Municipal Solid Waste:
Effect of State-Level Programs on Waste
Reduction and Increased Recycling
Matthew V. Brooks
T his study examines the effectiveness of state waste management policies in reducing municipal solid waste and
increasing recycling. None of the general programs currently in practice effect the amount of waste generated.
However, programs such as bottle bills and curbside pickup, and factors such as landfill and incinerator tipping fees
and population density show a strong relationship to increased recycling.
Municipal solid waste, or MSW, consists of most non-industrial, non-hazardous solid wastes. In general, those components of
MSW that are not recycled are either landfilled or burned. Over the past ten years, the nationwide average has well exceeded a
ton of MSW generated annually per person. Though recycling rates have risen slightly, the national population, as well as the per
capita rate of waste production, has also risen. Many programs, such as bottle bills, tonnage taxes, backyard composting, and envi-
ronmentally preferred purchasing, exist on the state level to reduce waste and encourage recycling. The purpose of this study is
to determine which of the current practices are most effective and which are not worth further pursuit.
Topic Background and Literature
The two general methods that exist to reduce the amount of solid waste requiring disposal are recovery and source reduction.
Over the past few decades the focus has been on basic recovery, or recycling. Abert, Alter, & Bernheisel lay out the early economic
basis of waste recovery in the 1974 paper “The Economics of Resource Recovery.” The paper separates recoverable waste into the
categories of pure raw materials and organic materials.1 “Front end” recovery, which involves mechanical extraction of usable ele-
ments, results in new raw materials. This method corresponds to the modern view of recycling. The paper also refers to “back
end” recovery, which consists of composting or burning organic material in order to recover usable material or energy. Composting
remains an important component of recycling while incineration has recently raised air pollution concerns.
Many observers have argued that disposal costs can provide a strong driver for increased recycling. A 1983 Worldwatch paper
claims that recycling reduces the need for raw materials, lowers energy use, and reduces the costs of waste disposal.2 Market dis-
tortions, Chandler argues, are what prevent recycling from being more pervasive. The lack of accounting for externalities, subsi-
dized energy, and tax incentives for production eliminates the economic advantages of recycling. In addition, solid waste disposal
costs can often come out of general taxes, reducing individual incentives to cut waste.
In Recycling Solid Waste, Duston makes similar claims about the ability of cost to drive increased recycling. He argues that
the avoided costs of disposal provide 75 percent of the benefits of recycling.3 He also points to rising disposal costs and increas-
ing scarcity of landfill area as potential future drivers for recycling. In this model there are significant economic gains to be made
by recycling. The model predicts that rises in the cost of alternative methods of disposal should increase the incidence of recy-
cling. However, the idea that increased cost of disposal leads directly to recycling is not unanimous.
In a 1996 study entitled “Household Responses to Pricing Garbage by the Bag,” Fullerton and Kinnaman observed the house-
hold response to implementation of a per-bag fee on garbage. By measuring the weight of garbage at pickup both four weeks prior
to and four weeks after the implementation of the fee, the study determined that the drop in the weight of garbage was rather
small.4 Through subsequent surveys, the study estimated that at most one third, and possibly none, of the modest weight drop was
attributable to a reduction in waste production. Increased recycling accounted for slightly over a third of the weight drop, with
illegal dumping comprising the remaining amount.
The conclusion of the Fullerton and Kinnaman study is that direct unit pricing of garbage does not represent an efficient
method of reducing garbage weight and increasing recycling.4 Overall, the administrative costs, as well as the social costs incurred
by illegal dumping, outweighed the marginal increase in recycling. However, the study addresses the effects of unit costs at the
individual level. It leaves open the question of whether increased costs can provide a significant driver for towns or cities to
reduce the amount of waste they must dispose.
Previous research has also addressed factors other than direct costs that contribute to waste reduction and recycling rates.
In 1991, Dobbs conducted a study of disposal taxes and user charges; consumers were required to pay a fee for the eventual dis-
posal of the item and those who dispose of the item must pay.5 He found that the optimal solution was a disposal tax coupled with
Reports Volume 12, Spring 2005 MURJ 37
• Recycled content standards
• Curbside pickup programs
• Stated recycling goals
In addition, the study includes two potential factors that
are not controlled by state policy. The first is landfill and incin-
erator tipping fees. These fees are charged to municipalities by
landfill and incinerator operators, and range from $10 to $75
Table 1: Data descriptors. dollars per ton. As landfilling and incineration are the primary
alternatives to recycling, higher tipping fees might push com-
munities to find better ways to recycle. The second potential
factor outside of the direct control of state policy is population
Data on bottle bills are from the Container Recycling
Institute. Data on composting, tonnage taxes, EPP, and recy-
cled content standards are from telephone interviews con-
Table 2: Frequency of waste and recycling pro-
ducted with representatives from the solid waste divisions of
each of the 50 states. For a given state, each of these five vari-
ables either has a value of 1 if the state implemented the pro-
a negative user charge. This is the deposit-refund system that
gram or 0 if it did not. Virtually all instances of these programs
many states now have on aluminum cans and various other
had been in place for well over the 10-year span of this study. In
products. In theory, the disposal tax reduces the incentive to
theory, each of these programs should not only result in either
generate waste, while the negative user charge, or refund,
greater recycling (bottle bills, EPP, or recycled content) or
increases the incentive to properly deal with waste.
lower waste (composting and tonnage taxes), but should also
Sigman compares lead recovery policies and finds virgin
promote better awareness of recycling as an alternative method
materials taxes and deposit-refund systems to be the most effi-
of disposing of garbage.
cient at increasing recycling.6 Recycled content standards and
subsidies for recycled material production were less effective. The other two policies from the list are curbside pickup and
Palmer, Sigman, & Walls agree in a study entitled “Reducing stated recycling goals. I measure curbside pickup of recyclables
Municipal Solid Waste,” which examines the relative efficacy of as a percentage of the total population served, which I average
deposit-refund systems, advance disposal fees, and recycling over the duration of the study. The rationale behind curbside
subsidies.7 The study finds deposit-refund to be less costly to pickup of recyclables is that it takes some of the onus off of the
implement than the alternatives. consumer, thus making it much more likely that high numbers
Despite the strong support for deposit-refund systems, will participate. Stated recycling goals are represented by a
there is no consensus on which programs provide the best percentage of total municipal waste, and are from the tele-
incentives to increase recycling. In a more recent study, phone interviews. State governments typically did not strictly
Kulshreshthra & Sarangi analyze the theoretical model behind enforce these goals.
deposit-refund and claim that the practice can lead to under- The dependent variables in the study are municipal solid
recycling.8 Ekvall is also pessimistic about recycling initiatives, waste production and recycling rates. To measure municipal
and argues that setting minimum recycling standards can also solid waste, I simply use figures published in BioCycle’s annual
result in under-recycling.9 In an attempt to add fuel to the report on state waste production, averaged over the period
debate, this study empirically examines a large range of pro- from 1994 to 2002 (the most recent report, in 2004, used 2002
grams designed to stimulate recycling and reduce the amount data). Recycling levels are a percentage of total municipal solid
of waste placed in landfills and incinerators. waste generated in each state, and are also from BioCycle.
Research Design Data and Results
In this study, I analyze the relative effects of different The data for this study falls into two categories. Some of the
approaches to controlling solid waste and increasing recycling. variables, such as recycling rate and population density, occur
I examine the following policies, some of which have been pur- as a percentage or some other value. Table 1 summarizes these
sued by many states over the past 10 years: variables, which are often averages over the time period of the
study. Other variables, such as the existence of a bottle bill,
• Bottle bills
were difficult or impossible to quantify and thus appear simply
• Backyard composting programs
as a policy that was or was not implemented in each state.
• Disposal tonnage taxes
These variables appear in Table 2, along with the number of
• State sponsored environmentally preferred purchasing states that adopted the policy. As mentioned earlier, virtually
all instances of these policies stretch back more than 10 years.
While some of the independent variables theoretically
should reduce the amount of waste, none had a significant
effect on the per capita MSW, upon analysis. In fact, the policy
Table 3: Correlation of variables. of recycled content standards was the only one to have an effect
38 MURJ Volume 12, Spring 2005 Matthew V. Brooks Reports
that was statistically different from zero. The extremely small
sample size of states adopting the policy casts even this result
By contrast, the effect of certain variables on recycling rate
was strong. Table 3 displays some of the stronger correlations
between the independent variables and the rate of recycling.
The correlation of these variables with MSW is provided for
comparison. Figures 1 and 2 illustrate the relationship of curb-
side service and tipping fees with recycling rate. One notable
characteristic of the data is the relationship between density
and recycling rate. Though they demonstrate some correlation,
Figure 3 clearly suggests a non-linear relationship. To correct Figure 1. Recycling rate (% of MSW) and curbside pickup
service (% of population served).
this, I logged the density value, which is then appended to
Table 3 and displayed in Figure 4. Using the log of the original
variable accounts for the likelihood of the effect of each addi-
tional person per square mile decreasing as the density
increases. Indeed, the correlation is much higher for logged
density, and Figure 4 demonstrates a more linear relationship.
Finally, I use single and multivariate regressions to esti-
mate the size of the effect of these variables on recycling rate.
Table 4 displays the regression results. As shown by the single-
variable regressions, a 1% increase in the proportion of the pop-
ulation served by curbside pickup led to an expected increase
in the recycling rate of 0.2%. For tipping fees, a $1 increase led
to an expected recycling increase of 0.3%. Both of these results
were significant at the 99% confidence level. However, the
curbside pickup variable covaries with the density, which sug-
gests the use of multivariate regression. Upon examination of Figure 2. Recycling rate (% of MSW) and average tipping fee
Column 4 of Table 4, it appears that some combination of the
density and curbside service affects the recycling rate.
However, the effect of tipping fees, while smaller, is still signif-
icant at the 95% confidence level.
The bottle bill was the only one of the non-quantified vari-
ables that showed a strong relationship to recycling rate. The
presence of a bottle bill led to an expected increase in the recy-
cling rate of 9%. This result was significant at the 99% confi-
dence level. None of the other variables showed a relationship
statistically different from zero.
One possible explanation for the lack of effect on MSW of
the many and varied state policies is that the amount of waste
per capita is enormous and includes many elements, such as Figure 3: Recycling rate (% of MSW) and population density.
construction debris, that are not affected by small-scale poli-
cies. Also, policies that encourage recycling but not reduction
do not affect MSW, as recycled waste is still considered part of
the total waste stream. Despite these problems, one might still
expect economic factors such as tipping fees and tonnage taxes
to encourage communities to reduce the amount of waste gen-
erated. However, tonnage taxes were typically quite insignifi-
cant compared to the raw costs of disposal. Meanwhile, it is
possible that higher tipping fees, rather than waste reduction,
may have simply caused diversion of part of the waste stream
When we examine the results of the recycling analysis, this
does seem to be the case. Higher tipping fees resulted in
increased recycling rates. This suggests that states might fur- Figure 4. Recycling rate (% of MSW) and log of population
ther encourage recycling by artificially increasing the cost of density (natural log used).
Reports Matthew V. Brooks Volume 12, Spring 2005 MURJ 39
the country has lagged behind those who have had success with
the policy. The results of this study clearly suggest that bottle
bills should be universal. As with all of the policies the study
examines, deposit-refund does not promise to reduce the
amount of waste generated. However, there is evidence that
states can encourage the better disposal of waste. As the field
of waste management progresses, perhaps a more optimistic
set of choices will arise.
The original inspiration for this paper was the recent boom
in tradable carbon emissions permits. I wondered whether sim-
ilar market-based controls could work in waste management.
Table 4: Recycling regression results. The focus of the paper shifted, as few of these types of programs
currently exist. However, in recent years there has been greater
disposal. A non-trivial tonnage tax would accomplish this. speculation on the future of market controls in waste manage-
Though charging for disposal raised concerns with Fullerton & ment. Some governments have begun to use techniques such as
Kinnaman, among others, the tonnage tax applies to municipal- virgin materials taxes and takeback laws (for example,
ities rather than individuals—so illegal activity such as dump- Germany requires auto manufacturers to dispose of cars when
ing would be less convenient. they are no longer usable) to manipulate the market. As more
In addition to the tip fees, density and curbside service governments adopt innovative ways to reduce waste at both dis-
seemed to have some influence on recycling rates. This is likely posal and at the source, new research will be necessary to
due to increased convenience. For example, those that live in determine the effects.
an apartment building might be able to easily separate their
recyclables into different containers, which are then picked up Further Reading
by the municipality. By contrast, those inhabiting more sparse Thomas Kinnaman and Don Fullerton are leading experts
areas would have to do separation and disposal by themselves. in the field of solid waste management. They have collaborated
Increasing the level of curbside service, where possible, would on several papers on the economics of solid waste at the house-
complement the benefits of population density. hold and community level. For a theoretical, systems-based
Finally, the deposit-refund system showed a very strong approach to waste management, try Aspinwall & Cain, The
influence on recycling rates. As the average rate among the Changing Mindset in the Management of Waste; Hawken,
states was below 25%, the 9% expected increase resulting from Lovins & Lovins, Natural Capitalism; and Braden Allenby,
a bottle bill represents a significant improvement. As of 2004, Industrial Ecology. These authors move beyond waste to
only 10 states had adopted a bottle bill, and most of these did “materials management,” or a more cyclical and less linear
so in the 1980s, or even earlier. It is surprising that so much of approach to material use.
References 5. Dobbs IM. Litter and Waste Management: Data Sources
Disposal Taxes versus User Charges. Can J
1. Abert JG, Alter H, Bernheisel F. The BioCycle magazine
Econ 1991; 24:221—227.
Economics of Resource Recovery from
Container Recycling Institute:
Municipal Solid Waste. Science 1974; 6. Sigman HA. A Comparison of Public
183:1052—1058. Policies for Lead Recycling. Rand J Econ
1994; 26:452—478. Environmental Protection Agency:
2. Chandler WU. Materials Recycling: The
Virtue of Necessity. Worldwatch Paper #56. 7. Palmer K, Sigman HA, Walls M. The Cost
Washington, D.C.: Worldwatch Institute, of Reducing Municipal Solid Waste. J Telephone interviews with state offices
1983. Environ Econ Manage 1997; 33:128—150.
3. Duston TE. Recycling Solid Waste: The 8. Kulshreshtha P, Sarangi S. “No Return, No
First Choice for Private and Public Sector Refund”: An Analysis of Deposit-Refund
Management. Westport, CT: Quorum Systems. J Econ Behav Organ 2001;
Books, 1993. 46:379—394.
4. Fullerton D, Kinnaman TC. Household 9. Ekvall T. A Market-Based Approach to
Responses to Pricing Garbage by the Bag. Allocation at Open-Loop Recycling. Resour
The Am Econ Rev 1996; 86:971—984. Conserv Recy 2000; 29:91—109.
40 MURJ Volume 12, Spring 2005 Matthew V. Brooks Reports