Ecological Economics-Principles and Applications-3

72 • The Containing and Sustaining Ecosystem: The Whole
resource as stock-flow or fund service is a function of its use. A bicycle for

sale in a bike store is a stock-flow, while the bicycle you ride every day is
a fund-service.
Box 4-1 Stock-flow and Fund-service Resources

In the academic literature, there are many distinct definitions for stocks,
flows, funds, and services. To make it clear, we are discussing the spe-
cific definitions given here. Future references will be to stock-flow and
fund-service resources.
Stock-flow resources
• Are materially transformed into what they produce (material cause).
• Can be used at virtually any rate desired (subject to the availability of
fund-service resources needed for their transformation), and their
productivity is measured by the number of physical units of the prod-
uct into which they are transformed.
• Can be stockpiled.
• Are used up, not worn out.
Fund-service resources
• Are not materially transformed into what they produce (efficient
cause).
• Can be used only at a given rate, and their productivity is measured
as output per unit of time.
• Cannot be stockpiled.
• Are worn out, not used up.
The stock-flow and fund-service concepts are important when analyz-

ing human production, and probably more so when focusing on the
goods and services provided by nature. Note that “material cause” is al-
ways stock-flow in nature, and “efficient cause” is always fund-service.
THINK ABOUT IT!
Think about a specific ecosystem—or better yet, go visit one, and take
along a field notebook. Make a list of three stock-flow resources pro-
vided by (or found in) that ecosystem and three fund-service re-
sources. (Note that you will need to be very specific about the use of
each resource. For example, drinking water is a stock-flow, while water
for swimming is a fund-service.) Check off the attributes of stock-flow
and fund-service for each (see Box 4.1). See Chapter 2 in the Work-
book for Problem-Based Learning that accompanies this text for more
on stocks, funds, excludability and rivalness.
Chapter 4 The Nature of Resources and the Resources of Nature • 73
I Excludability and Rivalness

Excludability and rivalness are also crucial concepts for economic analy-
sis, and rivalness is in fact related to the stock-flow, fund-service distinc-
tion. Although conventional economists first introduced these concepts,
they rarely receive the attention they deserve. We believe they are im-
portant enough to be described in some detail both here and in Chap-
ter 10.
Excludability is a legal principle that when enforced allows an owner
to prevent others from using his or her asset. An excludable resource is
one whose ownership allows the owner to use it while simultaneously
denying others the privilege. For example, in modern society, when I own
a bicycle, I can prohibit you from using it. In the absence of social insti-
tutions enforcing ownership, nothing is excludable. However, the charac-
teristics of some goods and services are such that it is impossible or highly
impractical to make them excludable. While someone could conceivably
own a streetlight on a public street, when that streetlight is turned on,
there is no practical way to deny other people on the street the right to use
its light. There is no conceivable way that an individual can own climate
stability, or atmospheric gas regulation, or protection from UV radiation,
since there is no feasible institution or technology that could allow one
person to deny all others access. When no institution or technology exists
that makes a good or service excludable, it is known as a nonexcludable
resource.
Rivalness is an inherent characteristic of certain resources whereby
consumption or use by one person reduces the amount available for
everyone else. A rival resource is one whose use by one person precludes
its use by another person. A pizza (a stock-flow resource) is clearly rival,
because if I eat it, it is no longer available for you to eat. A bicycle (a fund-
service resource that provides the service of transportation) is also rival,
because if I am using it, you cannot. Although you can use it after I am
done, the bicycle has worn out a bit from my use and is not the same as
it was. A nonrival resource is one whose use by one person does not af-
fect its use by another. If I use the light of a streetlight when riding my
bike at night, it does not decrease the amount of light available for you to
use. Similarly, if I use the ozone layer to protect me from skin cancer, there
is just as much left for you to use for the same purpose. It is possible to
deplete the ozone layer (through the emission of chlorofluorocarbons, for
example), but depletion does not occur through use. Nonrival resources
are not scarce in any conventional sense. They do not need to be allocated
or distributed, and their use does not directly affect scale. Rivalness is a
physical characteristic of a good or service and is not affected by human
institutions. As we will discuss at length in Chapter 10, however, institutions
can make nonrival resources such as information excludable, artificially

enclosing them in the world of scarcity and allocation.
Note that all stock-flow resources are rival, and all nonrival goods are
fund-service. However, some fund-service goods are rival. For example,
my bicycle is a fund that provides the service of transportation, but it is
rival; the ozone layer is a fund that provides the service of screening UV
rays, but it is nonrival.
As you will see when we turn to allocative mechanisms in subsequent
chapters, the concepts of rivalness and excludability are very important.
THINK ABOUT IT!

For the list of resources you made earlier, answer the following
questions:
Is the resource rival or nonrival? In general, can you think of any stock-
flow resources that are nonrival? Can you think of any fund-service re-
sources provided by nature that are rival?
Is the resource excludable or nonexcludable? (Note that excludability

may differ depending on the specific value in question.) If it is non-
excludable, can you think of an institution or technology that could
make it excludable? Do you think it should be made excludable? Why
or why not?
Is the resource a market good or a nonmarket good?
In general, can you think of any stock-flow resources that cannot be

made excludable? Can you think of fund-service resources provided by
nature that can be made excludable?
Exercise 2.2 of the Workbook for Problem-Based Learning accompany-

ing this text expands on this and previous questions we explore in
"Think About It!"
I Goods and Services Provided by the

Sustaining System
To make this discussion of entropy, fund-services, stock-flows, exclud-
ability, and rivalness more concrete, and to really understand the implica-
tions for economic theory and policy, we must see how these concepts
apply to the specific scarce resources available to our economy: the goods
and services provided by nature. We undertake this task in the next two
chapters and conclude this one by simply introducing the scarce re-
sources.
For our purposes, we will present eight types of goods and services
Chapter 4 The Nature of Resources and the Resources of Nature • 75
provided by nature, divided for convenience into nonliving and living re-
sources. Clearly this is an enormous abstraction from the number and
complexity of resources our Earth actually does supply, but these cate-
gories illustrate why the specific characteristics of goods and services we
have described are of fundamental importance to economic policy.
1. Fossil fuels. For practical purposes, fossil fuels are a nonrenewable
source of low-entropy energy. They are also very important as ma-
terial building blocks.
2. Minerals. The Earth provides fixed stocks of the basic elements in
varying combinations and degrees of purity, which we will refer to
hereafter simply as minerals. This is the raw material on which all
economic activity and life itself ultimately depend. Rocks in which
specific minerals are found in relatively pure form we refer to as
ores. Ores in which minerals are highly concentrated are a nonre-
newable source of low-entropy matter. We will refer to mineral re-
sources and fossil fuels together as nonrenewable resources and
the first five goods and services in this list as abiotic resources (see
Chapter 5).
3. Water. The Earth provides a fixed stock of water, of which fresh-
water is only a miniscule fraction. All life on Earth depends on
water, and human life depends on freshwater.
4. Land. The Earth provides a physical structure to support us that is
capable of capturing the solar radiation and rain that falls upon it.
Land as a physical structure, a substrate, or a site has economic
properties unrelated to the productivity of its soil, and is thus dis-
tinct from land as a source of nutrients and minerals. To capture this
distinction, we will refer to land as a physical structure and location
as Ricardian land.18 The quantity and quality of soil available on a
given piece of Ricardian land will be grouped with minerals, dis-
cussed below.
5. Solar energy. The sustaining system provides solar energy, the ulti-
mate source of low entropy upon which the entire system depends.
6. Renewable resources. Life is able to harness solar energy to organize
water and basic elements into more useful structures (from the
human perspective) that we can use as raw materials in the eco-
nomic process. Only photosynthesizing organisms are capable of
achieving this directly, and virtually all other organisms, including
humans, depend on these primary producers. These biological re-
sources are traditionally referred to as renewable resources, but
they are renewable only if extracted more slowly than the rate at
18Ibid., p. 232.
which they reproduce. Clearly, species can be exploited to extinc-

tion, so, as we shall see, biological resources are exhaustible in a way
that mineral resources are not.
7. Ecosystem services. Living species interact to create complex ecosys-
tems, and these ecosystems generate ecosystem functions. When
functions are of use to humans, we refer to them as ecosystem
services. Many of these ecosystem services are essential to our
survival.
8. Waste absorption. Ecosystems process waste, render it harmless to
humans, and, in most cases, again make it available to renewable re-
source stocks as a raw material input. This is really a specific type
of ecosystem service but one whose economic characteristics make
it worth classifying on its own. We refer to these last three goods
and services as biotic resources (see Chapter 6).
We refer to all the structures and systems that provide these goods and
services as natural capital. In the following chapters, we will examine
these resources in the light of entropy, fund-services, stock-flows, exclud-
ability, and rivalness.
BIG IDEAS to remember

I Laws of Thermodynamics I Fund-service resource
–Conservation of matter- I Excludable and
energy nonexcludable resources
–The law of increasing I Rival and nonrival resources
entropy I Eightfold classification of
I Stock-flow resource resources
CHAPTER
5
Abiotic Resources
I n this chapter, we closely examine the five abiotic resources intro-
duced in Chapter 4: fossil fuels, minerals, water, land, and solar en-
ergy. Our goal is to explain how the laws of thermodynamics, the
distinction between stock-flow and fund-service resources, and the con-
cepts of excludability and rivalness relate to these resources, in order to
better understand the role they play in the ecological-economic system.
We will also assess the extent to which substitutes are available and the
degree of uncertainty associated with each resource. As we will see,
however, abiotic resources are fundamentally different from each other,
and it is their even greater dissimilarity from biotic resources that binds
them together more than their similarity to each other. Perhaps the most
important distinction is that biotic resources are simultaneously stock-
flow and fund-service resources that are self-renewing, but human ac-
tivities can affect their capacity to renew. Abiotic resources are either
nonrenewable (fossil fuels) or virtually indestructible (everything else).
The differences between abiotic resources probably deserve more em-
phasis than their similarities, and we’ll start with a brief summary. Fossil
fuels and mineral resources are frequently grouped together under the
classification of nonrenewables. The laws of thermodynamics, however,
force us to pay attention to an important difference: The energy in fossil
fuels cannot be recycled, while mineral resources can be, at least partially.
Water is one of the most difficult resources to categorize, precisely because
it has so many different forms and uses. Fossil aquifers (those that are not
being recharged) are in some ways similar to mineral resources—once
used, the water does not return to the ground, and while it cannot be de-
stroyed, it can become less useful when polluted by chemicals, nutrients,
or salt. Rivers, lakes, and streams, in contrast, share similarities with bi-
otic resources: They are renewable through the hydrological cycle, driven
77
by solar energy, and they can exhibit stock-flow and fund-service proper-
ties simultaneously. However, human activity cannot affect the total stock
of water to any meaningful extent, while we can and do irreversibly de-
stroy biotic resources. Similarly, land as a physical substrate, a location
(hereafter referred to as Ricardian land), cannot be produced or destroyed
in significant amounts by human activity (with the exception of sea level
rise induced by anthropogenic climate change), and solar energy flows are
not meaningfully affected by humans at all, although we can affect the
amount of solar energy that moves in and out of the atmosphere. We now
examine these resources in more detail.
Fossil Fuels
Perhaps the simplest resource to analyze is fossil fuels, or hydrocarbons,
upon which our economy so dramatically depends. It would take an esti-
mated 25,000 hours of human labor to generate the energy found in a
barrel of oil. The fact that the fossil fuel economy and market economy
emerged simultaneously in England over the course of the eighteenth cen-
tury is almost certainly no coincidence. The magic of fossil fuels may be
more important than the magic of the market in generating today's living
standards.1 In 1995, crude oil supplied about 35% of marketed energy in-
puts into the global economy, followed by coal at 27% and natural gas at
23%. In all, 86% of the energy in the economy comes from fossil hydro-
carbons.2 In geological terms and as far as humans are concerned, fossil
fuels are a fixed stock. For a variety of reasons, however, it is extremely
difficult to say precisely how large that stock is.
For practical purposes, we are concerned only with recoverable sup-
plies. But what does recoverable mean? Clearly, hydrocarbons are found in
deposits of varying quality, depth, and accessibility, and there are different
costs associated with the extraction of different deposits. In economic
terms, we can define recoverable supplies as those for which total extrac-
tion costs are less than the sales revenues. However, fossil fuel prices fluc-
tuate wildly, and recoverable supplies defined in this way show similarly
chaotic variation through time. We could also define recoverable supplies
in entropic terms, in which case a hydrocarbon is recoverable if there is a
net energy gain from extraction; that is, it takes less than a barrel of oil to
recover a barrel of oil. This measure must include all the energy costs, in-
1M. Savinar, “Life after the Oil Crash.” http://www.lifeaftertheoilcrash.net/Research.html. This
Web site provides full references for the calculations. D. Pimentel and M. Pimentel. Food, Energy,
and Society. 3d edition. (Boca Raton, FL: CRC Press, 2007). Account for the entropic energy loss
of converting gasoline to work in an internal combustion engine, in which case it would take only
5,000 hours of labor.
2Energy Information Administration, Washington, DC. International Energy Outlook, 2009.
Chapter 5 Abiotic Resources • 79
cluding those of exploration, machinery, transportation, decommission-

ing, and so on. While technological change can reduce these, there is a
certain irreducible limit to the energy costs of extracting fossil fuels. It
takes 9.8 joules of energy to lift 1 kilogram 1 meter, and no amount of
technology can change that basic fact.
As we deplete the most accessible hydrocarbon supplies first, over time
it will take more and more energy to recover remaining supplies. In other
words, the energy return on investment (EROI), which is “the ratio of gross
fuel extracted to economic energy required directly and indirectly to de-
liver the fuel to society in a useful form,” declines over time.3 In entropic
terms, the energy cost of oil and natural gas extraction in the United States
increased by 40% from 1970 to the 1990s.4 During the 1950s in the U.S.,
every barrel of oil invested in exploration led to the discovery of about 50
more. By 1999, the ratio was about one to five. A sustainable society prob-
ably needs an EROI of at least three to one.5 Still, under either the eco-
nomic or the entropic definition of recoverable, estimates of recoverable
reserves change constantly. Largely this is the result of new discoveries,
but it also results from dramatically different methods for calculating
“proven” supplies between different companies and different countries,
with frequent changes often based on political or economic motives.6 Pe-
troleum geologists can assign reasonable probabilities to different esti-
mates of total stocks, however.
Box 5-1 Estimating Oil Stocks

Every year, the world consumes in the neighborhood of 25 billion barrels
of oil (Gbo). Yet at the end of most years, reported reserves of oil are
greater than they were at the start, and there is a fairly wide range of es-
timates as to what those reserves actually are. The increase is possible
as long as new oil discoveries are greater than oil consumed, but that
has not occurred in decades. For example, in 1997 the world used about
23 Gbo and discovered 7 Gbo, yet estimated reserves increased by 11
Gbo. How do we explain this anomaly?
When geologists estimate the quantity of oil in any given field, they
assign a probability to the estimate. For example, in the late 1990s
3C. Cleveland, R. Costanza, C. Hall, and R. Kaufmann, Energy and the US Economy: A Bio-
physical Perspective, Science 225:297 (1984).

4C. Cleveland and D. Stern, “Natural Resource Scarcity Indicators: An Ecological Economic
Synthesis.” In C. Cleveland, D. Stern, and R. Costanza, eds., The Economics of Nature and the Na-
ture of Economics. Cheltenam, England: Edward Elgar, 2001.
5C. A. S. Hall, S. Balogh and D. J. R. Murphy, What Is the Minimum EROI That a Sustainable
Society Must Have? Energies 2:25–47(2009).

6C. J. Campbell and J. H. Laherrère, The End of Cheap Oil, Scientific American, March 1998.
geologists estimated that the Oseberg field in Norway would supply 700
million barrels of oil with 90% certainty (known as probability 90, or
P90) and 2.5 billion with 10% certainty (known as P10). Different corpora-
tions and countries generally use some number within the P10–P90
range when stating their reserves, and they are often purposefully vague
about what number they use. Higher reported reserves can increase
stock prices, provide greater access to credit, and for OPEC countries, in-
crease their quotas. As oil fields are exploited, geologists can use the in-
formation acquired to make better estimates about how much they
contain. Based on this information and other factors (e.g., moving from
P90 to P50 estimates), countries frequently revise their reserve esti-
mates from existing fields, often upward. In the absence of major new
discoveries or technological breakthroughs in the late 1980s, six OPEC
countries alone revised their estimates upward by 287 Gbo, 40% more
than all the oil ever discovered in the U.S.!
When calculating global oil reserves, it makes the most sense to sum
the P50 estimates across countries, but even this is no easy task. In ad-
dition, revised estimates from existing reserves are not new discoveries
and should not be counted as such.a
aC. J. Campbell and J. H. Laherrère, The End of Cheap Oil, Scientific American, March
1998.
THINK ABOUT IT!

Economists argue that price reflects scarcity. Do you think the price of
oil is a good indicator of how much oil is left in the ground? Why or
why not?
Regardless of what the stocks of fossil fuels are, however, they are
stocks that can be extracted as flows, and the rate of flow is determined
largely by human efforts. If we had adequate infrastructure, we could the-
oretically extract all entropically recoverable fossil energy stocks in a single
year, or we could make them last 1000 generations. How long recoverable
stocks will last, therefore, is determined as much by how fast we extract
them as by how much there actually is. We almost certainly will never ex-
haust fossil fuel stocks in physical terms, because there will always remain
some stocks that are too energy-intensive or too expensive to recover.
From this point of view, fossil fuel stocks are nonrenewable but not ex-
haustible.
As we extract fossil fuels, we will logically extract them from the most
accessible and highest-quality known reserves first, where net energy
gains are highest.7 These stocks essentially offer the lowest-entropy re-
7Note that the largest and most accessible reserves are also the most likely to be discovered
first.
source. Therefore, as we continue to extract fossil fuels over time, we

can expect not only a quantitative decrease but also a qualitative decline
in stocks. For example, the first oil to be extracted actually pooled on
the surface and erupted in geysers from wells with no pumping. But as
stocks diminish, it takes more and more energy to extract energy; ever-
larger fractions of a barrel of oil are needed as energy inputs to retrieve
a barrel of oil as output, until we have reached entropic exhaustion.
Of course, resource exhaustion is only one component of fossil fuel
use. Used fuel does not disappear; it must return to the ecosystem as waste.
Acid rain, global warming, carbon monoxide, heat pollution, and oil spills
are unavoidably associated with the use of fossil fuels. On a small scale,
some of these wastes could be readily processed by natural systems, but
on the current scale, they pose serious threats. Indeed, the growing accu-
mulation of waste products from fossil fuel use and the negative impacts
these have on planetary ecosystems is probably a far more imminent
threat to human welfare than depletion; the sink will be full before the
source is empty.
We must reiterate here that ecosystems, via the primary producers they
sustain, themselves capture solar energy, and humans make direct use of
much of the energy they capture. If waste products from fossil fuel use di-
minish the ability of these ecosystems to capture energy, there are more
energy costs to fossil fuel extraction than the direct ones discussed above.
These costs are, however, several degrees of magnitude more difficult to
measure—and therefore that much more likely to be ignored (Figure 5.1).
THINK ABOUT IT!
Many people are concerned by the United States’ dependence on oil
imports from a number of politically unstable regions and countries
(e.g., the Middle East, Nigeria, Venezuela, Colombia). Proposed solu-
tions to this problem have included increased domestic drilling and
extraction, greater energy efficiency, and the development of renew-
able energy sources. What do you think are the pros and cons of each
approach?
The basic equation here is:

net recoverable energy from oil = (initial total stock of entropically
recoverable reserves) – (oil already consumed) –
(energy cost of extraction) –
(loss of solar energy due to induced loss of capacity to capture)
Net energy from fossil fuels must account for the damage fossil fuel use
causes to the ability of the sustaining system to capture solar energy, a
fund-service resource. This lost capacity is measured as energy-flow/time,
Figure 5.1 • Net recoverable energy from fossil fuels.
and we must account for the total amount of energy not captured from the
time the damage occurs to the time the fund-service recovers.8
What points can we draw from this discussion of fossil fuels? First,
once fossil fuels are used, they are gone forever—they are rival goods.
While a seemingly trivial point, this has important implications for eco-
nomic policy, as we will show in Chapter 11. Second, while fossil fuel
stocks are finite, they are a stock-flow resource that can be extracted vir-
tually as quickly as we wish, limited only by existing infrastructure,
knowledge of stock locations, and the energy costs of extraction. We have
control of the spigot and have been opening it a bit wider every decade.
Eventually the reservoir must run dry. This is in stark contrast to flows of
solar energy, as we pointed out in Chapter 4.
Third, our current populations and economic systems depend for sur-
vival on the use of fossil fuels. Fossil fuels not only supply 85% of our en-
8Estimates of oil already consumed, probable reserves, and oil yet to be discovered are from
Campbell and Laherrère, ibid.

ergy needs, much of which is used to produce food, they also provide the
raw materials for a substantial portion of our economic production, in-
cluding ubiquitous plastics and, even more importantly, the fertilizers,
herbicides, and pesticides that help provide food for nearly 7 billion
people. At this point, we do not have the technologies available to support
7 billion people in the absence of fossil fuels.
While we may be able to substitute renewable energy for fossil fuels, it
is highly uncertain that we can achieve this before the negative impacts of
fossil fuel waste products force us to stop using them or the fuels them-
selves are depleted.
THINK ABOUT IT!
The U.S. and Canada have vast deposits of shale oil and tar sands, re-
spectively. Both of these are fossil fuels but of fairly low quality, requir-
ing more energy to extract and process than conventional fossil fuels
and creating more associated waste. Do you think these resources
present possible solutions to our energy problems? Can you dig up
any information on their energy returns to investment and waste out-
puts?
Mineral Resources
Though typically grouped together with fossil fuels in economics text-
books and labeled nonrenewable resources, minerals differ in important
respects from fossil fuels. Like fuels, minerals can be analyzed in terms of
stocks and flows. We know the total stock is finite, and according to the
First Law of Thermodynamics, this imposes a physical limit on their con-
tribution to the material growth of the economy. Again, technology can in-
crease the efficiency with which we extract minerals from ore, but there
exists an entropic limit to efficiency. Valuable mineral deposits occur in
varying degrees of purity, and, like fuels, the degree of purity can be
looked at as a measure of low entropy. Highly concentrated ores are highly
ordered low entropy.9 It is much easier to extract their mineral content,
and they are much more valuable. As our growing economy depletes these
most valuable ores first, we must move on to ores of lower and lower pu-
rity, incurring higher and higher processing costs.
As in the case of oil, we are not exactly certain of the total stock of any
particular mineral, but geologists assign reasonable probabilities to differ-
ent estimates. Even the most efficient process conceivable will require
9Even if we do not accept the notion of entropy in materials, concentrated ores require much
less low-entropy energy to process.

some energy to extract minerals from an ore, and the less pure the ore, the
more energy will be needed. Currently, mining accounts for about 10% of
global energy use.10 However, unlike fossil fuels that cannot be burned
twice, materials can be recycled (though this, too, requires energy). There-
fore, we must think in terms of nonrenewable subterranean stocks as well
as aboveground stocks, which accumulate as the subterranean ones are
depleted. Still, we cannot avoid the laws of entropy even here, and use
leads to dissipation through chemical and physical erosion; therefore,
100% recycling of any material may be impossible.
There is considerable debate over the impossibility of 100% recycling,
as well as the implications. Georgescu-Roegen argues that because solar
energy can provide a substitute for fossil fuels and nothing can provide a
substitute for minerals, mineral depletion is actually more of a concern
than fossil fuel depletion, and its inevitability means that a steady-state
economy11 is impossible (see Box 3.2). In contrast, Ayres claims that even
if all elements in the Earth’s crust were homogeneously distributed (the
material equivalent of “heat death” mentioned above), a sufficiently effi-
cient solar-powered extraction machine would enable us to extract these
elements,12 presumably at a rate that would provide enough raw materi-
als to maintain the machine and still leave a material surplus. This sce-
nario implicitly assumes that damage caused by extracting all the
resources from the Earth’s crust in the first place, and their consequent re-
turn to the ecosystem as waste, would not irreparably damage the Earth’s
ability to capture solar energy and sustain life.
Alternatively, we may be able to master the art of creating polymers
from atmospheric CO2, which could provide substitutes for many of the
minerals we currently use. If such polymers were biodegradable and sim-
ply returned to the atmosphere as CO2 we would presumably be able to
achieve 100% recycling (though in this case we may not want to, at least
not before atmospheric CO2 stabilizes at preindustrial levels). Of course,
none of these propositions can currently be proven empirically. Nonethe-
less, it appears that mineral deposits are sufficiently large, and recycling
has the potential to become sufficiently efficient, that with careful use,
minimizing waste and appropriate substitution where possible, we could
sustain a steady-state economy for a very long time.
Figure 5.2 depicts both the accumulation of extracted minerals into
10P. Sampat, From Rio to Johannesburg: Mining Less in a Sustainable World. World Summit
Policy Brief #9. Online: http://www.worldwatch.org/worldsummit/briefs/20020806.html. (World

Watch).
11N. Georgescu-Roegen, The Entropy Law and the Economic Process, Cambridge, MA: Harvard
University Press, 1971.

12R. U. Ayres, The Second Law, the Fourth Law, Recycling and Limits to Growth, Ecological
Economics 29:473–484 (1999).

Figure 5.2 • The cumulative extraction of subterranean stock and aboveground

stock of minerals over time. The distance between the two curves is a measure
of entropic dissipation.
aboveground low-entropy stocks embodied in artifacts in use by (or avail-

able to) the economy (solid line) and the cumulative depletion (extrac-
tion) of subterranean stocks (dashed line) over time. We assume that
initial rates of mineral extraction are low, but they increase with economic
growth and greater knowledge of reserve location. Eventually, however,
stocks become scarcer, the costs of extraction become greater than the
benefits, and extraction ceases. The point where this occurs is labeled “In-
ground stocks exhausted” on the graph, and cumulative depletion ceases.
In the absence of entropy, and if 100% recycling were possible and prac-
ticed, the two lines on the graph would be identical. In the real world,
some portion of aboveground stocks dissipates into waste every year. The
rate of increase in the aboveground stocks is equal to net annual mineral
extraction minus entropic dissipation; that is, aboveground stocks are
equal to minerals currently in use plus those that can be recycled.
There are two important categories of waste. Much waste is in the form
of products that have stopped working, become obsolete, or simply gone
out of fashion and are discarded while still in a relatively ordered state.
They are not recycled because it is either cheaper or more convenient to
extract virgin mineral flows from the Earth. For our purposes, this waste
returns to the subterranean stock, though with higher entropy than the
ore from which it was initially extracted.13 Eventually, as we deplete the
13Some of this material will be in a highly ordered state and have lower entropy than the same
amount of mineral in the form of an ore. Georgescu-Roegen distinguishes “garbo-junk” (a bald tire
useless as a tire but recyclable) from “pure waste” (the dissipated rubber particles that are not re-
cyclable). For practical purposes, however, large stocks of ore presumably still have an overall
lower entropy; otherwise waste material would be processed before the ore.
most concentrated ores, it becomes cheaper to start mining the lowest-

entropy waste. For example, slag heaps near old silver mines have been
mined again with newer methods. But the slag heaps resulting from the
second mining will be harder to mine.
Another type of waste results from entropy in the form of mechanical
or chemical erosion of the material in question. Pennies eventually wear
out through use—an atom rubbed off here, another there. Other metals
rust away. Hence, gross subterranean stocks are never depleted (First Law
of Thermodynamics). They simply become stocks of higher and higher
entropy, akin to bound energy, and are no longer of use to humans (Sec-
ond Law of Thermodynamics).
We have some control over the creation of waste in the form of dis-
carded goods, but virtually none over the effects of entropy. The entropic
limit to extraction in this case occurs when the extraction process con-
sumes more material than it can provide. As we stated earlier, some peo-
ple assert that this never happens, while others assert it will happen soon
enough to make a steady-state economy a pipe dream; we take the mid-
dle road.
As more minerals are brought to the surface and put to use, entropy
acts on a larger stock. As more subterranean stocks are extracted, the re-
mainder becomes more difficult to find and extract. Therefore, even before
in-ground stocks are exhausted, the rate of dissipation of aboveground
stocks must become greater than the net extraction of new material, and
aboveground stocks begin to decline. However, even after reaching the
entropic limit to extraction, we are still likely to have a large stock of ma-
terial in the economy that can be reused and recycled. Over time, of
course, it must gradually erode away, atom by atom. In Figure 5.2, the dis-
tance between the lines depicting cumulative extraction and aboveground
stocks measures cumulative dissipation, and eventually the entire above-
ground stock must succumb to entropy. This process is probably slow
enough that we could achieve a steady state through material recycling for
a very, very long time. However, as is the case with oil, the threat to us is
probably more from the impacts of the waste itself than from the exhaus-
tion of mineral resources. We’ll put this discussion off until we get to the
section on waste absorption capacity in Chapter 6.
What points can we draw from our discussion? First, mineral resources
are rival goods at a given point in time. If I am using a hunk of steel in my
A resource is nonrival between car, it is not available for you to use. But through recycling, most of these
generations if the use by one resources could be made available for someone else to use in the future.
generation does not leave less
Thus, we can think of mineral resources as rival goods within a generation but
of the resource for future
as partially nonrival between generations, depending on how much is
generations.
wasted and how much recycled. Fossil fuels are rival both within and be-
tween generations. Second, stocks of low-entropy mineral ores are finite
but can be extracted at virtually any rate we choose. In contrast to fossil

fuels, we have control not only over the spigot of extraction but also over
the drain by which extracted materials return to the ecosystem as waste.
We open the spigot wider almost every year and do very little to close the
drain, but if we shut the drain as much as possible (though it will always
leak some), the open spigot matters less to future generations. Third, we
could not sustain existing populations or levels of economic production
in the absence of these minerals. While it would clearly be impossible to
develop substitutes for all minerals, thus far it has been reasonably easy to
develop substitutes for specific minerals as they become scarce, and it may
be possible to keep this up for some time to come.
Water
Earth is a water planet. Though the stock of water is finite, fully 70% of
the Earth’s surface is covered in water. Freshwater, however, is far less
abundant, accounting for less than 3% of the total, of which less than one-
third of 1% is in the form of readily exploited lakes (0.009%), rivers
(0.0001%), and accessible groundwater (0.31%). Another 0.01% is found
in the atmosphere, 0.31% is deep groundwater, and over 2% is in the
polar ice caps and glaciers.14 Humans are composed mostly of water, and
in addition to drinking, we depend on it for agriculture, industry, hydro-
electricity, transportation, recreation, and waste disposal and for sustain-
ing the planet’s ecosystem services. Water for different uses has different
relevant characteristics that make generalizations difficult.
Water for drinking, irrigation, industry, and waste disposal is clearly a
stock-flow resource, but a unique one. In contrast to fossil fuels and min-
eral deposits, many water resources are renewable as a result of the hy-
drologic cycle. However, for all practical purposes, many aquifers are
“fossil” water, with negligible recharge rates. Many other aquifers are being
mined; that is, the rate of water extraction is greater than the rate of re-
plenishment. Even many rivers around the world, including the Colorado
and the Rio Grande in North America, the Amu-Dar’ja and Syr-Dar’ja
rivers that once fed the Aral Sea in Central Asia, and at times the Yellow,
Hai, and Huai rivers in northern China, are so heavily used (primarily for
irrigation) that they never reach the sea.
At first glance, flowing water might appear to be a fund-service re-
source. In any stream or river at any given time, water is flowing at a spe-
cific rate, and the proper unit of measurement is volume/time (volume per
unit of time), as is the case for fund-service resources. Dams, however,
14P. Gleick, The World’s Water: The Biennial Report on Freshwater Resources, Washington, DC:
Island Press, 2002.

allow us to stockpile flowing water for later use, which is a characteristic

of stock-flow resources, and water is “used up” by drinking, irrigation, in-
dustry, and waste disposal but never “wears out.”
Perhaps the best way to look at flowing water is to distinguish it from the
hydrologic cycle. The water itself is a stock-flow resource that is rapidly re-
newed by the service (provided by solar energy) of the hydrologic cycle. Hy-
droelectricity is produced not by water but rather by the energy transferred
to water by the hydrologic cycle—it is solar energy stored in water. Solar en-
ergy is generally a fund service, but when stored in water, it can be either a
stock-flow or a fund-service resource. When mechanical energy in the water
is converted to electric energy by a microhydropower plant that depends on
river flow, it is essentially a fund-service resource. However, damming of the
river allows the energy to be stockpiled by converting mechanical energy to
potential energy, which is a stock-flow resource.
When used for transportation, recreation, or sustaining all other eco-
systems on the planet, water functions as a fund-service resource. Atmos-
pheric moisture, as part of the hydrologic cycle, is essentially a fund-service
resource.
Like biotic resources, water can be a stock-flow and fund-service re-
source simultaneously. Unlike biotic resources, however, humans cannot
meaningfully affect the total stock of water on the planet. We can and do re-
duce the stock of usable water, and while it is possible to restore the usabil-
ity of water, there are no substitutes available for its most important uses.
As one would expect from its dual nature as a stock-flow, fund-service
resource, water can be rival or nonrival depending on its use; stock-flow
uses are rival, and fund-service uses are nonrival. However, as flowing
water is recycled through the hydrologic cycle, it is intergenerationally
nonrival. Excludability varies dramatically depending on existing institu-
tions, though for all practical purposes rainfall is nonexcludable by na-
ture.15
Ricardian Land
Ricardian land—land as a physical substrate and location, distinct from
its other productive qualities—is also a fund that provides the service of a
substrate capable of supporting humans and our infrastructure and of
capturing solar energy and rain (Ricardian land does not include soil or
the nutrients in the soil). A hectare of land may be capable of producing
1000 tons of wheat over 100 years, but one cannot produce that wheat
15The seeding of clouds with silver nitrate can produce rainfall in a specific location, but for
practical purposes this is irrelevant.

from the same land in an appreciably shorter period, nor would it be pos-
sible to accumulate land’s capacity as a substrate.
The services provided by land are certainly excludable, and at any
given point in time, they are also rival. For example, if used for farming,
land provides the service of a substrate for crops. If one farmer uses that
service, no one else can in the same time period. Economists often use the
term “depletable” as a synonym for “rival,” but the case of land suggests
that this is inappropriate.16 Using Ricardian land does not deplete it.
While rival within a generation, it is intergenerationally nonrival and ab-
solutely nondepletable.
THINK ABOUT IT!
Why do you think we distinguish between Ricardian land as a physical
substrate and the more conventional definition of land that includes
the soil and its mineral content? Who or what creates value in Ricar-
dian land? What makes land in one place more valuable than a similar
piece of land elsewhere? Who or what creates value in fertile topsoil?
Solar Energy
The last abiotic producer of goods and services we will discuss is the sun.
It bathes the Earth in 19 trillion tons of oil equivalent (toe) per year—
more energy than can be found in all recoverable fossil fuel stocks—and
will continue to do so for billions of years.17 Why the fuss over the con-
sumption of the Earth’s fossil fuels?
While the flow of solar energy is vast, it reaches the Earth at a fixed
rate in the form of a fine mist and hence is very difficult to capture and
concentrate. Most of the sunlight that strikes the Earth is reflected back
into space.18 Over the eons, life has evolved to capture enough of this en-
ergy to maintain itself and the complex ecosystems that life creates. It
would appear that the “order” of the global ecosystem over billions of
years has reached a more or less stable thermodynamic disequilibrium. A
better term is “meta-stable,” meaning that the global ecosystem fluctuates
around a steady state rather than settling into one without further varia-
16When “depletable” is used in this sense, it means that one person’s use depletes the resource
in question. Hence, the ozone layer is nondepletable because if I use it to protect me from skin can-
cer, it is still there for someone else to use. It is certainly possible to deplete the ozone layer with
chemicals, but that is not a case of depletion caused by use.
17Unless otherwise cited, estimates of energy use and availability are from World Energy
Council, 2007 Survey of Energy Resources, London: World Energy Council, 2007. Online:
http://www.worldenergy.org.
18N. Georgescu-Roegen, Energy and Economic Myths: Institutional and Analytic Economic Essays,
New York: Pergamon Press, 1976.

tion.19 Virtually all energy captured from the sun is captured by chloro-
phyll. In the absence of the evolution of some alternative physiological
process for capturing sunlight, it would seem that our planet cannot sus-
tain more low entropy than it currently does for any extended period. Yet
through the use of fossil fuels, Americans are able to consume 40% more
energy than is captured by photosynthesis by all the plants in the coun-
try. We also directly use over half of the energy captured by plants.20
As fossil fuels run out, we will need an alternative source of low entropy
to maintain our economy at its current level of thermodynamic disequilib-
rium. The sun unquestionably radiates the Earth with sufficient energy to
meet our needs, but how do we capture it? Global gross commercial energy
consumption is about 11.3 billion toe (~14.5 TW) per year. Biomass, hy-
droelectricity, wind, photovoltaics, wave, and ocean thermal energy are all
forms of solar energy we could potentially capture. Biomass is widely
touted as a substitute for fossil fuels, but converting all of the net primary
productivity (NPP) of the United States to liquid fuel would still not meet
our liquid fuel needs. Hydroelectricity currently provides 19% of global
electricity, but even fully developed it could not supply 60%. Wind cur-
rently supplies little energy (about 72,000 MW-h in 2006), but it is a
promising alternative: At current installation rates, capacity is doubling
every 3.5 years.
Photovoltaics and wave/ocean thermal technologies still play very
minor roles. With all of these technologies, however, large energy invest-
ments are needed to produce the infrastructure needed to capture solar
energy, and in many cases (e.g., photovoltaics), the energy returns on in-
vestment may be negligible. At the same time, human activity decreases
the surface area of the planet covered in plant life and disrupts the ability
of plants to capture sunlight. The net effect is likely to be an annual de-
crease in the amount of solar energy the Earth captures and hence a de-
crease in the complexity of the systems it is capable of maintaining. Figure
5.1 earlier illustrates the loss of solar energy capture that can be attributed
to waste from fossil fuels.
While solar energy will bathe the Earth in more energy than humans
will ever use, for practical purposes it is a fund-service resource that ar-
rives on the Earth’s surface at a fixed rate and cannot be effectively stored
for later use.21 No matter how much solar energy one nation or land-
19E. Laszlo, Vision 2020, New York: Gordon and Breach, 1994.
20D. Pimentel and M. Pimentel, Land, Energy and Water: The Constraints Governing Ideal U.S.
Population Size, Negative Population Growth, Forum Series, 1995. Online: http://www.npg.org/
forum_series/land_energy&water.htm.
21Solar energy can be stored in fossil fuels, in batteries, or in the form of hydrogen for later
use by humans, but this energy cannot subsequently be used to power photosynthesis, the most
important function of solar energy.
owner captures, there is no less left for others to capture, and it is inher-
ently nonexcludable.22
Summary Points
Table 5.1 summarizes some of the policy-relevant characteristics of these
five abiotic resources. Why are these details important to ecological eco-
nomic analysis, and what message should you take home from this chap-
ter? The stock-flow/fund-service distinction is important with respect to
scale. We have control over the rate at which we use fossil fuels, mineral
resources, and water. As the economy undergoes physical growth, it must
use ever-greater flows from finite stocks. Because fossil aquifers and fuels
are irreversibly depleted by use, and mineral resources may be irreversibly
dissipated through use, the finite stock of these resources imposes limits
on total economic production over time. Limits to growth may not be
apparent until the stock is seriously depleted, and once gone, it is gone
forever. Funds, in contrast, provide services at a fixed rate over which we
have no control (though one thing that distinguishes biotic fund services
from abiotic ones is that we can damage or even destroy them). Fund-
services therefore limit the size of the economy at any given time, but they
do not limit total production over time.
Table 5.1
SELECTED POLICY-RELEVENT CHARACTERISTICS OF ABIOTIC RESOURCES

Rival
Abiotic Stock-Flow or Can Be Made Between
Resource Fund-Service Excludable Rival Generations Substitutability
Fossil Fuels Stock-flow Yes Yes Yes Modest at margin,
(nonrecyclable) but possibly
substitutable
over time
Minerals Stock-flow Yes Yes Partially High at margin,

(partially ultimately
recyclable) nonsubstitutable
Water Context- Context- Context- Stocks, yes; Nonsubstitutable

(solar recycling) dependent dependent dependent funds and for most
recycled, no important uses
Ricardian Land Fund-service Yes Yes No Nonsubstitutable

(indestructible)
Solar Energy Fund- No No, for No Nonsubstitutable

(indestructible) service practical purposes
22Future space-based solar technologies may change this but are irrelevant at present.
Water sources are a complex mix of stock-flow and fund-service. But

even the stock-flow uses of water are completely recyclable—in particu-
lar, running water is so closely linked to the fund-service of the solar-
powered hydrologic cycle that it acts much like a fund-service, imposing
limits on the output of the economy only at a given point in time.
Substitutability is also relevant to scale. If we can develop a substitute
for a resource, then the constraints it imposes on scale are less rigid. How-
ever, developing substitutes generally relies on technology, and technol-
ogy takes time to develop. In addition, truly innovative technologies are
impossible to accurately predict; we could predict one only if we already
knew what it would look like, in which case it would not be truly inno-
vative.
Rivalness is relevant primarily to distribution, both within and between
generations. All abiotic resources are rival except for water in some of its
forms and uses, and solar energy (for practical purposes). One person’s
use of these rival resources means they are not available for others to use,
and we must be concerned about distribution within a generation. People
can use the nonrival resources of solar energy and water in its fund-
service functions without leaving less for anyone else and without affect-
ing scale, and all else being equal, we should therefore let anyone use
them. When a good is nonrival between generations, we needn’t worry
about excessive use within a generation. When addressing distribution,
we must remember that all natural resources are produced by nature, not
humans, although the value of Ricardian land is generally produced by so-
ciety as a whole.
Excludability is relevant primarily to allocation. The market cannot al-
locate nonexcludable goods, and other allocative mechanisms are needed.
However, in the case of sunlight and rainfall, allocation by human insti-
tutions is simply not feasible.
Big ideas from Chapter 4 that Aboveground and

recur in Chapter 5 subterranean mineral stocks
Ricardian land Entropic dissipation
Energy return on investment Rival within versus between
Recoverable reserves generations
P10, P50, and P90 reserve Garbo-junk versus pure
estimates waste
Net recoverable energy from Unique characteristics of
fossil fuels water and solar energy
CHAPTER
6
Biotic Resources
B iotic resources include the raw materials upon which economic pro-
duction and human life depend, the ecological services that create a
habitat capable of supporting human life, and the absorption capacity that
keeps us from suffocating in our own waste. As nonrenewable resources
are exhausted, human society will come to rely more and more on the self-
renewing capacity of biotic resources. It is therefore critically important
that we understand the nature of these resources.
As we turn our attention from abiotic resources to biotic ones, we must
address a quantum increase in complexity, inevitably accompanied by a
quantum increase in ignorance and uncertainty. One level of complexity
arises from the intrinsic value we give to living systems. Abiotic resources
are almost entirely considered means to various ends, where one of the
foremost ends is the sustenance of life, the maintenance of biotic resources.
Biotic resources not only enhance human well-being directly, they are also
considered by many to be an end in their own right, especially in the case
of sentient creatures. Biotic resources are also physically complex in two
ways. First, the processes responsible for the sustained reproduction of in-
dividuals, populations, or species are highly complex and poorly under-
stood. Second, individuals, populations, and species interact with other
individuals, population, and species, as well as abiotic resources, to create
an ecosystem. Ecosystems are extraordinarily complex and dynamic,
changing over time in inherently unpredictable ways. The differences be-
tween these two types of physical complexity bear closer examination.
I Ecosystem Structure and Function

Ecologists look at ecosystems in terms of structure and function, corre-
sponding to the two types of physical complexity mentioned above. This
93
distinction is very relevant to economic analysis. Conventional natural re-

source economics is essentially the economics of ecosystem structure. En-
vironmental economics focuses on certain ecosystem functions. In reality,
structure and function are mutually interdependent, and we need an eco-
nomics that effectively integrates both. Certainly we must understand the
distinctions and interactions between the two if we are to incorporate
them into economic analysis.
Ecosystem structure refers to the individuals and communities of
plants and animals of which an ecosystem is composed, their age and spa-
tial distribution, and the abiotic resources discussed in Chapter 5.1 Most
ecosystems have thousands of structural elements, each exhibiting vary-
ing degrees of complexity. Scientists have learned that when enough sep-
arate elements are thrown together into a complex system, a sort of
spontaneous order results. One property of such systems is their tendency
to generate emergent phenomena, which can be defined as properties of
the whole that could not be predicted from an understanding of the indi-
vidual parts, no matter how detailed that understanding. Complex sys-
tems are also characterized by highly nonlinear behavior, which means
that we cannot predict the outcomes of large interventions based on an
understanding of smaller ones. For example, removing 40% of a species
stock from an ecosystem may have a qualitatively different impact than re-
moving 20%—that is, not just twice the known impact of removing 20%.
In an ecosystem, the structural elements act together to create a whole
that is greater than the sum of the parts. We refer to these emergent phe-
nomena in ecosystems as ecosystem functions,2 and they include such
things as energy transfer, nutrient cycling, gas regulation, climate regula-
tion, and the water cycle. As is typical of emergent properties, ecosystem
functions cannot be readily explained by even the most extensive knowl-
edge of system components.3 Variability, ignorance, and uncertainty play
an extremely important role in the analysis of ecosystem structure and a
far greater role in the analysis of ecosystem function. We have a very lim-
ited understanding of exactly how ecosystem functions emerge from the
complex interactions of ecosystem structure and thus a difficult time
1It may seem strange to include such things as fossil fuels and mineral deposits as elements of
ecosystem structure, but we must not forget that humans are part of the global ecosystem, and
these resources affect our ability to thrive.
2Whether a particular element of an ecosystem is part of structure or part of function depends
on perspective. Organelles are part of the structural components of a cell that enable the cell to
function. Cells are structural components of an individual that enable the individual to function.
In the same way, individuals are part of the structure of a population, a population is part of the
structure of a local ecosystem, an ecosystem is part of the structure of a landscape, and a land-
scape is part of the structure of the global ecosystem.
3E. Odum, Ecology: A Bridge Between Science and Society, 3rd ed., Sunderland, MA: Sinauer,
1997.
Box 6-1 Risk, Uncertainty, and Ignorance
Whenever we do not know something for sure, we are uncertain, but there
are different types of uncertainty. When I throw dice, I cannot say in ad-
vance what the outcome will be, but I do know the possible outcomes and
their probabilities. This type of uncertainty is referred to as risk. Pure un-
certainty occurs when we know the possible outcomes but cannot assign
meaningful probabilities to them. Ignorance or absolute uncertainty oc-
curs when we do not even know the range of possible outcomes.
In economics, Frank Knight pointed out that risk is a calculable or in-
surable cost, while pure uncertainty is not. In his view profit—the differ-
ence between revenue and calculable risk-adjusted costs—is a return for
willingness to endure pure uncertainty. However, Knight was discussing
the case where the entrepreneur bore the costs of failure and reaped the
rewards of success. In economic decisions regarding exploitation of
ecosystems, it is often the entrepreneur who reaps the rewards, while
society bears the costs.a
Discoveries in quantum physics and chaos theory suggest that uncer-
tainty and ignorance do not result simply from a lack of knowledge but
are irreducible, inherent properties in certain systems. For example,
chaos theory shows that even in a deterministic (i.e., nonrandom) sys-
tem, extremely small differences in initial conditions can lead to radi-
cally different outcomes. This has been popularized as the butterfly
effect, in which a butterfly flapping its wings over Japan can create a
storm in North America.
Change in highly complex systems is characterized by ignorance, es-
pecially over long time spans. We cannot predict evolutionary change in
organisms, ecosystems, or technologies. For example, while we can pre-
dict that computers will continue to get faster and cheaper, we cannot
predict what the next big technology will be 50 years from now. Leading
experts are often notoriously wrong even when predicting the future of
existing technologies. Bill Gates reputedly once predicted that no one
would ever need more than 540 kilobytes of computer memory.
Estimating stocks of natural resources or reproductive rates for culti-
vated species is basically a question of risk. Estimating reproductive rates
for wild species is a question of uncertainty, since we cannot accurately pre-
dict the multitude of factors that affect these reproduction rates, but we do
know the range over which reproduction is possible. Estimating ecological
thresholds, conditions beyond which ecosystems may flip into alternative
states, is a question of pure uncertainty, since we have limited knowledge
of ecosystems and cannot predict the external conditions that affect them.
Predicting the alternate state into which an ecosystem might flip when it
passes an ecological threshold, and how humans will adapt, are cases of
absolute ignorance involving evolutionary and technological change.
aF. H. Knight, Risk, Uncertainty, and Profit, Boston: Houghton Mifflin, 1921; Library of
Economics and Liberty, Feb. 21, 2002. Online: http://www.econlib.org/library/Knight/

knRUPl.html.
predicting and managing the impacts of human actions on these func-

tions. Therefore, a great deal of uncertainty attends decision making in-
volving ecosystem functions. How we choose to treat uncertainty in
economic analysis is ultimately a normative (ethical) decision, yet another
source of complexity. One of the most important issues concerning any
analysis of biotic resources is the degree of uncertainty involved.
Concrete examples always help clarify a concept. To illustrate the links
between structure and function, and the implications of complexity, let’s
focus on a wet tropical forest, the terrestrial ecosystem that exhibits the
greatest biodiversity of any yet studied. The forest is composed of indi-
vidual plants (part of ecosystem structure). Each plant alone has little im-
pact on climate, nutrient cycling, and habitat provision and may even be
unable to reproduce. However, when we bring together hundreds of mil-
lions of plants, as in the Amazon or Congo basin, these and other ecosys-
tem functions emerge.
The forest canopy filters out about 98% of the sunlight at ground level,
dramatically reducing daytime temperatures. It traps air and insulates, in-
creasing night temperatures under the canopy and maintaining high and
constant humidity. Trees absorb the energy of tropical storms, aerate the
soil to allow water absorption, and slow water flows—all of which prevent
soil and nutrients from being washed out of the system. Trees create the
microclimate and habitat essential to the soil fauna that help recycle nu-
trients, facilitating their reabsorption by the system.
On a regional scale, the water retained by forest structure is absorbed
and returned to the atmosphere through evapotranspiration, increasing
humidity over the forest. Greater humidity increases the frequency of
rainstorms. Estimates for the Amazon forest suggest that it generates up to
50% of its own rainfall, enabling the water-dependent species there to
thrive. Without the increased absorption capacity of the soil and the evap-
otranspiration it facilitates, rainfall would simply drain into the rivers and
be flushed from the system forever.
On an even larger scale, forests absorb up to 90% of the solar energy
that strikes their canopies. Much of this is released through evapotranspi-
ration and carried high up into the atmosphere, where it is carried into the
temperate zones, helping stabilize the global climate (a function provided
by carbon sequestration as well).
The species and populations in the forest cannot survive without a sta-
ble climate and a steady nutrient flow. Loss of forest structure can degrade
forest function to the point where the forest spontaneously declines,
creating a positive feedback loop with potentially irreversible and cata-
strophically negative consequences. Numerous models suggest continued
deforestation in the Amazon could lead to dramatic declines in rainfall, in-
creased susceptibility to fires (such as those that occurred in 1997 in the
Chapter 6 Biotic Resources • 97
Amazon, Indonesia and Mexico), and spontaneous degradation of the re-

maining forest.4
In other words, ecosystem structure interacts to create ecosystem func-
tions, and the structural elements depend on these functional attributes
for their own survival. Owing to the complex nature of the whole system,
as structural elements of an ecosystem are lost, in most cases we cannot
say for sure to what extent ecosystem functions will be affected. Similarly,
as ecosystem functions change in response to human impacts or non-
anthropogenic change, we cannot say for certain what the impact will be
on ecosystem structure.
THINK ABOUT IT!
In Chapter 4, we asked you to make a list of stock-flow and fund-
service resources provided by a local ecosystem. Which of these
resources are elements of ecosystem structure? Which are elements
of ecosystem function? Do you see any links between these
classifications?
Roughly speaking, conventional and natural resource economics has

focused on ecosystem structure, while conventional environmental eco-
nomics has focused on certain elements of ecosystem function, with a
major emphasis on waste absorption capacity and the monetary valuation
of other functions. In reality (as many conventional economists are fully
aware), ecosystem structure and function are mutually interdependent,
and conclusions based on the analysis of one dimension may not apply to
the multidimensional case. With this caveat in mind, we now turn our at-
tention to specific categories of biotic resources.
Three basic categories of biotic resources deserve attention. First are re-
newable resources, the elements of ecosystem structure that provide the
raw materials for economic processes. Second are ecosystem services,
defined as the ecosystem functions of value to humans and generated as
emergent phenomena by the interacting elements of ecosystem structure.
Third is waste absorption capacity, an ecosystem service that is suffi-
ciently distinct from the others to warrant separate treatment.
I Renewable Resources
For simplicity, we can treat biological resources as material stock-flow re-
sources, that is, as elements of ecosystem structure. Like nonrenewable re-
sources, biological stocks can be extracted as fast as humans desire, but
4D. Nepstad et al., Interactions Among Amazon Land Use, Forests and Climate: Prospects for
a Near-Term Forest Tipping Point. Philosophical Transactions of the Royal Society B: Biological Sciences
363:1737–1746(2008).
they are capable of reproduction. Figure 6.1 depicts a renewable natural

resource in stock-flow space: the x-axis depicts the stock, or amount of re-
source that exists, and the y-axis depicts the flow. The flow in this case can
be the rate of reproduction (or biomass increase) that is likely for any
given stock, or the rate of extraction (harvest). A 45-degree dashed line
shows the theoretically maximum rate at which we can extract a given
stock (i.e., we can extract the entire stock at one time; stock = flow). Ac-
tual extraction rates must lie on or below this line. We have also drawn a
curve that shows the growth rate of each level of stock, which is also the
sustainable yield curve. The sustainable yield is the net annual reproduc-
tion from a given stock; for every population of a resource, there is an as-
sociated average rate of population increase, and that increase represents
a sustainable harvest that can be removed every year without affecting the
base population.
We must caution here that there is a great deal of uncertainty concern-
ing the position of this sustainable yield curve. Not only do we not know
at precisely what rate a given population will reproduce, we are also un-
certain of the exact population of any given species, though this is more
true for animals than for plants, since plants sit still for the census takers
while animals do not. While with careful study and census techniques we
can assign reasonable probabilities to population estimates for renewable
resource stocks (risk), there is greater uncertainty of a qualitatively differ-
ent type concerning reproduction rates, particularly because these rates
depend on a host of “external” factors such as rainfall, abundance of pred-
ator and prey species, disease, and so on. In addition, habitat destruction
and degradation, pollution, climate change, and other human impacts can
profoundly affect the entire curve, shifting it dramatically over time. Thus,
in any given year, the actual rate of increase from a given population stock
may be wildly different from the average.
As most people are at least vaguely aware, stocks of plants and ani-
mals in nature cannot grow forever. Instead, populations reach a point
where they fill an available niche, and average death rates are just
matched by average birth rates. Populations “stabilize” around an equi-
librium, known as the carrying capacity. (We use the term stabilize
loosely, because populations fluctuate in the short term depending on
weather conditions, predator-prey cycles, etc. and in the longer term de-
pending on a wide variety of factors. To paraphrase John Maynard
Keynes, in the very long term, all species go extinct.) At carrying capac-
ity, there is just enough food and habitat to maintain the existing popu-
lation, and the rate of growth in biomass is zero (point K in Figure 6.1).
Obviously, the rate of growth of a stock is also zero when the stock has
Figure 6.1 • The growth, or sustainable yield, curve.
been driven extinct (the origin in Figure 6.1). In between these two
points, things get interesting.
The growth, or sustainable yield, curve for a renewable natural resource
indicates the increase in stock over one time period for any given stock.
The y-axis can measure growth or harvest, that is, flows from or to the ex-
isting stock depicted on the x-axis. Any harvest up to the total stock is the-
oretically possible. A harvest at any point on the sustainable yield curve,
such as S, is just equal to the growth in the stock and hence has no net im-
pact on stocks. Harvests above the sustainable yield curve deplete the re-
source, and harvests below the curve lead to an increase in the stock, as
indicated by the arrows. For example, a harvest at R will reduce the stock
to S′′, and a harvest at S′ will allow the stock to increase to R′′. MVP rep-
resents the minimum viable population, which is the level below which a
species or stock cannot sustain itself even in the absence of harvest. The
phrase critical depensation refers to the spontaneous decline of a popula-
tion or ecosystem that has fallen below the minimum viable population or
size.5 K represents carrying capacity. The graph suggests extremely rapid
growth rates—about 30% per time period at maximum sustainable yield
(MSY), which is the average maximum catch that can be removed under
existing environmental conditions over an indefinite period without caus-
ing the stock to be depleted, assuming that removals and natural mortality
are balanced by stable recruitment and growth. While this may be appro-
5The first edition of this textbook used the term critical depensation level instead of minimum
viable population. Critical depensation is useful when referring to ecosystems that cannot sustain
themselves when stocks fall below a certain level, since it makes little sense to talk about size of
an ecosystem as a population. However, minimum viable population, in general, is a more intuitively
obvious phrase.
priate for small, rapidly reproducing species, growth rates for many eco-
nomically important species are on the order of 1% per year or less.
THINK ABOUT IT!
Prior to the arrival of humans in North America, where on the graph in
Figure 6.1 would you locate American bison populations? Ten thou-
sand years later, approximately where on the graph do you think bison
harvests occurred? After the introduction of the horse, where do you
think they occurred? After the introduction of the rifle? After the settling
of the frontier by Europeans? What impact would the conversion of the
Great Plains to agriculture have on the sustainable yield curve for
bison?
What happens when we remove some fish, for instance, from a popu-
lation at carrying capacity? In terms of the graph, we represent this as a
harvest at point Q. As there is zero net productivity at our starting point
K, harvest Q reduces the stock of fish by the quantity Q-K, and we find
ourselves at stock R′′. At the lower population stock, there are fewer fish
competing for available food, shelter, and breeding grounds, and the re-
maining fish get more of each than they would under more crowded con-
ditions. This greater resource abundance per individual leads to increased
growth rates and fertility. With less competition for spawning areas, a
higher percentage of eggs are laid in desirable locations, thereby increas-
ing recruitment.6 In addition, most species’ growth is fastest in their youth
(as a percentage of biomass) and slower as they age. If we harvest larger,
older fish, the remaining population has a higher percentage of recruits
and potentially faster net growth.
At stock R′′, we could continue annual harvests forever at point R′, just
equal to the annual rate of increase of the stock. Any harvest below the
sustainable yield curve would be less than the annual growth rate. Flows
can accumulate into stocks, and the population would increase. Any har-
vest above the sustainable yield curve would reduce the stock even fur-
ther. The arrows above and below the curve indicate the direction of
change in stock for harvests in each of these regions. For example, a har-
vest at point R would reduce the population to stock S′′. At S′′, per-capita
resource abundance would be even greater than at R′′, increasing net an-
nual growth and sustainable harvest from R′ to S.
Over a certain range, lower population stocks can generate higher sus-
tainable yields, but this obviously cannot go on forever. Eventually, the
breeding population is insufficient to sustain high yields. Insufficient
6Note that a harvest at point R decreases the stock in the subsequent period by an amount
equal to R-R′ (as measured on the y-axis), and a harvest at point S′ increases the next period’s stock
by S-S′. Thus, R-R′ and S-S′ as measured on the y-axis equal S′′-R′′ as measured on the x-axis.
Minimum Viable Population, Maximum
Box 6-2 Sustainable Yield, and Uncertainty
Using the notion of maximum sustainable yield to help communicate
ideas in ecological economics is a far cry from using the concept as a
tool in resource management. In reality, the MSY will vary dramatically
from year to year in response to climatic cycles such as the El Niño
Southern Oscillation (ENSO), changes in populations of predator and
prey species of the species of interest, changes in pollution levels, and
a broad array of other ecological changes and cycles. Under the most
stable conditions, natural variability can mask the effects of economic
exploitation, and the scale of human impacts is rapidly changing the
global ecosystem. Science relies on replication and controls, and nei-
ther of these is possible when dealing with a unique species in a highly
complex and rapidly changing ecosystem. We cannot scientifically esti-
mate MSY accurately enough for use in resource management.a
Where the minimum viable population lies, and whether or not one
exists for a given species or population, is similarly marked by extreme
uncertainty. When a population becomes small enough, it is more sus-
ceptible to stochastic events and the negative impacts of inbreeding.
For the North American passenger pigeon, probably once the most
numerous bird on Earth, numbering in the billions but driven extinct in
a matter of decades, it appears the minimum viable population was
quite high, perhaps because of its colonial nesting habit. At the other
extreme, the Mauritius Kestrel has rebounded from a known population
of 6 in 1974 to a present population of over 600, though it would have
almost certainly gone extinct without substantial conservation efforts,
and inbreeding may make it highly susceptible to disease or other sto-
chastic shocks. There is currently a debate over whether or not the blue
whale and some populations of North Atlantic cod are below their mini-
mum viable population. Obviously, experiments to scientifically deter-
mine minimum viable populations could not be replicated, as the first
trial could wipe out the species in question.
On the other hand, uncertainty as to where the MSY is does not
mean that we are not overshooting it, nor does it relieve policy makers
of the responsibility to decide acceptable levels of offtake. As a general
rule, the higher the uncertainty, especially in the presence of irre-
versibility (extinction), the more conservative (i.e., lower) the
allowed offtake should be.
aD. Ludwig, R. Hilborn, and C. Walters, Uncertainty, Resource Exploitation, and Conser-
vation: Lessons from History, Science 260:17, 36 (1993).

numbers of eggs lead to insufficient recruits, reducing yield in spite of su-

perabundant resources per individual within the population. This means
that at some point there is a maximum sustainable yield, MSY on Figure
6.1. Again, we caution that the MSY may vary dramatically from year to
year, and there is probably no way we can accurately estimate it in any
given year. Although it is useful as a pedagogical device, it has very little
value as a calculating device for setting annual quotas.7 If harvests con-
tinue to be greater than annual yield, the population will continue to fall.
Eventually, fish will become too scarce to find each other for breeding, and
other ecological mechanisms (many only poorly understood) can break
down. This means that at some population above zero, we may reach a
minimum viable population, at which point the rate of growth is zero.
Below this point the population enters into spontaneous decline, that is,
death rates exceed birth rates.
Note that the sustainable yield curve enters into negative numbers
below the minimum viable population. This means that in order to sus-
tain the population, a negative harvest is needed; that is, recruits have to
be placed into the system every year just to maintain the existing stock.
Harvest is still possible anywhere below the 45-degree line, but this will
simply lead to even more rapid extinction (or extirpation) of the popula-
tion. Unfortunately, we do not know what the minimum viable popula-
tion is and can only make very rough (and highly contentious) estimates
of where the maximum sustainable yield occurs.
The same basic concepts explained here apply to plant species or even
plant communities. For example, what happens when we clear some
trees from a virgin forest for timber? Light, water, and nutrients become
available to the other trees already there, accelerating their growth, and
space becomes available for new seedlings to sprout. This process can
initially speed up the growth rate and increase the sustainable yield of a
forest. Soon, however, sources of new seeds become more distant from
the cleared land, and recruitment slows. Nutrients are lost from the soil
as trees are removed. Trees of the same species are too far apart to cross-
pollinate, resulting in sterile seeds or the problems of inbreeding. The
sustainable yield begins to fall. And as we described above, the removal
of ecosystem structure can dramatically affect ecosystem function, with
the potential of further reducing the capacity of the forest to reproduce
itself. Thus, like animal populations, a stock of forest will show a maxi-
mum sustainable yield and a minimum viable size.
We will return to this analysis in Chapter 12 when we examine the mi-
croeconomics of biotic resource allocation.
7D. Ludwig, R. Hilborn, and C. Walters, Uncertainty, Resource Exploitation, and Conserva-
tion: Lessons from History, Science 260:17, 36 (1993).

Cassowaries, Minimum Viable Populations,

Box 6-3 and Critical Depensation of Ecosystems
In the rainforests of northeastern Australia, up to 100 species of large-
seeded fruit trees depend almost entirely on a single bird species for
distribution. This bird, the cassowary, is a large ratite, an ostrich-like
bird that lives in the forest. It is the only animal known in the region ca-
pable of swallowing and transporting very large seeds, up to 2 kg of
which can be found in a single scat. Evidence suggests that some seeds
must pass through the digestive tract of a cassowary before they can
germinate.
Cassowaries need large home territories to survive, especially in the
highland forests. As forests are cleared in a patchwork pattern, few areas
remain that can sustain a viable cassowary population. Without cas-
sowaries, many trees in the region will be unable to disperse, and some
may not even be able to germinate. Eventually, these species are likely
to go extinct. Other plants and animals depend on these species, and
they too will go extinct, igniting a chain reaction of extinction in species
that may in turn depend on them. The net result could be a dramatic
change in forest composition, leading to a qualitatively different ecosys-
tem. The entire process could take a very long time. It might not be no-
ticed until centuries after it is too late.a
Such examples of mechanisms for critical depensation are just a few
of the possibilities that have been proposed. Again, we must emphasize
that we really have little idea where maximum sustainable yields or criti-
cal depensation points lie. Ignorance, uncertainty, and variability are our
constant companions in the real world.
aJ. Bentrupperbaumer, Conservation of a Rainforest Giant, Wingspan 8(Dec.): 1–2.
(1992). Also extensive personal communications.
I Ecosystem Services
In our discussion of ecosystem structure and function, we explained why
forests need the functions generated by forests to survive, but we also
hinted at the presence of extensive benefits that ecosystem functions pro-
vide for humans. We call an ecosystem function that has value to human be-
ings an ecosystem service. For example, forested watersheds help maintain
stable climates necessary for agriculture, prevent both droughts and
floods, purify water, and provide recreation opportunities—all invaluable
services for watershed inhabitants. But ecosystems provide many more
services, of course. Unfortunately, we are unsure exactly how ecosystem
structure creates ecosystem services, and we are often completely unaware
of the services they generate. For example, prior to the 1970s, most peo-
ple were unaware that the ozone layer played a critical role in making our
planet habitable.8 If we also take into account the tightly interlocking na-
ture of ecosystems, it’s safe to say that humans benefit in some way from
almost any ecosystem function.
We just described forests as a stock of trees that generates a flow of
trees. Now we want to look at the forest as a creator of services; as such,
it is very different from a stock of trees. A stock of trees can be harvested
at any rate; that is, humans have control over the rate of flow of timber
produced by a stock of trees. Trees can also be harvested and used imme-
diately or stockpiled for later use. Ecosystem services are fundamentally
different. We cannot use climate stability at any rate we choose—for ex-
ample, drawing on past or future climate stability to compensate for the
global warming we may be causing today. Nor can we stockpile climate
stability for use in the future. Nor does climate stability become a part of
what it produces. If timber is used to produce a chair, the timber is em-
bodied in that chair. If climate stability is used to produce a crop of grain,
that grain in no way embodies climate stability. Furthermore, climate sta-
bility is not altered by the production of a crop of grain (unless perhaps
the grain is grown on recently deforested land, but still it is the deforesta-
tion and not the grain that affects climate stability).
Intact ecosystems are funds that provide ecosystem services, while
their structural components are stocks that provide a flow of raw materi-
als. However, recall that stock-flow resources are used up, and fund-
service resources are worn out. But when ecosystems provide valuable
services, this does not “wear them out.” The fact is, however, that ecosys-
tems would “wear out” if they did not constantly capture solar energy to
renew themselves. The ability of ecosystem fund-services to reproduce
themselves distinguishes them in a fundamental way from manmade
fund-services. Depreciating machines in a factory do not automatically re-
produce new machines to replace themselves.
Examples of ecosystem services provided by a forest may help clarify
the concept. Costanza et al. describe 17 different goods and services gen-
erated by ecosystems.9 Forests provide all of these to at least some degree.
Of these, food and raw materials are essentially stock-flow variables,
though their ability to regenerate is a fund-service. The remaining fund-
service variables included are described in Table 6.1.
8As further evidence of the extreme uncertainty concerning ecosystem function and human
impacts upon it, in 1973 physicist James Lovelock, famous for the Gaia hypothesis, to his later re-
gret stated that fluorocarbons posed no conceivable hazard to the environment. M. E. Kowalok,
Common Threads: Research Lessons from Acid Rain, Ozone Depletion, and Global Warming, En-
vironment 35(6):12–20, 35–38 (1993).
9R. Costanza et al., The Value of the World’s Ecosystem Services and Natural Capital, Nature
387:256, Table 2 (1997).

I Table 6.1
EXAMPLES OF SERVICES PROVIDED BY ECOSYSTEMS
Ecosystem Service Examples from Forests
Gas regulation Trees store CO2 and growing trees create O2; forests can clean SO2 from the atmosphere.
Climate regulation Greenhouse gas regulation; evapotranspiration and subsequent transport of stored heat
energy to other regions by wind; evapotranspiration, cloud formation, and local rainfall;
effects of shade and insulation on local humidity and temperature extremes.
Disturbance regulation Storm protection, flood control (see water regulation), drought recovery, and other
aspects of habitat response to environmental variability controlled mainly by vegetation
structure.
Water regulation Tree roots aerate soil, allowing it to absorb water during rains and release it during dry
times, reducing risk and severity of both droughts and floods.
Water supply Evapotranspiration can increase local rainfall; forests can reduce erosion and hold stream
banks in place, preventing siltation of in-stream springs and increasing water flow.
Waste absorption Forests can absorb large amounts of organic waste and filter pollutants from runoff; some
capacity plants absorb heavy metals.
Erosion control and Trees hold soil in place, forest canopies diminish impact of torrential rainstorms on soils,
sediment retention diminish wind erosion.
Soil formation Tree roots grind rocks; decaying vegetation adds organic matter.
Nutrient cycling Tropical forests are characterized by rapid assimilation of decayed material, allowing little
time for nutrients to run off into streams and be flushed from the system.
Pollination Forests harbor insects necessary for fertilizing wild and domestic species.
Biological control Insect species harbored by forests prey on insect pests.
Refugia or habitat Forests provide habitat for migratory and resident species, creating conditions essential
for reproduction of many of the species they contain.
Genetic resources Forests are sources for unique biological materials and products, such as medicines,
genes for resistance to plant pathogens and crop pests, ornamental species.
Recreation Ecotourism, hiking, biking.
Cultural Aesthetic, artistic, educational, spiritual, and scientific values of forest ecosystems.
THINK ABOUT IT!

Of the ecosystem services in Table 6.1, which are rival and which are
excludable? Which would be impossible to make excludable?
Again we emphasize that the precise relationship between the quantity

and quality of an ecosystem fund and the services it provides is highly un-
certain and is almost certainly characterized by nonlinearities, thresholds,
and emergent properties. We can say with reasonable confidence that the
larger an ecosystem fund and the better its health, the more services it is
likely to generate. As we deplete or degrade a complex ecosystem fund,
we really cannot predict what will happen with any reasonable probabil-
ity. Since we have defined service as an anthropocentric concept, we do
know that it can be dramatically affected by human presence and use and
not just by abuse. For example, a highly degraded forest in an urban set-
ting may offer more water regulation and more recreational and cultural
services (as measured by benefits to humans) than a pristine forest remote
from human populations. Forests near orchards or other insect-pollinated
crops may offer far more valuable pollination services.
Perhaps even more critical for the economic problem of efficient allo-
cation of ecosystem services is their spatial variation. To use an example
already described, large tropical forests can regulate climate at the local
level, the regional level, and the global level. Flood control and water pu-
rification provided by forests may benefit only select populations border-
ing local rivers and floodplains, and the provision of habitat for migratory
birds may benefit primarily populations along the migratory pathways.
Ecosystem services have some other characteristics that make them ex-
tremely important economically. Probably most important, it is unlikely
that we can develop substitutes for most of these services, including the
provision of suitable habitat for humans. We scarcely understand how
these services are generated, and we are not aware of all of them. At the
cost of some $200 million, a billionaire named Edward Bass initiated the
Biosphere Two project in Arizona to see if he could develop substitutes for
these services sufficient to sustain only eight people. The project failed.
Imagine creating substitutes for billions of people! In addition, most
ecosystem services are nonrival—if I benefit from a forest’s role in reduc-
ing floods, providing habitat for pollinators, or regulating atmospheric
gases, it does not diminish the quantity or quality of those services avail-
able to anyone else. Many ecosystem services (though certainly not all) are
nonexcludable by their very nature as well.
The Relationship Between Natural Capital Stocks and Funds

In review, the structural elements of an ecosystem are stocks of biotic and
abiotic resources (minerals, water, trees, other plants, and animals), which
when combined together generate ecosystem functions, or services. The

use of a biological stock at a nonsustainable level in general also depletes
a corresponding fund and the services it provides. Hence, when we har-
vest trees from a forest, we are not merely reducing the stock of trees but
are also changing the capacity of the forest to create ecosystem services,
many of which are vital to our survival. The same is true for fish we har-
vest from the ocean, except we know even less about the ecosystem serv-
ices produced by healthy oceanic ecosystems.
The relationship between natural capital stock-flow and fund-service
resources illustrates one of the most important concepts in ecological eco-
nomics: It is impossible to create something from nothing; all economic
production requires a flow of natural resources generated by a stock of
natural capital. This flow comes from structural components of ecosys-
tems, and the biotic stocks are also funds that produce ecosystem services.
Therefore, an excessive rate of flow extracted from a stock affects not only
the stock and its ability to provide a flow in the future but also the fund
to which the stock contributes and the services that fund provides. Even
abiotic stocks (i.e., elements and fossil fuels) can be extracted and con-
sumed only at some cost to the ecosystem. In other words, production re-
quires inputs of ecosystem structure. Ecosystem structure generates
ecosystem function, which in turn provides services. All economic pro-
duction thus has an impact on ecosystem services, and because this im-
pact is unavoidable, it is completely internal to the economic process.
I Waste Absorption Capacity

But this is only half the story. The laws of thermodynamics ensure that raw
materials once used by the economic system do not disappear but instead
return to the ecosystem as high-entropy waste. They also ensure that the
process of producing useful (ordered) products also produces a more than
compensating amount of disorder, or waste. Much of this waste can be as-
similated by the ecosystem. Indeed, waste assimilation and recycling are
ecosystem services on which all life ultimately depends. However, as a
fund-service, waste absorption occurs only at a fixed rate, while conver-
sion of stock-flow resources into waste occurs at a rate we can choose.
Waste absorption capacity is a sink for which we have control over the
flow from the faucet but not over the size of the drain. The removal of
ecosystem structure also affects the ability of the ecosystem to process
waste. If we discharge waste beyond the ecosystem’s capacity to absorb it,
we can reduce the rate at which an ecosystem can absorb waste, which
makes the waste accumulate more quickly. In time, the waste buildup will
affect other ecosystem functions, though we cannot always predict which
services will be affected and when.
A specific example can help illustrate these points. When we first begin
to dump wastes, such as raw sewage and agricultural runoff, into a pris-
tine lake, they will be heavily diluted and cause little harm. Higher waste
loads may threaten humans who use the lake with intermittent health
problems from bacteria and noxious chemicals contaminating the sewage,
and water becomes unsuitable for drinking without prior treatment. In-
creasing nutrients allow bacterial and algal populations to thrive, increas-
ing the ability of the system to process waste but reducing a number of
other ecosystem services. Fish will begin to accumulate noxious com-
pounds present in the waste stream and become inedible. Pollution-
sensitive species will be extirpated. Yet more waste may make the water
unsuitable for drinking even after extensive processing, and eventually it
will become too contaminated for industrial use. Excess nutrients eventu-
ally lead to eutrophication, where algal and bacterial growth absorbs so
much oxygen during the night10 and during the decay process that fish,
amphibians, and most invertebrate species die out. Birds and terrestrial
animals that depend on the lake for water and food will suffer. With even
greater waste flows, even algae may fail to thrive, and we have surpassed
the waste absorption capacity of the system. Waste begins to accumulate,
further decreasing the ability of algae to survive and leading to a more
rapid accumulation of waste even if the waste flow is not increased any
more. The system collapses.
Prior to the point where waste flows exceed the waste absorption ca-
pacity, a reduction in flows will allow the system to recuperate. After that
point, it may not. Similar dynamics apply to other ecosystems. If the
ecosystem in question provides critical life-support functions, either lo-
cally or globally, the costs of exceeding the waste absorption capacity of
an ecosystem are basically infinite, at least from the perspective of the hu-
mans it sustains.
In general, ecosystems have a greater ability to process waste products
from biological resources and a much more limited capacity to absorb
manmade chemicals created from mineral resources. This is because
ecosystems evolved over billions of years in the presence of biological
wastes. In contrast, products such as halogenated cyclic organic com-
pounds and plutonium (two of the most pernicious and persistent pollu-
tants known) are novel substances with which the ecosystem has had no
evolutionary experience and therefore has not adapted.
In contrast to many ecosystem services, waste absorption capacity is
rival. If I dump pollution into a river, it reduces the capacity of the river
10While growing plants are net producers of oxygen and absorbers of CO , they also require
2
oxygen for survival. During the day, photosynthesis generates more oxygen than the plants con-
sume, but at night they consume oxygen without producing any. Average oxygen levels may be
higher, but the lowest levels determine the ability of fish and other species to survive.
to assimilate the waste you dump in. It is also fairly simple to establish in-
stitutions that make waste absorption excludable, and many such institu-
tions exist.
The bottom line is that the laws of thermodynamics tell us that natural
resources are economic throughputs. We must pay close attention to
where they come from and where they go.
Table 6.2 summarizes some of the important characteristics of the three
biotic resources. We will discuss these characteristics and examine their
policy relevance in greater detail in Chapter 12 and Part VI.
The points to take away from this chapter deserve reiteration. First, hu-
mans, like all animals, depend for survival on the ability of plants to cap-
ture solar energy in two ways: directly as a source of energy and indirectly
through the life-support functions generated by the global ecosystem,
which itself is driven by the net primary productivity of plants. There are
no substitutes for these life-support functions. Second, every act of eco-
nomic production requires natural resource inputs. Not only are these in-
puts being used faster than they can replenish themselves, but when these
structural elements of ecosystems are removed, they diminish ecosystem
function. Third, every act of economic production generates waste. Waste
has a direct impact on human well-being and further diminishes ecosys-
tem function. While the removal of mineral resources may have little di-
rect impact on ecosystem function, the waste stream from their extraction
and use is highly damaging to ecosystems and human well-being in the
long run. As the economy expands, it depletes nonrenewable resources,
displaces healthy ecosystems and the benefits they provide, and degrades
remaining ecosystems with waste outflows.
Biotic resources are unique because they are simultaneously stocks and
funds, and their ability to renew themselves is a fund-service. This means
I Table 6.2
ECONOMIC CHARACTERISTICS OF BIOTIC RESOURCES
Can Be
Biotic Stock-Flow or Made Rival Between
Resource Fund-Service Excludable Rival Generations Substitutability
Renewable Stock-flow Yes Yes Depends on High at margin,
Resources rate of use ultimately
nonsubstitutable
Ecosystem Fund-service For most, no For most, no No Low at margin,

Services nonsubstitutable
Waste Absorption Fund-service Yes Yes Depends on Moderate at margin,

Capacity rate of use nonsubstitutable
that ultimately economic scale is determined by the amount of fund-

services provided in a given year, where one of those fund-services is the
ability of renewable natural resources to renew. Biotic resources have a
particularly large impact on scale because they ultimately have no substi-
tutes, and we cannot survive without them.

I Ecosystem structure I Carrying capacity
I Ecosystem function I Minimum viable population
I Ecosystem services I Critical depensation
I Stock-flow and fund-service I Maximum sustainable yield
resources I Waste absorption capacity
I Risk, uncertainty, ignorance
CHAPTER
7
From Empty World
to Full World
G iven the undeniable importance of entropy to the economic process
and the resulting fact that “sustainable economic growth” is an oxy-
moron,1 how do we explain the unwavering devotion to continuous eco-
nomic growth by economists, policy makers, and the general public in the
face of ecological and natural resource limits? Apparently, people believe
that the economic system faces no limits to growth or that the limits are
far off. The laws of thermodynamics ensure that there are limits to growth.
Now we must briefly address the question of how close those limits are.
Certainly for most of human history, including the time when modern
economic theory was being developed, human populations and levels of
resource use were quite low. Material and energetic limits to growth ap-
peared so far off that it seemed sensible to ignore them and concentrate on
developing a system that efficiently allocated the much scarcer labor, cap-
ital, and consumer goods. But since the development of market economies
and neoclassical explanations thereof, both human populations and per-
capita levels of resource use have been increasing exponentially. The suc-
cess of the market system reduced the relative scarcity of market goods
and increased that of nonmarket goods and services provided by the
1To reiterate, we do not believe that sustainable growth of the “psychic flux” of satisfaction
(what we would call development, not growth) is an oxymoron as long as that flux is not produced
by ever-increasing natural resource consumption. However, as far as we know, economic growth
as measured by GNP has never occurred without increased throughput. Even when each unit of
GNP requires fewer resources, the net outcome has always been greater throughput. While this
need not be the case, the free market economy seems poorly suited for promoting the activities
that provide improved human well-being without increasing throughput.
111
Box 7-1 How Close Are We to a Full World?

Exponential growth occurs when a system keeps growing at a certain
rate. For example, from 1900 to 2000, the per-capita material output of
the global economy grew at about 2.3% per year. One can calculate the
doubling time of a given growth rate by dividing 72 by that growth rate.
This means that per-capita output doubled more than three times dur-
ing the twentieth century. Over the same period, the human population
has increased from 1.6 billion to 6.1 billion, almost a fourfold increase.
The total material output increased more than 36 times in the twenti-
eth century.a How many more times can our material output double?
Our situation may have parallels to a well-known riddle. If the area
of a petri dish covered in bacteria doubles every hour, you inoculate
the dish at noon on day one and it is completely full at noon two days
later (and thereafter the population crashes because it has exhausted
its food source and inundated the petri dish with waste), when is the
dish half full? The answer, of course, is at 11 A.M. on the final day. At 9
A.M., 7⁄8 of the resources available for continued growth are still pres-
ent. The question right now for humans is: How close is it to noon?
Humans, of course, are very different from bacteria, and the Earth is
different from a petri dish. Humans can control their rate of reproduc-
tion and, to an extent, the quantity of resources they use. The Earth
hosts numerous ecosystems capable of providing renewable resources
and processing wastes. However, human adaptation to resource
scarcity requires taking time to develop new technologies, new institu-
tions, and new ways of thinking—perhaps a great deal of time. Essen-
tially, the closer it is to noon, the less time we have to develop and
implement the necessary changes to show that we actually are sub-
stantially different from bacteria in a petri dish. (See Figure 7.1.)
aCalculations by author from data found in Chapter 5 of J. B. Delong, Macroeconomics,
Burr Ridge, IL: McGraw-Hill Higher Education, 2002.
Figure 7.1 • How close are we to a full world?

Chapter 7 From Empty World to Full World • 113
sustaining system. What follows is a very quick assessment of how full the
world is and how close we are to resource exhaustion.
THINK ABOUT IT!
The world is always “full” of some things and “empty” of others. In the
“full world,” what is it that the world is relatively full of? Relatively
empty of? How is the fullness with respect to some things related to the
emptiness with respect to others?
I Fossil Fuels
As fossil fuels run the world economy and are among the most well stud-
ied of the resources required to sustain us, we will assess their limits first.
At first glance, exhaustion hardly seems imminent. Economists tell us that
price is a measure of scarcity, yet the price of crude oil averaged $24 per
barrel between 1899 and 1999 (in 2008 dollars), when the price was only
$23.60.2 However, as we mentioned earlier, we can extract fossil fuels at
virtually any rate we want, and it is the scarcity of flow that determines
prices, not the scarcity of stocks (a point we return to in Chapter 11). In
the regions with the vastest reserves, installed extraction capacity, more
than the size of underground stocks, determines flow rates. Best estimates
suggest that if we continue to extract oil at the same rate, we will exhaust
probable stocks in about 40 years, yet the Energy Information Adminis-
tration estimates that global demand for oil will increase by nearly 40%
over the next 30 years.3 As we said above, the net energy returns to fossil
fuel exploration are declining dramatically. The same is true for new dis-
coveries, which peaked in 1962 at 40 billion barrels per year4 and fell to
6 billion barrels per year in the 1990s. Consumption currently (2008)
stands at 31 billion barrels per year, exceeding new discovery rates by a
factor of 2 to 6.5 Although the rate of increase in global oil consumption
began to decline after 1973, the world still used over twice as much oil
2British Petroleum, Statistical Review of World Energy 2009. Online: http://www.bp.com.

3Energy Information Administration, Annual Energy Outlook 2006, Washington, DC, February
2006.
4J. J. MacKenzie, Oil as a Finite Resource: When Is Global Production Likely to Peak? World
Resources Institute, 2000. Online: http://www.wri.org/wri/climate/jm_oil_000.html.

5When a new oil field is discovered, it is very difficult to say exactly how much oil exists. Also,
some sources report increased estimates of recoverable oil from a previously discovered source as
a new discovery, while others do not, hence the discrepancy in estimates. MacKenzie (ibid.) cites
a 2:1 ratio for 1996, and L. F. Ivanhoe cites a 6:1 ratio for major discoveries during the 1990s.
Hubbert Center Newsletter #2002/2, M. King Hubbert Center for Petroleum Supply Studies, Pe-
troleum Engineering Department, Colorado School of Mines, Golden, CO, 2002.
Figure 7.2 • A. Hubbert curve for oil discovery. The bars represent the average
amount of crude oil discovered worldwide during each 5-year period from 1912
to 1992. The line known as the Hubbert curve is the weighted average of global
oil discovered from 1915 to 1992. (Source: Adapted from L. F. Ivanhoe, King Hub-
bert, updated. Hubbert Center Newsletter #97/1. Online: http://hubbert.mines
.edu/news/v97n1/mkh-new2.html.)
since 1973 than for all of human history prior to 1973.6 What is the net
result of all this?
M. King Hubbert, while working as a petroleum geologist for Shell Oil
Company, developed a theory of nonrenewable resource extraction, graph-
ically depicted in the Hubbert curve. Figure 7.2 shows a Hubbert curve
for oil discoveries using actual data, and Figure 7.3 shows a Hubbert
curve for oil production that includes estimates of future production. Hub-
bert hypothesized that peak production must follow peak discovery with a
time lag. In 1954, Hubbert used this theory to predict that oil production
in the U.S. would peak between 1967 and 1971—a prediction that was
treated with considerable skepticism. In reality, it peaked in 1970. Apply-
ing Hubbert’s methods, leading industry experts in the 1990s predicted
that oil production would peak sometime between 2003 and 2020, fol-
lowed by a decline.7 Sophisticated analyses of oil prices accounting for
both scarcity and information effects suggested that oil prices would rise
6The area under the curve in Figure 7.3 shows total oil production, which is almost identical
to consumption. You can see that the area under the curve from 1973 to the present is nearly two
and a half times greater than that from 1869 to 1973.
7C. J. Campbell and J. H. Laherrère, The End of Cheap Oil, Scientific American, March 1998.
Figure 7.3 • A Hubbert curve for oil production. Global production of oil, both
conventional and unconventional (solid black line), recovered after falling in
1973 and 1979. But a more permanent decline is less than 10 years away, ac-
cording to the authors’ model, based in part on multiple Hubbert curves
(dashed lines). A crest in the oil produced outside the Persian Gulf region now
appears imminent. (Source: Adapted from C. J. Campbell and J. H. Laherrère,
The End of Cheap Oil, Scientific American, March 1998.)
suddenly and sharply.8 Despite the highest global economic growth rates
in decades and unprecedented demand for oil, oil production essentially
stagnated from late 2004 to mid-2007 while prices doubled. Though pro-
duction began increasing slightly from mid-2007 to July 2008, prices again
doubled, leading to a simultaneous peak in production and prices. The
onset of a global recession dragged prices and production back down, but
in 2009 oil prices had begun to rise again even as the recession worsened.
While oil exhaustion may not be imminent, we believe that oil production
has already plateaued, and while fluctuations will occur, the trend in com-
ing years will be toward steadily declining output and rising prices.
Given the vast supplies of solar power available as a substitute, does it
matter if we exhaust oil supplies? Developing solar energy as a substitute
will take considerable time. Also, because solar energy strikes the Earth as
a fine mist, large areas of land are needed to capture that energy in sig-
nificant quantities. With current technologies and without disrupting
agriculture, forestry, or the environment, the amount of solar energy that
could be captured in the U.S. would meet only 20–50% of our current en-
ergy demands. And we could not get around this problem simply by using
8D. B. Reynolds, The Mineral Economy: How Prices and Costs Can Falsely Signal Decreasing
Scarcity, Ecological Economics 31(1):155–166 (1999).

lower-wattage light bulbs. Food production and transportation in the

United States currently consumes at least three times as much hydrocar-
bon energy as it provides in carbohydrate energy.9 One grain-fed steer
“consumes” some 284 gallons of petroleum in the process of becoming
our dinner.10 Returning to animal traction to power our farms would re-
quire additional land devoted to fodder to feed the draft animals. We can-
not sustain economic growth without cheap fossil fuels, but far more
important, in the absence of radical changes in agricultural technology, we
will not even be able to sustain food production.
I Mineral Resources
Mineral resources are also growing scarcer. As noted earlier, the richest,
most available ores are used first, followed by ores of decreasing quality.
We previously used hematite ore from the Mesabi range in Minnesota,
which is about 60% pure iron. That ore is now exhausted, and we must
use taconite ore, at about 25% pure iron.11 The situation with other ores
is similar. At least metal can be recycled, and other materials are adequate
substitutes. If we consider topsoil as a mineral resource, the situation
looks more serious. Rates of topsoil depletion in the U.S. are currently 100
times the rate of formation.12 Globally, experts estimate that 40% of agri-
cultural land is seriously degraded, and this number is as high as 75% in
some areas.13 Currently, the most widely used substitute for declining soil
fertility is petroleum-based fertilizers.
I Water
Among the more threatening of the imminent shortages is that of fresh-
water. While water is the quintessential renewable resource, thanks to the
hydrologic cycle, global water consumption has tripled over the last 50
9D. Pimentel and M. Pimentel, Land, Energy and Water: The Constraints Governing Ideal U.S.
Population Size, 1995. Online: http://www.npg.org/forum_series/land_energy&water.htm. Nega-

tive Population Growth, Inc. Forum Series. Georgescu-Roegen rightly pointed out that the notion
of producing carbohydrates from hydrocarbons was absurd, and when this was first presented as
an option, he also accurately predicted that we would move in the other direction first. N.
Georgescu-Roegen, The Entropy Law and the Economic Process, Cambridge, MA: Harvard University
Press, 1971. However, in the short term direct conversion of oil to food might be more efficient
than Western agriculture.
10M. Pollan, Power Steer, New York Times Magazine, March 31, 2002.
11J.Hanson, Energetic Limits to Growth. Online: http://www.dieoff.com/page175.htm. Also
appeared in Energy Magazine, Spring 1999.
12D. Pimentel and M. Pimentel. Land, Energy and Water: The Constraints Governing Ideal
U.S. Population Size, 1995. Online: http://www.npg.org/forum_series/land_energy&water.html.

13World Resources Institute, People and Ecosystems: The Fraying Web of Life, Washington, DC:
WRI, 2000.
years, and it continues to climb. Humans are pumping rivers dry and
mining water from aquifers faster than it can be replenished. While global
climate change may lead to a wetter climate overall as increased evapora-
tion leads to increased rainfall, increased evaporation will also dry out the
land much more quickly. Many climatologists believe the net result will be
intense downpours interspersed with severe drying. In addition, global
climate change is likely to affect where water falls, leading overall to
greater risks of both flooding and drought.14
The dominant use of water (70%) is agriculture, so a water shortage
will probably translate into hunger before thirst. The estimated water
deficit (extraction of water greater than the recharge rate) in northern
China is 37 billion gallons, which produces enough food to feed 110 mil-
lion people.15 The Ogallala aquifer in the United States has turned the
arid Western plains into a breadbasket. Water levels have been in steady
decline since the 1950s,16 and diminishing rainfall in the American
West17 is likely to increase demand for aquifer water while reducing dis-
charge rates.
The use of river water for irrigation has already led to one of the
planet’s worst environmental catastrophes in the Aral Sea. On every con-
tinent, important aquifers are falling at rates between 2 and 8 m per
year.18 Currently, nearly one billion people lack access to potable drink-
ing water,19 less than one-third of the world’s population enjoys abundant
water supplies,20 and some studies suggest that nearly 50% of the world’s
population will be living in water shortage areas by 2025.21 The World
Bank warns that continued reduction in aquifers could prove cata-
strophic.22 Fortune magazine suggests that water shortages will make
14 C. J. Vörösmarty, P. Green, J. Salisbury, and R. B. Lammers, Global Water Resources: Vul-
nerability from Climate Change and Population Growth, Science 289:284–288 (July 14, 2000).
15L. Brown, Water Deficits Growing in Many Countries: Water Shortages May Cause Food
Shortages. Earth Policy Institute, Eco-Economy Updates, August 6, 2002.

16V. L. McGuire, Water-Level Changes in the High Plains Aquifer, Predevelopment to 2005 and
2003 to 2005. Reston, VA: U.S. Geological Survey, 2007.

17E. Cook, C. Woodhouse, C. M. Eakin, D. Meko, and D. Stahle. Long-Term Aridity Changes
in the Western United States. Science 306(5698):1015–1018 (2004).

18L. Brown, Water Deficits Growing in Many Countries. Earth Policy Institute, Eco-Economy
Updates, August 6, 2002. Online: http://www.earth-policy.org/Updates/Update15.html.

19WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation. Millennium
Development Goals Assessment Report 2008: Country, Regional and Global Estimates on Water
and Sanitation. New York: UNICEF; Geneva: WHO, 2008.
20Vörösmarty et al., op. cit.
21L. Burke, Y. Kura, K. Kassem, C. Revenga, M. Spalding, and D. Mcallister, Pilot Analysis of
Global Ecosystems: Coastal Ecosystems, Washington, DC: WRI, 2000.
22Brown, op. cit.
water the oil of the twenty-first century, “the precious commodity that de-
termines the wealth of nations.”23
Projections concerning future water supplies are highly uncertain.
First, we lack adequate data.24 Second, consumption patterns and tech-
nology can dramatically change the demand for water. Third, as men-
tioned, climate change can have serious impacts on the hydrologic cycle,
increasing evaporation rates and changing rainfall patterns.25
I Renewable Resources
The fact that we live in a full world is even more obvious when it comes
to “renewable” resource stocks. For virtually every renewable stock of sig-
nificance, the rate of extraction is limited by resource scarcity, not by a
lack of adequate infrastructure. It is a shortage of fish, not fishing boats,
that has stagnated fish harvests over the last few years. The Food and Agri-
culture Organization (FAO) of the United Nations estimates that 11 of the
world’s 15 major fishing areas and 69% of the world’s major fish species
are in decline and in need of urgent management. For instance, cod
catches dropped by 69% from 1968 to 1992. West Atlantic bluefin tuna
stocks dropped by more than 80% between 1970 and 1993.26 Similarly,
it is a shortage of trees, not chainsaws, that limits wood production. As
commercially valuable species are depleted, we turn to harvesting others
that were formerly considered trash. As a result, for both fish and timber,
the number of commercially valuable species has increased dramatically
over recent decades.
Many economists cite this ability to substitute one species for another
as evidence that there are no limits to potential harvests. However, when
one fish species is exhausted because too many boats are going after too
few fish, the whole fishing fleet is available to deplete any new stocks we
identify. Having virtually exhausted rapidly reproducing species such as
cod, we now pursue species such as orange roughy, which may take as
long as 30 years to reach sexual maturity. We run the risk of harvesting
23N. Currier, The Future of Water Under Discussion at “21st Century Talks.” United Nations
Chronicle XL(1) (2003). Online: http://www.un.org/Pubs/chronicle/2003/webArticles/013003_fu-

ture_ of_water.html.
24K. Brown, Water Scarcity: Forecasting the Future with Spotty Data, Science 297(5583): 926–
927 (August 9, 2002).

25Vörösmarty et al., op. cit.
26FAO of the U.N. Focus: Fisheries and Food Security, 2000. Online: http://www.fao.org/focus
/e/fisheries/challeng.htm.
such species to extinction before we even acquire sufficient data to esti-

mate their sustainable yields.27
While there is serious cause for concern for the resource exhaustion of
raw material inputs into the economy, these may pale in significance when
compared to the dangers presented by the depletion and destruction of
ecosystem services. Ecosystem services are destroyed directly by the har-
vest of their structural components, primarily the renewable resources of
which they are composed, and less directly by waste emissions. Forest
cover is currently being depleted in the poorer countries at the rate of
about 140,000 km2 per year,28 and if the World Trade Organization’s ef-
forts to liberalize trade in forest products go forward as planned, the rate
of deforestation is expected to increase.29 The Ramsar convention on wet-
lands is an intergovernmental treaty providing a framework for the con-
servation of wetlands and their resources, yet 84% of the wetlands
supposedly protected by the treaty are threatened.30 While we understand
marine ecosystems less than terrestrial ones, it seems unimaginable that
healthy fish populations do not play a vital role in these ecosystems and
the scarcely understood mechanisms by which they provide ecosystem
services. For example, biodiversity appears to enhance ecosystem pro-
ductivity and stability along with other ecosystem services, and continued
loss of oceanic biodiversity may lead to the total collapse of marine fish-
eries by 2048.31 Virtually all other ecosystems confront similar threats
through depletion of their component stocks.
I Waste Absorption Capacity

People have worried about resource exhaustion at least since the time of
Malthus, but concern over the excessive accumulation of waste is more re-
cent. Every economic activity produces waste. As humans overwhelm the
waste absorption capacity of ecosystems at local and global levels, we suffer
in two ways. First, accumulating toxins have direct negative effects on hu-
mans. Second, pollutants damage ecosystems and degrade the ecosystem
27For example, one study found a 60–70% decline in total biomass of one stock of orange
roughy in less than 10 years of fishing. P. M. Smith, R. I. C. C. Francis, and M. McVeigh “Loss of
Genetic Diversity Due to Fishing Pressure,” Fisheries Research 10(1991):309–316.
28World Resources Institute, People and Ecosystems: The Fraying Web of Life, Washington, DC:
WRI, 2000.
29P. Golman, J. Scott, et al. Our Forests at Risk: The World Trade Organization’s Threat to Forest
Protection (Oakland, CA: Earthjustice, 1999).

30M. Moser, C. Prentice, and S. Frazier, “A Global Overview of Wetland Loss and Degrada-
tion,” Proceedings of the 6th Meeting of the Conference of Contracting Parties of the Ramsar Con-
vention (1996), vol. 10. Online: http://www.ramsar.org/about_wetland_loss.htm.
31B. Worm et al., “Impacts of Biodiversity Loss on Ocean Ecosystem Services,” Science
314(5800)(2006):787–790.
services on which we depend. Accumulating evidence suggests we are

overwhelming the waste absorption capacity of the planet for several
classes of wastes.
The most prominent category of waste in the news today is CO2 emis-
sions. In spite of an impressive ability of ecosystems to absorb CO2, there
is irrefutable evidence that it is currently accumulating in the atmosphere
and near consensus in the scientific community that this has already con-
tributed to global climate change. International recognition of the seri-
ousness of the problem has led to international discussions, but at the
time of this writing, the world’s worst emitter of greenhouse gases has re-
fused to participate in international accords. Even if the United States did
participate, the reductions proposed under the Kyoto protocol would fail
to limit CO2 emissions to the waste absorption capacity of the environ-
ment, and would therefore at best merely slow the rate of global warm-
ing.32 In the absence of major changes in human behavior, global
warming will have dramatic impacts on global ecosystems. This is partic-
ularly true because so many remaining ecosystems are islands in a sea of
humanity, and the species they contain will be unable to leave their islands
in response to changing conditions.
Waste emissions from mineral resources also pose serious threats.
Heavy metals are highly toxic to humans. As these metals are elements,
there is no waste absorption capacity per se; once in the environment or
in our aquifers, they remain indefinitely. These elements are normally
highly diluted in nature or out of reach of living systems; humans have ex-
tracted and purified them and released them into the environment in dan-
gerously high concentrations. Many of them tend to bioaccumulate; when
ingested, they are not released, so predators retain all that has been con-
sumed by their prey. Many fish species have dangerously high levels of
mercury and other metals, which cause human birth defects and worse
when consumed, not to mention their impacts on other species.
Nuclear wastes are also elements and far more toxic than the other
heavy metals. Nuclear wastes do break down, but not on a human time
scale. Plutonium, one of the most toxic substances known, has a half-life
of 24,300 years. At minimum, we must sequester such waste for ten times
that long—nearly fifty times as long as civilization has existed.
Halogenated hydrocarbons are another class of particularly dangerous
manmade mineral wastes. Chlorofluorocarbons (CFCs) are the best
known, and they are now banned. However, many countries continue to
use hydrochlorofluorocarbons (HCFCs). While HCFCs have lower ozone
32Intergovernmental Panel on Climate Change, Climate Change 2007: Synthesis Report. Sum-
mary for Policymakers. Intergovernmental Panel on Climate Change. Cambridge, UK: Cambridge
University Press, 2007.
depleting potential than CFCs, China and India have been increasing their
use by as much as 35% annually. As a result, the greatest recorded de-
crease in the ozone layer occurred in 2006.33 Ozone depletion threatens
not only human health but also global plant and animal life. The Antarc-
tic ozone hole poses a particularly serious threat to phytoplankton pro-
duction in the southern seas. In addition to its key role at the bottom of
the oceanic food chain, phytoplankton may play an important role in se-
questering carbon dioxide, and its depletion may contribute to global
warming.34
Other halogenated hydrocarbons are classified as persistent organic
pollutants (POPs). International negotiators are currently calling for a ban
on the most notoriously harmful POPs. These chemicals are now found in
every ecosystem on Earth. Among their negative traits, some of them seem
to mimic hormones and are capable of affecting the reproductive capacity
of many species. As their name implies, POPs will continue to persist in
the environment for many years to come, in spite of the ban. In the mean-
time, industry is busy introducing new chemicals, many with a very sim-
ilar structure to the most toxic ones, at the rate of over 1000 per year. We
often do not become aware of the negative impacts of these chemicals for
years or even decades. And while it may be possible to perform careful
studies about the damage caused by a single chemical, outside of the lab-
oratory, ecosystems and humans will be exposed to these chemicals in
conjunction with thousands of others.35
Pollution in some areas is becoming so severe that it threatens human
health, ecosystem function, and even large-scale climate patterns. For ex-
ample, a recent study has shown that a 3-km-thick layer of pollution over
South Asia is reducing the amount of solar energy striking the Earth’s sur-
face by as much as 15% in the region yet preventing heat from the energy
that does pass through from leaving. In addition to threatening hundreds
of thousands of premature deaths, the pollution cloud is likely to increase
monsoon flooding in some areas while reducing precipitation by as much
as 40% in others.36
33K. Bradsher, The Price of Keeping Cool in Asia: Use of Air-Conditioning Refrigerant Is Widen-
ing the Hole in the Ozone Layer, New York Times. United Nations Environmental Program (UNEP),
2006 Antarctic Ozone Hole Largest on Record. New York: UNEP, 2006.
34R. C. Smith, B. B. Prezelin, K. S. Baker, R. R. Bidigare, N. P. Boucher, T. Coley, D. Karentz,
S. MacIntyre, H. A. Matlick, D. Menzies, M. Ondrusek, Z. Wan, and K. J. Waters. Ozone Deple-

tion: Ultraviolet Radiation and Phytoplankton Biology in Antarctic Waters, Science 255:952–959
(1992).
35A. P. McGinn, Why Poison Ourselves? A Precautionary Approach to Synthetic Chemicals,
World Watch Paper 153, Washington, DC: World Watch, 2000.

36P. Bagia, Brown Haze Looms Over South Asia, Science 13 (2002).

Ecological Economics-Principles and Applications-3

Uploaded by

Copyright:

Available Formats

Ecological Economics-Principles and Applications-3

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ecological Economics-Principles and Applications-3

Uploaded by

Copyright:

Available Formats

72 • The Containing and Sustaining Ecosystem: The Whole

resource as stock-flow or fund service is a function of its use. A bicycle for

Box 4-1 Stock-flow and Fund-service Resources

The stock-flow and fund-service concepts are important when analyz-

I Excludability and Rivalness

can make nonrival resources such as information excludable, artificially

THINK ABOUT IT!

Is the resource excludable or nonexcludable? (Note that excludability

Is the resource a market good or a nonmarket good?

In general, can you think of any stock-flow resources that cannot be

Exercise 2.2 of the Workbook for Problem-Based Learning accompany-

I Goods and Services Provided by the

which they reproduce. Clearly, species can be exploited to extinc-

BIG IDEAS to remember

1M. Savinar, “Life after the Oil Crash.” http://www.lifeaftertheoilcrash.net/Research.html. This

cluding those of exploration, machinery, transportation, decommission-

Box 5-1 Estimating Oil Stocks

physical Perspective, Science 225:297 (1984).

Society Must Have? Energies 2:25–47(2009).

THINK ABOUT IT!

source. Therefore, as we continue to extract fossil fuels over time, we

The basic equation here is:

Figure 5.1 • Net recoverable energy from fossil fuels.

Campbell and Laherrère, ibid.

less low-entropy energy to process.

Policy Brief #9. Online: http://www.worldwatch.org/worldsummit/briefs/20020806.html. (World

University Press, 1971.

Economics 29:473–484 (1999).

Figure 5.2 • The cumulative extraction of subterranean stock and aboveground

aboveground low-entropy stocks embodied in artifacts in use by (or avail-

most concentrated ores, it becomes cheaper to start mining the lowest-

but can be extracted at virtually any rate we choose. In contrast to fossil

Island Press, 2002.

allow us to stockpile flowing water for later use, which is a characteristic

practical purposes this is irrelevant.

New York: Pergamon Press, 1976.

SELECTED POLICY-RELEVENT CHARACTERISTICS OF ABIOTIC RESOURCES

Minerals Stock-flow Yes Yes Partially High at margin,

Water Context- Context- Context- Stocks, yes; Nonsubstitutable

Ricardian Land Fund-service Yes Yes No Nonsubstitutable

Solar Energy Fund- No No, for No Nonsubstitutable

Water sources are a complex mix of stock-flow and fund-service. But

BIG IDEAS to remember

Big ideas from Chapter 4 that Aboveground and

I Ecosystem Structure and Function

distinction is very relevant to economic analysis. Conventional natural re-

Economics and Liberty, Feb. 21, 2002. Online: http://www.econlib.org/library/Knight/

predicting and managing the impacts of human actions on these func-

Amazon, Indonesia and Mexico), and spontaneous degradation of the re-

Roughly speaking, conventional and natural resource economics has

they are capable of reproduction. Figure 6.1 depicts a renewable natural

Figure 6.1 • The growth, or sustainable yield, curve.

vation: Lessons from History, Science 260:17, 36 (1993).

numbers of eggs lead to insufficient recruits, reducing yield in spite of su-

tion: Lessons from History, Science 260:17, 36 (1993).

Cassowaries, Minimum Viable Populations,

(1992). Also extensive personal communications.

387:256, Table 2 (1997).

Biological control Insect species harbored by forests prey on insect pests.

Recreation Ecotourism, hiking, biking.