CHAPTER 8: NATIONAL INCOME AND ENVIRONMENTAL ACCOUNTING

Chapter Focus Questions

• How do traditional national income accounting measures
present a distorted view of well-being?
• How can traditional measures be adjusted to better reflect the
importance of natural capital and environmental quality?
• What is the potential for alternative “green” measures of national
welfare?

8.1 GREENING THE NATIONAL INCOME ACCOUNTS

Taking natural capital and environmental quality seriously affects the way we
evaluate measures of national income and well-being. Can we say that a nation with a
higher per-capita income is necessarily better off than a similar country with a lower per-
capita national income? Of course, the overall well-being of a nation is dependent upon
many factors besides income levels, including health, education levels, social cohesion,
and political participation. But most importantly for our purposes, a nation’s well-being
is also a function of natural capital levels and environmental quality.

Standard measures of gross national product (GNP) or gross domestic

product (GDP) are commonly used to measure a country's level of economic activity
and progress in development.1 (See the appendix to this chapter for an introduction to
national income accounting.) Macroeconomic analyses and international comparisons
are based on these measures, and they are widely recognized as important standards
of economic progress.

Many analysts have pointed out that these measures can give a highly
misleading impression of economic and human development. To be fair, GDP was
never intended to be an accurate measure of a nation’s well-being. But politicians and
economists often place disproportionate importance on GDP, and act as if maximizing
such measures is the primary objective of public policy. But maximizing GDP may
conflict with goals such as maximizing well-being, promoting social equity, or protecting
the environment.

While GDP accurately reflects the production of marketed goods and services, it
fails to provide a broader measure of social welfare. Some of the common critiques of
standard accounting measures include:

1
The difference between GNP and GDP concerns whether foreign earnings are included. GNP includes
the earnings of a nation’s citizens and corporations regardless of where they are located in the world.
GDP includes all earnings within a country’s borders, even the earnings of foreign citizens and
corporations. GDP is the more common measure when comparing international statistics.
 2

• Volunteer work is not accounted for. Standard measures don’t count the
benefits of volunteer work, even though such work can contribute to social well-
being as much as economic production.
• Household production is not included. While standard accounting measures
include the paid labor from such household activities as housekeeping and
gardening, these services are not counted when they are unpaid.
• No consideration is made for changes in leisure time. A nation’s GDP will
rise if, ceteris paribus2, total work hours increase. However, no accounting is
made for the loss of leisure time.
• Defensive expenditures are included. An example is expenditures on police
protection. If police expenditures are increased to counter a rise in crime levels,
the increased spending raises GDP, but no consideration is made for the
negative impacts of higher crime rates.
• The distribution of income is not considered. Two nations with the same
GDP per capita may have significantly different income distributions and,
consequently, different levels of overall well-being.
• Non-economic contributors to well-being are excluded. GDP does not
consider the health a nation’s citizens, education levels, political participation, or
other social and political factors that may significantly affect well-being levels.

In our study of environmental issues, we must add another major criticism of

standard accounting measures—they fail to account for environmental degradation and
resource depletion. This issue can be important especially in developing nations, which
depend heavily on natural resources. If a nation cuts down its forests, depletes its soil
fertility, and pollutes its water supplies, this surely makes the nation poorer in some very
real sense. But national income accounts will merely record the market value of the
timber, agricultural produce, and industrial output as positive contributions to GDP. This
may lead economic policy-makers to view the nation's development in an unrealistically
rosy light—at least until the effects of the environmental damage become apparent,
which in some cases may be decades (see Box 8.1).

If we are measuring social welfare with, so to speak, the wrong ruler, we may
obtain policy prescriptions that could actually make a nation worse off, rather than better
off. Economic growth alone does not necessarily represent true economic
development, and may even lower human well-being if it is accompanied by growing
inequity and environmental degradation. The attempt to define better measures of
development has led to new proposals to adjust or replace traditional accounting
measures in order to take into account resource and environmental factors. In this
chapter we will discuss the estimation and application of several of these alternatives.

2
Ceteris paribus, a Latin phrase, means “other things equal” and is used by economists to make clear
what assumptions are used as the basis of an analysis.
 3

Box 8.1. The Rise and Fall of GDP

“For decades, academics and gadflies have been critical of [GDP], suggesting that it is
an inaccurate and misleading gauge of prosperity. [It] has not only failed to capture the
well-being of a 21st-century society but has also skewed global political objectives
toward the single-minded pursuit of economic growth.

Environmental and sustainability indicators offer a few good examples of how big the
challenge is. A relatively easy first step, … would be to build in a “depletion charge” to
G.D.P. for the natural resources — oil, gas, timber and even fisheries — that a country
transforms into dollars. At the moment, we don’t do this; it’s as if these commodities
have no value until they are extracted and sold. A charge for resource depletion might
not affect G.D.P. in the United States all that much; the country is too big and too
thoroughly based on knowledge and technology industries for the depletion costs of
things like coal mining and oil drilling to make much of an impact. On the other hand, in
countries like Saudi Arabia and China, G.D.P. might look different (that is to say, lower)
if such a charge were subtracted from their economic outputs.

But environmental accounting gets more difficult. “We can put monetary values on
mineral stocks, fisheries and even forests, perhaps,” [Columbia University professor
Geoffrey] Heal says. “But it’s hard to put a monetary value on alteration of the climate
system, loss of species and the consequences that might come from those.” On the
other hand, Heal points out, you have to decide to measure something difficult before
you can come up with a technique for measuring it.

To Heal, making a real and rapid effort at calculating these costs and then posting the
information is imperative. According to Heal, we have no sense of how much “natural
capital” — our stocks of clean air and water and our various ecosystems — we need to
conserve to maintain our economy and our quality of life. “If you push the world’s
natural capital below a certain level,” Heal asks, “do you so radically alter the system
that it has a long-term impact on human welfare?” He doesn’t know the answer. Yet,
he adds, if we were to pass that point — and at present we have no dials to indicate
whether we have — then we couldn’t compensate for our error through technological
innovation or energy breakthroughs. Because by then it would be too late.

Source: Gertner, 2010.

 4

Efforts to develop “greener” accounting measures are relatively new. Interest in

inclusion of the environment in national accounting began in the 1970s and 1980s, with
several European countries estimating physical accounts for natural resources such as
forests, water, and land resources.3 In 1993 the United Nations published a
comprehensive handbook on environmental accounting, which was revised in 2003 and
further systematized in 2012.4 The 2003 System of Environmental and Economic
Accounts (commonly referred to as SEEA-2003) considers four basic approaches to
environmental accounting:5

1. Measuring the relationships between the environment and the economy in

both directions.6 This approach seeks to quantify the ways various economic
sectors are dependent upon natural resources, as well as the way the
environment is impacted by different economic activities. For example, one
might seek to estimate how much air pollution results when different industrial
sectors increase their production levels. These accounts combine monetary data
with information on the flow of materials, pollution, and energy in an economy. A
key motivation for this approach is to determine how closely economic activity is
linked to material inputs and pollution outputs.
2. Measuring environmental economic activities. This approach measures
expenditures on environmental protection and the impact of economic policies,
such as taxes and subsidies, to reduce environmental damages.
3. Environmental asset accounts. This approach collects data on the levels of
various types of natural capital, such as forests, minerals, and groundwater. As
we’ll discuss later in this chapter, these accounts (also called natural resource or
satellite accounts) can be kept in either physical units or monetary terms.
4. Adjusting existing accounting measures to account for natural capital
degradation. This approach seeks to monetize the damages associated with
the depletion of natural resources and environmental quality degradation, as well
as identify defensive expenditures made in response to, or in order to avoid,
environmental damages. This approach essentially takes existing national
accounting measures and makes a monetary deduction to represent
environmental damages.

Note that these approaches aren’t necessarily mutually exclusive—we could

theoretically implement all of them simultaneously. While many countries have adopted
one or more of these accounts, no country has fully implemented the SEAA-2003
provisions. We should also note that all these approaches either adjust or complement
standard accounting measures, such as GDP. In this chapter we will focus mainly on
the last two of these approaches. In addition, we will consider proposals for entirely

3
See Hecht, 2007, for a history of environmental accounting.
4
United Nations, et al., 2003; European Commission, et al., 2012.
5
Smith, 2007.
6
This approach is referred to as “physical flow accounts” or “hybrid accounts”.
 5

new national welfare measures that seek to provide a fundamentally different

perspective on measuring national welfare.

Before we delve into specific measures, it is important to note that there is no

universally-accepted approach to environmental accounting. While various measures
have been developed and implemented, there is no uniform standard for alternative
national accounting. We will consider the future of environmental accounting at the end
of the chapter.

8.2 ENVIRONMENTALLY-ADJUSTED NET DOMESTIC PRODUCT

Perhaps the most basic approach to green accounting is to start with traditional
measures and make adjustments that reflect environmental concerns. In current
national income accounting, it is commonly recognized that some of each year’s
economic production is offset by the depreciation of manufactured, or fixed, capital such
as buildings and machinery.7 In other words, while economic activity provides society
with the benefits of new goods and services, each year the value of previously-
produced assets declines, and this loss of benefits should be accounted for. Thus
national accounting methods produce estimates of net domestic product (NDP), which
starts with GDP and then deducts the annual depreciation value of existing fixed capital.

For example, in 2010 the GDP of the United States was $14.7 trillion. But the
depreciation of fixed capital in this year amounted to $1.9 trillion.8 Thus the NDP of the
United States in 2010 was $12.8 trillion.

Taking this logic a step further, we realize that each year the value of natural
capital may also depreciate as a result of resource extraction or environmental
degradation. In some cases, the value of natural capital could increase as well if
environmental quality improves. The net annual change in the value of natural capital in
a country can simply be added or subtracted from NDP to obtain what has been called
environmentally-adjusted NDP (EDP). So we would obtain EDP as:

EDP = GDP – Dm – Dn

where Dm is the depreciation of manufactured capital and Dn is the depreciation

of natural capital. This measure requires estimating natural capital depreciation in

7
Depreciation is simply a measure of the loss of capital value through wear-and-tear. For accounting
purposes, it can be calculated using a “straight-line” formula according to which, for example, a new
machine is estimated to lose 10% of its original value each year over ten-year period, or using more
complex valuation methods.
8
Estimates of fixed capital depreciation are obtained from tax records. Businesses are not taxed on the
value of their fixed capital depreciation – thus they have a strong incentive to claim this deduction.
 6

monetary terms, rather than physical units such as biomass volume or habitat area.
The methods discussed in Chapter 6 can theoretically be used to estimate such values,
but obviously estimate all types of natural capital depreciation in monetary terms is a
daunting task that would require many assumptions. Thus the estimates of EDP that
have been produced focus on only a few categories of natural capital depreciation.

One of the earliest attempts at green accounting estimated EDP for Indonesia
over a 14-year period, 1971-1984.9 This pioneering analysis deducted the value of
depreciation for three categories of natural capital: oil, forests, and soil. The values of
GDP and EDP over this time period are displayed in Figure 8.1.10

Figure 8.1: Indonesian GDP Adjusted for Resource Depreciation, 1971-1984

Source: Repetto et al., 1989.

The results present several important points that will continue to be relevant as
we proceed through this chapter:

1. Natural capital depreciation can amount to a significant portion of GDP.

According to this analysis, EDP is normally about 20% lower than GDP. In
other words, natural capital depreciation offset about 20% of total economic
9
Repetto, et al., 1989.
10
The analysis actually refers to EDP as “NDP*,” which they called “adjusted net domestic product.” But
to avoid confusion with the more common usage of the term “net domestic product” – only deducting for
fixed capital depreciation – we call their environmentally-adjusted values EDP.
 7

production. Thus GDP presents an overly-positive assessment of social

welfare.

2. Measuring the growth of GDP to illustrate changes in social welfare may

not produce accurate results. Over the time period covered in Figure 8.1,
GDP grew at an annual rate of 7.1%. However, EDP only grew at an annual
rate of 4.0%. So this case demonstrates that only looking at GDP to
determine the trend in national welfare may lead policy makers to conclude
that growth is robust. But accounting for environmental degradation shows
that much of the apparent growth was at the expense of the environment.

3. Monetization of natural capital needs to be approached carefully. In

Figure 8.1 there is a noticeable spike in EDP in 1974. Does this indicate an
appreciation of natural capital and an environmental improvement? Not
necessarily – this spike is mainly a result of a dramatic increase in world oil
prices which resulted from the 1973-1974 Arab oil embargo, rather than a
change in the actual oil reserves in Indonesia. Similarly, in some years the
total volume of timber decreased but since the market price went up the
overall value of timber resources increased. However, this masks the
physical degradation of timber resources. So if we measure the value of
natural capital at market prices, we may lose important information regarding
the actual physical stock of those resources.

A more recent attempt to measure EDP in Sweden looked at a broader set of

natural resource categories, including soil erosion, recreation values, metal ores, and
water quality.11 The results found that EDP in Sweden was about 1-2% lower than NDP
for 1993 and 1997. The author notes that while the overall adjustment may seem
relatively minor, the analysis didn’t consider all potential environmental damages, such
as climate change and loss of biodiversity. Also, looking at the effects of environmental
degradation on the overall economy fails to recognize that some sectors are particularly
affected, such as agriculture, forestry, and fisheries.

Another study estimated the value of changes in forest resources in India in

2003.12 Based on timber and firewood market prices, the results indicated that while the
overall stock of timber decreased, EDP was actually slightly higher than NDP. Again,
this illustrates the potential distortionary effect of looking only at adjustments in
monetary terms without looking in more detail at the actual physical environment.

11
Skånberg, 2001.
12
Gundimeda, et al., 2007.
 8

BOX 8-2: INCORRECT ACCOUNTING LEADS TO INCORRECT POLICIES

If economists accept conventional GDP estimates, then their policy

recommendations are likely to be wrong in the case of natural resource dependent
economies. Output estimates may be exaggerated by 20% or more and true
estimates of capital formation may turn out to be nil or negative. Factor productivity
estimates are thrown into question when neither the products nor the inputs are
measured correctly. Capital/output ratios will be incorrect if they ignore rapid liquidation
of natural capital. Sophisticated macroeconomic models based on such data will give
highly questionable results for guiding long-term development.

International trade will tend to align domestic with international prices. But
international prices are often distorted by agricultural subsidies, political and military
interventions, and the failure to internalize externalities. As a result, natural resources
are likely to be sold below full environmental cost.

The impact of natural capital depletion will be especially large in estimates of

national saving and investment. Estimates of “genuine saving” by the World Bank
indicate that many countries' net saving and capital formation may in fact be negative, a
clear indicator of unsustainability.

The export of natural capital also distorts exchange rates, and creates a bias
against non-resource-exporting sectors, including manufacturing. Methods used to
estimate exchange rate overvaluation will not be reliable when proceeds from the
unsustainable export of natural assets finance an import surplus. In this case, an
apparent stability of the domestic price level will be illusory, masking significant damage
to non-resource exporting sectors which must compete with artificially cheap imports.
In the balance of payments accounts, a trade deficit may be concealed, or appear to be
a surplus, since the proceeds of natural capital exports are recorded incorrectly in the
current account.

“Greening the national accounts is more important for economic than for
environmental policy . . .especially for those countries whose natural resources are
rapidly eroding, and the erosion is counted misleadingly in GDP as value added. Once
the accounts are greened, macroeconomic policies need to be re-examined.”

Source: Salah El Serafy, “Green Accounting and Economic Policy.” Summarized in Harris et al., 2001, pp.
33-36.
 9

8.3 ADJUSTED NET SAVING

In addition to GDP, traditional national accounting methods also estimate saving

and investment rates. These accounts provide some insight into how much a nation is
saving for its future. Starting with gross savings, by governments, businesses, and
individuals, net domestic saving is obtained after adjustments for borrowing and fixed
capital depreciation. Thus net domestic saving could be positive or negative. For
example, in 2010 the U.S. had a negative net domestic saving rate of—1.1% of national
income.

We can propose that how a country manages its natural resources and
environmental quality also provides information about whether it is saving for the future,
or causing depletion that may make future generations worse off. Just as in the
calculation of EDP, we can adjust net domestic saving to incorporate a nation’s
management of its natural resources. The World Bank has developed such a measure,
called Adjusted Net Saving (ANS).13 Unlike standard measures of national saving,
ANS

takes the broader view that natural and human capital are assets upon which the
productivity and therefore the well-being of a nation rest. Since depletion of a
non-renewable resource (or over-exploitation of a renewable one) decreases the
value of that resource stock as an asset, such activity represents a disinvestment
in future productivity and well-being.14

An ANS analysis, particularly appropriate for developing countries, may show

that what appears to be a development “success story” can conceal serious natural
capital depletion and in some cases even a negative adjusted net saving rate.

ANS is normally calculated as a percentage of national income, although it could

also be expressed in monetary units. The calculation of ANS is summarized in Figure
8.2. ANS is obtained using the following steps:15

• Start with gross national saving

• Make a deduction to account for the depreciation of fixed capital to obtain net
national saving.
• Adjust for education expenditures. Unlike standard measures, ANS considers
expenditures on education to be investments in the future of a society.16 So

13
Adjusted Net Savings is also called Genuine Savings.
14
Bolt, et al., (2002), p. 4.
15
In addition to the steps presented in the text, some calculations of ANS also include a deduction for
particulate matter emissions.
16
Gross saving already includes fixed capital education expenditures, such as spending on buildings and
buses. However, wages and salaries are not included, nor are spending on books and other educational
supplies. ANS adds in these non-fixed capital expenditures.
 10

expenditures on education are added to net national saving to reflect investment

in human capital.
• Adjust for energy resource depletion. A deduction is made for the depletion of
non-renewable fossil fuels—oil, coal, and natural gas. The deduction is
calculated as the total market value of the resource minus its extraction cost.
• Adjust for metal and mineral depletion. A deduction is made for the extraction
of non-renewable mineral resources, including copper, gold, lead, nickel,
phosphate, and several other resources. The deduction is again calculated as
the total market value of each mineral minus its extraction cost.
• Adjust for net forest depletion. Unsustainable depletion of a nation’s forest
resources is considered a disinvestment in the future. As forests are renewable
resources, it is possible that a country could actually increase its forest
resources. Thus net forest depletion is calculated as the annual value of
extraction for commercial uses such as timber and fuelwood combined with an
estimate of the net change in forest area.
• Adjust for carbon dioxide damages. Carbon dioxide emissions represent a
disinvestment in a nation’s future as they contribute to climate change damages.
A nation’s annual emissions are multiplied by an assumed damage of $20 per
ton of carbon.17

Figure 8.2: Calculation of Adjusted Net Saving

Source: World Bank, 2012

17
Some analysts would consider this a low value for carbon damages (see e.g. Ackerman and Stanton,
2011). We will consider this issue in Chapter 18.
 11

The World Bank has calculated ANS rates for most countries of the world. In
Table 8.1 we see the results for several countries. For most countries, the
environmental adjustments are relatively minor. For example, we see that the ANS
rates of France and the United States are primarily a result of their respective net
national saving and rates and education expenditures. But the environmental
adjustments can be quite significant in some countries.

The Republic of Congo, Saudi Arabia, Indonesia, and Russia offset relatively
robust net nation savings by depleting their energy resources. So based on traditional
saving measures, these countries may appear to be investing heavily in their future, but
once we account for their extraction of non-renewable fossil fuels, the ANS measure
suggests they are actually disinvesting in their future. Chile is an example of a country
that may be overly-dependent upon non-renewable minerals for its wealth. Uganda has
a significant deduction for forest depletion – about 5% of national income.

Net
Gross Forest
National Fixed Capital Education Energy Mineral Depleti Carbon
Country Saving Depreciation Expenditure Depletion Depletion on Damage ANS
Chile 24.23 -12.86 3.60 0.26 -14.32 0.00 0.31 0.08
China 53.89 -10.08 1.80 -6.74 -1.70 0.00 -1.26 35.92
Congo,
Rep. 26.68 -14.08 2.25 -71.19 0.00 0.00 -0.16 -56.50
France 18.74 -13.86 5.05 -0.03 0.00 0.00 -0.10 9.80
India 38.17 -8.49 3.17 -4.86 -1.42 -0.78 -1.16 24.64
Indonesia 22.25 -10.66 1.15 -12.60 -1.38 0.00 -0.61 -1.85
Russia 32.78 -12.39 3.54 -20.47 -1.00 0.00 -0.85 1.62
Saudi
Arabia 48.33 -12.46 7.19 -43.51 0.00 0.00 -0.62 -1.06
Uganda 12.63 -7.42 3.27 0.00 0.00 -5.06 -0.15 3.27
United
States 12.60 -13.96 4.79 -1.93 -0.11 0.00 -0.31 1.07

Table 8.1: Adjusted Net Saving Rates, Selected Countries in

Percent of GDP, 2008.
Source: World Bank, 2012

The World Bank has also tracked ANS rates over time. Figures 8.3 and 8.4
present the results for several country aggregates. We see in Figure 8.3 that ANS in
high income countries has generally been decreasing over the last couple of decades.
Meanwhile, ANS in South Asia (which includes countries such as India, Bangladesh,
and Pakistan) has shown a clear upward trend in the last decade. This reflects high
levels of investment in these countries, but does not indicate that environmental
 12

depletion has declined. ANS rates in the Middle East and North Africa have fluctuated
considerably, depending on oil extraction relative to domestic investment.

Figure 8.3: Adjusted Net Saving, 1982-2008, World Bank Country Aggregates

Figure 8.4: Adjusted Net Saving, 1982-2008, World Bank Country Aggregates
Source: World Bank, 2012
 13

Figure 8.4 shows similar variation among other country groups. ANS rates are
particularly high in East Asia (which includes countries such as China, Thailand,
Indonesia, and Vietnam). This is because of very high savings and investment rates,
but in many of these countries resource and environmental depreciation is also high
(see Box 8.2). ANS rates in Latin America have been moderate – between 5% and
10% - over the last couple of decades. Finally, ANS rates in Sub-Saharan Africa have
declined in recent years and have actually turned negative, with resource depletion
being significant in many of these countries.

Box 8.2. Environmental Accounting in China

In 2004 China’s State Environmental Protection Agency (SEPA) announced that it

would undertake a study to estimate the cost of various types of environmental damage.
The initial findings released in 2006 indicated that environmental costs equaled about
3% of China’s GDP. The report was widely criticized because it failed to include
numerous categories of environmental damage such as groundwater contamination.
Shortly afterwards, Zhu Guangyao, the deputy chief of SEPA, released a separate
report which concluded that environmental damage was closer to 10% of China’s
GDP—a value similar to what many observers were expecting.

In a 2007 report jointly produced by the World Bank and SEPA, the health and non-
health costs of air and water pollution alone were estimated to be 5.8% of China’s GDP.
(World Bank and SEPA, 2007)

The results indicate that much of China’s recent economic growth has been partially
offset by increased resource depletion and pollution. Recognizing the costs of
environmental damage, the Chinese government set targets in 2006 for such variables
as energy consumption per unit of GDP, releases of major air pollutants, and total forest
cover. China’s investment in pollution control and renewable energy is growing rapidly.
However, the Chinese government’s effort to develop green GDP measures have
abated somewhat in recent years, and some of the targets that were set in 2006 were
not met.

Further analysis of the cost of pollution and resource depletion in China can help the
government implement policies that achieve true human development.

Past policies and decisions have been made in the absence of concrete
knowledge of the environmental impacts and costs. [New], quantitative
information based on Chinese research under Chinese conditions [can] reduce
this information gap. At the same time …substantially more information is
needed in order to understand the health and non-health consequences of
pollution, particularly in the water sector. (World Bank and SEPA, 2007, p. xix)
 14

8.4 THE GENUINE PROGRESS INDICATOR

EDP and ANS adjust traditional national accounting measures to account for
natural capital depreciation and environmental damage. But just like GDP, neither of
these alternatives purport to measure social welfare. So another approach to greening
the national accounts is to think about how to create a measure of social welfare if one
were starting from scratch. Perhaps the most ambitious attempt to-date to design a
replacement to GDP is the Genuine Progress Indicator (GPI).18

One critique of GDP is that it includes all economic activity as a positive

contribution to welfare. For example, all expenditures by the U.S. government
Superfund for cleaning up toxic waste sites are contributions to GDP. The medical
costs of treating diseases caused by air or water pollution are similarly added to GDP.
If coastal homeowners or businesses whose property is damaged by an oil spill sue for
damages, the legal expenditures involved as well as the cleanup costs also contribute
to GDP. By this logic, the more pollution damage and resulting cleanup expense a
nation experiences, the better off it is. Clearly this is irrational. Thus the GPI
differentiates

…between economic activity that diminishes both natural and social capital and
activity that enhances such capital. [The GPI is] designed to measure
sustainable economic welfare rather than economic activity alone. In particular, if
GPI is stable or increasing in a given year the implication is that stocks of natural
and social capital on which all goods and services flows depend will be at least
as great for the next generation while if GPI is falling it implies that the economic
system is eroding those stocks and limiting the next generation’s prospects.19

Like the previous measures discussed in this chapter, the GPI is measured in
monetary units. The starting point of the GPI is personal consumption, based on the
rationale that it is consumption that directly contributes to current welfare.

In the United States, about 70% of GDP is comprised of personal consumption,

with the remainder being government consumption, investment, and net exports. The
GPI then adds to personal consumption several goods and services that are considered
to increase social welfare, some of which are not counted in GDP. The next step in
calculating GPI is to deduct factors which are considered to decrease social welfare.
Some of these deductions account for defensive expenditures—these are expenses
associated with cleaning up pollution or attempting to repair or compensate for other
environmental or social damage. In standard accounting, all such expenditures simply
add to GDP.

18
An earlier version of the GPI was called the Index of Sustainable Economic Welfare (ISEW).
19
Talberth, et al., (2006), p. 1-2.
 15

The various steps in calculating the GPI are: 20

• Weighing consumption by income inequality. Personal consumption is

adjusted to reflect the degree of income inequality in a society.
• Add in the value of household labor and parenting. GDP only includes paid
household and parenting work, such as housecleaning and daycare services.
The GPI estimates the market value of unpaid household labor and parenting.
• Add in the value of higher education. This component of the GPI reflects the
external benefit society receives from well-educated citizens—a positive
externality estimated to be $16,000 annually.
• Add in the value of volunteer work. GDP excludes the value of volunteer
work, even though society clearly derives benefits from these services. The
value of volunteer work hours is estimated using a market wage rate.
• Add in the service value of consumer durables. This category is meant to
capture the annual benefits consumers obtain from long-lasting goods such as
motor vehicles, appliances, and furniture.
• Add in the service value from highways and streets. The GPI excludes most
government spending, such as military expenditures, because it considers these
to be responses to various threats to living standards rather than enhancements
to consumer welfare. However, the ability to use public highways and streets is
assumed to provide consumers with direct benefits.
• Subtract the cost of crime. As crime detracts from social welfare, the GPI
counts costs associated with crime as a deduction—unlike GDP, which would
count these costs as positive additions. The cost of crime includes the costs of
prisons and defensive expenditures such as buying locks and alarms.
• Subtract the loss of leisure time. GDP may increase simply because people
work longer hours. However, the associated loss of leisure time is not
considered in GDP. Based on estimates of total working hours, the GPI
calculates the reduction of leisure time since1969.
• Subtract the cost of underemployment. Underemployed people includes
those who have become discouraged and given up looking for a job, people
working part-time who would prefer a full-time job, or people who are willing but
unable to work because of circumstances such as an inability to afford childcare.
• Subtract the cost of consumer durables. As discussed above, the GPI counts
the annual service value of consumer durables. To avoid double counting, the
annual expenditures on durable goods are subtracted out.
• Subtract the cost of commuting and auto accidents. While GDP counts the
costs of commuting as positive contributions, the GPI considers commuting costs
and lost time as deductions, as well as deaths and injuries from auto accidents.

20
These steps describe the calculation of the GPI for the United States. The GPI has been estimated for
other countries, and for some U.S. states, using similar methods and data.
 16

• Subtract the cost of household environmental defensive expenditures. The

cost of such products as air filters and water purifications systems don’t increase
welfare but simply serve to compensate for existing pollution.
• Subtract the costs of pollution (air, water, and noise). Relying on studies
using the valuation methodologies discussed in Chapter 6, the GPI estimates the
economic damages from each type of pollution.
• Subtract the value of lost wetlands, farmlands, and forests. The GPI
subtracts for losses of natural capital including reductions in ecosystem services,
lost recreation opportunities, and declining non-use values.
• Subtract the costs of depleting nonrenewable energy sources. While GDP
counts the market value of extracted nonrenewable energy sources as positive
contributions, it fails to consider that a diminishing stock of resources imposes a
cost on future generations. The GPI attempts to estimate this implied cost.
• Subtract the damages from carbon dioxide emissions and ozone depletion.
As we’ll discuss in Chapter 18, numerous economists have attempted to estimate
the damages associated with carbon emissions. The GPI multiples an estimate
of the marginal damage from a ton of CO2 by the cumulative tons emitted. Even
though production of CFCs in the U.S. has been virtually phased out as a result
of the 1987 Montreal Protocol (see Chapter 16), ozone damage continues as a
result of past emissions.
• Adjust for net capital investment and foreign borrowing. Net investment
(gross investment minus depreciation) is assumed to increase social welfare,
while net depreciation or foreign borrowing is assumed to decrease social
welfare.

As we might expect with all these adjustments, the GPI differs significantly from
GDP in magnitude and trends. The detailed results for U.S. GPI in 2004 are presented
in Table 8.2. We see that the largest positive adjustments to inequality-adjusted
personal consumption are the value of household work and parenting and the benefits
of higher education. But the additions are more than offset by the various deductions,
most importantly the deductions for nonrenewable energy depletion and carbon
emissions. Thus the GPI is significantly less than personal consumption, with the
implications that the various adjustments result in an overall reduction in social welfare.
 17

TABLE 8-2: Genuine Progress Indicator, United States, 2004

Component of GPI Value (billions of dollars)

Personal consumption 7,589
Personal consumption after inequality adjustment 6,318
Value of household work and parenting + 2,542
Value of higher education + 828
Value of volunteer work + 131
Service value of consumer durables + 744
Service value of highways and streets + 112
Costs of crime — 34
Loss of leisure time — 402
Costs of underemployment — 177
Cost of consumer durables — 1,090
Costs of commuting and auto accidents — 698
Costs of environmental defensive expenditures — 21
Costs of pollution — 178
Value of lost wetlands, farmland, and forests — 368
Costs of nonrenewable energy depletion — 1,761
Damages from carbon emissions and ozone depletion — 1,662
Adjustment for capital investment and foreign borrowing + 135
Genuine Progress Indicator 4,419

Source: Talberth et al, 2006

Comparing the relative trends in GDP and the GPI, we see in Figure 8.5 that
GDP per capita steadily increase from 1950 to 2004. While the GPI grew along with
GDP up to about the mid-1970s, since that time the GPI has remained relatively
constant. This implies that the gains in economic production have been approximately
offset by negative factors such as the loss of leisure time, pollution, and the depletion of
natural capital. Relying on the GPI, instead of GDP, would obviously present
significantly different policy recommendations, focusing more on reducing
environmental damages, preserving natural capital, and developing renewable energy
resources.
 18

Figure 8.5: Comparison of GDP and GPI per Capita, United States, 1970-2004
Source: Talberth, et al., 2006.

GPI estimates have been developed for countries other than the United States,
including Germany, Australia, China, and India. The GPI has also been applied at the
sub-national level. For example, a 2009 analysis of the Auckland region in New
Zealand showed that, unlike the case of the United States, the GPI grew at nearly the
same rate as the region’s GDP during 1990-2006 (Figure 8.6).21 However, even in this
case environmental losses grew at a faster rate than the GPI—rising 27% during this
period while the GPI rose 18%. But the positive contributions to the GPI, in particular
the growth of personal consumption, were enough to more than offset the
environmental losses. So we need to recognize that a growing GPI can still occur
despite increasing environmental damages.

This finding is further illustrated in Figure 8.7, which shows the economic, social,
and environmental components of the GPI for the U.S. state of Maryland over 1960-
2010.22 We see that while the economic contributions to the GPI rose steadily, the net
social contributions increased only slightly and the environmental costs more than
doubled.

21
McDonald, et al., 2009.
22
Posner and Costanza, 2011; http://www.green.maryland.gov/mdgpi/mdgpioverview.asp
 19

Figure 8.6: New Zealand’s Auckland Regional GPI vs. GDP, 1990-2006
Source: McDonald, et al., 2009

Figure 8.7: Components of the GPI for Maryland, 1960-2010

Source: http://www.green.maryland.gov/mdgpi/mdgpioverview.asp
 20

This demonstrates a potential problem with any index which reduces all
economic, social, and environmental factors into a single value. Looking only at the
overall index may fail to reflect important positive and negative trends that offset each
other. Thus we should always refer to disaggregated results, such as the data from
Figure 8.7, in order to achieve a more complete understanding of the changes occurring
in a society and the potential policies that may be necessary to increase social welfare.

Like EDP and ANS, the GPI requires converting various environmental factors
into a single metric—dollars. While this raises numerous methodological issues, as
discussed in Chapter 6, we may also question whether disparate environmental
resources and natural capital can be directly compared. Other approaches to
measuring national well-being have been developed that avoid the use of a monetary
metric, but consider different aspects of the quality of life rather than using a dollar
value.

8.5 The Better Life Index

While indices such as the GPI provide useful information, and have been used by
some policymakers, it currently seems unlikely that their adoption will become
widespread among nations. More attention is paid to indices and measures published
by international organizations such as the World Bank and United Nations. The most
referenced quality-of-life index is probably the United Nations’ Human Development
Index (HDI).

The HDI is calculated based on three components of well-being: life expectancy,

education, and income. A report on the HDI is produced every year, with rankings and
policy recommendations. In 2011 the countries with the highest HDI scores were, in
order: Norway, Australia, the Netherlands, the United States, and New Zealand.23 The
HDI is highly, although not perfectly, correlated with GDP. For example, of the 30
countries in 2011 with the highest HDI scores, all but one was also ranked in the top 40
by national income per capita. But there are some significant differences. For example,
Panama has about the same GDP/ capita as Namibia, and Vietnam has about the same
GDP/capita as Angola. But Panama has a much higher HDI scores than Namibia, and
Vietnam has a much higher HDI score than Angola. This is because both panama and
Vietnam life expectancy and literacy measures than Namibia and Angola. So in some
cases the HDI provides significantly more information than income alone.

A much more comprehensive attempt to assemble data on well-being in different

nations is the Better Life Initiative launched by the Organization for Economic
Cooperation and Development (OECD).24 Their 2011 report, “How’s Life?” describes

23
United Nations, 2011.
24
The OECD is a group of the world’s more developed nations, now including some developing nations
such as Mexico.
 21

the construction of the Better Life Index (BLI). 25 The report recognizes that well-being
is a complex function of numerous variables. While material living conditions are
important for well-being, so is quality of life and environmental sustainability. Further,
the distribution of well-being across a society is important. The report argues that we
need “better policies for better lives”:

Better policies need to be based on sound evidence and a broad focus: Not only
on people’s income and financial conditions, but also on their health, their
competencies, on the quality of the environment, where they live and work, their
overall life satisfaction. Not only on the total amount of the goods and services,
but also on equality and the conditions of those at the bottom of the ladder. Not
only on the conditions “here and now” but also those in other parts of the world
and those that are likely to prevail in the future. In summary, we need to focus on
well-being and progress.26

The BLI considers well-being to be a function of 11 dimensions:

1. Income, Wealth, and Inequality: The two main variables used for this
dimension are disposable household income and net financial wealth.27 The BLI
also considers the degree of inequality in income and wealth.
2. Jobs and Earnings: The three main variables comprising this dimension are
the unemployment rate, the long-term unemployment rate, and average earnings
per employee.
3. Housing Conditions: Sufficient housing is important to provide security,
privacy, and stability.
4. Health Status: The BLI includes life expectancy and a subjective evaluation of
one’s overall health status.
5. Work and Life Balance: The BLI measures the proportion of employees
working long (50 or more) hours per week, the time available for leisure and
personal care, and the employment rate for women with school-age children.
6. Education and Skills: This is measured as the percentage of the adult (25-64
years) population that has a secondary degree, and students’ cognitive skills
based on standardized testing.
7. Social Connections: This dimension is measured by people’s responses to a
standardized question asking whether they have friends or relatives they can
count on in times of need.
8. Civic Engagement and Governance: This dimension is based on data on voter
turnout and a composite index that measures citizen input into policy making.

25
OECD, 2011.
26
OECD, 2011, p. 3.
27
In addition to the main variables discussed here, most of the dimensions also consider secondary
variables. For example, the dimension of income and wealth also includes data on household
consumption and subjective evaluation of material well-being.
 22

9. Environmental Quality: The main variable used to measure environmental

quality is air pollution levels, specifically particulate matter. Secondary
environmental variables include an estimate of the degree to which diseases are
caused by environmental factors, people’s subjective satisfaction with their local
environment, and access to green spaces.
10. Personal Security: This dimension focuses on threats to one’s safety. It is
measured using homicide and assault rates.
11. Subjective Well-Being: This dimension measures people’s overall satisfaction
with their lives as well as reported negative feelings.

The results for each dimension are standardized across countries resulting in a
score from 0-10. While the BLI includes many components, it is designed to produce an
overall well-being index. But how do we assign weight to the various components?
One basic approach is to simply weigh each of the 11 dimensions above equally. But it
seems likely that some dimensions may contribute more to well-being than others. The
BLI report makes no specific recommendations for weighing the different dimensions.
An interesting feature of the BLI is that a website allows users to select their own
weights for each of the dimensions. The OECD is collecting users’ input and will use
this information to gain a better understanding of the factors that are most important for
measuring well-being.

The BLI has been measured for the 34 OECD nations, with plans to expand it to
Brazil, China, India, Indonesia, Russia, and South Africa. Even for the OECD nations,
some results have to be estimated because of a lack of consistent data. Improving the
standardization of data collection and reporting is one of the objectives of the Better Life
Initiative.

Based on equal weighing of each dimension, Figure 8.8 shows how selected
countries rank. We see that Australia, Canada, and Sweden are the top three
countries. The United States ranks 7th among OECD nations, performing well in terms
of housing and income but ranking lower in terms of work-life balance and health.
Realize that the equal weighing of each dimension reduces the importance of income
levels relative to most other national accounting approaches, such as the GPI and EDP.
As far as environmental rankings, the lowest pollution is found in Sweden and New
Zealand, and the highest pollution, among the countries evaluated, in Chile, Turkey, and
Poland.
 23

Figure 8.8: Better Life Index Values for Selected Countries

Source: OECD, 2011

BLI thus provides a comprehensive view of the many factors that influence well-
being. Income is not presented as the starting point, but as one component of many.
BLI indicators can be used to design policies that improve well-being. One of the
criteria used to choose the BLI variables is their policy relevance. Several of the
dimensions, such as education, housing, and environmental quality, can be directly
improved with effective policies, although the linkage between other dimensions (such
as subjective well-being) and policies needs further study. While the main focus of BLI
is not on environment and resource issues, its measures of environmental quality could
be expanded or given greater weight in future.

BLI calculations also indicate data collection needs in various countries. The
development of a consistent statistical agenda would improve the validity of the results
across OECD countries, and provide a basis for extending the results to other nations.
At least one nation, Bhutan, has created its own measure, Gross National Happiness,
which measures some of the same dimensions as BLI (see Box 8.3).
Box 8.3. Bhutan’s Gross National Happiness
24
Perhaps no country has advocated the need to devise alternatives to GDP as much as
the small Himalayan nation of Bhutan. In 1972, King H.M. Jigme Singye Wangchuck
introduced the concept of Gross National Happiness (GNH) to provide an alternative
development philosophy to simply maximizing economic growth. He sought to achieve
progress toward GNH by focusing on four policy objectives: equitable economic
development, environmental preservation, cultural resilience, and good governance.
(Braun, 2009)

While initially just a guiding concept, in recent years the Centre for Bhutan Studies
(CBS) has sought to operationalize GNH (CBS, 2011). The Centre has defined GNH
to encompass nine domains:

• Psychological well-being
• Standard of living
• Good governance
• Health
• Education
• Community vitality
• Cultural diversity and resilience
• Time use
• Ecological diversity and resilience

In 2010 the Centre conducted an extensive survey of over 7,000 Bhutanese households
to assess the country's GNH. Each domain was addressed by asking several
questions. For example, for the ecological domain respondents were asked questions
such as how concerned they were about air pollution, water pollution, waste disposal,
flooding, and soil erosion. Based on "sufficiency" thresholds set by CBS, the responses
determine whether each household is sufficient in each of the nine domains. The
results indicate that 41% of Bhutanese households have sufficiency in at least six
domains, and are thus considered happy. Bhutanese have the most sufficiency in
health, and then in ecology and psychological well-being. Sufficiency is greater in urban
areas, among the young, and among those with a formal education.

Bhutan, unlike most other nations, appears to not only be implementing an alternative to
GDP, but also using these results to guide future policies in a democratic manner.

Gross National Happiness seems to promote democracy in that it facilitates the

process of citizens voicing their opinions on various dimensions of their lives to
the Bhutanese government. The GNH survey and the index that the CBS
constructs from it open a channel of communication between the government and
society at large. People’s voices on an array of domains reflected in the GNH
index are the practical guiding forces for policy making in Bhutan.

Source: Braun, 2009, p. 35.

 25

8.6 Environmental Asset Accounts

An important issue to consider when evaluating any “green” national accounting

approach is how its results can be used to assess the environmental sustainability of a
society. As discussed in Chapter 7, we can define different levels of sustainability,
which we identified as “weak” and “strong” sustainability. (Recall that these terms refer
to different definitions, and do not imply that one is preferable to the other). How well do
the indicators introduced so far in this chapter reflect sustainability?

Any index which monetizes various environmental factors and combines the
results with traditional monetary aggregates, such as GDP, implicitly assumes some
degree of substitutability among natural capital and economic production. For example,
the GPI could remain constant if an increase in pollution damage is offset by an
increase in personal consumption. Thus the GPI, along with other aggregate indices
like EDP and ANS, can be considered as appropriate metrics to address weak
sustainability, but not stronger forms of sustainability.28

If we are instead interested in achieving strong sustainability, we need to

concern ourselves with the preservation of natural capital. A further distinction
emphasized by some analysts is between “strong sustainability” and “very strong
sustainability”. Strong sustainability seeks to maintain the overall level of natural capital
but allows the substitutability of different types of natural capital, at least for non-critical
resources. Very strong sustainability seeks to maintain the levels of various types of
natural capital, allowing for substitutability only within each category of natural capital.

The indicators discussed so far in this chapter are not necessarily designed to
provide information on stronger forms of sustainability. Still, a few of them do provide
some insight into strong sustainability objectives. The environmental components of the
GPI, for instance, provide information on natural capital depletion, although not the
overall level of natural capital.

An alternative approach is to maintain national accounts that track the levels of

different types of natural capital. The SEEA-2003 provides guidance on the
maintenance of environmental asset accounts (or natural resource accounts), in
both physical and monetary terms. These accounts are based on defining various
natural capital categories, such as timber resources, mineral resources, agricultural
land, and groundwater. The accounts may have different degrees aggregation. For
example, the account for mineral resources might include a separate account for each
mineral, or be even further disaggregated based on mineral quality, degree of
accessibility, or location. The units would vary for different accounts based on the
resource in question. So mineral accounts might be measured in tons, forest accounts

28 Dietz and Neumayer, 2006.

 26

in hectares of forest cover or board-feet of timber, groundwater accounts in acre-feet of

water, and so on.

The two main strengths of environmental asset accounts in physical units are:

1. They provide a detailed picture of a country’s natural capital levels and

trends over time. A particular focus can be on ensuring that levels of critical
natural capital are maintained.

2. They provide a means for assessing very strong sustainability. Since

each category of natural capital is quantified in a separate account, policy
makers can determine whether the levels of each are being maintained.

Environmental asset accounts can also be expressed in monetary units. In most

cases, this simply involves multiplying a physical unit estimate by the market price per
unit. For example, if a society has a standing timber stock of 500,000 board-feet of
lumber and the market price is $5.00 per board-foot, then the asset value of their timber
is $2.5 million. Environmental asset accounts in monetary terms offer the benefit of
comparability, both among different types of natural capital and to traditional economic
aggregates such as GDP. Unlike accounts in physical units, environmental asset
accounts in monetary units can be used to give an overall measure of sustainability
because gains and losses in different categories can be compared.

This is illustrated in Figure 8.11. For simplicity, assume there are only two
natural resource assets in a society—timber and agricultural land. In Year 1 the society
has a stock of 500,000 board-feet of timber and 6,000 hectares of agricultural land. At
the market prices indicated in Figure 8.11, the total value of the environmental assets in
the society is $8.5 million in Year 1. In the next year, the society harvests some its
timber stock but brings some additional land into agricultural production, as shown in
the figure. If we were only keeping asset accounts in physical units (i.e., in this
example, board-feet of timber and hectares of land), we would not be able to assess
whether this society has maintained its overall level of natural capital. But we see in
Figure 8.11 that the value of its natural assets has actually increased by $500,000,
indicating that the overall value of natural capital is being sustained.

Comparing different assets in monetary units has both advantages and

disadvantages. Suppose that the price of timber increased in Year 2 to $7.00 per
board-foot. Even though the stock of timber was reduced by 100,000 board-feet, the
value of the stock in year 2 would be $2.8 million (= 400,000 board feet x $7.00). Even
though the physical stock of timber was reduced, its market value increased relative to
Year 1. So if we were only looking at the monetary units, we could wrongly conclude
that the society’s stock of timber had increased due to factors such as increased
planting or conservation. This demonstrates that we need to be wary of the effect of
changing prices on the value of a society’s natural assets. This is particularly
 27

problematic for mineral and oil assets because the price of these commodities can
fluctuate considerably.

Another problem with the monetary value approach is that the estimates in
Figure 8.11 do not consider the loss of ecosystem services from harvesting timber. In
addition to the loss of timber, there may have been a loss of wildlife habitat, erosion
control, carbon storage, and other services. Ideally, assessing strong sustainability by
aggregating various asset accounts should consider non-market benefits as well as
market values. But estimating non-market values, such as ecosystem services and
non-use values, can be problematic, as discussed in Chapter 6. Thus any attempt to
assess strong sustainability based on monetary values is likely to be incomplete or
dependent upon numerous controversial assumptions.

Year 1 Year 2
Decrease in Capital Stock
Forest 
Resources

Board‐Feet of 
Standing Timber 500,000 400,000
Price per Board ‐Foot $5.00 $5.00
Timber Asset Value $2,500,000 $2,000,000
Increase in Capital Stock
Agricultural Land 
Resources

Hectares of Land 6,000 7,000

Price per Hectare $1,000 $1,000
Agricultural Asset Value $6,000,000 $7,000,000

Total Environmental  $8,500,000 $9,000,000

Asset Value

Figure 8.11: Example of Natural Resource Accounts

Several countries have started to maintain environmental asset accounts. The

United Kingdom’s Office for National Statistics provides estimates for the following three
categories of natural resources:29

29
Office for National Statistics, 2011.
 28

• Oil and gas reserves—these accounts are maintained in both physical and
monetary units.
• Forest account—this account includes the total area under forest cover, as well
as an estimate of the market value of standing timber. The report mentions other
benefits of forests, including recreation and wildlife habitat, but it makes no
attempt to quantify these benefits.
• Land account—this account maintains the total area of 19 categories of habitat
including woodlands, grasslands, marsh, open water, and built-up areas. Data
over time track changes, with some habitats increasing over time and others
decreasing.

Other countries that have prepared environmental asset accounts include

Australia, Canada, Denmark, and Norway. Perhaps the most extensive system of
environmental accounts, measured in physical units, are maintained by Sweden (see
Box 8.4).

Compared to the other indicators discussed in this chapter, environmental asset

accounts provide a means for assessing “strong” and “very strong” sustainability. If we
maintain these accounts solely in physical units, we can assess very strong
sustainability. If we convert physical units to monetary values, we can assess strong
sustainability, but only to the extent that we can accurately value different types of
natural resources and environmental services in money terms.

8.8 The Future of Alternative Indicators

As we’ve seen in this chapter, there are numerous proposals to address the
deficiencies of traditional national accounting approaches in order to account for the
environment and/or to better reflect social welfare, the ultimate goal of economic
analysis. Most of these indicators provide some guidance on sustainability objectives
as well. However, their implementation has been limited.

The current state of environmental information around the world is, by most
accounts, unacceptable. Environmental statistics are scattered among too many
organizations. They are not coherent with one another, let alone with other types
of statistics. They are incomplete and not consistent over time. This situation
greatly restricts national and international capacity to develop and monitor
progress toward environmental policy goals. (Smith, 2007, p. 598)
Box 8.4. Environmental Accounts in Sweden
29

In 2003 the Swedish government adopted sustainable development as an overall

objective of government policy. In order to monitor progress toward sustainability
objectives, an extensive database of environmental indicators is published on the
Internet by Statistics Sweden (see “Web Links” at the end of the chapter). The
government recognizes that:

no generally accepted set of indicators for sustainable development has been

worked up yet. …[But] Sweden is engaged in an ongoing effort to improve its
environmental accounting, monitoring of environmental objectives, public health,
green key ratios and index for development in the segregated districts of its
metropolitan areas. (Ministry of Sustainable Development, 2006, p. 69)

Currently, categories of environmental indicators include:

• Material flow statistics

• Chemical indicators
• Water accounts
• Waste
• Environmentally-related subsidies
• Emissions to air

Tracking trends over time have revealed some positive outcomes, along with the need
for improvement in other areas. Analysis of these trend indicate where policies could be
most effective in reducing environmental impacts.

The indicators show that although the state of several issues appears good in an
international perspective, there are trends that run counter to [sustainable
development] goals. These [include], for example, the climate change issue,
where the decrease of emissions needed for the year 2050 is not evident. More
energy efficiency and more non-fossil fuels are likely to be needed to bring about
[further progress]. It is noted that for some of the areas where emission trends
are most conspicuous, namely shipping, air traffic and goods transports, there
are economic instruments lacking. (Statistics Sweden, 2007, p. 4)
 30

While the SEEA-2003 provides guidance on various ways to approach

environmental accounting, it indicates no particular preference for one approach over
another. Instead it provides a menu of options from which a given country can choose
to implement some but not others. We remain a long way away from a universally-
accepted approach to environmental accounting that is adopted by the majority of
countries.

Recognizing the limitations of GDP and the need to develop indicators that
incorporate social and environmental factors, in 2008 French President Nicolas Sarkozy
created the Commission on the Measurement of Economic Performance and Social
Progress. The Commission was chaired by Nobel Prize-winning economist Joseph
Stiglitz and the chair advisor was another Nobel Laureate economist Amartya Sen.
Other members of the Commission included numerous prominent economists. The
goals of the Commission were:

to identify the limits of GDP as an indicator of economic performance and social

progress, to consider additional information required for the production of a more
relevant picture, to discuss how to present this information in the most appropriate
way, and to check the feasibility of measurement tools proposed by the
Commission.30

In September 2009 the Commission produced its report, nearly 300 pages in
length. The Commission noted that policies promoting economic growth, as measured
by GDP, may be unsuccessful in increasing well-being by failing to account for other
factors such as environmental degradation.

…traffic jams may increase GDP as a result of the increased use of gasoline, but
obviously not the quality of life. Moreover, if citizens are concerned about the quality
of air, and air pollution is increasing, then statistical measures which ignore air
pollution will provide an inaccurate estimate of what is happening to citizens’ well-
being. Or a tendency to measure gradual change may be inadequate to capture
risks of abrupt alterations in the environment such as climate change.31

The Commission concluded that it is necessary to shift from an emphasis on

measuring economic production to measuring well-being. It also distinguished between
current well-being and sustainability. Whether current well-being can be sustained
depends upon the levels of capital (natural, physical, human, and social) passed on to
future generations.

The Commission hoped that its report would spur additional research on the topic
of alternative indicators and encourage nations to investigate which indicators could

30
Stiglitz et al., 2009, http://www.stiglitz-sen-fitoussi.fr/en/index.htm.
31
Stiglitz et al., 2009, p. 8.
 31

provide the best information for measuring well-being and sustainability. Already
several nations have taken action.32 In the UK, the Office of National Statistics has
been directed to conduct a survey asking people what indicators they think should be
used to measure well-being. In Germany a commission on “Growth, Prosperity, and
Quality of Life” has been established. Other countries attempting to reform national
accounting include Canada, South Korea, Italy, and Australia. In the United States, the
“State of the USA Project” has been funded by the National Academy of Sciences to
develop a Key National Indicator System that:

will assemble the highest quality quantitative measures and related data, and will be
presented on the Web in a simple and straightforward way so that interested people
can assess whether progress is being made, where it is being made, by whom and
compared to what.33

Perhaps the most comprehensive attempt to-date to respond to the

Commission’s recommendations has been the Better Life Index discussed earlier. The
OECD report on the Better Life Index notes that

The work of the Commission has been critical in giving impetus to our path-finding
work on measuring progress and to a range of initiatives around the world aimed at
developing better indicators of peoples’ lives.34

The research agenda now appears focused on developing a range of indicators

that are most relevant to measuring well-being and sustainability. Some environmental
variables are rather obvious, such as measuring air pollution levels and carbon
emissions. But the measurement of a broader range of environmental impacts, such as
biodiversity and ecosystem services, requires further research. It also remains to be
seen whether each country will rely upon their own chosen set of indicators or if a
particular menu of indicators will become universally accepted. Another important
objective is to develop consistent methods for measuring different variables, such as
measuring carbon emissions and administering surveys to collect subjective data.

Improvement of data collection and international agreement on relevant indices

may lead to better measures of “green” national income accounts, and better ways to
measure progress in terms of well-being and sustainability rather than simply marketed
economic production. But measuring well-being and sustainability is only a first step
towards determining and implementing polices to promote social and environmental
progress. The chapters that follow examine the implications of environmental analysis
and policy for a range of different areas, including population, agriculture, renewable
and non-renewable resources, pollution control, and climate change, concluding by
returning to the overall issue of sustainable development.

32
Press, 2011.
33
http://www.stateoftheusa.org/about/mission/.
34
OECD, 2011, p. 3
 32

SUMMARY

Standard measures of national income such as Gross National Product (GNP)

and Gross Domestic Product (GDP) fail to capture important environmental and social
factors. This can result in misleading measurements of national well-being, potentially
ignoring important environmental problems. A variety of methods can be used to
correct the GNP/GDP measure, or to provide alternatives.

Estimates of natural capital depreciation measure the depletion of natural

resources such as oil, timber, minerals, and agricultural soils. Figures for these losses
are subtracted from the standard measures of national income and investment. The
results for many developing nations indicate a substantial impact of natural resource
depletion and environmental degradation.

For developed nations, expenditures on pollution control and clean-up, as well as

the cumulative impacts of long-lived pollutants, are significant factors. It is also possible
to estimate the value of environmental services such as water purification, nutrient
recycling, flood control, and provision of wildlife habitat. Systematic calculation of such
factors can give a measure of sustainable economic welfare, which often differs
significantly from GNP/GDP.

The application of modified national income accounting has wide-ranging policy

implications. Nations which gain a large proportion of their export earnings from
resource exports may be overestimating their economic progress. Natural resources
may be being sold below their true costs, leading to a net loss for the nation despite an
apparent trade surplus.

Social as well as environmental conditions affect calculations of national income.

Questions of human development including educational expenditures and measures of
equity are often interrelated with issues of environmental degradation. Despite the
evident importance of these factors, there is no consensus on how to include them in
national accounts. An alternative approach is to maintain satellite accounts, measuring
social and environmental indicators separately from GNP/GDP. International
institutions have moved towards more extensive reporting of such data, giving a basis
for more accurate assessments of true national well-being.
 33

KEY TERMS AND CONCEPTS

adjusted net saving (ANS) gross national product (GNP)

aggregation green accounting
better life index (BLI) human development index (HDI)
critical natural capital natural capital
defensive expenditures natural capital depreciation
environmental asset accounts net domestic product (NDP)
(natural resource accounts) net domestic savings (NDS)
environmental services net investment
environmentally-adjusted satellite accounts
net domestic product (EDP) strong sustainability
genuine progress indicator (GPI) system of environmental and
gross domestic product (GDP) economic accounts (SEEA)
gross national happiness (GNH) weak sustainability

KEY TERMS AND CONCEPTS FROM APPENDIX

constant dollars
depreciation
gross domestic product (GDP)
gross domestic product (GDP) per capita
gross investment
gross national product (GNP)
product, spending and income approaches to calculating GDP
net investment
net domestic product (NDP)
purchasing power parity (PPP)
real GDP
value-added method
 34

DISCUSSION QUESTIONS

1. What kinds of problems arise from the focus on standard GNP/GDP

measures in discussing economic policy? How do these problems differ for highly
industrialized nations like the United States and developing nations like Indonesia?

2. What are the main approaches which can be used to correct GNP/GDP for
natural resource depletion and environmental damage? What are some of the
difficulties and controversies which arise in calculating these adjustments to GDP?

3. Do you think that a revised national income measure would be an

improvement over current GNP concepts, or would it be better to keep GNP and
resource/environmental considerations separate by using natural resource
accounts?

4. What are some of the policy implications of using a revised measure which
takes into account environmental and resource depreciation? How might the use
of revised measures affect such policy areas as macroeconomic policy, trade
policy, and resource pricing policy?
 35

EXERCISE

1. Suppose you have been hired by the developing nation of Equatoria to calculate
their environmentally-adjusted net domestic product (EDP). Assume for simplicity
that only three adjustments need to be made to account for natural capital
depreciation and pollution damages: timber capital, oil capital, and carbon dioxide
damages. You have been given the following data:

Economic Data
Gross domestic product: $40 billion
Depreciation of manufactured capital: $6 billion

Timber Data
End-of-year timber stocks (board-feet): 2.0 billion
Start-of-year timber stocks (board-feet): 2.4 billion
End-of-year timber price ($/board-foot): $6
Start-of-year timber price ($/board-foot): $4

Oil Data
End-of-year oil stocks (barrels): 500 million
Start-of-year oil stocks (barrels): 550 million
End-of-year oil price ($/barrel): $60
Start-of-year oil price ($/barrel): $50

Carbon Data
CO2 emissions (tons): 75 million
Damage per ton of CO2 emissions: $20

For timber and oil, you will need to calculate the value of depreciation, or
appreciation, as the change in the total market value of the resource during the
year, where total market value is the physical quantity times the resource price.
What is the EDP for Equatoria? Would you recommend that Equatoria use EDP to
measure its progress toward sustainability objectives? Why or why not? Are there
any other recommendations you would make to policy makers in Equatoria?
 36

REFERENCES

Ackerman, Frank, and Elizabeth Stanton, 2011. “The Social Cost of Carbon,”
The Environmental Forum 28(6) (November/December): 38-41.

Bolt, Katharine, Mampite Matete, and Michael Clemens, 2002. Manual for Calculating
Adjusted Net Savings, Environment Department, World Bank.

Braun, Alejandro Adler, 2009. “Gross National Happiness in Bhutan: A Living Example of
an Alternative Approach to Progress,” Wharton International Research Experience,
September 24, 2009.

Centre for Bhutan Studies, 2011. http://www.grossnationalhappiness.com/.

Dietz, Simon, and Eric Neumayer, 2006. “Weak and Strong Sustainability in the SEEA:
Concepts and Measurement,” Ecological Economics 61(4): 617-626.

European Commission, Food and Agriculture Organization, International Monetary Fund,

Organisation for Economic Co-operation and Development, United Nations, and World
Bank, 2012. System of Environmental-Economic Accounting: Central Framework.

Gertner, Jon, 2010. “The Rise and Fall of the G.D.P,” The New York Times, May 13,
2010.

Gundimeda, Haripriya, Pavan Sukhdev, Rajiv K. Sinha, and Sanjeev Sanyal, 2007.
“Natural Resource Accounting for Indian States—Illustrating the Case of Forest
Resources,” Ecological Economics 61(4): 635-649.

Harris, Jonathan M., Timothy A Wise, Kevin P. Gallagher, and Neva R. Goodwin, eds.,
2001. A Survey of Sustainable Development: Social and Economic Dimensions,
Washington, D.C.: Island Press.

Hecht, Joy E., 2007. “National Environmental Accounting: A Practical Introduction,”

International Review of Environmental and Resource Economics 1(1): 3-66.

McDonald, Garry, Vicky Forgie, Yanjiao Zhang, Robbie Andrew, and Nicola Smith, 2009.
A Genuine Progress Indicator for the Auckland Region, Auckland Regional Council and
New Zealand Centre for Ecological Economics.

Ministry of Sustainable Development (Sweden), 2006. “Strategic Challenges: A Further

Elaboration of the Swedish Strategy for Sustainable Development,” Government
Communication 2005/06:126.
 37

Office for National Statistics, 2011. UK Environmental Accounts 2011. Statistical

Bulletin, June 29, 2011.

Organisation for Economic Cooperation and Development (OECD), 2011. “How’s Life?
Measuring Well-Being,” OECD Publishing.

Posner, Stephen M. and Robert Costanza, 2011. “A Summary of ISEW and GPI Studies
at Multiple Scales and New Estimates for Baltimore City, County, and the State of
Maryland,” Ecological Economics 70:1972-1980.

Press, Eyal, 2011. “The Sarkozy-Stiglitz Commission's Quest to Get Beyond GDP,” The
Nation, May 2, 2011.

Repetto, Robert, et al., 1989. Accounts Overdue: Natural Resource Depreciation in

Costa Rica, Washington D.C.: World Resources Institute.

Skånberg, Kristian, 2001. “Constructing a Partially Environmentally Adjusted Net

Domestic Product for Sweden 1993 and 1997,” National Institute of Economic Research,
Stockholm, Sweden.

Smith, Robert, 2007. “Development of the SEEA 2003 and its Implementation,”
Ecological Economics 61(4): 592-599.

Statistics Sweden, 2007. “Sustainable Development Indicators Based on Environmental

Accounts.”

Stiglitz, Joseph E., Amartya Sen, and Jean-Paul Fitoussi, 2009. Report by the
Commission on the Measurement of Economic Performance and Social Progress.

Talberth, John, Clifford Cobb, and Noah Slattery, 2007. The Genuine Progress Indicator
2006: A Tool for Sustainable Development, Redefining Progress.

United Nations, 2011. Human Development Report 2011, Sustainability and Equity: A
Better Future for All, United Nations Development Programme, New York, NY.

United Nations, European Commission, International Monetary Fund, OECD, and World
Bank, 2003. Integrated Environmental and Economic Accounting 2003.

World Bank and State Environmental Protection Agency, P.R. China, 2007. “Cost of
Pollution in China,” Rural Development, Natural Resources and Environment
Management Unit, East Asia and Pacific Region, The World Bank, Washington, D.C.
 38

WEBSITES

1. http://www.beyond-gdp.eu/index.html The website for “Beyond GDP,” an

initiative to develop national indicators that incorporate environmental and social
concerns. The project is sponsored by the European Union, the Club of Rome,
the WWF, and the OECD.
2. http://go.worldbank.org/3AWKN2ZOY0 The World Bank’s Adjusted Net
Saving website, which includes detailed data at the country level.
3. http://www.green.maryland.gov/mdgpi/index.asp The website for the state of
Maryland’s calculation of its Genuine Progress Indicator.
4. http://www.oecdbetterlifeindex.org The website for the OECD’s Better Life
Index. Note that you can adjust the weights applied to each dimension to create
your own version of the BLI.
5. http://www.mir.scb.se/Eng_Default.htm The website for environmental
accounts in Sweden.
 39

CHAPTER 8 APPENDIX:
BASIC NATIONAL INCOME ACCOUNTING

In this chapter we have discussed several modifications and alternatives to

traditional national income accounting. Standard accounting measures, such as gross
national product (GNP) and gross domestic product (GDP), are widely-accepted
estimates of the health of a national economy. However, there are numerous technical
and conceptual limitations to these measures. To understand the impetus for adjusting
or replacing these measures, some background knowledge of how they are calculated
and interpreted is useful. If you have not taken an introductory macroeconomics course
or need to refresh your knowledge, this appendix will help you work through the
concepts presented in the chapter.

National income accounting was first developed in the United States in the 1930s
to provide policymakers with information on the overall level of economic activity in the
country. We should realize that national income accounting was not designed to
estimate the welfare of society—only the aggregate level of economic production. Also,
at the time the accounts were being designed environmental degradation was not an
important issue.

For many years, the official measure of national economic activity in the United
States was gross national product. GNP is defined as the final market value of all new
goods and services produced by the citizens of a country over a period of time (typically
one year). GNP includes goods and services produced by U.S. citizens and
corporations in foreign countries but not goods and services produced within U.S.
borders by foreign citizens and corporations.

In the early 1990s the U.S. switched to gross domestic product as its official
measure to conform with international standards developed by the United Nations.
GDP measures the value of goods and services produced within the national
boundaries of a country regardless of the producer’s nationality. Thus GDP would
exclude production by U.S. citizens and corporations in foreign countries. In practice
there is little quantitative difference between GNP and GDP. In 2011 the values differed
only by about 1% in the U.S.

It is important to note that GNP and GDP measure only the final value of goods
and services. Intermediate values are excluded to avoid double-counting. For
example, consider some of the steps involved in producing this textbook. First a lumber
company harvested wood and sold the wood to a paper mill. The paper mill then
produced paper and sold it to a printing company. The printing company then printed
the text under contract with the publisher. The publisher then sold the book to a retail
store for final sale to you. If we add up the prices paid by the paper mill, printing
company, publisher, retail store, and you, we’ll end up with a value much higher than
the price you paid for the book. The more intermediate production steps taken to
produce an item, the higher the sum of all the prices paid. So all the intermediate steps
are not counted and only the final price you paid is included in GNP.
 40

Since it may be difficult in practice to distinguish intermediate from final goods,

the accounting method generally used to compute GNP/GDP is the value-added
method, in which the extra value added at each step of the production process is
counted. In the textbook example, the value added for the paper mill is the value of its
output minus the cost of inputs purchased from the lumber company. The sum of the
values added at all stages of production is equal to the value of the final good.

GNP and GDP only count the production of new goods. If you purchased this
book second-hand from a store or other student, then it would not be included in the
national account. The sale of used products does not contribute to current economic
production.

Calculating the Value of Gross Domestic Product

As you might imagine, calculating the total value of all goods and services
produced in a national economy is not a simple task. Economists use a variety of data
sources to estimate aggregate production including data from tax returns, surveys of
businesses and households, and government records. There are three ways to obtain
an estimate of GDP: the product approach, the spending approach, and the income
approach. The product approach simply adds up the dollar value of all final goods and
services produced in the economy. The spending approach adds up the expenditures
of consumers, businesses, governments, and institutions for final goods and services.
The income approach adds up the earnings of everyone in the economy, including
wages, profits, investment income, and rental income.

In a simplified closed economy with no depreciation or inventories and where all

business earnings are paid out as income, all three approaches should produce the
same value.35 In other words, if an economy produces $7 billion worth of goods and
services we can conclude that $7 billion was spent to purchase these goods and
services and $7 billion in income was earned to make these purchases. Of course, the
real economy is more complex and the different approaches may produce different
values. For example, not everything produced in a year may be sold during that year.
Economists have devised adjustment methods so that the different approach should
produce the same values, but even so there remain statistical discrepancies arising
from the complexity of the data or missing information.

The national accounts divide the economy into four sectors: businesses,
households and institutions, governments, and the foreign sector. Using the product
approach, we would add up the final goods and services produced by all businesses,
households and institutions, and governments. As you might guess, using the product
approach we discover that the business sector produces most of the marketed final
goods and service in the economy (75% in 2010 for the U.S. economy). Household
production, as defined in the national accounts, includes rental values and paid work
within households such as maid services, child care, and gardeners. However, similar

35
By a closed economy we mean one without any imports or exports.
 41

household work that is not marketed, such as people cooking their own meals or
cleaning their own homes, is not included in GDP. This is one common criticism of
GDP. For example, essentially the same service is produced whether a household
cleans their own home or hires someone to do it, but only in the latter case is the value
of the service included in GDP.

Using the spending approach, we need to consider the foreign sector in

calculating GDP. We add demands by foreigners for goods produced in the U.S.
(exports) and subtract demand by U.S. residents and institutions for foreign goods
(imports). Spending by businesses on machinery, buildings, and other goods is called
gross investment. Governments also purchase goods and services and make
investments. Using the spending approach, we find that households and institutions
purchase most of the final goods and services produced in the economy (about 71%).
The spending approach can be summarized by the equation:

Y = C + I + G + (X-M)

where Y represents GDP, C is consumer expenditures, I is business investment,

G is government spending on goods and services, and (X-M) is net exports (exports
minus imports).

Calculating GDP using the income approach, we only include income received
for production that occurs within the national boundaries. The income approach
includes corporate profits and rental income as well as wages and salaries. Most of the
U.S. national income, about 55% in 2010, is paid to workers as wages and salaries.

Adjusting for Depreciation, Population Growth, and Inflation

One reason why GDP is not the best measure of national income is that a portion
of investment in capital equipment such as factories and machinery simply replaces
worn-out capital. Since capital that wears out or becomes obsolete decreases national
wealth, the depreciation of this capital should be counted as a deduction from GDP.
Gross investment minus depreciation is called net investment. If we deduct capital
depreciation from GDP we get a measure called net domestic product (NDP). The
depreciation of fixed capital amounts to about 10-15% of GDP in the United States.

Of course, politicians and economists hope that the economy expands over time
and GDP increases. But an increase in GDP does not necessarily indicate greater
wealth for the citizens of a nation. GDP could increase simply because the nation has a
higher population. We can account for population growth (or decline) in national
accounting by calculating GDP per capita, equal to GDP divided by population. Data
on GDP per capita also allows us to compare economic production across different
nations. For example, the GDP of the United States is much greater than the GDP of
Sweden, but when we adjust for population size we find that GDP per capita is higher in
Sweden than in the U.S.
 42

The other factor we need to control for when comparing GDP values across time
is inflation. Remember that GDP is based on market prices and GDP could grow simply
because market prices have risen. So when comparing GDP data from different years
we need to use constant dollars. For example, suppose we know that the general
level of prices in 2012 was twice as high as it was in 1990. So if we wanted to compare
GDP for these two years we could compare them using 2012 dollars by doubling the
GDP from 1990. Or we could compare them using 1990 dollars by dividing the GDP for
2012 by half. The first method gives us real GDP in 2012 dollars, while the second
gives us real GDP in 1990 dollars.

The GDP for the U.S. has grown tremendously over recent decades. As seen in
Table 1, GDP has increased by a factor of 51 between 1950 and 2011 if we don’t
consider any adjustments. Making an adjustment for population, we find that economic
production per person has increased by about a factor of 25. But most of this increase
is due to inflation. When we adjust for differences in price level by calculating real GDP
per capita in 2011 dollars, we discover that economic production per person has
actually increased by a factor of 3.2. This still suggests a large increase in the standard
of living for the average American, but a much less significant increase than would be
implied looking at the unadjusted aggregate GDP data.

Table 1. Historical Gross Domestic Product Data, United States

Unadjusted Unadjusted GDP

U.S. GDP per Capita GDP per capita in
Year (billion $) (dollars) 2011 dollars
1950 294 1,929 14,920
1960 526 2,914 17,747
1970 1,038 5,064 23,586
1980 2,788 12,270 29,105
1990 5,801 23,252 36,476
2000 9,952 36,170 46,214
2011 15,094 48,409 48,409
Source: U.S. Bureau of Economic Analysis and U.S. Census Bureau websites

Comparing GDP for Different Countries

A final adjustment that is made when comparing GDP data across nations is to
adjust for purchasing power parity (PPP). Even if we use currency exchange rates to
put all nations’ GDP per capita in U.S. dollars, we should still adjust for differences in
what a dollar can purchase in different countries. For example, a U.S. dollar converted
into Chinese currency will buy a lot more in China than it will in the United States. As
mentioned above, Sweden has a higher GDP per capita than the U.S., but when we
adjust for purchasing power parity, GDP per capita is higher in the U.S. than in Sweden
because of the relatively high prices in Sweden.

Of course GDP per capita varies widely by country. In 2011 the World Bank
 43

classified 36 countries as “low-income”, with a per capita gross national income (GNI, a
measure similar to GDP) of less than $1,025 annually. A total of about 800 million
people in 2011 lived in countries classified as low income, mainly in African countries.
There were 108 middle-income countries – those with a GNI per capita between $1,025
and $12,475. These countries include the majority of the world’s population, about 5
billion people, in countries such as China, India, Brazil, Mexico, Russia, and Indonesia.
Finally, there were 70 high-income countries, with a per capita GNI above $12,475.
These countries, including the U.S., Japan, Australia, and the nations of Western
Europe, had a total population of about a billion people in 2011.

National income accounting data illustrate the varying economic conditions of

people in different nations. We can use the data to compare rates of economic
development and to determine income inequality between countries. But we need to be
careful about interpreting national accounting data. GDP measures only the aggregate
level of economic production; it does not measure social welfare. If GDP per capita
goes up only because people are working longer hours, we can’t conclude that they are
happier. Also, GDP per capita could go up only because the wealthy members of
society are becoming wealthier. GDP data tell us nothing about the level of economic
inequality in a country. This and other known problems with GDP make it important to
be aware of the limits of GDP as a measure of well-being—even before we consider the
environmental and resource issues discussed in this chapter.