Guidance note on the treatment of renewable energy resources as

Guidance note on the treatment of

renewable energy resources as

assets

2025 SNA Revision

April 2022

Summary of recommendations

This guidance note recommends recognition of renewable energy resources under certain

conditions as natural resource assets in the revised SNA. The absence of renewable energy

resources from the current SNA asset boundary is a concern from the point of view of assessing

the environmental sustainability of economic activity. Since fossil fuel resources are included

within the asset boundary, there is an imbalance in the way in which the national accounts treat

non-renewable and renewable energy resources. Given the climate-related consequences of

fossil fuel use, this imbalance risks sending distorted signals to decision makers regarding the

relative economic importance of carbon-intensive, non-renewable energy sources versus less

climate-damaging renewable sources. Renewable energy resources already account for 30% of

global electricity production and are the fastest growing source, according to the

International

Energy Agency. Moreover, governments are beginning to capture the associated rents. They

have long done so in the case of hydroelectric resources and are starting to do so in the case of

newer renewables, like wind in the United Kingdom and Canada. As they expand in the future,

the overall value of renewable energy resources will become too economically important to

ignore in the national accounts. Doing so would permit an imbalance between the value of the

assets recognized in the capital account and the income measured in the production account.

It is important to be clear as to what “renewable energy resources” are and how they are related

to the topic of interest here, renewable energy assets. In keeping with the

renewable energy

specifications of the United Nations Framework Classification for Resources (UNFC; United

Nations Economic Commission for Europe, 2020), it is taken that renewable energy resources

comprise the cumulative quantities of kinetic, heat or radiative 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).

The physical

unit of measure for these resources is the joule in the SI system (other units that could be used

include BTUs, tonnes of oil equivalent or watt-hours).

Not all renewable energy resources qualify as economic assets however. In further keeping with

the general definition of an asset, only those renewable energy resources that are viable for use

in economic production under prevailing technological and economic conditions qualify as

assets. These resources correspond to those that would be classified to the UNFC category

“commercial projects”. More specifically, these are the renewable energy resources that provide

inputs into renewable energy production facilities where extraction and sale is economic on

the basis of current market conditions and realistic assumptions of future market conditions; all

necessary approvals/contracts have been confirmed or there are reasonable expectations that

all such approvals/contracts will be obtained within a reasonable timeframe; economic viability is

not affected by short-term adverse market conditions provided that longer-term forecasts remain

positive; extraction is currently taking place; or, implementation of the development project is

underway; or, sufficiently detailed studies have been completed to demonstrate the feasibility of

extraction by implementing a development project.

The stock of renewable energy assets in a country at a given point in time is, then, the

cumulative quantity of renewable energy resources “harvestable” by the commercial renewable

energy projects in existence at the time. The UNFC recommends limiting the quantification of

renewable energy resource stocks to by considering lifetimes for existing commercial projects.

Lifetimes can be determined from the design basis of the facilities or based on industry

practice/benchmarks. The fact that renewable energy assets can be characterized in physical

terms permits use of the standard accounting concept of “price times quantity” as the basis for

valuing the assets, where their price is the rent that arises during their use in a production

process.

Given the above, it is clear that a remote river with no hydroelectric generation facilities on it,

nor any under such facilities under construction, is not a renewable energy asset. Nor is an air

current in which no wind turbine currently stands (or is imminently being installed) or a ray of

incident solar radiation that does not fall on an existing or imminent energy capturing device

such as a photovoltaic cell or a mirror in a concentrated solar plant. This is consistent with the

treatment of other natural resource assets in the SNA and SEEA-CF. For example, the SNA and

SEEA-CF recognize timber in a forest (another renewable resource) as an asset only in

instances where that timber may be commercially logged at a profit under existing technological

and economic conditions. Remote forests with no potential for logging do not qualify as assets.

Similar criteria are applied to defining other renewable and non-renewable natural resources as

assets in the SNA and SEEA-CF.

In this guidance note, we review the accounting concepts and methods relevant to treating

renewable energy resources as assets and make several recommendations for their application

in the context of the revised SNA. These recommendations are rooted in both the theory and

practice of national economic and environmental accounting as laid out in the SNA and the

SEEA-CF. Where best practices are unclear or simply unrealistic, our recommendations are

guided by the need to provide practical, useful information to allow management of a country’s

natural resource assets.

Neither the SNA nor the SEEA-CF defines a complete and internally consistent approach to the

treatment of renewable energy resources as assets. The SNA has little to say about these

resources explicitly but what is written implies they do not qualify as economic assets because

ownership rights cannot be enforced. The SEEA-CF treats them in detail, considering their

asset value to be captured in associated land values. We find, however, that this treatment fails

to adequately address renewable energy generation that 1) is not associated with land (offshore

wind, solar and ocean resources); 2) exists under ownership rights clearly separated from land

(hydroelectric and most geothermal resources); or 3) is associated with land that has no

economic value and does not appear in the national accounts (hydroelectric and most utility-

scale solar/wind resources). In the few cases where renewable energy resource use is in fact

bundled with land that is both owned and valued in the national accounts (for example, wind

turbines on agricultural land), the SEEA-CF’s approach assumes that land markets accurately

“price in” the value of associated resources. While the SEEA-CF’s argument has prima facie

appeal, we find the assumptions underlying it do not stand up fully under examination and that

is unlikely (impossible in many instances) that existing land values in the SNA will capture

renewable energy asset values. We find as well that a review of the empirical literature on the

relationship between renewable energy production supports this view. Available studies do not

point to a consistently positive relationship between renewable energy production and land

values. Evolutions in market participants’ knowledge and policy contexts both within and among

countries appears to play a significant role in determining the size – and even the direction – of

the relationship between land values and renewable energy production.

We recommend instead that a separate asset category for renewable energy resources

be created within the revised SNA (and, ultimately, the SEEA-CF) and that the value of

these assets be partitioned between their governments (their legal owners) and

renewable energy companies, since both governments and renewable energy companies

may be considered economic owners of the resources.

We consider the risk of double

Here we refer to the guidance on this issue laid out in the Guidance note on Accounting for the Economic

Ownership and Depletion of Natural Resources prepared by Peter van de Ven and Mark de Haan.

counting between the existing land asset and this new asset category to be low on both

economic and accounting grounds; specifically, land markets today likely fail to completely

internalize renewable energy resource values (where such resources are bundled with land)

and there is rivalry between use of some land (e.g., farmland) for renewable energy generation

and other uses. Furthermore, national accountants can employ practical means in compiling

national balance sheets that avoid of double counting.

If, as we recommend, renewable energy resources are to be recognized as assets in the

revised SNA, a method must be found to value them. The SEEA-CF and (somewhat less

clearly) the SNA recommend a residual value method for natural resource assets in

general. In this, asset value is taken to be equal to the present value of the future stream

of rent flowing from the resource. Rent, for its part, is calculated as the difference

between resource revenues (less specific subsidies received plus specific taxes paid)

and production costs, including returns to labour and produced capital. We recommend

that the same approach be applied to the valuation of renewable energy resources in

most instances in the revised SNA, noting that pilot empirical work by the World Bank

(Smith et al., 2021) demonstrates that this approach yields results that are both plausible

and defensible in the context of broader economic trends.

The validity of the residual value method rests on an assumption of renewable energy markets

approximating long-run competitive equilibrium and we acknowledge that markets in many

countries – especially in the developing world – do not meet this standard. However, data from

the OECD suggest considerable movement toward competitiveness since deregulation of

electricity markets began, at least in developed countries. Heavy subsidization of renewable

energy production and consumption remains, however, and this poses a clear theoretical

challenge to the residual value method. Where the residual value method is inappropriate

due to subsidization or other market distortions, an alternative approach, known as the

“least-cost alternative” method is recomended. This approach attempts to identify rents by

comparing the cost of electricity generation with and without renewable resources. This

technique has been applied with varying sophistication to the valuation of hydroelectric

resources.

Our primary rationale for recommending the residual value method in most instances is

consistency: the method is widely applied in country practice and by the World Bank

and

UNEP

to other environmental assets that policy-makers must evaluate against renewable

energy resources. We consider the potential pitfalls of applying this approach, notably the

payment of subsidies on renewable energy production, consumption and equipment

manufacture. While care must be taken, we note that heavy subsidization is not unique to the

renewable energy sector.

See the World Bank’s Changing Wealth of Nations series of reports.

See the World Bank’s Inclusive Wealth series of reports.

1 Introduction

In this guidance note (GN), we address the treatment of renewable energy resources as assets

in the System of National Accounts 2008 (SNA; European Commission et al., 2008) and in the

related System of Environmental-Economic Accounting 2012 – Central Framework (SEEA-CF;

United Nations et al., 2014a). We find that current guidance in the SNA and the SEEA-CF does

not define a fully satisfactory approach to measuring renewable energy resources. The SNA

currently excludes them from its natural asset boundary and offers only a limited discussion of

them. The SEEA-CF’s treatment, while more extensive, is found to require re-examination in

certain aspects. In particular, the SEEA-CF’s view that renewable energy resource values are

already captured in the value of the associated land does not address renewable energy

generation that 1) is not associated with land (offshore wind, solar and ocean resources); 2)

exists under ownership rights clearly separated from land (hydroelectric and most geothermal

resources); or 3) is associated with land that has no economic value and does not appear within

the scope of land resources measured in the SNA (hydroelectric and most utility-scale

solar/wind resources). Only the value of privately owned land used for renewable energy

generation (e.g., solar and wind installations on private land) is liable to reflect the value of the

renewable energy resources.

The absence of renewable energy resources from the SNA’s asset boundary is a concern from

the point of view of assessing the environmental sustainability of economic activity. Fossil fuel

resources are included within the SNA’s asset boundary, leading to an imbalance in the way in

which the system treats energy resources. Given the climate-related consequences of fossil fuel

use, this imbalance risks sending distorted signals to decision makers regarding the relative

economic importance of carbon-intensive, non-renewable energy sources versus less climate-

damaging renewable sources.

To address this concern, we propose creation of a new asset category for renewable energy

resources in the revised SNA (and in any future revision of the SEEA-CF) and the partitioning of

the value of the assets between their legal owners (governments) and economic owners

(renewable energy companies). We argue that valuation of these resources can proceed in

most instances via the “residual value” method recommended in the SEEA-CF (and the SNA)

for the valuation of other natural resource assets. An alternative approach, known as the “least-

cost alternative” method, may also be applicable, particularly in countries where electricity

markets are very far from equilibrium.

The significant and growing importance of renewables means there is an economic case for

their inclusion within the revised SNA’s asset boundary. Available evidence, limited as it may

be, suggests that renewable energy assets – especially hydroelectric resources – may already

be worth trillions of dollars worldwide. Extrapolating from partial findings for Canada (Gillen and

Wen, 2000), Canada’s hydroelectric resources alone may be worth $US380 billion in current

dollars. This would place them on par with the value of Canada’s large fossil fuel reserves. As

solar and wind energy expand and nascent technologies like geothermal and wave energy are

developed, the overall value of renewable energy assets will likely become too economically

important to ignore in the national accounts. Doing so would permit an imbalance between the

This note has been prepared by Robert Smith of Midsummer Analytics. We wish to thank Karen Wilson and Patrick

O’Hagan, both formerly of Statistics Canada, for sharing their knowledge and insights regarding the national accounts

and its treatment of natural resource assets. We are grateful as well to Grzegorz Peszko, Glenn-Marie Lange,

Albertine Potter van Loon, Shun Chonabayashi, Stefanie Onder and Catherine Van Rompaey of the World Bank for

their helpful direction and comments during its preparation. Any errors or omissions remain entirely our own.

value of the assets recognized in the capital account and the income measured in the

production account.

Our discussion in this GN is focused on geothermal, hydroelectric, solar and wind resources

and primarily on their use to generate electricity. We realize this ignores other important benefits

of these resources; for example, the use of geothermal and solar resources directly as sources

of heat. We realize as well that other renewable energy resources are of economic importance,

most notably biological resources (fuelwood and other biomass) but also ocean energy (waves

and tides). Our restricted focus is partly pragmatic – to keep the GN to a reasonable length –

but also reflects the fact that there is great attention focused today on renewable electricity

generation. Renewable resources offer the possibility of an emissions-free source to meet the

world’s growing need to provide homes, factories, communications and, increasingly,

transportation networks with electricity. We would note, however, that the arguments we present

with respect to geothermal, hydroelectric, solar and wind electricity apply equally to other

renewable energy sources.

The remainder of the GN proceeds as follows. In Section 2 we discuss the treatment of

renewable energy resources as assets, starting from what is proposed in the SNA and SEEA-

CF. Section 3 is devoted to a discussion of the options considered for the inclusion of renewable

energy assets in the SNA and our recommended approach, which is grounded in theory but with

due regard for practicality. Section 4 deals with the conceptual aspects of our recommended

approach, while Section 5 deals with its practical aspects. Several appendices provide

additional information where necessary.

2 Existing material

In this section, we discuss the current treatment of renewable energy resources as economic

assets. This requires discussion of the way assets are defined in general terms in SNA and the

SEEA-CF. We follow this with more detailed discussions of their treatments of natural resources

broadly and of renewable energy resources specifically. The SEEA-CF treatment of renewable

energy assets, in particular, is reviewed thoroughly.

2.1 Renewable energy resources as economic assets

2.1.1 Assets in general terms in the SNA and SEEA-CF

Assets are socially defined entities that can, and do, evolve over time. Both the asset boundary

(the criteria that separate assets from non-assets) and the domains over which it extends

depend on specific economic and institutional arrangements. The evolution of asset definitions

is of central importance to any effort to account for renewable energy resources, because these

resources are emerging as important parts of economic life across the globe. Some, like run-of-

river hydropower, have been used for millennia. Others, such as geothermal, solar and wind

electricity, have only emerged recently as widespread inputs to production as technology and

consumer preferences change.

When natural resources serve no purpose in economic production, no reason exists for them to

be recognized as assets. Conversely, when resources become productive (that is, used in

economic activity), they can be recognized as assets for national accounting purposes (see, for

example, the case of the radio spectrum discussed in Section 2.1.3). The emergence of newly

productive economic resources is often followed by legal recognition of associated property

rights, a pre-condition for the resources to be viewed as assets. This is already occurring in the

instance of renewable energy resources. For example, Danish authorities have awarded

damages to the owner of a wind energy farm due to losses caused by construction of another

farm upwind, implying that the wind is a resource with benefits to which one individual’s access

is not to be unduly restricted by another (Diamond, 2015). In the United States, governments

have begun to develop regulations to ensure access to sun and wind for the production of

energy by restricting development on neighbouring properties (Diamond and Crivella, 2011;

Landis, 2019).

According to the SNA, an asset is an entity over which ownership rights are enforced “by some

unit, or units, and from which economic benefits are derived by their owner(s) by holding or

using them over a period of time.” (SNA ¶1.46). Key to this definition is the notion of an

economic benefit, which is defined in the SNA as a benefit, measurable in monetary terms, from

the use of an entity in the context of a market activity (production, consumption or accumulation)

or from holding the entity as a store of monetary value (SNA ¶3.19). For something to be

considered an asset, then, any benefits it provides must flow in the context of productive

activity. This excludes entities that provide benefits outside the scope of human productive

actives from consideration as assets. This approach means that economic assets are “revealed”

by these activities (often market activities); anything outside them may be excluded.

Also key to the SNA asset definition is the notion of ownership. The SNA is explicit in noting that

ownership need not be private for an entity to qualify as an asset (SNA ¶1.46). Collective

ownership by all members of a country is acceptable, allowing the SNA to define natural

Brief descriptions of renewable energy resources in physical terms are provided in Appendix 1.

resources – like oil reserves – owned by governments on behalf of all citizens as assets.

Collective ownership does not extend beyond the national level, however, since the focus of the

SNA is on accounting for the economies of nation states; this explains the SNA’s rejection of the

high seas as an economic asset.

The SEEA-CF follows the SNA almost completely in its basic asset definition, with one

difference. Unlike the SNA, which focuses only on assets that provide economic benefits, the

SEEA-CF extends its asset boundary to include “all resources that may provide benefits to

humanity”, opening the door to inclusion of resources that provide both economic and non-

economic benefits (SEEA-CF ¶5.14). However, resources of the latter type are measured only

in physical terms in the SEEA-CF and are not referred to as “economic” assets. Only assets that

provide economic benefits are measured in monetary terms in the SEEA-CF. For example, in

physical terms, all land within a country lies within the asset boundary of the SEEA-CF, while, in

monetary terms, some land may have zero economic value and hence be excluded from

consideration as an economic asset.

2.1.2 Natural resources as assets in the SNA and SEEA-CF

In keeping with its general definition of assets, the SNA recognizes as assets only natural

resources over which ownership rights can be – and are – enforced. The specific natural

resources recognized as assets in the SNA are:

• land (including soil and associated surface water)

• mineral and energy resources found on and under the earth’s surface (including

underwater)

• biological resources (trees, plants and animals) that grow under natural conditions (as

opposed to those, like farm animals or plantation forests, that grow under managed

conditions)

• surface and groundwater, so long as it is regularly used for extraction

• electromagnetic (radio) spectrum used for telecommunications purposes.

As with its basic definition of assets, the SEEA-CF largely mirrors the SNA in its recognition of

natural resources as assets, though it treats some resources – especially land – differently. The

SEEA-CF places land in a separate category from other natural resource assets, seeing it as an

asset only from the perspective of its use for the provision of space (¶1.49). “Soil resources” are

a separate asset unto themselves in the SEEA-CF. In contrast, the SNA considers “land” to

comprise both the space it provides as well as the soil underlying it. The other assets in the

SEEA-CF natural resource category are, as in the SNA: mineral and energy resources,

biological resources and water resources. Interestingly, however, the SEEA-CF does not

recognize the radio spectrum as a natural asset, arguing that it is “not part of the biophysical

environment” (SEEA-CF ¶5.36, footnote 48).

When it comes to the question of ownership of natural resource assets, the SNA is clear that the

general principle upon which asset ownership is to be determined is economic ownership:

“assets appear on the balance sheet of the unit that is the economic owner

”. The SNA goes on,

The classic example is remote public land, such as a wild forest, that provides no economic benefits. Such land is

generally not included on national balance sheets. Australia is an exception to this; a measure of the value of “other

land” (government-owned land that is not used for residential, commercial or other economic purposes) is included on

Australia’s national balance sheet (Cadogan-Cowper and Comisari, 2009).

The economic owner of an asset is an entity that has agreed by way of contract to accept the risks and rewards of

using the asset in production in return for an agreed amount to be paid to the asset’s legal owner. The legal owner is

the unit entitled in law to the benefits of the asset’s use (SNA ¶2.47).

however, to note that “when a natural resource is the subject of a resource lease, the asset

continues to appear in the balance sheet of the lessor [e.g., a government] even though most of

the economic risks and rewards of using the asset in production are assumed by the lessee

[e.g., a resource company]” (SNA ¶13.3). There is, then, an inconsistency in the SNA’s

approach to asset ownership; ownership is attributed to the economic owner unless the asset in

question is a natural resource, in which case ownership is attributed to the legal owner. The

SNA explains this by stating "…there is no wholly satisfactory way in which to show the value of

the [natural resource] asset split between the legal owner and the extractor, [so] the whole of

the resource is shown on the balance sheet of the legal owner” (SNA ¶13.50). The SNA

acknowledges that this treatment is not “wholly” satisfactory. One concern is that the value of

any natural asset recorded on a government’s balance sheet (as the legal owner) would not be

in line with the actual rent earned by the government on that asset except in cases where

royalty payments succeed in extracting all resource rent from resource companies. This is

rarely, if ever, the case. Thus, the valuation of the asset on the national balance sheet would

not, in fact, reflect the actual economic value of the asset to the government.

The SEEA-CF, for its part, states that “the economic value of mineral and energy resources

should be allocated between the extractor and the legal owner” (SEEA-CF preface ¶33) and

that “the allocation of assets and the resulting estimates of institutional sector net worth should

reflect the expected future income streams for each unit from the extraction of the resources”

(SEEA-CF ¶5.223). There is no more explicit statement in the SEEA-CF regarding the

recommended approach to attributing ownership of natural resource assets. Based on these

two excerpts, the SEEA would seem to diverge with SNA treatment, suggesting that the value of

natural resource assets should be divided between the legal and economic owners rather than

being attributed wholly to the former.

A clear and explicit treatment of this issue is found in the

Guidance note on Accounting for the

Economic Ownership and Depletion of Natural Resources prepared by Peter van de Ven and

Mark de Haan. We agree with the approach to sharing of ownership of natural resources

laid out in that note and recommend that it be considered as the guidance for treating the

ownership of renewable energy resources as well. We would note that the guidance of van

de Ven and de Haan is consistent with the approach taken by Statistics Canada (2015), the only

statistical agency worldwide to have developed sectored and quarterly balance sheet accounts

for natural resources.

2.1.3 Renewable energy resources as assets in the SNA and SEEA-CF

The SNA says little regarding the specific treatment of renewable energy resources as assets. It

simply states, as noted earlier, that entities “over which no property rights can be exercised” do

not qualify as assets, using the high seas and atmosphere as examples. This suggests that both

solar and wind resources would not be recognized as assets within the SNA, since they are

closely linked to the atmosphere. What the SNA actually intends with respect to solar and wind

resources is unclear, however, as neither is mentioned anywhere in the text. It is worth recalling

that the SNA does recognize the radio spectrum used by telecommunications companies as a

natural resource asset (SNA ¶10.185). The reasons for this are not fully spelled out but appear

related to 1) the unprecedented demand for access to the spectrum created by the arrival of 3G

cellular telephone technology in the early 2000s and 2) the fact that use of the spectrum is rival

(users can disrupt and degrade one another’s signals) but not physically excludable (no user

can physically prevent another’s use). In response, governments expanded regulation of access

to the spectrum through auctioning of cellular communications licenses beginning in the late

Statistics Canada, Table 36-10-0580-01, National Balance Sheet Accounts. Available here.

1990s. This generated very large public revenues (Jilani, 2015). Acknowledging this, the

drafters of the SNA

agreed to include the radio spectrum within its asset boundary.

Interestingly, they argued that “land, mineral deposits and the spectrum are similar types of

assets” and consequently classified the spectrum as a natural resource asset (SNA ¶17.317).

This is an example of an entity previously deemed to have no economic value and over which

ownership rights were not exercised subsequently meeting both SNA asset boundary

requirements through government’s decision to exercise public ownership rights prompted by

changing economic circumstances. The potential parallels with renewable energy resources are

clear.

As with solar and wind resources, the SNA is silent on geothermal and hydroelectric resources.

It does, however, acknowledge that water “regularly” used for extraction can be considered a

natural resource asset. Assuming the temporary diversion of water through electric power

turbines constitutes regular extraction, it is plausible that water in a hydroelectric power

reservoir could be considered an asset in the SNA. Similarly, extraction of hot water from an

underground geothermal reservoir may be sufficient for the SNA to recognize those reservoirs

as assets (though it is unlikely that extraction of heat from dry, hot bedrock would qualify).

Again, the SNA’s intentions are unclear, as neither resource is mentioned explicitly.

In contrast to the SNA, the SEEA-CF is explicit and detailed in its discussion of renewable

energy resources as assets. The SEEA-CF recognizes that energy from renewable sources is

important in many countries and increasingly seen as an alternative to fossil fuels and nuclear

power. Renewable energy resources recognized in the SEEA-CF include, in addition to the four

considered in this note, wave/tidal power and undefined “other” sources. The SEEA-CF argues

that these resources

“cannot be exhausted in a manner akin to fossil energy resources and, unlike biological

resources, they are not regenerated. Thus, in an accounting sense, there is no physical

stock of renewable sources of energy that can be used up or sold” (SEEA-CF ¶5.226).

The SEEA-CF therefore limits physical measurement of these resources to measurement of the

flows of energy produced from them; no measurement of the stock of the resources in physical

terms is proposed. Further, physical measurement of renewable energy production is limited to

the amounts actually produced by currently installed generation infrastructure. No account is

taken of the potential amounts of energy that could be produced from renewable sources if

investment and technology were to change in the future. This is consistent with the SEEA-CF’s

exclusion of sub-soil energy resources that are not currently under active development from

consideration as natural resource assets.

Though the SEEA-CF argues that the concept of a physical stock does not apply to renewable

energy resources, it does acknowledge that the resources have value unto themselves,

recognizing that a resource does not have to be measurable in physical terms in order to have a

monetary value. The SEEA-CF argues that the value of renewable energy resources should be

captured in the value of the land associated with renewable energy generation facilities:

“Opportunities to earn resource rent based on sources like wind, solar and geothermal should

be expected to be reflected in the price of land” (SEEA-CF ¶5.228).

Thus, the asset value of

The SNA was undergoing a major revision around the same time as the spectrum issue came to the fore.

The viewpoint that renewable energy resources cannot be “regenerated” is puzzling, as solar, wind, hydroelectric

and geothermal resources would all seem to be so. The first three will regenerate so long as the sun shines. The

fourth will regenerate so long as the earth’s core remains molten. In both cases, the processes of regeneration are

expected to last for billions of years.

For its part, the specialized SEEA-Energy handbook devoted to accounting for energy resources (United Nations,

2019) states that “the sun and the wind are not considered to be environmental assets” for its purposes (¶2.24) and

wind power should, according to the SEEA-CF, be captured in the value of land where windmills

are sited or where they might be one day. Similarly, the value of solar and geothermal resource

assets should be reflected in the value of the associated land, even though it is not clear in the

case of geothermal resources (particularly deep-earth geothermal) what would constitute the

associated land. In the case of hydro resources, the SEEA-CF argues it is more relevant to

consider the value in relation to the water used to generate the energy than to an area of land.

Thus, in the case of hydropower, it is the value of the water resource that would capture the

value of the hydro asset according to the SEEA-CF.

2.1.4 Exploring the assumptions underlying the SEEA-CF’s treatment of

renewable energy resources

While the SEEA-CF’s argument that renewable energy resource values will arise “due to the

scarcity of the sites used for energy generation” (SEEA-CF ¶5.310), has prima facie appeal, the

assumptions underlying it deserve further examination.

The SEEA-CF’s argument seems predicated

on the notion that if two parcels of otherwise

identical land differ only in their use (or potential use) to generate renewable energy, their

market price should reflect this difference (with the one with higher renewable energy

generation or potential being the more highly valued). Sites for renewable energy production are

not infinite in supply, especially not for geothermal and hydroelectric resources, both of which

depend on specific attributes of the earth’s crust or its hydrologic features. Thus, it would be

reasonable to expect areas of land with high geothermal or hydroelectric potential to command

relatively higher prices than those without, other things equal. Sites suitable for solar and wind

production are also in limited supply, even if the sun shines and the wind blows everywhere.

Variations in both the degree of sunshine/wind speed and the physical characteristics of sites

(angle to the sun or wind, obstructions, prohibitions, existing land uses) impact suitability – and

therefore value – for solar/wind production.

Several assumptions are implicit in the SEEA-CF’s argument around the value of renewable

energy resources in relation to land:

• markets for renewable energy production are in something close to long-term

competitive equilibrium

• property rights to land include the rights to the economic benefits flowing from any

associated renewable energy generation, and

• land associated with renewable energy production has a positive and measurable

economic value and that this value is captured, at least in principle, in the national

accounts.

Each of these assumptions is considered in turn below.

repeats the SEEA-CF position that renewable energy resource values should be captured in the value of the

associated land (¶2.50).

We say “seems predicated” because the basis for the SEEA-CF’s argument is not spelled out explicitly anywhere in

the text.

Arguably, this is not true in the case of nations, like Saudi Arabia, with vast areas of desert with largely

undifferentiated potential for solar power production and no other meaningful potential for economic use. Even in

such cases, however, the principles of valuation would apply – but the supply of suitable land would so far outstrip

demand that no opportunity cost would arise.

2.1.4.1 Renewable energy markets are close to equilibrium

For the value of land to reflect the value of any associated renewable energy assets, markets

for renewable energy must be in something close to long-term competitive equilibrium. For long-

term equilibrium to obtain, buyers and sellers of land require clear and reliable information to

account for renewable energy generation potential in their assessments of land value. The

points on which land buyers/sellers require clear and reliable information include, inter alia:

• the physical potential of the land in question for renewable energy production

• how the use of the land for renewable energy production might affect its value for

other purposes (e.g., how installation of a wind turbine on farmland might affect its

value as potential residential development site)

• the current and future demand for different forms of renewable energy

• the legal and regulatory frameworks for renewable energy, including property rights

to the economic benefits of renewable energy production, government support for

renewable energy production and capture of the associated resource rent, and

• the capital and operating costs of renewable energy production.

The extent to which clear and reliable information on these points is actually available to

renewable energy investors is difficult to determine. Moreover, this availability likely differs

considerably from country to country. In jurisdictions with relatively long histories of renewable

energy production, such as Denmark and Holland, it may be the case that there is sufficient

certainty on the above points for land transactions to reflect the value of renewable energy

production. In jurisdictions where renewable energy production is newer and/or markets are in

rapid evolution (which includes most of the world), it seems likely that uncertainty exists on

many of these points, meaning that the economic benefits of renewable energy resources will

be, at best, imperfectly factored into land prices. As an example of the uncertainty investors

face, public perception of the desirability of renewable energy infrastructure – especially wind

turbines and solar panels – is fickle. While citizens in many countries support renewable energy

(PEW Research, 2018; European Commission, 2017), they can also object when projects are

developed “in their backyards”. Opposition is driven by socioeconomic concerns, visual and

sound impacts and environmental concerns (Bidwell, 2016; Enevoldsen and Sovacool, 2016)

and has resulted in planned developments being scrapped, such as the Cape Wind project in

Nantucket Sound, Massachusetts (Smith, 2007) and the White Pines

wind project in Ontario.

Uncertainty regarding legal and regulatory environments can create considerable distortions in

markets. A rational investor would want clarity regarding the allocation of the economic benefits

of renewable energy production before making a large land purchase. Since those benefits

would be expected to play out over decades, the challenges posed for the investor by

uncertainties are multiplied. In another example, a solar energy investor in California had its

panels blocked by neighbouring redwood trees. After a lengthy lawsuit, the owner of the solar

array won out and the trees had to be removed (Borenstein, 2011). This suggests that property

rights to renewable energy resources are recognized but are still a matter of some debate, even

in California with its long history of production. Appendix 2 explores another source of

uncertainty – what governments might do in the long run regarding solar and wind property

rights.

The foregoing is not intended to suggest that renewable energy markets are chaotic and that

long-run market equilibrium is but a distant dream. In fact, as discussed in Section 4.1.3 below,

some renewable electricity markets – and electricity markets more generally – have moved in

recent years toward long-run competitive equilibrium even if they cannot all be considered to

have arrived there yet. Rather, the point is simply that enough uncertainty remains in many

renewable energy markets to question whether the SEEA-CF’s assertion that land prices will

capture renewable energy resource rents is valid. Our view is that it may be in a few instances

but not in most. This view is supported by the empirical literature, which we review briefly below

for both renewable energy production on farmland and on private residences.

Studies on the relationship between farmland prices and renewable energy demonstrated a

mixed set of results, with some studies pointing to a positive relationship, others to a negative

relationship and still others to ambiguous results. Haan and Simmler (2018), in a study for

Germany, find that German feed-in tariff’s for wind power led to capitalization of 18% of wind

turbine profits into land prices.

Lai et al. (2019), in a spatially explicit study for Taiwan covering 2013-2017, find increased value

of farmland in parts of their study area due to installation of solar panels, both for farmland used

for solar power production and for surrounding farmland not yet used for such production.

However, farmland in other parts of their study area showed decreased value. Lai et al.

speculate this is because of differences in the installed solar capacity in different areas, with

areas where full capacity has been reached showing negative effects and vice versa. Lee et al.

(2021), in another study for Taiwan covering the period 2012-2019, considered the impact of

Taiwan’s "Two-Year Solar PV Promotion Plan” on farmland prices. They found that farmland

prices fell after implementation of this policy, which, they hypothesize, is due to the Taiwanese

government's declaration that only agriculturally inefficient farmland can be used for solar farms.

Lehn and Bahrs (2018a), in a study for North Rhine-Westphalia in Germany for the year 2013,

find a statistically, though not economically, significant relationship between wind power

production and farmland value. They conclude that agricultural characteristics of farmland (such

as livestock density) and proximity of land to population centres were the main factors driving

farmland value. In a related, spatially explicity study, again for North Rhine-Westphalia for 2013,

Lehn and Bahrs also find a statistically signification relationship between farmland value and

wind power production but only for farmland in the 90th percentile of value; for farmland in lower

value quantiles, no statistically significant relationship was found.

Myrna et al. (2019), in a study for Saxony-Anhalt, Germany covering the period 2007-2016,

show a small (0.4%) increase in farmland value for a doubling in installed wind turbine capacity.

Sardaro et al. (2019) consider the impact of wind turbine production on farmers when the

turbines are owned by a separate power utility rather than by the farmers. They find that

renewable energy production has a negative value on farmland in these instances. Seifer et al.

(2020), in another study for Germany that placed considerable emphasis on information

asymmetries in farmland markets (which they characterise as “thin”), find that wind power is not

a statistically significant variable in farmland prices. They note their findings are inconsistent

with other studies (e.g., Haan and Simmler, 2018) but note these other studies referred to

periods during which feed-in tariffs large and economically attractive to famers whereas their

study includes transactions for 2014-2017 and German wind-energy tariff rates became variable

in 2017.

As for studies considering the impacts on residential property prices from solar power

production, several were found and all showed either small, positive or no net increases in

property values. Qui et al. (2017), in a study for Arizona, USA, found a gross increase in house

price of $28,000 (equivalent to 17% of median prices), which translated to a net increase of

about $5,000 after the average cost of solar rooftop installation was considered. They found no

price premium for houses with solar hot water heaters. In a study for Hawaii, USA, Wee (2016)

found similarly that the net effect of solar rooftop installations on property values after

accounting for installation costs was about $5000. Dastrup et al. (2012) in a study for California,

USA, found essentially no net price increase after accounting for installation costs. In another

study for California, USA, Hoen et al. (2012) concluded that “homes with PV systems sold for a

premium over comparable homes without PV systems” but only yielding a “near full return on

investment.” (emphasis added). In a larger, multi-state study for the USA, Hoen et al. (2017)

found that the price premium paid for houses with solar electric rooftop systems essentially

exactly offset the average costs of installing those systems.

In a study for Perth, Australia, Ma et al. (2015) found that the premium received on residential

properties seemed to cover both the costs of installation of the systems plus, in cases where the

system qualified for preferred feed-in tariff (FiT) rates from government incentive programs, the

future value of those benefits. A more recent study for Queensland, Australia (Lan et al., 2020)

found that the average net price premium (exclusive of installation cost) was 21,403 AUD for

properties with systems operating under a generous FiT rate of 44 cents/kWh but just 5,600

AUD when the FiT rate dropped to 8 cents/kWh. Both these Australian studies demonstrate the

important of renewable energy policy contexts for determining the impact of RE production on

property prices. They suggest that where FiT policies exist, property buyers will take those into

account when purchasing a property with an existing rooftop solar system. They will pay a

premium for those properties that reflects not just the cost of installation of the system, but also

the value of the future payments made under the FiT regime. In cases where FiT policies are

not in place, buyers seem in some cases to value the savings in energy costs solar systems

offer (by paying small net premiums for properties with systems) and in other cases not to value

these savings (paying premiums that cover only the cost of installation of the system itself). This

suggest that buyers do not uniformly understand the economic benefits of solar rooftop

ownership.

Overall, the farmland and residential property price studies we reviewed support the SEEA’s

contention that renewable energy production can positively influence land values where the

benefits of that production accrue to the owners of the land. However, they also support our

contention that even on private land, the size (and even direction) of this change is

unpredictable today because markets are not yet in equilibrium. These studies point to the

importance of real-world buying and selling decisions, which are influenced by both policy

contexts and market participants’ knowledge, both of which continue to evolve within and

among countries. The simulations of the impact of policy reforms on the value of renewable and

non-renewable energy wealth in Angola and South Africa in Smith et al. (2021

) further

demonstrate how policies can destroy or increase the resource rents accruing to renewable

energy resources.

Thus, our view remains that evolving market conditions mean that land values cannot yet be

taken as reliable guides to the value of renewable energy assets exploited by private

landowners, even if they should in theory.

2.1.4.2 Property rights and renewable energy resources

For the value of land to positively reflect the value of any associated renewable energy assets, it

must be the case that the land owner also owns the rights to the economic benefits associated

with the renewable energy resources. Where ownership rights accrue to a party other than the

land owner, the study by Sardaro et al. (2019) suggest land values can actually fall as a result of

renewable energy production.

The only case in which it would seem to be true that the benefits of renewable energy

production clearly accrue to land owners is that of solar and wind energy production on privately

owned land. When a farmer erects a wind turbine on his farm or a homeowner installs solar

panels on her roof, for example, the economic benefits of the energy production will be

conferred on him/her simply by the fact of his/her ownership of the associated land. As the

literature review above demonstrates, empirical evidence supports the notion that land values

can increase in such instances, though they do not uniformly do so today due to market

imperfections.

In the case of renewable energy production on public land/water, it is true that the rights to both

the land/water and the economic benefits of the energy production vest in the same unit (in this

case, the government on behalf of the public). However, as we argue more fully in Section

2.1.4.3, in those cases the land in question is likely not traded in markets and has no economic

value (at least, no value that is observed within the national accounts). There is, therefore, no

recorded value that renewable energy production on public land might influence, even if the

property rights to the resource are clear in this case.

The case of geothermal resources is the most problematic from the perspective of property

rights, as rights to these resources in most countries – as with the rights to sub-soil resources in

general – are assumed not by the owner of the land above them but by the government on

behalf of all citizens (see Appendix 3). The United States is something of an exception to this,

as sub-soil resource rights there are legally conferred by default on the owners of the

associated surface land. However, it is common in the United States for rights to the surface

and mineral estates to have been severed at some point in the past, meaning that the rights to

the two are frequently owned by different parties today. Thus, in most countries, including the

United States, it is not clear that the presence of geothermal resources on or under a given

piece of land would have any positive impact on the value of the land. Indeed, a case could be

made that the opposite is true. To the extent that geothermal energy production disrupts surface

land uses and/or raises concern for groundwater quality, the exploitation of geothermal

resources could cause land values to fall, not rise.

2.1.4.3 Land value is already captured in the national accounts

For the value of land to reflect the value of any associated renewable energy assets, it goes

without saying that the land in question must have a non-zero economic value in the absence of

renewable energy production and that its value must be captured, at least in principle, in the

national accounts. Otherwise, there is no observed land value for renewable energy assets to

impact. This is germane in two instances: renewable energy production on land with no other

practical use; and production that effectively occurs in the absence of land (for example, off-

shore wind farms and deep-well geothermal). Only in the case of renewable energy production

on privately owned land does an observable land value exist that could be plausibly influenced

by the presence of renewable energy assets. Privately owned land always has a value because

it has, by definition, at least one economic use.

The value of all private land is captured in the

national accounts and markets should be able to “price in” its value for renewable energy

Note that the argument here pertains to large-scale exploitation of geothermal resources and not to their small-

scale use for heating/cooling purposes in residential- or commercial-building ground-source geothermal systems. The

property rights to this energy likely do accrue to the land owner. Though here again it seems unlikely that the value of

the land on which the building is sited will rise simply because the owner decides to install a ground-source

geothermal system. Land below essentially any building can be used as a source of heat/cooling, so neither

differential rent nor scarcity rent could be expected to arise.

Even private land that is not actively used in any kind of production process nearly always has an economic use as

a store of value for its owner. An exception to this might be very remote areas of land held privately for the purposes

of ecological preservation. The economic value of such areas, even as a store of value, may be zero. For the present

purposes, such areas can be ignored.

production (notwithstanding the issues associated with property rights and market equilibria just

noted).

Publicly owned land is different, especially public land that is found in its “natural” state (e.g.,

forests, rivers or deserts) and has no practical economic use (for example, because of

remoteness or lack of any plausible economic use). Such land is not considered an asset in the

context of the national accounts, as no value can be observed for it. The argument that

renewable energy asset values will be captured in land values is not plausible, then, for

production that occurs on public land with no other economic use. This includes essentially all

rivers used for hydroelectricity production and any public lands used exclusively for solar, wind

or geothermal production. In the absence of renewable energy production, these areas have no

economic value. In such cases, the SEEA-CF notes “the value of the land will, in theory, be

equal to the net present value of the future income stream [from the renewable energy

production]” (SEEA-CF ¶5.229). This recognition that land used only for renewable energy

production is equal in value to the value of the renewable energy resources themselves

simply another way of stating that such land has no economic value unto itself; all its value

arises from its use to produce renewable energy.

Why the value that arises from renewable energy production on otherwise valueless land should

be attributed to the land rather than to the renewable energy assets themselves is not clear and

the SEEA-CF offers no explicit justification. It is the equivalent of attributing the value of

standing timber resources to the land on which the trees grow and not to the trees themselves,

something the SEEA-CF recommends against. The SEEA-CF recommends instead that timber

resources (in fact, all natural resources other than renewable energy) be recognized as assets

in and of themselves. Thus, the approach suggested for renewable energy assets would seem

inconsistent with the main thrust of the SEEA-CF’s arguments regarding natural resource

valuation.

This issue is of considerable practical importance, since renewable energy

production is increasingly sited on land that has no alternative economic use. Large solar

energy farms (utility-scale projects), which account for most of the growth in the world’s installed

capacity (IEA, 2017), can be tens of square kilometres in size and tend to be built in

inhospitable areas such as deserts. If the SEEA-CF renewable energy resource

recommendations were followed, the value of the renewable energy resources captured by

these large farms would be attributed to land in the national accounts.

If the main SEEA-CF

An interesting case (not discussed in the SEEA-CF) is that of speculation in land with no obvious economic value

other than renewable energy production. It may be the case that private entrepreneurs, anticipating future growth in

demand for solar or wind energy, will purchase large tracts of public land in remote areas with high potential for

renewable energy production but no other apparent economic value (such as deserts). The value of such land would,

in principle, appear in the national accounts. The extent of such speculation is unknown, but it seems unlikely to be

widespread given the large areas in question (renewable energy farms in remote areas can be tens of kilometres

squared in size). Governments may be reluctant to sell off large areas of public land to speculators unless plans for a

renewable energy farm have already been put in place and approved, in which case the land would likely be sold to

the firm developing the farm rather than to a speculator and it would be quickly brought into production rather than

lying “fallow” waiting for a farm to be proposed.

The valuation of natural resource assets by measuring the present value of the future rent they generate is the

standard approach in the SEEA-CF. It is recommended for the valuing of fossil fuel, mineral, timber and other natural

resource assets.

Section 5.8.4 of the SEEA-CF discusses valuation of timber resources. There the asset of economic significance is

considered the timber itself and not the land on which the trees grow, though it is recognized that timber rent may

include a small share that should be attributed to the land on which the timber stands. The SEEA-CF recommends

that this share be estimated and deducted for the purpose of deriving the estimate of resource rent on timber

resources. It does not, however, recommend that the entire timber rent be considered to arise from the land asset, as

it does in the case of renewable energy resources.

It is worth noting that the value of such inhospitable land for solar energy production might, in fact, be zero

according to standard economic theory, as the opportunity cost of its use for solar farms may be zero.

approach to valuing natural resources were followed instead, they would be attributed to a

separate solar energy asset (with, perhaps, a small amount attributed to the land asset, as

discussed further in Section 4.1.2).

A final concern with the SEEA-CF’s argument around renewable energy assets comes in the

case of off-shore production, such as the off-shore wind farms that are of growing importance.

As with most publicly owned territory, no economic value is attributed to off-shore areas in the

national accounts and, therefore, the value of off-shore renewable energy production cannot be

captured in the value of a country’s off-shore territory. Recognizing this, the SEEA-CF states

that “by convention, the value of income streams from [off-shore] sources are attributed to the

value of land” (SEEA-CF ¶5.231). Again, it is unclear why this should be the case. If it is the

case, the question of which land the value of off-shore energy resources should be assigned to

arises. As noted above, the approach elsewhere in the SEEA-CF suggests the opposite

treatment of off-shore resources: assigning their value to the resources themselves and

recording the resources as distinct assets in the national accounts. This is what the SEEA-CF

recommends be done for timber and other renewable resources, as well as for all non-

renewable resources.

2.2 Summary

To summarize, the SNA provides limited guidance as to the treatment of renewable energy

resources as assets. The SEEA-CF offers considerably more guidance, arguing that solar, wind,

geothermal and hydroelectric energy asset values are captured in the value of the associated

land. While having some prima facie appeal, this argument appears implausible in many

instances. The conditions in which the value of such assets could be expected to be reflected in

observed land values are limited to the production of solar and wind energy only and that only

on land that is 1) privately owned; 2) has a positive economic value for something other than

solar/wind energy production; and 3) is located where renewable energy markets could be

assumed to be in long-run equilibrium. We believe such instances are likely few in number and

that the values involved are not large in comparison to the overall value of renewable energy

resources. In many countries, solar and wind energy markets are nascent and rapidly evolving,

and, so, do not yet in our view approach the long-run equilibria in which private land values

could be reasonably expected to accurately reflect the potential for renewable energy

production, a view supported by our review of recent empirical literature. Even in countries with

long histories of renewable energy production, it is not clear that the SEEA-CF approach is

always appropriate; for example, it would not apply to the 32% (and growing) of Denmark’s wind

energy capacity that was installed off-shore in 2016 nor would it apply to the massive, utility-

scale solar farms rapidly developing in remote regions of China and elsewhere.

With respect to the most important renewable energy resource today, hydroelectric power, the

SEEA-CF’s argument does not seem appropriate at all. Hydroelectric dams and generating

stations are almost exclusively built on publicly owned rivers that have no measured value as

economic assets. There is, thus, no possibility that the value of hydroelectric resources could be

captured in any recorded value of these public waterbodies. The SEEA-CF’s recommendation

that the public waterbodies be given a value equal to calculated value of the hydroelectric

resources themselves (SEEA-CF ¶5.491) is inconsistent with its recommended approach to

valuation of other natural resource assets and it implies that the hydro resources have value

unto themselves.

Denmark, for example, saw its share of off-shore wind power capacity increase from just 2% of total capacity in

2000 to 32% in 2016 (Danish Energy Agency, 2018).

For their part, geothermal resources are publicly owned in most countries, with property rights

separate from the land found above them. There is no reason, given this, to expect the value of

the land above the resources to be positively influenced by their presence (and the opposite is

plausible). Even in the United States, where sub-soil asset property rights vest by default with

land owners, many land parcels have had their associated sub-soil resource rights severed and,

therefore, land values may not change even if geothermal resources were discovered.

Given the above, the application of the SEEA-CF’s approach to the valuation of renewable

energy resources risks missing much of the value of these increasingly important resources. For

example, none of the value of the 56% of Canada’s total electricity generating capacity that was

accounted for by hydroelectric resources in 2016

would be captured in any existing land (or

water) value on Canada’s national balance sheet accounts. The only way for this value to be

included in Canada’s national accounts would be to be explicitly calculate it as the present value

of the future stream of rent from the resource. The SEEA-CF acknowledges this, but argues that

this calculated value should be considered part of the value of water and not of the hydroelectric

resource itself, a treatment that is out of line with its approach to other natural resource assets.

Statistics Canada, Table 25-10-0022-01, Installed plants, annual generating capacity by type of electricity

generation. Available here.

3 Options considered

In this section, we first discuss define “renewable energy resources” and “renewable energy

assets” in concrete terms and then present our recommended approach to the inclusion of

renewable energy assets in the revised SNA. We then discuss two approaches to the valuation

of these assets; our recommended approach and an alternative that may be appropriate in

instances where energy markets are far from equilibrium.

3.1 The nature of renewable energy assets

It is important to be clear as to what “renewable energy resources” are and how they are related

to the topic of interest here, renewable energy assets. In keeping with the

renewable energy

specifications of the United Nations Framework Classification for Resources (UNFC; United

Nations Economic Commission for Europe, 2020), it is taken that renewable energy resources

comprise the cumulative quantities of kinetic, heat or radiative energy recoverable from in

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).

The physical unit of measure for these resources is the joule in the SI system (other units that

could be used include BTUs, tonnes of oil equivalent or watt-hours).