Content uploaded by F.C. Mcmichael
Author content
All content in this area was uploaded by F.C. Mcmichael on Feb 16, 2015
Content may be subject to copyright.
944 / JOURNAL OF ENVIRONMENTAL ENGINEERING / OCTOBER 1999
M
UNICIPAL
S
OLID
W
ASTE
R
ECYCLING
I
SSUES
By Lester B. Lave,
1
Chris T. Hendrickson,
2
Member, ASCE, Noellette M. Conway-Schempf,
3
and Francis C. McMichael,
4
Member, ASCE
A
BSTRACT
:Municipal solid waste (MSW) recycling targets have been set nationally and in many states.
Unfortunately, the definitions of recycling, rates of recycling, and the appropriate components of MSW vary.
MSW recycling has been found to be costly for most municipalities compared to landfill disposal. MSWrecycling
policy should be determined by the cost to the community and to society more generally. In particular, recycling
is a good policy only if environmental impacts and the resources used to collect, sort, and recycle a material
are less than the environmental impacts and resources needed to provide equivalent virgin material plus the
resources needed to dispose of the postconsumer material safely. From a review of the existing economic
experience with recycling and an analysis of the environmental benefits (including estimation of external social
costs), we find that, for most communities, curbside recycling is only justifiable for some postconsumer waste,
such as aluminum and other metals. We argue that alternatives to curbside recycling collection should be ex-
plored, including product takeback for products with a toxic content (such as batteries) or product redesign to
permit more effective product remanufacture.
INTRODUCTION
The United States is a ‘‘throwaway’’ society whose total
and per capita waste has been increasing for more than 40
years. The average American produces about 4.4 lbs of mu-
nicipal solid waste (MSW) each day, resulting in roughly
210,000,000 tons/year for the nation (Statistical 1998). Most
MSW goes to landfills. From examination of landfill wastes,
Rathje and Murphy (1992) found that the composition of land-
fill mass deposited in the 1980s was roughly 40% paper, 8%
organics, 8% plastic, 11% metal, 6% glass, 12% construction/
demolition waste, and 15% additional, unclassified waste.
Some people dislike landfills because they are a nuisance.
The closing of landfills, the threat of running out of space in
landfills, and the waste of resources have alarmed others when
MSW is sent to landfills.
The almost universal aversion to landfills comes from the
history of city dumps that smelled, looked terrible, were in-
fested with rats and other pests, and posed risks to health.
Sanitary engineers responded by designing modern landfills
that pose few of these problems. Modern landfills have a min-
imum odor nuisance, do not have pests, and pose few prob-
lems after they are closed. With rules mandating daily cover,
clay and rubber liners, clay caps, and leachate collection sys-
tems, modern landfills are a tribute to sanitary engineering.
Even with these improvements, landfills are still unpopular.
The traffic and other nuisances of even a modern landfill are
a bother to nearby residents. Methane emissions from landfills
can pose a safety hazard to nearby buildings and contribute to
urban ozone problems and global warning. In most commu-
nities, groups attempt to close current landfills and have made
it extremely difficult to site new ones. Dislike of landfills has
led to a popular revolt in states like Pennsylvania and Virginia
against taking MSW from other states, although interstate
1
Grad. School of Industrial Admin., Carnegie-MellonUniv.,Pittsburgh,
PA 15213.
2
Dept. of Civ. and Envir. Engrg., Carnegie-Mellon Univ., Pittsburgh,
PA.
3
Grad. School Industrial Admin., Carnegie-Mellon Univ., Pittsburgh,
PA.
4
Dept. of Civ. and Envir. Engrg., Carnegie-Mellon Univ., Pittsburgh,
PA.
Note. Associate Editor: Mark A. Tumeo. Discussion open until March
1, 2000. To extend the closing date one month, a written request must
be filed with the ASCE Manager of Journals. The manuscript for this
paper was submitted for review and possible publication on September
12, 1998. This paper is part of the Journal of Environmental Engineer-
ing, Vol. 125, No. 10, October, 1999. 䉷ASCE, ISSN 0733-9372/99/0010-
0944–0949/$8.00 ⫹$.50 per page. Paper No. 19243.
transfers of waste are the cheapest way of handling MSW in
some situations (Louis 1996).
In the popular press, the closing of many landfills in the
last decade and the opposition to opening new ones has led to
concern that we are running out of landfills. The number of
landfills has declined significantly because of new regulatory
requirements for improved design and management. However,
the decline in tipping fees in recent years is evidence that
landfill space meeting the new regulatory requirements is read-
ily available (Biocycle 1998).
The third objection that landfills waste precious resources
has led to two actions: energy recovery units and recycling.
MSW contains a great deal of energy that potentially could be
recovered. It also contains a great deal of valuable raw ma-
terials. Although energy recovery reduces the landfill problem
(it reduces the volume by 2/3) and extracts some of the raw
materials value in the MSW, it recovers only a tiny fraction
of the potential value in the materials.
Many states and cities have responded by requiring house-
holds to recycle; some have specific goals, such as requiring
50% recycling of MSW (Goldstein and Glenn 1997). How-
ever, little analysis underlies the recycling targets (Garrick
1998). There has been some analysis of whether MSW recy-
cling is beneficial, particularly recycling with curbside collec-
tion of recyclables (Denison 1996; Tierney 1996). Germany
required consumer product packaging to be recycled beginning
in 1991 (OECD 1996); and more recently, in Germany, auto-
mobile and electronics manufacturers have ‘‘volunteered’’ (un-
der threat of legislation) to take back their products and to
meet recycling targets for these products at the end of their
lives. These German initiatives are worth study, since they
should be viewed as a ‘‘social experiment’’ that can help to
enlighten our future policy.
This paper considers the life-cycle economic and environ-
mental impacts of MSW recycling. We seek to identify cost-
effective policies to achieve environmental and sustainability
goals for MSW. Similar to Haith (1998), we emphasize that
some recycling improves environmental quality and sustaina-
bility, whereas other recycling has the opposite effect. For ex-
ample, recovering aluminum beverage containers in an urban
area generally benefits the environment and lowers the use of
energy and other resources; in contrast, the minuscule amount
of aluminum in consumer packages is likely to require more
energy and other resources to separate and recycle than it
saves.
Before recycling can occur, the materials must be collected
from consumers—a reversal of the logistics system that dis-
tributed products to consumers. People familiar with the com-
JOURNAL OF ENVIRONMENTAL ENGINEERING / OCTOBER 1999 / 945
plexity of the current distribution system should not be sur-
prised at the difficulty of designing and operating a ‘‘reverse
logistics’’ system that is universal, cheap, and reliable. Curb-
side pickup is one of several reverse logistics systems with its
peculiar advantages and drawbacks. Other reverse logistics
systems include consumers taking recyclables to a central col-
lection point or returning them to the retailer as part of a de-
posit/refund system. The alternative systems have radically
different implications for the amount of work that consumers
must perform, the cost of collection and sorting, and the over-
all efficiency of the system.
WHAT SHOULD WE DO WITH PRODUCTS AT THE
END OF THEIR LIVES?
Despite efforts of the Environmental Protection Agency
(‘‘Documents’’ 1999) and the legislation and regulations man-
dating recycling programs, there is no consensus on what con-
stitutes MSW recycling, either on which postconsumer waste
is included in MSW or on how to measure the fraction of
material that is recovered for reuse.
When a consumer no longer wants to keep a product, any
of the following options may be possible. The product might
be
1. Reused (as with old furniture)
2. Remanufactured (as with copier machines or automobile
alternators)
3. Recycled into the same use in a ‘‘closed loop’’ (as with
asphalt pavements)
4. Recycled into a lower valued use (as with recycled plas-
tic molded into park benches)
5. Incinerated (as with burning paper to recover energy)
6. Landfilled (as with most MSW)
7. Discarded directly to the environment (as with littering)
Individuals and organizations differ on how many of these
categories should be included within the definition of ‘‘recy-
cling,’’ although most people would include Options 1–4. If
incineration productively recovered most of the energy in
MSW (Option 5), there would be a good case for including it
as recycling. Storing the waste in a landfill until it is recovered
might even be considered recycling; in a sense, a landfill can
be thought of as a giant storage bin of materials that could be
recovered in the future. The EPA has been working to stan-
dardize definitions and methods of calculating the proportion
recycled.
The definition of recycling distracts society from the real
issues: environmental quality and sustainability. The definition
matters only because recycling goals have been specified. Note
that the goal is not to increase recycling: it is to improve en-
vironmental quality and sustainability. Recycling, whatever the
definition, is one possible way to accomplish these goals.
Some laws declare, for example, that 50% of MSW must
be recycled without defining what is included in MSW.A strict
definition might include only the waste collected at curbside
from residences. However, this definition excludes important
consumer products such as batteries and automobiles, as well
as waste from residential construction and demolition. Wepre-
fer a broader definition that includes all postconsumer waste
that ordinarily is sent to a landfill. However, because little of
the demolition waste is a candidate for recycling, an arbitrary
requirement for recycling does not make sense.
A final issue is the controversy about measuring what is
recycled. Roughly 95% of automotive lead-acid batteries are
returned for recycling. Does this mean that 95% of automotive
batteries are recycled? We would answer no. Typically, all ma-
terials in the battery other than lead are discarded. Thus, 40%
of the battery weight is discarded. Of the lead in these batter-
ies, 95% is retrieved in the secondary smelter recovery pro-
cess. Thus, of all lead-acid batteries taken from cars, 54%
(0.95 ⫻0.6 ⫻0.95) of discarded automobile battery material
is recycled, and 90% (0.95 ⫻0.95) of the lead in these bat-
teries is recycled. In our judgment, the best measure of recy-
cling is the proportion of discarded products that are returned
to a productive use, or 54% in the case of lead-acid starting/
lighting/ignition batteries.
WHAT SHOULD BE THE OBJECTIVES OF MSW
RECYCLING?
Perhaps the most widespread goal for MSW recycling is to
increase recycling. Many people feel guilty about our profli-
gate lifestyle and feel that steps need to be taken to improve
environmental quality and sustainability. Recycling seems to
be an obvious response. We agree with the concern for the
environment and sustainability but do not regard recycling it-
self as a goal. Instead, the four primary goals of this study are
1. To save landfill space.
2. To save money from handling MSW. Governments face
fiscal difficulties and constant criticism for being ineffi-
cient in providing public services.
3. To increase environmental quality, by lowering dis-
charges of pollutants. In particular, the goal is to elimi-
nate dissipative emissions of hazardous and toxic mate-
rials to the environment, including greenhouse gases and
toxic materials sent to MSW landfills.
4. To increase the sustainability of the economy. This im-
plies minimizing the use of depletable resources such as
ores or petroleum and reducing the use of renewable re-
sources, such as lumber, to sustainable levels.
Low cost is important for environmental as well as fiscal rea-
sons. For example, petroleum to run collection trucks is just
as much a use of this resource as petroleum to make consumer
products; future generations will not have a barrel of petro-
leum to use for either purpose. The resources going to recy-
cling are an important aspect of which MSW alternative is
best at achieving the goals of environmental quality and sus-
tainability. Sound policy requires examining the full range of
alternatives to compare the resources, energy, and labor needed
for the entire life cycle of each alternative (Facing 1989). The
comparison between recycling and making new products must
be an evenhanded examination of the total use of energy and
nonrenewable resources and dissipative emissions.
We can state our conclusion as the ‘‘economic-environmen-
tal criterion’’: Recycling is good policy only if environmental
discharges and the resources used to collect, sort, and recycle
a material are less than the environmental discharges and re-
sources needed to provide an equivalent virgin material plus
the resources needed to dispose of the material safely.
For example, glass is made of sand and potash, neither of
which is in short supply. Glass is nontoxic: discharging it to
the environment poses no risk, save from cuts. For recycling
of glass to be a sensible environmental policy, the energy,
equipment, and labor associated with collection, separation,
and recycling of glass should be smaller than the energy,
equipment, and labor associated with producing the new glass,
including the energy, equipment, and labor required to collect
and landfill the postconsumer glass. If the resources associated
with collecting, separating, and recycling of glass are larger
than the resources associated with making new glass and with
landfilling the used glass, recycling does not help either en-
vironmental quality or sustainability.
A more general form of the economic-environmental crite-
rion applies to reuse, remanufacturing, and other programs for
dealing with MSW such as resource reduction: A program is
946 / JOURNAL OF ENVIRONMENTAL ENGINEERING / OCTOBER 1999
TABLE 2. Environmental Effects of 1,000 Metric Tons of Production for Different Commodities—Savings Available from Recycling
Category
(1) Primary aluminum
(2)
Blast furnaces
and steel mills
(3) Glass containers
(4) Logging
(5)
Electricity (kW⭈h) 21,000,000 1,000,000 100,000 60,000
Fuel use (metric tons) 1,700 850 470 800
Total energy use (TJ) 58 25 2 4
Water intake (gal.) 29,000,000 23,000,000 400,000 600,000
SO
2
emissons (metric tons) 46 27 1 0.9
Particulate emissions <10 m (metric tons) 1 1.2 0.09 0.02
Global warming potential (metric tons CO
2
equivalent) 4,500 2,200 120 230
Hazardous waste generated—RCRA
a
(metric tons) 110 26 3 7
Toxic air emissions—TRI
b
(metric tons) 2 0.2 0.1 0.5
External cost due to criteria air emissions (dollars) 220,000 11,000 700 8,000
Note: Economic effects are calculated throughout the U.S. economy using the U.S. Department of Commerce’s 500 ⫻500 commodity 1992 input/
output model; prices for various commodities typically vary considerably over time and space, and assumed prices for these calculations are $135/ton
for blast furnace and steel mills, $1,500/ton for aluminum, $50/ton for glass, $300/ton (or $500/thousand-board-ft) for logs. See EIOLCA 99.
a
RCRA = Resource Conservation and Recovery Act.
b
TRI = Toxics Release Inventory.
TABLE 1. Average Annual Curbside Recycling Costs in the
United States
Curbside recycling
(1)
Per
household
(dollars)
(2)
Per ton
(dollars)
(3)
Avoided MSW disposal cost (savings) (7) (31)
Recycling collection 27 123
Avoided MSW collection 0 0
Recycling processing 11 50
Total cost (sum of four categories) 31 142
Revenue from sale of recyclables (1997) 10 45
Net cost after sale of recyclables 21 97
Note: From Ackerman (1996) for costs and Berenyi (1997) for reve-
nue.
beneficial to the environment and sustainability only if it ac-
tually reduces energy, resource use, and pollution, taking ac-
count of the full life cycles of the program compared with its
alternatives.
A limitation to this statement of the economic-environmen-
tal criterion is that it neglects the fact that current products
were not designed to be recycled. As a consequence, many
cannot be recycled easily. For example, Lave et al. (1998)
showed that nylon carpet could be redesigned to improve the
implications of recycling for environmental quality and re-
source use.
Finally, recycling may also have ancillary benefits associ-
ated with community building and involvement. However,
these activities would be more rewarding if they were chan-
neled toward types of recycling (or other activities such as
park cleanups) that are undoubtedly environmentally benefi-
cial. Hence the need for pursuing the four goals articulated
above.
IS MSW RECYCLING PROFITABLE?
At one time, advocates claimed that recycling of MSW
would be profitable for municipalities. Recycling programs
were expected to more than pay for themselves. A few cate-
gories of postconsumer wastes can be recycled or reused prof-
itably; aluminum cans and automobiles are common examples.
For most categories of MSW, the costs of collection are likely
to exceed the revenue from sales. Based on national data, Ack-
erman (1996) estimated that curbside recycling cost $142/ton
even after a credit for avoided tipping fees (Table 1). Revenues
from the sale of some, but not all, recyclables might offset this
cost. Revenue for a typical bundle of MSW recyclables (in-
cluding metals, paper, and glass) was estimated at $140/ton in
1995 but only $45/ton in 1997 (Berenyi 1997). Combining the
cost of $142/ton and the 1997 revenue of $45 would result in
a revenue loss of $97/ton for municipalities. The composition
of recyclables is also important, with aluminum cans com-
manding revenue more than 10 times that of recycled paper.
However, at current price levels, curbside collection programs
for most recyclable materials cost more than landfilling and
must be justified on environmental grounds.
Separate collection of recyclables is particularly expensive,
because each residence is visited twice (Lave et al. 1994). A
collection truck that can carry regular MSW and recyclables
is preferable, because each residence gets a single pickup.
However, trash pickup is likely to become more expensive
because the truck will be delayed by any sorting and because
it must visit both the landfill and the recycling facility. Because
the truck will be collecting trash and recyclables in different
compartments, one compartment will fill first requiring the
truck to go to the recycling site and landfill even though the
other compartment(s) is partially empty. Having recycling
compartments that are too big or too small will increase col-
lection costs sharply. Drop-off points can reduce municipal
costs but may incur substantial private costs if they require
additional driving.
DOES MSW RECYCLING HELP THE ENVIRONMENT?
Denison (1996) reviewed several studies of overall environ-
mental impact of recycling MSW, concluding that recycling
saved energy and resource inputs. Denison evaluated the bun-
dle of household recyclables rather than each component; he
implicitly has the high value components subsidize the low
value components. Pearce (1995) found that the net benefit of
recycling is not always positive. Our analysis suggests that
recycling some of the components (e.g., aluminum) have a
much higher potential for recycling than do other materials
(e.g., glass). An analysis of environmental externalities for
curbside recycling in Milton Keynes, England, also found sig-
nificant differences in benefits for different components of
MSW (Craighill and Powell 1996).
Table 2 gives a direct indication of the environmental ben-
efits of avoided production due to recycling of different com-
modities. This table summarizes electricity use, fuel use, en-
ergy (including electricity and fuels), industrial water intake,
some conventional pollutant emissions, global warming poten-
tial, toxic air releases, and hazardous waste generation for
1,000 metric tons of different commodity productions. These
environmental effects are calculated by tracing all of the econ-
omy-wide supply chain requirements for the various commod-
ities using the 500 sector 1992 economic input-output model
JOURNAL OF ENVIRONMENTAL ENGINEERING / OCTOBER 1999 / 947
developed by the U.S. Department of Commerce coupled with
ancillary environmental impact calculations (Horvath and
Hendrickson 1997; Hendrickson 1998a; EIOLCA 1999). For
example, toxic air emissions are computed by multiplying the
level of activity in each of the 500 commodity sectors by the
average level of toxic air emissions per dollar of output. These
calculations show an upper bound on savings from recycling
by avoiding this primary production; the figures are an upper
bound because the resource costs of recycling are not included.
The final row in Table 2 represents a rough estimate of the
external environmental costs of this production. It is based
upon estimates of the social costs from air emissions of con-
ventional pollutants (Hendrickson et al. 1998b; Matthews
1999). Chung and Poon (1997) and Craighill and Powell
(1996) also made estimates of such external costs, for Hong
Kong and the United Kingdom, respectively. Included in these
costs are the estimated health effects related to ozone, partic-
ulate, and other conventional or ‘‘criteria pollutants.’’ The es-
timates are reported in thousands of social cost dollars, and so
a metric ton of primary aluminum is estimated to have an
external environmental cost due to air emissions of $220 (Ta-
ble 2). Comparing this number to the estimated cost of col-
lection ($142/ton), aluminum appears to be a good candidate
for recycling, even without counting the economic costs of
producing a ton of aluminum.
Our calculations find that avoiding primary aluminum pro-
duction has the greatest environmental benefit. Recycling alu-
minum is generally profitable because of the high price for
this scrap. Ferrous metals and logging have intermediate ben-
efits. Avoiding additional glass production has relatively small
environmental benefits for the various categories of environ-
mental emissions we analyzed, particularly because the num-
bers for glass are overestimates since they include the final
container processes that would also be incurred for recycled
glass.
A full analysis of the environmental effects would also in-
clude the environmental effects associated with collection,
sorting, and processing of recycled materials. These processes
require capital equipment (particularly trucks) and the use of
energy (for truck operation and sorting).
POLICY TEST FOR RECYCLING
Should materials be recycled or put in a landfill? The ques-
tion can be answered with the economic-environmental crite-
rion.
One form of the economic-environmental criterion is that
faced by companies. What should be done with the waste gen-
erated by a manufacturing plant, service center, or office? The
company would like to reduce its costs and so calculates
whether recycling is less costly than disposal. Consider, for
example, a stamping plant that turns out steel parts for auto-
mobiles, generating large quantities of steel scrap. This
‘‘prompt scrap’’ is of high quality and commands a relatively
high price. Automobile companies would laugh at the idea of
paying to landfill this scrap steel, because they get paid hand-
somely to recycle it. Similarly, many companies find that re-
cyclers will pay high prices for their scrap office paper.
A well-run company will recycle waste if it costs them less
than disposing of it; they should separate and collect the val-
uable materials for recycling and dispose of other materials.
This means that the market prices of scrap, landfill costs, and
separating and transport costs determine whether ‘‘waste’’ is
recycled or landfilled. Thus, the first form of the economic-
environmental criterion is to recycle only if the cost of col-
lection and separation is less than the cost of collection and
disposal. An environmentally conscious company might de-
cide to do more recycling than is implied by the economic-
environmental criterion. However, as markets get more com-
petitive, companies are forced to cut ‘‘unnecessary’’expenses,
but the companies need to be careful that the additional unit
recycled actually reduces environmental discharges and ma-
terials use.
This ‘‘private’’ form of the economic-environmental crite-
rion squarely faces the realities of companies. They are driven
by costs. They will recycle materials where the costs of col-
lection plus the tipping fee is greater than the cost of collecting
and sorting the recyclables less the revenue from selling the
recycled material. Although many companies would like to do
well, they are severely limited by competition or current
budgets.
Cities face tight budgets as well but may adopt a modifi-
cation of the private rule. For a municipality, the economic-
environmental criterion is modified slightly: The city seeks not
only to minimize its costs, it also seeks to avoid local envi-
ronmental nuisances. This means that a city might choose an
alternative that is somewhat more expensive, if this avoided a
nuisance.
Recycling only those materials that satisfy this first form of
the economic-environmental criterion is not fully satisfactory
in protecting the environment or working toward a sustainable
future. If there are externalities associated with extracting re-
sources, landfilling, or sorting recyclables, or if there is a lack
of foresight in managing resources, the private costs that are
faced by companies or cities neglect important dimensions of
the MSW decision. For example, the regulations governing
landfills might be inadequate, leading to future environmental
degradation. If so, the price of landfills will be ‘‘too low,’’ and
landfilling will damage environmental quality. The obvious
remedy is to change the landfill regulations so that environ-
mental quality will not suffer. Similarly, society may give too
little weight to the needs of future generations for raw mate-
rials. If so, raw materials will be priced too low, and compa-
nies will choose to do too little recycling. An obvious remedy
is to impose a tax to increase the prices of raw materials so
that more will be saved for future generations.
Another example might be inadequate environmental regu-
lations associated with mining coal, which is then used in pro-
ducing aluminum and steel. If so, the cost of producing steel
and aluminum would be too low, discouraging recycling of
these materials. An obvious remedy is to improve environ-
mental regulations concerning coal mining. Mandating steel
and aluminum recycling decreases the amount of coal that is
mined, but the coal is still mined in a way that damages the
environment.
Still, another example might be the profligate use of fossil
fuels leading to greenhouse gas emissions. If recycling is more
energy efficient, low recycling rates lead to ‘‘too much’’
greenhouse gas emissions. The externality could be internal-
ized either by a fee on greenhouse gas emissions (making pro-
duction of virgin materials more expensive) or a cap on the
total emission of greenhouse gases, which would mean the
production of virgin materials was ‘‘inadequate’’ for the needs
of the economy, thereby increasing the price of recycled ma-
terials. Generally, the externalities can be accounted for by
having regulatory agencies give direct orders to firms and con-
sumers that internalize the externalities. However, in an econ-
omy as large and complicated as that of the United States,
regulatory agencies do not have the knowledge or personnel
to figure out what actions will internalize the externalities. An
alternative is to use the market system by imposing taxes to
account for the externality. The use of market incentives has
greatly reduced the cost of achieving such environmental ob-
jectives as reducing the emissions of sulfur dioxide to prevent
acid rain and the emissions of chlorofluorocarbons to prevent
the destruction of stratospheric ozone (Schmallensee et al.
1998).
948 / JOURNAL OF ENVIRONMENTAL ENGINEERING / OCTOBER 1999
The second form of the economic-environmental criterion,
the ‘‘public’’ formulation, is to internalize all important exter-
nalities in prices or regulations so that prices and practice re-
flect the full environmental costs associated with each action,
including the availability of resources for future generations.
Once this is done, the private form of the recycling policy
prescription becomes an accurate social formulation of the
right decisions. Once the important externalities have been
controlled or internalized, materials should be recycled only if
the cost of collection and separation, less the revenue from
selling the recycled material, is less than the cost of collection
plus the tipping fee. The private version of the economic-en-
vironmental criterion helps to understand current recycling be-
havior. The social version of the economic-environmental cri-
terion helps to guide us toward the best social policy.
Does this social version of the economic-environmental cri-
terion help sustainability? It requires society to examine the
need of future generations for resources and to satisfy this need
either by explicitly preserving some resources for future gen-
erations or by raising the prices of raw materials through a
‘‘sustainability’’ tax. Such a tax may not be needed for metals
and other durable resources. Because landfills simply store
these materials, they are available whenever society decides to
‘‘mine’’ them.
For fossil fuels and other depletable resources, there is little
alternative to explicitly examining the needs of future gener-
ations. This analysis is difficult because technology changes
and the tastes of future generations are likely to change. For
example, planting oak trees in the past to enable the current
generations to have masts for sailing ships has not proven to
be much of a boon. The technology for energy production has
been changing rapidly. It is hard to know what future gener-
ations will desire and how much energy they will need to
provide a lifestyle that they will find at least as good as the
current generation finds its lifestyle.
CONCLUSIONS: IS MSW RECYCLING THE BEST
POLICY?
The goal of MSW recycling programs should not be to in-
crease MSW recycling. The goal should be to increase envi-
ronmental quality and the sustainability of the economy. Our
hopes concerning MSW recycling must be tempered by the
economic-environmental criterion: Recycling will benefit the
environment and sustainability only if the energy, resources,
and environmental discharges associated with recovering the
material are less than those associated with producing virgin
material. Curbside recycling of postconsumer metals can save
money and improve environmental quality if the collection,
sorting, and recovery processes are efficient. Curbside collec-
tion of glass and paper is unlikely to help the environment and
sustainability save in special circumstances.
Some alternative policies also deserve consideration as
MSW recycling options. Deposit/refund schemes offer an im-
portant option. In these systems, products earmarked for re-
cycling would require a consumer (or producer) deposit, with
a refund to the consumer when they are returned. For example,
each return of a nickel-cadmium battery would receive a re-
fund sufficient to make it attractive to undertake the return.
Aluminum cans and metal scrap are sufficiently valuable that
‘‘trash pickers’’ routinely search for these postconsumer
wastes even without deposit/refund schemes. An advantage of
these deposit/refund schemes is that products and materials can
be individually targeted for removal from the MSW stream.
Palmer et al. (1997) concluded that deposit/refund schemes
can be more efficient at waste reduction than recycling sub-
sidies. Although deposit/refund systems can recover the vast
majority of the product, the energy and resources required
could be large. For example, if consumers make a special trip
to return the recoverable materials, the energy required is
likely to exceed the energy saved by recovery.
Another policy that can be beneficial is product takeback
by manufacturers, particularly when remanufacturing and re-
use is available (Klausner 1998). This option attempts to pre-
serve the value of the original goods. In contrast, recycling
seeks to recover only the value of the raw materials. Product
takeback for small appliances, such as handtools, might have
significant benefits. In particular, the raw material value of
most complicated products such as computers is only a small
fraction of the product value. Also, manufacturers would have
incentives to alter designs to make remanufacture and use
more effective.
ACKNOWLEDGMENTS
Financial support from the Environmental Protection Agency under
Cooperative Agreement CR825188-01-0 is gratefully acknowledged.
APPENDIX. REFERENCES
Ackerman, F. (1997). ‘‘Recycling: Looking beyond the bottom line.’’
BioCycle, May, 67.
Berenyi, E. B. (1997). ‘‘Materials recovery facilities: A 1997 update.’’
BioCycle, September.
BioCycle. (1998). ‘‘The state of garbage in America.’’ Annu. Rep.,
BioCycle, 39(4), 32–43.
Craighill, A., and Powell, J. C. (1996). ‘‘Lifecycle assessment and eco-
nomic evaluation of recycling: A case study.’’ Resour., Conservation
and Recycling, 17, 75–96.
Chung, S.-S., and Poon, C.-S. (1997). ‘‘Quantifying externalities in solid
waste management in Hong Kong.’’ J. Envir. Engrg., ASCE, 123(3),
282–289.
Denison, R. A. (1996). ‘‘Environmental life-cycle comparisons of recy-
cling, landfilling and incineration: A review of recent studies.’’ Annu.
Rev. of Energy and the Envir., 21, 191–237.
‘‘Documents to aid measurement of recycling.’’ (1999). Environmental
Protection Agency, 具http://www.epa.gov/epaoswer/non-hw/recycle/
recmeas/document.htm典.
EIOLCA. (1999). 具http://www.eiolca.net/典.
Facing America’s trash: What next for municipal solid waste? (1989).
Ofc. of Technol. Assessment, GPO, Washington, D.C.
Garrick, L. (1998). ‘‘Economics and environmental costs and benefits of
recycling post-consumer waste: A systems view.’’ Green Design Initia-
tive Bibliography, Carnegie-Mellon University, Pittsburgh.
Goldstein, N., and Glenn, J. (1997). ‘‘The state of garbage in America.’’
BioCycle, May, 71–75.
Haith, D. A. (1998). ‘‘Materials balance for municipal solid-waste man-
agement.’’ J. Envir. Engrg., ASCE, 124(1), 67–75.
Hendrickson, C., Horvath, A., Joshi, S., and Lave, L. B. (1998a). ‘‘Intro-
duction to the use of economic input-output models for environmental
life cycle assessment.’’ Envir. Sci. and Technol.
Hendrickson, C., Horvath, A., and Matthews, H. S. (1998b). ‘‘External
environmental costs of transportation equipment, materials and ser-
vices.’’ Tech. Rep., Green Design Initiative, Carnegie-Mellon Univer-
sity, Pittsburgh.
Horvath, A., and Hendrickson, C. T. (1998). ‘‘A comparison of the en-
vironmental implications of asphalt and steel-reinforced concrete pave-
ments.’’ Transp. Res. Rec., Transportation Research Board, Washing-
ton, D.C.
Huysman, M. (1994). ‘‘Waste picking as a survival strategy for women
in Indian cities.’’ Envir. and Urbanizations, 6(2), 155–174.
Klausner, M. (1998). ‘‘Design and analysis of product takeback systems:
An application to power tools,’’ PhD dissertation, Carnegie-Mellon
University, Pittsburgh.
Lave, L., Conway-Schempf, N., Harvey, J., Hart, D., Bee, T., and Mac-
Cracken, C. (1998). J. Industrial Ecology, 2(1), 117–126.
Lave, L. B., Hendrickson, C. T., and McMichael, F. C. (1994). ‘‘Recy-
cling decisions and green design.’’ Envir. Sci. and Technol., 28(1),
18A–24A.
Louis, G. (1996). ‘‘Regional integrated solid waste management in the
Northeastern United States,’’ PhD thesis, Carnegie-Mellon University,
Pittsburgh.
Matthews, H. S. (1998). ‘‘An analysis of the economic effects of envi-
ronmental externalities.’’ Tech. Rep., Green Design Initiative,Carnegie-
Mellon University, Pittsburgh.
JOURNAL OF ENVIRONMENTAL ENGINEERING / OCTOBER 1999 / 949
OECD. (1996). ‘‘Extended producer responsibility in the OECD area.’’
Phase 1 Rep., Working Papers No. 66, Organization for Economic Co-
operation and Development, Paris.
Palmer, K., Sigman, H., and Walls, M. (1997). ‘‘The cost of reducing
municipal solid waste.’’ J. Envir. Economics and Mgmt. 33, 128–
150.
Pearce, D. W., and Turner, R. K. (1993). ‘‘Market-based approaches to
solid waste management. Resour., Conservation and Recycling, 8(1–
2), 63–90.
Rathje, W., and Murphy, C. (1992). ‘‘Rubbish! The archeology of gar-
bage,’’ Harper Perennial.
Schmalansee, R., Joskow, P. L., Ellerman, A. D., Montero, J. P., and
Bailey, E. M. (1998). ‘‘An interim evaluation of sulfur dioxide emis-
sions trading.’’ J. Economic Perspectives, 12(3), 53–68.
Statistical abstract of the United States. (1998). 1998 Ed., Bureau of the
Census, 具http://www.census.gov/prod/3/98pubs/98statab典.
Tierney, J. (1996). ‘‘Recycling is garbage.’’ New York Times Mag., June
30.
