Towards an Environmental Macroeconomics

Author(s): Herman E. Daly

Source: Land Economics , May, 1991, Vol. 67, No. 2 (May, 1991), pp. 255-259
Published by: University of Wisconsin Press

Stable URL: https://www.jstor.org/stable/3146415

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

University of Wisconsin Press is collaborating with JSTOR to digitize, preserve and extend
access to Land Economics

This content downloaded from

186.84.20.92 on Tue, 17 Aug 2021 00:37:04 UTC
All use subject to https://about.jstor.org/terms
 SPECULATIONS

Towards an Environmental Macroeconomics

Herman E. Daly

I. INTRODUCTION peter called "vision." One might say that

vision is what the "right brain" supplies
Environmental economics, as it is to the "left brain" for analysis. Whatever
taught
in universities and practiced in government
is omitted from the preanalytic vision can-
agencies and development banks, not is over-
be recaptured by subsequent analysis.
whelmingly microeconomics. The theoreti-
Schumpeter is worth quoting at length on
cal focus is on prices, and the big issue is
this point:
how to internalize external environmental
costs so as to arrive at prices that reflect
In practice we all start our own research from
full social marginal oportunity costs. Oncethe work of our predecessors, that is, we hardly
prices are right the environmental problem ever start from scratch. But suppose we did start
is "solved "-there is no macroeconomic from scratch, what are the steps we should have
dimension. There are, of course, very to take? Obviously, in order to be able to posit
good
reasons for environmental economics to be to ourselves any problems at all, we should first
have to visualize a distinct set of coherent phe-
closely tied to microeconomics, and it nomena is as a worthwhile object of our analytic
not my intention to argue against that con- effort. In other words, analytic effort is of neces-
nection. Rather I want to ask if there is not sity preceded by a preanalytic cognitive act that
a neglected connection between the envi- supplies the raw material for the analytic effort.
ronment and macroeconomics. In this book, this preanalytic cognitive act will
A search through the indexes of three be called Vision. It is interesting to note that
leading textbooks in macroeconomics vision of this kind not only must precede histori-
(Barro 1987; Dornbusch and Fischer 1987; cally the emergence of analytic effort in any
Hall and Taylor 1988) reveals no entries un- field, but also may re-enter the history of every
established science each time somebody teaches
der any of the following subjects: environ-
us to see things in a light of which the source is
ment, natural resources, pollution, deple- not to be found in the facts, methods, and results
tion. Is it really the case, as prominent of the pre-existing state of the science. (Schum-
textbook writers seem to think, that macro- peter 1954, 41)
economics has nothing to do with the envi-
ronment? What historically has impeded The vision of modern economics in gen-
the development of an environmental mac- eral, and especially of macroeconomics, is
roeconomics? If there is no such thing the familiar circular flow diagram. The
as environmental macroeconomics, should macroeconomy is seen as an isolated sys-
there be? What might it look like? tem (i.e., no exchanges of matter or energy
The reason that environmental macro-
economics is an empty box' lies in what
Thomas Kuhn calls a "paradigm," and Environment Department, World Bank, Washing-
what Joseph Schumpeter more descrip-ton, DC.
tively called a "preanalytic vision." As The views here presented are those of the author
and should in no way be attributed to the World Bank.
Schumpeter emphasized, analysis has to
start somewhere-there has to be some- 'The box is not entirely empty. Recent work on
correcting national income accounts, along with appli-
thing to analyze. That something is given
cations of input-output models to environmental prob-
by a preanalytic cognitive act that Schum-
lems, should be noted.

Land Economics * May 1991 * 67(2): 255-59

This content downloaded from

186.84.20.92 on Tue, 17 Aug 2021 00:37:04 UTC
All use subject to https://about.jstor.org/terms
 256 Land Economics May 1991

with its environment)

ion. Resources are seen as "necessary"
in which for
value circulates between firms and house- production, but the amount required can be
holds in a closed loop. What is "flowing in as little as one likes!
a circle" is variously referred to as produc- What is needed is not ever more refined
tion or consumption, but these have physi- analysis of a faulty vision, but a new vision.
cal dimensions, and the circular flow doesThis does not mean that everything built on
not refer to materials recycling, which in the old vision will necessarily have to be
any case could not be a completely closed scrapped-but fundamental changes are
loop, and of course would require energy likely when the preanalytic vision is al-
which cannot be recycled at all. What is tered. The necessary change in vision is to
truly flowing in a circle can only be abstractpicture the macroeconomy as an open sub-
exchange value-exchange value ab- system of the finite natural ecosystem (en-
stracted from the physical dimensions of vironment) and not as an isolated circular
the goods and factors that are exchanged. flow of abstract exchange value, uncon-
Since an isolated system of abstract ex- strained by mass balance, entropy, and fi-
change value flowing in a circle has no de- nitude. The circular flow of exchange value
pendence on an environment, there can is bea useful abstraction for some purposes.
no problem of natural resource depletion, It highlights issues of aggregate demand,
nor environmental pollution, nor any de- unemployment, and inflation that were of
pendence of the macroeconomy on natural interest to Keynes in his analysis of the
services, or indeed on anything at all out- Great Depression. But it casts an impene-
side itself (Georgescu-Roegen 1971; Daly trable shadow on all physical relationships
1985). between the macroeconomy and the envi-
Since analysis cannot supply what the ronment. For Keynes this shadow was not
preanalytic vision omits, it is only to be ex- very important, but for us it is. Once the
pected that macroeconomics texts would macroeconomy is viewed as an open sub-
be silent on environment, natural re- system, rather than an isolated system,
then the issue of its relation to its par-
sources, depletion, and pollution. It is as if
the preanalytic vision that biologists had ofent system (the environment) cannot be
animals recognized only the circulatoryavoided. And the most obvious question is
system and abstracted completely from the how big should the subsystem be relative
digestive tract. A biology textbook's index to the overall system?
would then contain no entry under "assimi-
lation" or "liver." The dependence of the II. THE MACRO-ECONOMICS OF
animal on its environment would not be evi- OPTIMAL SCALE
dent. It would appear as a perpetual motion
machine. Just as the micro unit of the economy
Things are no better when we turn to (firm or household) operates as part o
the advanced chapters at the end of most larger system (the aggregate or macro
macroeconomics texts, where the topic economy), so the aggregate economy i
is growth theory. True to the preanalytic likewise a part of a larger system, the nat
vision the aggregate production is written ral ecosystem. The macroeconomy is an
as Y = f(K,L), i.e. output is a function of open subsystem of the ecosystem and is to-
capital and labor stocks. Resource flows tally dependent upon it, both as a source for
(R) do not even enter! Neither is any waste inputs of low-entropy matter-energy and as
output flow noted. And if occasionally R is a sink for outputs of high-entropy matter-
stuck in the function along with K and L it energy. The physical exchanges crossing
makes little difference since the production the boundary between system and subsys-
function is almost always a multiplicative tem constitute the subject matter of envi-
form, such as Cobb-Douglas, in which R ronmental macroeconomics. These flows
can approach zero with Y constant if only are considered in terms of their scale or to-
we increase K or L in a compensatory fash- tal volume relative to the ecosystem, not in

This content downloaded from

186.84.20.92 on Tue, 17 Aug 2021 00:37:04 UTC
All use subject to https://about.jstor.org/terms
 67(2) Daly: Speculations 257

terms of the problem by providing

price of one the necessary
comp infor-
total flow relative to another. Just as stan- mation and incentive. It does that one thing
dard macroeconomics focuses on the vol- very well. What it does not do is to solve
ume of transactions rather than the relative the problems of optimal scale or of optimal
prices of different items traded, so environ- distribution. The market's inability to solve
mental macroeconomics focuses on the vol- the problem of just distribution is widely
ume of exchanges that cross the boundary recognized, but its similar inability to solve
between system and subsystem, rather than the problem of optimal or even sustainable
the pricing and allocation of each part of scale is not as widely appreciated.2
the total flow within the human economy An example of the confusion that can re-
or even within the nonhuman part of the sult from the nonrecognition of the inde-
ecosystem. pendence of the scale issue from the ques-
tion of allocation is provided by the
The term "scale" is shorthand for "the
physical scale or size of the human pres-
following dilemma (Pearce et al. 1989, 135).
ence in the ecosystem, as measured by pop-Which puts more pressure on the environ-
ment, a high or a low discount rate? The
ulation times per capita resource use." Op-
usual answer is that a high discount rate
timal allocation of a given scale of resource
is worse
flow within the economy is one thing (a for the environment because it
microeconomic problem). Optimal scale speeds
of the rate of depletion of nonrenew-
the whole economy relative to the eco-
able resources and shortens the turnover
system is an entirely different problem (a fallow periods in the exploitation of re-
and
macro-macro problem). The micro alloca-
newables. It shifts the allocation of capital
tion problem is analogous to allocating
and labor toward projects that exploit natu-
optimally a given amount of weight in ralaresources more intensively. But it re-
boat. But once the best relative location of
stricts the total number of projects under-
weight has been determined, there is still
taken. A low discount rate will permit more
the question of the absolute amount ofprojects to be undertaken even while en-
weight the boat should carry, even when couraging less intensive resource use for
optimally allocated. This absolute optimal
each project. The allocation effect of a high
scale of load is recognized in the maritime
discount rate is to increase throughput, but
institution of the Plimsoll line. When the the scale effect is to lower throughput.
watermark hits the Plimsoll line the boat is Which effect is stronger is hard to say, al-
full, it has reached its safe carrying capac- though one suspects that over the long run
ity. Of course if the weight is badly allo- the scale effect will dominate. The resolu-
cated the waterline will touch the Plimsoll tion to the dilemma is to recognize that two
mark sooner. But eventually as the abso- independent policy goals require two inde-
lute load is increased the watermark will pendent policy instruments-we cannot
reach the Plimsoll line even for a boat
whose load is optimally allocated. Opti-
mally loaded boats will still sink under 2This
too can be illustrated in terms of the familiar mi-
much weight-even though they may sink croeconomic tool of the Edgeworth box. Moving to
optimally! It should be clear that optimal
the contract curve is an improvement in efficiency
allocation and optimal scale are quiteofdis-
allocation. Moving along the contract curve is a
change in distribution which may be deemed just or
tinct problems. The major task of environ-
unjust on ethical grounds. The scale is represented by
mental macroeconomics is to designthe
andimensions of the box, which are taken as given.
economic institution analogous to the Plim-
Consequently the issue of optimal scale of the box
soll mark-to keep the weight, the absolute
itself escapes the limits of the analytical tool. A micro-
scale, of the economy from sinkingeconomic
our tool cannot be expected to answer a macro-
economic question. But so far macroeconomics has
biospheric ark.
not answered the question either-indeed, has not
The market of course functions only even asked it. The tacit answer to the implicit question
within the economic subsystem, where it to be that a bigger Edgeworth box is always
seems
does only one thing: it solves the allocation
better than a smaller one!

This content downloaded from

186.84.20.92 on Tue, 17 Aug 2021 00:37:04 UTC
All use subject to https://about.jstor.org/terms
 258 Land Economics May 1991

serve both optimal scale

tosynthesis and
by primary optim
producers, less th
tion with the single policy
energy used instrumen
in their own growth and repro-
discount rate (Tinbergen
duction. NPP is thus the basic 1952).
food re- T
count rate should be allowed to solve the source for everything on earth not capable
allocation problem, within the confines ofof a photosynthesis. Vitousek et al. (1986)
solution to the scale problem provided by calculate that 25 percent of potential global
a presently nonexistent policy instrument (terrestrial and aquatic) NPP is now appro-
that we may for now call an "economicpriated by human beings. If only terrestrial
Plimsoll line" that limits the scale of the NPP is considered, the fraction rises to 40
throughput. percent.3 Taking the 25 percent figure for
Economists have recognized the inde-the entire world it is apparent that two more
pendence of the goals of efficient allocationdoublings of the human scale will give 100
and just distribution and are in generalpercent. Since this would mean zero energy
agreement that it is better to let prices serve left for all nonhuman and nondomesticated
efficiency, and to serve equity with incomespecies, and since humans cannot survive
redistribution policies. Proper scale is awithout the services of ecosystems, which
third independent policy goal and requiresare made up of other species, it is clear that
a third policy instrument. This latter pointtwo more doublings of the human scale is
has not yet been accepted by economists, an ecological impossibility, although arith-
but its logic is parallel to the logic underly-metically possible. Furthermore, the terres-
ing the separation of allocation and distri-trial figure of 40 percent is probably more
bution. relevant since we are unlikely to increase
Microeconomics has not discovered in our take from the oceans very much. Total
the price system any built-in tendency to appropriation of the terrestrial NPP is only
grow only up to the scale of aggregate re- a bit over one doubling time in the future.
source use that is optimal (or even merely Perhaps it is theoretically possible to in-
sustainable) in its demands on the bio- crease the earth's total photosynthetic ca-
sphere. Optimal scale, like distributive jus- pacity somewhat, but the actual trend of
tice, full employment, or price level stabil- past economic growth is decidedly in the
ity, is a macroeconomic goal. And it is a opposite direction. If the above figures are
goal that is likely to conflict with the other approximately correct, then expansion of
macroeconomic goals. The traditional solu- the world economy by a factor of four (two
tion to unemployment is growth in produc- doublings) is not possible. Yet the Brundt-
tion, which means a larger scale. Fre- land Commission calls for economic expan-
quently the solution to inflation is also sion by a factor of five to ten. And the
thought to be growth in real output and a greenhouse effect, ozone layer depletion,
larger scale. And most of all the issue of and acid rain all constitute evidence that we
distributive justice is "finessed" by the have already gone beyond a prudent Plim-
claim that aggregate growth will do more soll line for the scale of the macroeconomy.
for the poor than redistributive measures.
Macroeconomic goals tend to conflict, and IV. HOW BIG SHOULD THE
certainly optimal scale conflicts with any ECONOMY BE?
goal that requires further growth, once the
optimum has been reached. Optimal scale of a single activity is no
strange concept to economists. Indeed m
III. HOW BIG IS THE ECONOMY?

Probably the best index of the scale3The

of definition of human appropriation underlyi
the human economy as a part of the thebio-
figures quoted includes direct use by human bei
sphere is the percentage of human (food,
appro- fuel, fiber, timber), plus the reduction from
potential due to degradation of ecosystems caused
priation of the total world product of humans.
photo- The latter reflects deforestation, desertif
synthesis. Net primary production (NPP) is
tion, paving over, and human conversion to less p
the amount of solar energy captured in pho- systems (such as agriculture).
ductive

This content downloaded from

186.84.20.92 on Tue, 17 Aug 2021 00:37:04 UTC
All use subject to https://about.jstor.org/terms
 67(2) Daly: Speculations 259

croeconomics is about little else. An activ- capital is just equal to the marginal cost to
ity is identified, be it producing shoes or human beings of sacrificed natural capital.
consuming ice cream. A cost function and All nonhuman species and their habitats
a benefit function for the activity in ques-are valued only instrumentally according
tion are defined. Good reasons are given forto their capacity to satisfy human wants.
believing that marginal costs increase and Their intrinsic value (capacity to enjoy their
marginal benefits decline as the scale of theown lives) is assumed to be zero.
activity grows. The message of microeco- 2. The biocentric optimum. Other spe-
nomics is to expand the scale of the activitycies and their habitats are preserved be-
in question up to the point where marginalyond the point of maximum instrumental
costs equal marginal benefits, a condition convenience, out of a recognition that other
which defines the optimal scale. All of mi-species have intrinsic value independent of
croeconomics is an extended variation of their instrumental value to human beings.
this theme. The biocentric optimal scale of the human
When we move to macroeconomics, niche would therefore be smaller than the
however, we never again hear about opti- anthropocentric optimum.
mal scale. There is apparently no optimal The notion of sustainable development
scale for the macro economy. There are does not specify which concept of optimal
no cost and benefit functions defined for scale to use. Sustainability is a necessary,
growth in scale of the economy as a whole.but not sufficient condition for optimal
scale and the further elaboration of an envi-
It just doesn't matter how many people
ronmental macroeconomics.
there are, or how much they each consume,
as long as the proportions and relative
prices are right! But if every micro activity
has an optimal scale then why does not the References
aggregate of all micro activities have an op-
timal scale? If I am told in reply that the Barro, Robert J. 1987. Macroeconomics. 2d ed.
reason is that the constraint on any one ac- New York: John Wiley and Sons.
tivity is the fixity of all the others and that Daly, H. E. 1985. "The Circular Flow of Ex-
when all economic activities increase pro- change Value and the Linear Throughput of
portionally the restraints cancel out, then I Matter-Energy: A Case of Misplaced Con-
will invite the economist to increase the creteness." Review of Social Economy
scale of the carbon cycle and the hydrologic43(3):279-97.
cycle in proportion to the growth of indus-Dornbusch, Rudiger, and Stanley Fischer. 1987.
Macroeconomics. 4th ed. New York:
try and agriculture. I will admit that if the
McGraw-Hill.
ecosystem can grow indefinitely then so
Georgescu-Roegen, Nicholas. 1971. The En-
can the aggregate economy. But, until the
tropy Law and the Economic Process. Cam-
surface of the earth begins to grow at a rate bridge: Harvard University Press.
equal to the rate of interest, one should not
Hall, Robert E., and John B. Taylor. 1988.
take this answer too seriously. The indiffer-Macroeconomics. 2d ed. New York: W. W.
ence to scale of the macroeconomy is due Norton.
to the preanalytic vision of the economy Pearce,
as David, et al. 1989. Blueprint for a Green
an isolated system-a view the inappro- Economy. London: Earthscan, Ltd., p. 135.
Schumpeter, Joseph. 1954. History of Economic
priateness of which has already been dis-
cussed. Analysis: New York: Oxford University
Press, p. 41.
Two concepts of optimal scale can be
Tinbergen, Jan. 1952. On the Theory of Eco-
distinguished, both formalisms at this
nomic Policy. Amsterdam: North-Holland
stage, but important for clarity. Press.
1. The anthropocentric optimum. The Vitousek, Peter M., Paul R. Ehrlich, Anne H.
rule is to expand scale, i.e., grow, to the Ehrlich, and Pamela A. Matson. 1986. "Hu-
point at which the marginal benefit to hu- man Appropriation of the Products of Photo-
man beings of additional manmade physical synthesis." BioScience 34(May):368-73.

This content downloaded from

186.84.20.92 on Tue, 17 Aug 2021 00:37:04 UTC
All use subject to https://about.jstor.org/terms