top-down approach recommended for sub-soil assets.20 There are a number of reasons for this recommendation. First, as discussed in Section 3.3.1 on the Residual value method for calculating resource rent, the national accounts (and most economic statistics) currently do not distinguish between renewable and non-renewable energy production as separate economic activities. This is expected to change with ISIC Revision 5, which will have a breakdown of electricity generation into two new classes (3511 and 3512 respectively) for electric power generation activities from non-renewable sources and renewable sources. But even with the implementation of ISIC Revision 5, compilers would still need a further disaggregation between different types of renewables (solar, wind etc.).
336. The overall intent in applying the bottom-up method lies in proxying national accounts concepts as best as possible. For instance, business survey data may provide figures for depreciation, but these may be based on historic costs and may need to be adjusted to reflect exchange value/written-down replacement cost required for the national accounts.
337. The bottom-up approach uses a range of data sources, where we will distinguish between companies and households. Large-scale renewable energy production is undertaken by companies, but households are important for solar energy (both PV and thermal), aerothermal, and geothermal energy.
338. For companies, the recommended approach consists of collecting micro-data of businesses engaged in different types of renewable energy production such as hydro, wind, solar (e.g. business surveys). Depending on the national market, it may be sufficient to focus on the biggest companies (e.g. the main wind farms) and impute unit resource rents for the smaller companies in Step 3 below based on their respective shares in electricity generation, assuming similar unit resource rents. A challenge may arise in case companies are active in different kinds of renewable energy (or even fossil energy). In such cases, it is recommended to apply a product-based approach (see Section 4.3.4, Basic approach), where for instance output is based on physical output multiplied with relevant energy prices. Energy and electricity markets are however quite complex and care should be taken to apply a valuation consistent with national accounts principles, see Box 4-5.
339. In order to apply the RVM, the starting point for corporations in Step 1 is gross operating surplus (GOS). Information about taxes and subsidies and depreciation is also relevant. Business surveys typically do not contain information about the value of stocks of fixed assets (e.g. equipment), which is needed for the estimation of user costs of produced assets. This would need to be obtained from other sources (such as business accounts) or estimated based upon information about gross investment in tangible goods (which is similar to the concept of gross fixed capital formation in national accounts).21
340. As explained in Section 3.3.1 the user costs consist of two elements: depreciation and net return to assets, specifically fixed capital. The value of depreciation (consumption of fixed capital in the 2008 SNA) is usually derived from a perpetual inventory model (PIM) and available as part of the national accounts but probably not in the required disaggregation that allows to distinguish between renewable and non-renewable energy generation. The asset life of the fixed assets could differ from the asset life of the natural resource, but in principle it should not be longer than the life of the natural resource.
Electricity markets are complex. Electricity is sold across two main markets: the wholesale market and the retail market. In the wholesale market, electricity generators sell electricity in bulk to large buyers, such as utilities, at prices influenced by supply conditions and demand forecasts. The wholesale market itself consists of forward markets (up to years ahead of physical delivery), the spot market (up to minutes), and the balancing market (real time to avoid black-outs).22
The spot price – the current price for electricity traded for immediate delivery – reflects the value of electricity at a given moment. Fluctuations in the spot price are influenced by real-time supply and demand, particularly for renewable sources, which are often weather-dependent. For instance, on a sunny or windy day, in a country where electricity mainly is used for lighting and heating, renewable electricity generation may surge, leading to lower spot prices due to excess supply. In such situations even negative prices for renewable electricity may occur. The marginal cost of the electricity producers usually drives price formation.
In the retail market, energy suppliers then sell electricity to end consumers (households and businesses), where the final price includes additional costs like distribution and retailing, as well as taxes and levies.
The design of electricity markets differs a lot between countries, ranging from highly centralised to decentralised markets, and from closed to open markets.
Government policies, such as feed-in tariffs (FiTs), also play a key role in price formation. FiTs ensure that renewable energy producers receive a guaranteed price for the electricity they generate, smoothing out the volatility of spot prices. Similarly, power purchase agreements (PPAs), which are long-term contracts between producers and buyers, set fixed prices for electricity over time. As a result, a significant part of renewable energy can be sold outside the competitive electricity markets. This poses a major challenge for estimating the value of output for purposes of the national accounts.
The national accounts measure output which for an electricity producer consist of the transactions in electricity produced. This value may be quite different from the average wholesale electricity spot price times physical generation, in case PPAs (or FiTs) are in place. The valuation of output for wholesale or retail traders generally consists of the wholesale and retail margins.
341. The net return to fixed capital is estimated by multiplying the value of fixed assets (row 22) with the rate of return, which is specified in cell H24. The rate of return is assumed to be 6% real in the example, but countries should apply their own rate of return, based on the recommendations provided in Section 3.3.1 by applying the “everything but” the activity method to be consistent with the estimation of subsoil assets.
342. After deducting both elements – depreciation and the return to fixed capital – from GOS and GMI for the derivation of resource rent, we obtain the resource rent.
343. Plants that are not yet producing renewable energy (e.g. a hydropower dam being constructed) are recorded under work-in-progress (AN122). As discussed in Section 3.3.1 the user cost of capital is restricted to fixed assets and therefore excludes a return to the work-in-progress.
344. For households, the 2025 SNA is clear (§7.27) that electricity and heat production by households should be recorded. Eurostat (2024[17]) provides more detailed guidance on the treatment. First, the output is recorded as production of an unincorporated enterprise owned by the household sector, either in ISIC 3512 Production of electricity from renewable sources or ISIC 6820 Rental and operating of own or leased real estate, with the second option “best justified [when] electricity production is small compared to total electricity production and small compared to the production of owner-occupied dwelling services.” (2024[17]). Electricity produced by so-called balcony solar panels, i.e. smaller panels usually not connected to the grid, may be excluded for practical reasons of data availability. Furthermore, both the production of electricity provided to other units and for own final use is included, the latter only when “significant, i.e. when it is quantitatively important in relation to the total supply of that good in a country” (2024[17]).
345. When it comes to valuation of electricity the following recommendations apply (Eurostat, 2024[17]). “When households sell own-account production of electricity to the grid, this amount of electricity should be valued with the price the household receives. The price to be used should be a weighted average of the different Feed-in-Tariffs (FiT) for households” (Eurostat, 2024[17]). The FiTs are usually contractual arrangements between the household and the energy company.23 Secondly, “the own final consumption of electricity produced by households themselves, should be valued at the basic price which incorporated producers receive for electricity intended for sale to households. This price should exclude any charges for transmission or distribution of electricity. As it has to be valued at basic prices, taxes less subsidies on products applicable to incorporated electricity producers need to be deducted” (Eurostat, 2024[17]). In case charges apply to households for delivering energy back to the grid, these charges need to be treated as intermediate consumption. Finally, “renewable energy installations, which fulfil the criteria of being fixed assets, should be classified as GFCF in electricity production, or as GFCF in dwellings” (Eurostat, 2024[17]).
346. While Eurostat (2024[17]) does not discuss the generation of heat, similar treatment would apply. It should be noted (as discussed in Stage 1) that biomass is not considered as a renewable energy resource, hence use of wood or charcoal for heating or cooking is excluded from the valuation of renewable energy resources.
347. In order to derive resource rent for unincorporated enterprises in the household sector, the following recommendations apply:
GMI will be broadly equivalent to profit (GOS for corporations) because remuneration of employees can be neglected as no labour is usually involved (after the installation of the solar panels or heat-pumps) in the production of electricity.
According to Eurostat (2024[17]) “Taxes less subsidies on production, if applicable at all for households, also need to be taken into account.” In case households are subsidised for the production of electricity (by government)24 this is to be recorded as D3 and treated as (specific) subsidies on products and hence deducted in the derivation of the resource rent.
The user costs of produced assets need to be estimated based on estimates of the value of the installations. Countries are recommended to include solar panels and heat pumps as two additional and separate assets in their perpetual inventory models (PIMs). The assumed service life should not necessarily be the same as the life of the resource. Unless more specific information is available, it is recommended to apply 25 years as default for solar panels with an assumed loss of 0.5% annually in efficiency.25 For other equipment used by households a default asset life of 15 years is assumed with an assumed loss of 1% annually in efficiency.26 The loss in efficiency can be used to estimate suitable depreciation profiles for these assets.
It is recommended to use an “everything but” the activity approach to estimate the return to fixed capital to be consistent with the estimation of resource rent for companies.
348. As many countries will be compiling estimates for household production of renewable energy the first time, it may be helpful to note that the estimation of output (and value added) of household production can be undertaken based on a variety of data sources: information collected from network operators, from energy companies (e.g. information obtained through smart meters) or through household surveys, (see also Box 4-6). The output is recorded in the production account; the income generated is recorded as GMI in the earned income account.
349. In line with the recommendations of Guidance Note WS.11, Variable Renewable Energy (VRE) grid integration costs such as cost made to connect wind turbines and solar panels to the network are likely to be modest and therefore may be excluded when estimating renewable energy resource rents.
Step 2: Project the physical asset account and physical output until end of the asset life of the resource
350. In the Workbook: Renewable energy (Year 1 worksheet, rows 34-42), we have included a physical asset account table because it is helpful for compilers when going through the steps recommended by this guide. Below the table there is a line for generation of energy (physical output), in row 45 of the worksheet.
351. It should be noted that there are some key differences between this physical asset account and the one for subsoil assets:
There is no “extraction” or "discoveries” lines, as these are not applicable for renewable energy.
The opening and closing stock of physical renewable energy assets is calculated as the current generation of energy (in megawatt (MW)) times the asset life and is therefore expressed in energy units (MW).
352. The choice of the asset life (of the natural resource) should not be based on the expected lifetime of the equipment used in capturing energy, as we are valuing the resource itself. Similar to the exploitation of other natural resources, it is reasonable to assume that equipment will be replaced when there are still economic benefits that can be derived from the natural resource.
353. It is not recommended to use an infinite life either due to uncertainties in technology and climate considerations. For example, water levels in Lake Mead in the United States have dropped significantly in recent years, reducing electricity generation by the Hoover Dam.27 The dam was completed in 1936, less than 100 years ago, and uncertainties as a result of exacerbating climate change will only increase.28
354. It is recommended to use the average (or median) licensed lifetime of projects as asset life of the resource, and if such data is not available to apply a default asset life of the resource of 30 years for wind and solar and 50 years for hydro.
355. In Table 4-14, we assume that in 2023 we have energy generation of 83 units per year, which is based on the last year and is assumed to remain constant (see Chapter 3). Alternatively, it is possible to use a specific projection for future energy generation when available. In order to use such a projection, in light of the UNFC (as discussed in Section 4.1), it is important that clear evidence is available of future renewable energy projects, hence the existence of policy goals for increasing renewable energy production alone is not sufficient.
356. When making projections, we only project energy generation, as by definition we do not have information about reappraisals or reclassifications. We also make an assumption (in cell B31 of the Year 1 worksheet) that the asset life of the resource is 50 years (e.g. for a hydro project), giving an opening stock of 4150 physical units (MW). This remains constant if the amount of energy generated remains constant.
Note: Cells in green indicate actual data; blue indicates modelled estimates; yellow indicates projections. Projections continue up to 2072.
Source: Workbook: Renewable energy, Year 1 worksheet.
For Steps 1-2, the following data are required:
Physical information about the amount of renewable energy generation for different technologies.
Data on renewable energy production is usually not collected/compiled by NSIs but rather by the Ministry of Energy or an Energy Agency.
For companies, micro-data of businesses engaged in different types of renewable energy production such as hydro, wind, solar (e.g. business surveys).
For households, information about FiTs, number (and value of) solar panels and heat pumps for instance through household surveys.
Data may also be collected from network operators, from energy companies (e.g. information obtained through smart meters). Information about the energy market (e.g. average wholesale electricity prices) may also be useful either to estimate output (by multiplying with physical data about generation), or to corroborate information from other sources.
Average (or median) licensed lifetime of wind and/or solar projects.
Step 3: Calculate the unit resource rent
357. The unit resource rent is the resource rent (from Step 1) divided by the physical amount of energy generated during the same accounting period. It can be considered a price measure. The unit resource rent needs to be calculated for several years, as this is required for the next step. This is done in row 49 of the workbook.
358. In the workbook it is assumed (as a convention) that the resource rent is generated in the middle of the accounting period and therefore reflects the average price level of the accounting period.29
359. In order to apply smoothing of unit resource rents in Step 4, we need to first bring the unit resource rent of the previous years to the same price level as the current accounting period (in this case 2023). This is done in the Year 1 worksheet by applying a price deflator in row 50 to the unit resource rent figure (row 49), obtaining – for each past year – the unit resource rent in mid-2023 prices (row 51). We use a fixed price deflator of 2% but countries should apply their own price index (which may differ from year to year).
Step 4: Smooth unit resource rents to address price volatility
360. It is assumed that the unit resource rent will remain constant unless specific policies have been implemented which would allow us to estimate a specific path of future unit resource rents. An example of such a specific policy would be if there is solid evidence that subsidies for renewable energy will change over the period of the projection (see Box 4-7).
A key input in projecting future unit resource rent for renewable resources is the level of taxes/subsidies. Guidance Note WS.11 proposed to assume a linear decline of subsidies from their current level to zero over a reasonable time horizon, and a linear increase of taxes from zero to 25% over the same time horizon.30 However, further research in support of these guidelines has indicated that there is large uncertainty regarding the future development of subsidies/taxes on renewables. In addition to FiTs, a wide variety of policy instruments has been developed which makes it difficult to predict the aggregate future development. Also, the arguably optimistic outlook that existed when the Guidance Note was written about the competitiveness of renewable energy seems to have changed somewhat in recent years with consulted experts questioning whether any taxes (rents) will be levied on renewables in the foreseeable future. Another issue that has become increasingly clear is that with a larger percentage of electricity being provided by renewables, the competitiveness of renewable energy not only depends on cost but also on a range of other factors such as the ability to address peak demands, flexibility in supply and their computability with the electricity system.31
If there is no evidence to support a different scenario, the standard approach recommends assuming that the (smoothed) unit resource rent stays constant i.e. it assumes that the current (net) subsidy levels remain constant when projecting resource rent into the future.
361. As discussed in Chapter 3, it is recommended to project future unit resource rents based on an average of actual unit resource rents for several years. Due to volatility in commodity prices for several natural resources, if we were to use only the unit resource rent of the last year, asset values would become highly volatile, which is hard to cope with in the national accounts. Moreover, current unit resource rent will likely not be a good predictor of future resource rents. The number of years used for smoothing will depend on the type of resource, but typically would range from three to ten years.
362. Under certain circumstances there may be good reasons not to smooth, for instance when futures markets provide a different signal compared with the long-term price trend or if there are expected to be changes in the regulatory regime.
Step 5: Project future resource rents
363. We now multiply the smoothed unit resource rent in mid-2023 prices (cell F55) by the projected electricity generation for the year in question (Year 1 worksheet row 45). This results in projections of future resource rents in mid-2023 (constant) prices in row 59.
364. Next, we project discounted future flows of resource rents using a discount factor for each projected year (Year 1 worksheet row 61). The discount factors are calculated from a real discount rate (cell B60). The opening stock is to be calculated (in Step 6) for the start of the accounting period (1 January) and the resource rents are assumed to arise in the middle of the accounting period as these activities occur mid-year on average, so we halve the discount factor in the first period (in this case 2023).
365. As the resource rent in future periods is expressed in constant prices, the discount rate used must be “real” (excluding inflation), as noted in Section 3.3.2 on Discounting future flows of resource rent. The Workbook: Renewable energy example uses the real discount rate of 2% that is recommended as the common, stable rate by the EGNC (see Chapter 3). The resulting discounted projections of future resource rents is shown in the Year 1 worksheet row 62.
366. Countries may prefer to use a real discount rate that is higher or lower than the common, stable rate agreed by the EGNC. As noted in Section 3.3.2, countries are free to set their own discount rates as long as they also include a valuation using the common agreed rate as part of sensitivity analysis. This is simple to do as part of Step 5: compilers need only change the figure in cell B62 from 0.02 (2%) to the desired rate.
367. Countries may also prefer to project resource rents including future price increases. If so, a nominal discount rate which includes price changes must be used. However, it is easier to assume that the price of the resource remains constant and apply a real discount rate, and this is the method recommended in this compilation guide.
Step 6: Calculate NPV for the opening stocks
368. Now we are able to estimate the opening stock value (in this case of the year 2023) by applying the NPV equation (see Section 3.3.2 on Net Present Value method).
369. In the Workbook: Renewable energy (Year 1 worksheet, cell F66), we sum the discounted future resource rents to give the opening stock of assets. We obtain an opening asset value as of 1 January 2023 of 1205. If a country were to change the discount rate from 2% to 4%, the resulting value would be 831. In this case, the value of 831 would be used by the country in its accounts, and the value of 1205 would also be reported (as part of sensitivity analysis).
Source: Workbook: Renewable energy workbook.
Step 7: Calculate NPV for the closing stocks
370. A year goes by, after which we redo compilation steps 1-6 using information now available (Year 2 worksheet in the Workbook: Renewable energy) in order to estimate, in Step 7, the opening stock value of the year 2024 (which gives us the closing stock value of the year 2023).
Note: Cells in green indicate actual data; blue indicates modelled estimates; yellow indicates projections. Projections continue up to 2073.
Source: Workbook: Renewable energy workbook, Year 2 worksheet.
371. Suppose during the accounting period we had net reappraisals of 29 physical units (for instance due to increases in efficiency of the solar panels), resulting in a closing stock of 4179 units in 2023 (Table 4‑15). The column for 2023 is now coloured green as the figures are all actuals. Using the same assumptions of constant production, we now have revised projections as shown in the table above, with projection period continuing into 2073.
372. For the year 2023 we now also have measured data on output and user cost of fixed assets (included in column F of the Year 2 worksheet, Step 1). We again estimate the resource rent, unit resource rent, but now expressed in mid-2024 prices. Again, we do smoothing, and we have an opening stock value for 2024 of 1334 (with a 2% real discount rate), which is also the 2023 closing stock value. This figure can be found in cell G71 of the Year 2 worksheet. The opening stock for 2023 (cell F67) is not re-calculated in the Year 2 worksheet, but instead taken from the Year 1 worksheet. This is because of the forward-looking (or ex ante) nature of balance sheets: their main purpose is to describe how the value of assets changes over time, for instance due to revaluation.
Step 8: Put together the monetary asset account
373. We now have the NPV of the renewable energy asset at the start of the accounting period (1205 as of 1 January 2023) and end of the accounting period (1334 as of 1 January 2024, or end of 2023). It is important to realise that these NPV estimates represent the value of the assets in current prices. The compilation of the monetary asset account in constant prices is discussed in Chapter 6.
374. Table 4‑16 shows how this can be used to calculate the price of the resource32 at the beginning and end of 2023, as well as the 2023 average price and average physical stock.
Source: Workbook: Renewable energy workbook, Year 2 worksheet.
375. In the monetary asset account (Table 4‑17), there is no depletion row for renewable energy resources. The other rows of the physical asset account retain the same labels and are calculated by multiplying the average price from Table 4‑16 by the estimates in physical units from Table 4‑15. Revaluation can be calculated by multiplying the average physical stock with the change in the price of the resource in situ. Revaluation will pick up both the effect of changes in the resource rent as well as changes in the extraction path to the extent they lead to changes in asset value. In the Workbook: Renewable energy, a “check” is included to ensure that the sum of opening stock value and all changes results in a closing stock equal to the opening stock estimate of the next year.
Note: Discoveries and depletion are not applicable for renewable energy.
Source: Workbook: Renewable energy workbook, Year 2 worksheet.
Stage 4 (renewable energy): Integration
376. The information from the monetary asset account for renewable energy assets will be used in the sequence of economic accounts (standard SNA presentation) (see Table 4-18).
377. Reclassifications (upward or downward) and catastrophic losses (for example extreme weather events destroying plants) are recorded as other changes in volume. Reappraisals (upward and downward) are also treated as other changes in volume. Revaluation is to be recorded in the revaluation account. Opening and closing stocks are part of the national accounts balance sheets (opening and closing balance sheet).
4.3.3. Specific issues
Return to Intellectual Property Products
378. The advent of renewable energy is primarily driven by technology. An issue that was discussed during the EGNC was whether the resource rent captures a return to Intellectual Property Products (IPP) embedded in the equipment used to capture renewable energy, such as wind turbines or solar panels. If so, this would necessitate estimating a return to the IPP and deduce this when calculating resource rent.
379. However, as explained in 2025 SNA (§11.97) it is reasonable that the product embodying the knowledge has an increased price relative to a similar product without this embodied knowledge. User cost (depreciation and net return to capital) related to this equipment used to capture renewable energy are thus implicitly already included in the resource rent.
380. It could be argued that rents from renewable resources are embodied in the market prices of the fixed assets used to abstract value, when such assets are scarce. In case evidence exists that rents are internalised in this way (for instance if the RVM method generates negative resource rents), countries could make adjustments to split this market price between the fixed asset and the natural resource.
Taxes and subsidies
381. Section 3.4 on Taxes and subsidies explains in general terms what types of support are to be included or excluded when calculating resource rent. Essentially, resource rent excludes taxes on products (D21) and other taxes on production (D29) that are not specific; and includes subsidies on products (D31) and other subsidies on production (D39) unless they are specific. Here we will discuss several examples in more detail.
Support for manufacturers. Guidance Note WS 11 discusses the situation where support is provided to manufacturers of solar and/or wind energy equipment (e.g. capital transfers D9). It concludes that these should not be taken into account when estimating resource rent as “Market forces will tend to ensure that renewable energy producers pay fair market value for their equipment and these subsidies are not especially large in the global context. Cost structures of renewable energy producers are not likely to be any more distorted by manufacturing subsidies than are the structures of any capital-intensive sector of the economy” (SNA Update, 2023[6]).
Concessionary loans. Guidance Note WS.11 states that “a concessionary loan received by a solar electricity producer to finance purchase of solar panels would be considered a specific subsidy.” However, the AEG conclusion regarding debt concessionality (based on Guidance Note F.15) is to “never record a transfer element for concessional lending in the sequence of economic accounts, except for concessional loans provided by employers to employees (to be recorded as current transfers).” The 2025 SNA states:
Institutional units may lend to other units under conditions in which the contractual interest rate is intentionally set below the market interest rate that would otherwise apply. The degree of “concessionality” can be enhanced with grace periods, frequencies of payments and a maturity period favourable to the debtor. Since the terms of a concessional loan are more favourable to the debtor than market conditions would otherwise permit, concessional loans effectively include a transfer from the creditor to the debtor. In the sequence of economic accounts, adjustments are only made for concessional loans provided by employers to employees, whereby the difference between the market interest rate and the concessional rate is recorded as remuneration of employees. (2025 SNA, §14.112)
This implies that concessional loans are not treated as subsidies and therefore would not be deducted in calculating the resource rent.
Costa Rica stands out internationally for its renewable electricity matrix. During the decade from 2014 to 2023, 97.9% of the country's electricity production was generated from renewable sources, with hydro energy resources being its main source (72.8%), followed by geothermal (12.5%), wind (12%), solar and biomass (0.6%). The remaining 2.1% of electrical energy was produced from thermal sources.
According to information from the national accounts, published by the Central Bank of Costa Rica,33 the country's electricity industry accounted for 2% of total value added in 2021.
Due to the importance of hydro, geothermal and wind energy sources within the country's electricity generation (99.4% of electricity generation is from renewable sources), these sources are included within the preliminary valuation of renewable energy resources in Costa Rica.
For the construction of the monetary asset account for the year 2021, information on production, intermediate consumption, compensation of employees, and other taxes less subsidies on production was used. This information was taken from the estimates made by the National Accounts Unit of the Central Bank of Costa Rica. It is important to note that the national accounts make estimates for several economic activities, including the activity of electricity supply. This activity includes information in monetary terms of the electrical energy generated by all sources without distinguishing between electrical energy from renewable and non-renewable sources.
Based on the information published by the Costa Rican Electricity Institute34 on the production of renewable energy in physical units per year and with information from the national accounts, each component of the production and the generation of income account was estimated, obtaining as a balance item the gross operating surplus for each of the hydro, wind and geothermal resources.
Costa Rica is making important efforts focused on having a valuation of its fixed assets. As an approximation of the value of the renewable energy fixed assets, we have taken the information estimated by the national accounts related to the gross fixed capital formation of electric energy assets. To determine the value corresponding to the gross fixed capital formation of renewable energy assets, a disaggregation was made according to the installed capacity by type of energy.
From the value of fixed assets, fixed capital consumption was estimated by applying an asset life of fixed assets of 50 years for hydro power assets and 40 years for wind and geothermal power assets. The return to capital of the fixed assets was estimated using a 6% rate of return to capital.
Deducting from the gross operating surplus, the components of fixed capital consumption and the return to fixed capital, the resource rent associated with the production of renewable energy was obtained.
From the resource rent and the amount of energy generated per year in physical terms, the unit resource rent is estimated, which is used to obtain the projected resource rent flows during the asset life of the different assets.
A discount rate of 2% was used to convert the projected resource rent flows into an estimate of the total value corresponding to the current period; this rate expresses the asset owner's preference to receive the income in the present rather than in the future.
Given the projected resource income flows, the value of the asset as of 1 January 2021, and 1 January 2022, is estimated by the net present value. The estimated values represent the initial and final stock of the monetary account of renewable energy assets for the year 2021.
Changes in the stock of renewable energy resources have been valued at the average resource price. The average price is estimated using the net present value at current prices and the initial stock in physical terms for the years 2021 and 2022. The upward reappraisals during 2021 were due to an increase in installed capacity of 1.6% compared to the previous year, while the revaluation was due to a decline in the general level of electricity prices (-13%).
The monetary renewable energy asset account presented in Table 4‑19 shows aggregated information for hydro, wind and geothermal energy resource assets. This is a preliminary account; it is expected that future publications will present disaggregated information for the three different renewable energy resources included in this table.
Note: * Preliminary information
Source: Central Bank of Costa Rica
BPS-Statistics Indonesia regularly publishes Natural Capital Accounts in a publication entitled Indonesia System of Integrated Environmental-Economic Accounting (BPS, 2023[18]). This publication presents data on environmental asset accounts in both physical and monetary units. The compilation activities are undertaken based on the BPS Regulation No 3 of 2023 concerning the Compilation Guidelines of Natural Resources and Environmental Accounts, which is in accordance with the System of Environmental-Economic Accounting (SEEA).
Renewable energy plays an important role in reaching the target of Net Zero Emission (NZE) in Indonesia. More than half of the greenhouse gas emissions released to the atmosphere come from the energy sector, which is mainly caused by fuel combustion activities. As the demand of energy is expected to increase over time, the strategy toward NZE will be highly dependent on the shifting of energy supply from non-renewable energy resources to non-emitting renewable energy resources. Indonesia is estimated to have a huge potential of renewable energy-based power plants. In accordance with the commitment to reduce greenhouse gas emissions, the Government of Indonesia has set a target for renewable energy in the primary energy supply of 23% in 2025.
The Government of Indonesia has demanded that renewable energy resources have its own classification in the environmental asset accounts, in which the value of renewable energy resources could be differentiated from the value of land or the value of water resources. Hereto, BPS (2024[19]) has conducted an experimental study to value its renewable energy resources, summarised here.
Data sources and methodology
Indonesia has a state-owned company which is specialised in electricity supply activities, namely PT Perusahaan Listrik Negara (PLN). However, the electricity generation activities were not only carried out by PLN, but also by Independent Power Producers (IPPs) and Private Power Utilities (PPUs). Nonetheless, both IPP and PPU power plants sell their electricity to PLN as PLN controls the electricity distribution network. Meanwhile, there are also some off grid power plants, which usually operate in isolated islands and rural areas, but its production only constituted 7.63% of total electricity production in 2022.
The study used Electricity Statistics from PLN as the main data source to estimate the monetary value of renewable energy resources in Indonesia. It presents data on the electricity production and operating cost by type of power plant. While the data on electricity production covered five types of renewable energy resources, the operating cost data was limited to hydroelectric, geothermal, and solar energy.
The source data were not able to differentiate the revenue of electricity sales by type of power plant which produced them. Therefore, the value of output for each type of renewable energy power plant was calculated by multiplying the quantity of produced electricity and the highest benchmark price for purchasing electricity. Those prices were regulated in the Presidential Regulation Number 112 of 2022 concerning the Acceleration of Renewable Energy Development for Electricity Supply.
The valuation of renewable energy resources was carried out by using the NPV approach of future resource rents. The asset life of hydroelectric resources was set to 50 years because the use of a lifetime beyond 50 years has a small impact on the result of NPV calculation. Meanwhile, the asset life of geothermal and solar energy resources was set to 25 years as the future revenues and costs of such power plants were assumed to be less certain than for hydroelectric power plants.
Results
Table 4‑20 shows the results that were obtained. The valuation of renewable energy resources in Indonesia only managed to obtain a monetary value of hydroelectric and geothermal resources for electricity generation purposes only. It excluded the direct use of geothermal as well as other types of renewable energy power plant due to limited data availability. Also bioenergy used as fuel was not taken into consideration.
In 2022, the hydroelectric resources in Indonesia were estimated to have a monetary value around IDR 113 884 billion. Meanwhile, the monetary value of geothermal energy resources was IDR 106 986 billion. By considering the monetary value of non-renewable energy resources, comprising of coal, oil, and natural gas; the share of the monetary value of Indonesia renewable energy resources in 2022 was only 1.07%.
Conclusions
Even though the potential of renewable energy resources in Indonesia is enormous, the monetary value of renewable energy resources in Indonesia was highly dependent on the installed capacity of renewable energy power plants as well as on the quantity of electricity generation. The high operating cost may also influence the derivation of resource rent from the residual value method. For solar energy resources in Indonesia, the resource rent had a negative value due to high consumption of fixed capital. The net present value of renewable energy resources was also impacted by the choice of asset life and discount rate for each type of renewable energy resources.
The valuation of renewable energy resources might be better carried out by applying a bottom-up approach or site-by-site basis. The calculation based on macro data would not be able to take into account the remaining lifespan of renewable energy generation equipment of certain power plants. An in-depth study to the electricity generation establishment was recommended in order to obtain sufficient data, particularly on operating cost, to derive resource rent and to apply NPV method on the valuation of other types of renewable energy resources, such as wind and biomass energy.
4.3.4. Modifications to the standard approach
382. The approach described in Section 4.3.2 is the default. As countries differ in their data availability and resources for conducing valuation of natural resource, this section describes a Tier 1 (basic) approach that would typically be followed in case of limited data availability and/or resources, as well as Tier 3 (advanced) methods requiring more detailed data availability and more resources.
Basic approach
383. A Tier 1 method would consist of applying a product-based application of the RVM for deriving resource rent in Step 1, instead of the bottom-up method that is used for the default approach. The product-based method does not use national accounts data or business statistics data. Rather, it estimates all elements required for calculating resource rent (e.g. output, costs, user cost of fixed capital, taxes/subsidies) from a range of other data sources including commodity market prices, industry reports on cost to revenue ratios etc.
384. A good example of this method is the World Bank methodology used for the Changing Wealth of Nations (World Bank, 2021, p. 363[14]) which estimates resource rent for renewable energy for hydro, wind and solar for the 15 largest producers. The method is based upon the following steps and assumptions (Smith, Ilas and Peszko, 2021, p. 16[15]):
Multiply annual quantities of electricity generated for each technology/resource type with relevant market price.
For hydro it is assumed that the price received is the average country-wide spot price in a given year. For solar and wind, prices were estimated from a range of data sources.
Estimate subsidies received by electricity producers.
Hydro was assumed not to receive any subsidies as it a well-established technology. In the absence of readily available data on subsidies, the amount of (net) subsidies for solar and wind was estimated as the difference between actual revenues (e.g. the FiT) and the average electricity spot price.
Estimate operation and maintenance costs.
In the absence of readily available data, assumed as fixed proportion of value of investment flows: assumed 1% for onshore wind; 1.3% for solar; 1.75% for hydroelectric generation and 2% for offshore wind and Concentrated Solar Power.
Country specific investment flows were obtained by multiplying yearly physical additions to capacity by average annual investment costs (from a range of sources).
Estimate cost of produced capital.
Country specific economy wide real rate of returns were applied (e.g. 4% for OECD countries).
In order to measure depreciation, for hydro an asset life of 50 years was assumed, for solar and wind 25 years.
The stock value of produced capital of wind and solar related infrastructure was calculated by assuming that no investment was done before 1990; the stock in later years then being equal to accumulated investments minus depreciation. For hydro a more complex approach based on OECD (2009[20]) was followed according to which the value of the stock in the base year is obtained “by dividing the value of investment in the base year by the sum of the asset’s deprecation rate plus the long-term growth rate of real GDP.” (Smith, Ilas and Peszko, 2021[15])
Obtain the resource specific resource rent by deducting subsidies, operation and maintenance costs and user costs from revenue.
385. When applying a Tier 1 product-based approach, it is recommended to use national data whenever possible instead of global data sources or generic assumptions.
Advanced methods
386. If projections of future energy generation (quantities) are available and/or specific projections of specific taxes and subsidies, such results could be used to refine projections of future resource rents in Step 4 of the default method and thereby improve the estimates in the monetary asset account (Step 8).
Non-renewable mineral and energy resources (Section 4.2)
The SNA asset boundary is restricted to “Class A: commercially recoverable resources”, as defined in SEEA CF based on the UN Framework Classification of Resources.
A materiality threshold may be applied with a focus on valuation of (individual) subsoil assets that contribute more than 5% of output in ISIC Section B: Mining and quarrying and for which the long-term average contribution of mining and quarrying to GDP is at least 0.1 %. When a reasonable estimate can be made of how much of the asset value is missed, it is recommended to gross up the asset value. A default list of subsoil assets is proposed to assess which resources are available in the country and are above the materiality threshold in terms of output.
For valuing subsoil assets, the preferred valuation method is to use observable market transactions. When such markets are very thin or do not exist at all, and valuation based on rights to use natural resource is not adequate, the recommended valuation method is to apply the net present value of future resource rents, calculated with the Residual Value Method (RVM).
When using the RVM to calculate resource rents, a top-down approach is recommended.
When making projections of future extraction, extraction should be assumed to continue at the same level as in the (recent) past, unless a specific extraction profile is available. The key restriction is that the total projected extraction over the asset life at the start of the accounting period should be equal to the total physical opening stock (Class A) as included in the physical asset account for the resource in question.
It is recommended to assume that unit resource rents remain constant in the future unless specific policies have been implemented which would allow to estimate a specific path of future unit resource rents.
Stranded subsoil assets should be recorded as (downward) reappraisals in the physical asset account. There can be different types of situations that require a different recording in the national accounts:
In case of a change in relative prices, this could have both a volume effect (lower reserves considered as economically viable to exploit) and a price effect (the reserves extracted generate a lower price). It is recommended to split these two impacts when compiling the monetary asset account.
In case of a change in the extraction path: if the total amount of reserves that is extracted remains the same, this is not a volume effect but (due to the effect of discounting) a value effect, which should be recorded as a revaluation.
In case there is more (or less) of the resource accessible (e.g. due to legal changes), this should be recorded as an “other change in the volume of assets”.
A Tier 1 (basic) method would consist of applying a product-based application of the RVM for deriving resource rent, instead of the top-down activity-based method that is used for the default approach. The product-based method does not use national accounts data but estimates all elements required for calculating resource rent from a range of data sources. When applying the product-based approach, it is recommended to use national data whenever possible instead of the global data sources.
Several Tier 3 methods exist such as applying more advanced modelling of future prices and/or extraction rates or compiling the value of mineral and energy resources at a disaggregated (ideally individual deposit) level.
Renewable energy resources (Section 4.3)
Renewable energy resources are shown separately in the 2025 SNA classification of assets as AN322. Renewable energy assets to be included in the national accounts correspond to a subset of renewable energy resources, namely those resources “viable for use in economic production under prevailing technological and economic conditions”, similar to Class A for subsoil assets.
The scope for renewable energy resources includes resources that generate electricity and/or heat or cool air. Biomass or other biofuels (e.g. renewable municipal waste, biogas, liquid biofuels) are excluded from scope. In scope are geothermal energy resources including hot surface water, aerothermal energy (e.g. captured by heat-pumps), solar thermal, and wave and tidal resources.
A materiality threshold may be applied with a focus on valuation of those (individual) renewable energy resources that contribute more than 5% of the national energy supply. The easiest way to assess this is by looking at the generation of energy by source (in physical units) based on energy statistics. When a reasonable estimate can be made of how much of the asset value is missed, it is recommended to gross up the asset value. A default list of renewable energy resources is proposed to assess which resources are available in the country and are above the materiality threshold.
For valuing renewable resource assets, the preferred valuation method is to use observable market prices for transactions. Under some circumstances, the valuation of renewable energy will already be captured in associated land values; for instance, in the case of onshore wind energy; or in the value of dwellings equipped with solar panels. In such cases a hedonic pricing method could be followed, which may provide a direct estimate of the renewable energy asset value. However, as the use of hedonic pricing methods is highly resource intensive, requiring extensive micro-data, it is recommended as an advanced method. In some instances, rights to access the natural resource exist but valuation based on rights may only be feasible in certain circumstances. Therefore, in most cases, the recommended valuation method is to calculate asset values as the Net Present Value (NPV) of future resource rents, calculated with the Residual Value Method (RVM).
When using the RVM to calculate resource rents, a bottom-up approach is recommended, until the national accounts distinguish between renewable and non-renewable energy production as separate economic activities (ISIC revision 5 introduces this distinction at class level) which would allow for the top-down approach.
The Least Cost Alternative method for valuing renewable energy resources is considered not suitable for use in compiling national accounts.
The user costs of produced assets required for the RVM need to be estimated based on estimates of the value of the installations. Countries are recommended to include solar panels and heat pumps as two additional and separate asset classes in their perpetual inventory models (PIMs). The assumed service life used in PIMs should not necessarily be the same as the life of the natural resource. Unless more specific information is available, it is recommended to apply 25 years as default for solar panels with an assumed loss of 0.5% annually in efficiency. For other equipment used by households a default asset life of 15 years is assumed with an assumed loss of 1% annually in efficiency. The loss in efficiency can be used to estimate suitable depreciation profiles for these assets.
Variable renewable energy grid integration costs such as costs made to connect wind turbines and solar panels to the network are likely to be modest and therefore may be excluded when estimating renewable energy resource rents.
It is not recommended to use an infinite asset life when applying the NPV method for valuing renewable energy resources due to uncertainties in technology and/or climate considerations. Instead, it is recommended to use the average (or median) licensed lifetime of projects as asset life of the resource, and, if such data is not available, to apply a default asset life of the resource of 30 years for wind and solar, and 50 years for hydro.
If there is no evidence to support a different scenario, the standard approach recommends assuming that the (smoothed) unit resource rent stays constant i.e. it also assumes that the current (net) subsidy levels remain constant when projecting resource rent into the future.
A Tier 1 (basic) method would consist of applying a product-based application of the RVM for deriving resource rent, instead of the bottom-up activity-based method that is used for the default approach. The product-based method does not use national accounts data but estimates all elements required for calculating resource rent from a range of data sources. When applying the product-based approach, it is recommended to use national data whenever possible instead of the global data sources.
A Tier 3 method would consist in using projections of future energy generation (quantities) and/or specific projections of specific taxes and subsidies when projecting future resource rents.
References
[19] BPS (2024), “Valuation of Renewable Energy Resources in Indonesia”, Paper prepared for the Group of Experts on National Accounts Twenty-third session, Geneva, 23-25 April 2024.
[18] BPS (2023), Indonesia System of Integrated Environmental-Economic Accounting 2018-2022, https://www.bps.go.id/en/publication/2023/12/29/f66305e39dc383840a2a313b/sistem-terintegrasi-neraca-lingkungan-dan-ekonomi-indonesia-2018-2022.html.
[9] European Commission (2014), Manual on the changes between ESA 95 and ESA 2010, https://ec.europa.eu/eurostat/web/products-manuals-and-guidelines/-/ks-gq-14-002.
[17] Eurostat (2024), Methodological Note - Recording the Production of Electricity by Households in National Accounts.
[1] Eurostat (2003), Subsoil asset accounts for oil and gas – Guidelines for the set of standard tables, Revised version, January 2003.
[4] IMF (2017), Guide to analyze natural resources in national accounts, https://www.imf.org/external/pubs/ft/qna/pdf/na.pdf.
[12] Liu, G. (2023), Testing the split of economic ownership for petroleum resources in Norway, Documents 2023/24, Statistics Norway, https://www.ssb.no/en/nasjonalregnskap-og-konjunkturer/nasjonalregnskap/artikler/testing-the-split-of-economic-ownership-for-petroleum-resources-in-norway.
[10] Liu, G. and S. Midttun (2024), Compiling physical asset accounts for petroleum resources in Norway, Statistics Norway (Documents 2024/7), http://www.ssb.no/nasjonalregnskap-og-konjunkturer/nasjonalregnskap/artikler/compiling-physical-asset-accounts-for-petroleum-resources-in-norway/_/attachment/inline/430a7d74-d926-4337-8e9c-c668c22d116c:0c6b7c7611f79de53d7a31a1465abe61f0df72f2/NOT2024-07.pdf.
[13] Liu, G. and S. Midttun (2024), “Including Petroleum Resources as Asset in the Norwegian National Accounts”, Paper prepared for the 38th IARIW General Conference August 26-30, 2024.
[11] Liu, G. and S. Midttun (2024), Is it necessary and feasible to estimate the asset value from oilfield level?, Statistics Norway (Documents 2024/8), http://www.ssb.no/en/nasjonalregnskap-og-konjunkturer/nasjonalregnskap/artikler/compiling-physical-asset-accounts-for-petroleum-resources-in-norway/_/attachment/inline/430a7d74-d926-4337-8e9c-c668c22d116c:0c6b7c7611f79de53d7a31a1465abe61f0df72f2/NOT2024-07.pdf.
[20] OECD (2009), Measuring Capital - OECD Manual 2009: Second edition, OECD Publishing, Paris, https://doi.org/10.1787/9789264068476-en.
[5] Pionnier, P. and S. Yamaguchi (2018), “Compiling Mineral and Energy Resources According to the System of Environmental-Economic Accounting (SEEA) 2012: A Contribution to the Calculation of Green Growth Indicators”, OECD Green Growth Papers, No. 2018/0, OECD Publishing, Paris.
[15] Smith, R., A. Ilas and G. Peszko (2021), Valuation of Renewable Energy Resources in the Context of the Changing Wealth of Nations Experimental Results Final report, World Bank., https://documents1.worldbank.org/curated/en/099164002232311006/pdf/P17727804fa8610fd0a59c03a5b40122b05.pdf.
[6] SNA Update (2023), WS.11 Guidance note on the treatment of renewable energy resources as assets, https://unstats.un.org/unsd/nationalaccount/SNAUpdate/GuidanceNotes.asp.
[8] UNECE (2024), “Final report, Item 3 of the provisional agenda Improvement of measures of consumption of fixed capital ECE/CES/GE.20/2024/6”, in Directors of Macroeconomic Statistics Task Force on fixed assets and estimation of consumption of fixed capital under European System of Accounts 2010, presented at the Conference of European Statisticians / Group of Experts on Statisticians / Group of Experts on National Accounts, Twenty-third session Geneva, 23-25 April 2024.
[7] UNECE (2020), United Nations Framework Classification for Resources - Update 2019, https://unece.org/sites/default/files/2023-10/UNFC_ES61_Update_2019.pdf.
[16] UNFC (2016), Specifications for the application of the United Nations Framework Classification for Fossil Energy and Mineral Reserves and Resources 2009 to Renewable Energy Resources, Geneva.
[3] United Nations (2019), System of Environmental-Economic Accounting — Accounting for Energy, https://seea.un.org/sites/seea.un.org/files/documents/seea-energy_final_web.pdf.
[2] United Nations et al. (2014), System of Environmental-Economic Accounting 2012 — Central Framework, United Nations, https://seea.un.org/sites/seea.un.org/files/seea_cf_final_en.pdf.
[14] World Bank (2021), The Changing Wealth of Nations. Managing assets for the future, International Bank for Reconstruction and Development / The World Bank, https://www.worldbank.org/en/publication/changing-wealth-of-nations.
← 1. It should be noted that at the time of finalising the guide, some of the codes associated with the classification hierarchies of the 2025 SNA were still under review and may undergo change. Users are advised to consult the final published version of the SNA when it becomes available.
← 2. The 2025 SNA also includes specific guidance on the measurement of household production of electricity and heat (§7.154-7.155) which may lead to changes in the estimation of output.
← 3. The SEEA CF distinguishes five broad classes, while the SEEA Energy (2019) separately identifies uranium (and other nuclear fuels) for its importance in energy production.
← 4. Lignite is also known as brown coal.
← 5. The asset category Other could consist of assets such as stone, sand or clay, salt or peat which may be significant in certain countries.
← 6. The SEEA CF (§5.189) explains that in case of subsoil assets: “reappraisals ... pertain only to known deposits ... based on changes in geologic information, technology, resource price or a combination of these factors .... Catastrophic losses are rare in relation to most mineral and energy resources …. Flooding and collapsing of mines do occur but the deposits continue to exist and can, in principle, be recovered: the issue is one of economic viability of extraction rather than actual loss of the resource itself. An exception to this general principle concerns oil wells that can be destroyed by fire or become unstable for other reasons, leading to significant losses of oil resources …. Reclassifications may occur if certain deposits are opened or closed to mining operations owing to government decisions concerning the access rights to a deposit.”
← 8. In case a bottom-up approach is preferred, the specific recommendation is to use actual labour costs in the resource rent calculation (rather than opportunity cost of labour), following the SEEA Central Framework (CF) recommendation.
← 9. Guidance Note WS 10 alludes to this same issue when it mentions the importance of paying attention to a.o. “Constraints imposed on mineral production at the micro level by initial investments in physical capital”. (§18)
← 10. It would be possible to make a different choice (e.g. that it falls at the end of the accounting period) as this is merely a convention. However, in order to standardise the approach, a decision had to be made, and this was the recommendation of the EGNC (and is consistent with 2025 SNA (§18.117) which mentions that for flow variables the desired valuation point is usually the mid-point of the period).
← 11. The price being estimated is price in the ground (in situ) before any processing has taken place.
← 12. The term “mineral” in this context includes all non-renewable energy and mineral resources, not only minerals.
← 13. If capitalisation of mineral exploration and evaluation still needs to be undertaken, the following applies: intermediate costs (e.g. of surveying) would be recorded as gross fixed capital formation (GFCF). If exploration is undertaken for own purposes, any labour costs (compensation of employees) would need be recorded as additional output (used by GFCF). Intuitively, the capitalisation increases output and therefore GOS, at the same time there is also an increase in user costs of produced assets (as the fixed capital stock increases with the IPP). The resource rent will reflect the net effect.
← 15. Referred to in these guidelines as the “everything but” the activity approach. Mainland Norway consists of all domestic production activity except exploration of crude oil and natural gas, transport via pipelines and ocean transport.
← 16. While no depletion costs are recorded, ecological constraints are taken into account in the choice of asset life especially for resources like hydro, geothermal, and others that may face long-term sustainability limits.
← 17. The SEEA Energy (UN 2019) contains a breakdown of renewable sources that is very similar, only wave and tidal are separately identified. The International Renewable Energy Agency (IRENA) commonly distinguishes between renewable hydro, wind, solar energy, bioenergy (e.g. solid biofuels, biogas, renewable municipal waste), geothermal energy, and marine energy.
← 18. This may change in the near future, for instance the SIEC (Standard International Energy Product Classification is at the time of writing these guidelines under revision; the International Renewable Energy Agency (IRENA) has also prepared a new system to classify different energy sources.
← 19. For instance, in the EU context detailed protocol exist for the estimation of the share of renewable energy in total electricity or heat end-use (SHARES tool).
← 20. An exception may arise in case a country’s energy supply predominantly is generated from renewable energy resources.
← 23. In addition to FiTs, a range of other policy instruments may exist such as Feed-in-Premiums (that pay a difference compared to the electricity market price).
← 24. These may be in addition to the remuneration households receive through the FiT.
← 26. Based on UNECE (2014) recommendation for AN1139 Other machinery and equipment, specifically CPA 27: electrical equipment.
← 28. A practical advantage of using a finite asset life is that the total physical stock also remains finite.
← 29. It would be possible to make a different choice (e.g. that it falls at the end of the accounting period) as this is merely a convention. However, in order to standardise the approach, a decision had to be made and this was the recommendation of the EGNC (and is consistent with 2025 SNA (§18.117) which mentions that for flow variables the desired valuation point is usually the mid-point of the period.
← 30. The rationale for the 25% was based on data from the Canadian balance sheets being the only country at the time of writing that applied the split-asset approach (SNA Update, 2023, p. 36[6]).
← 31. This is why in addition to LCOE (as discussed in Annex A) currently VALCOE (value adjusted LCOE) is used as a metric designed to also take additional issues such as flexibility and grid stability into account.
← 32. The price being estimated is price in the ground (in situ) before any processing has taken place.