BioCycle September 2004, Vol. 45, No. 9, p. 30
Study illustrates the dynamics of managing the solid waste stream, and how increases in collection costs impact generation rates and other variables.
Shanna Hallas-Burt and John M. Halstead

MOST local governments in the United States are facing severe fiscal management crises. Stresses caused by national and regional economic downturns, a steady decline in state and federal support, and opposition to traditional revenue sources like the local property tax have all negatively affected cash flows. Nonetheless, local government continues to bear the responsibility for providing solid waste management, education, police and fire protection, and other services. This has forced public managers to consider new policies which incorporate not only innovative means of waste disposal, but also innovative approaches to funding this expensive service.
In 2001, the New Hampshire Solid Waste Task Force found that transportation costs, tight labor markets, unpredictable fuel prices, and industry concentration resulted in average tipping fees higher than at any time in the previous ten years. The task force found that one of the most successful programs initiated to both reduce generation rates and increase recycling rates was the Pay-As-You-Throw (PAYT) program.
INCENTIVES, OLD AND NEW
Typically, municipal governments have funded local programs and services through taxation. Services and programs are budgeted, approved or accepted through a town meeting or ballot and the tax rate is set accordingly. In this manner, each household pays a certain amount, a portion of which then goes toward solid waste disposal costs. This leaves every household with a marginal (extra) cost of zero for every additional unit of solid waste they produce – a system with no financial incentives for reducing the waste stream even when free recycling is offered.
Around 1988, a number of cities across the United States – including Illinois, Pennsylvania, New York and New Jersey – implemented extensive and successful programs. Since then, the use of PAYT has grown and today more than 4,000 communities employ such a program, according to M.L. Miranda and colleagues, who studied 21 cities and found reductions in MSW disposal at landfills ranging from 17 percent to 74 percent. These cities also experienced an increase in the rate of recycling. A later study found that in seven of nine communities in the states of California, Michigan and Illinois, MSW waste generation decreased by 20 percent after a PAYT program was instituted, while in the remaining two communities, waste decreased by 50 percent and 38 percent. Factors influencing reduction included higher unit pricing fees, smaller container size, accompanying yard waste collection programs, and free recycling. Eight of the nine communities had significant increases in the recycling rates, ranging from 30 percent to a doubling in the recycling level. In Tompkins County, New York, it was found that a PAYT program combined with a mandatory curbside recycling program increased recycling from 22 to 58 percent (depending on the material).
Several studies have examined diversionary means such as illegal disposal, burning or increased composting. These means of diversion in New Hampshire were not included in this study because of the lack of accurate state level statistics; however, the rural nature of New Hampshire certainly leaves illegal dumping or burning as a potential problem that needs further investigation. When she looked closer at the nine towns from her 1998 study, Miranda reported that seven cited problems with illegal dumping or burning of trash. A four-week observation immediately following implementation of a PAYT program in Charlottesville, North Carolina found that 5.33 percent of households disposed of garbage illegally, with illegal disposal accounting for 28 percent of the overall reduction in MSW.
Regarding costs, two analysts in 1999 concluded that MSW was reduced by approximately 1.6 lbs. per household per day at a savings of $0.06 per household per day. Another study indicated that with a $1.31 per 32-gallon price for PAYT, MSW will decrease by 20 percent (183 lbs.) annually per capita. Table 1 summarizes price elasticities found in previous studies.
THE NEW HAMPSHIRE EXPERIENCE
Our study constructed and tested three basic regression models. The first model tested the hypothesis that the marginal price of a PAYT service (measured by price per bag) is inversely related to waste generation. We also examined other variables which might affect waste generation such as property taxes, population, per capita income level, and whether the town has curbside trash pickup or ordinances on mandatory recycling programs. The second model was designed to test the effect of the existence of a PAYT program on MSW generation rates. To test this hypothesis, we used the same variables as in the first statistical models, but instead of price included a variable describing whether a town has a PAYT program or not. Finally, using the variables from both regression analyses, a logit model was used to predict a community’s likelihood of adopting PAYT.
Analyses focused on the year 2000 due to the availability of United States Census data. Thus, all towns used in this analysis had adopted the PAYT program prior to 2000. Choice of explanatory variables used in the estimation was based on previous studies. The average price of a normal garbage bag was used for the unit price variable for towns that do not have a PAYT program. A price of $.0375 cents per 15-gallon bag (based on Wal-Mart prices) was used for the price variable in towns without a PAYT program.
Annual mean per capita waste disposal rates in New Hampshire are 0.488 tons/person. This works out to 2.67 lbs. per person, per day. The number captures the amount of waste reported by the MSW industry in tipping fees per ton. It does not capture the total amount of waste generated, which may include recycled and composted materials. This makes it hard to compare with the 4.46 lbs per person, per day national average cited by the EPA in 1998.
Data were collected for all 235 towns in New Hampshire from state sources. Information on the 34 towns which currently have PAYT programs, and details regarding each program including marginal cost to homeowners in cents per gallon, were obtained from the State of New Hampshire Governor’s Recycling Program. Of the 235 incorporated municipalities in New Hampshire, 186 were included in the analysis. Two of the 35 municipalities that had adopted PAYT programs were dropped because they adopted their programs in 2001 and two were dropped because they did not have MSW generation data for 2000. This resulted in 31 PAYT communities included in the analysis.
DATA ANALYSIS AND RESULTS
The regression analysis used both double log and linear forms. In order to determine the effect of price of waste disposal on average per capita generation rates across towns, the analysis included average annual per capita municipal solid waste production as the dependent variable and marginal price per 15 gallon bag as one of a series of independent variables. Towns with and without PAYT were included in this analysis.
Several variable coefficients that were statistically significant in other studies were not statistically significant in our New Hampshire study. For example, this study found that income was not a statistically significant influence, and that neither mandatory recycling nor curbside pick-up had any statistical significance in this study.
Average household size, existence of a capital improvement plan, and marginal price to dispose of waste were statistically significant influences. Household size may imply that households with more people tend to buy in bulk resulting in less packaging. Towns that have a Capital Improvement Plan may exhibit characteristics of a more forward-thinking policy-making regimen and may be more apt to adopt a new program such as PAYT.
The results suggest that marginal price for MSW disposal can lower annual per capita MSW generation rates. Using conversion factor where 15 gallons of household MSW weighs 10 pounds, results indicate that a one percent increase in the price of disposal per gallon will lead to a decrease in annual per capita municipal solid waste of approximately 10.53 gallons, or 7.0 pounds per capita per year. The average per capita MSW generation rate in New Hampshire, derived from our database is 0.488 tons or 976 lbs. Applying this to a community of 26,000 (approximately the size of Portsmouth, New Hampshire) would result in an overall reduction in household MSW of 91 tons per year.
Own price elasticities of demand were calculated for both the linear and logarithmic coefficients. For the linear model, own-price elasticity at mean variable values was – 0.31. Elasticity for the logarithmic form was – 0.633. Thus own price elasticity of demand for waste disposal is relatively inelastic and comparable to those found by previous studies.
To explore the possibility that simply the existence of a positive marginal price (or the program itself) influenced the average per capita MSW generation across the sample, models were estimated using only existence of PAYT as a dichotomous variable rather than price per gallon. The coefficient for existence of a PAYT program was significant at the 90 percent level in the linear form and at the 99 percent level in the logarithmic form. According to our results, towns with PAYT programs currently being implemented produce 532 gallons less waste per capita than towns that do not have the PAYT program. Using our conversion, 532 less gallons per capita equals 354.6 pounds per capita of MSW, or 0.18 tons. This equates to a reduction in per capita generation of about 37 percent, which is within the range found in other studies.
In order to predict the likelihood of a town adopting a PAYT program based on the variables collected for this study, a logit model was run using PAYT as the dependent variable. Average household size, property tax, and per capita MSW generation had coefficients statistically significant at either the 90 or 99 percent levels. According to the analysis, the larger the average household size in a town, the less likely that town was to adopt a PAYT program. Additionally, the lower the per capita waste generation rate, the more likely a town was to adopt a PAYT program. This raises the issue of causality. It has already shown been that PAYT and average household size do, in fact, have statistically significant negative influences on per capita waste generation rates. This logit analysis questions whether a town with low per capita MSW generation rates is more likely to adopt a PAYT program or whether the existence of the PAYT program is the reason the per capita MSW generation rate has a negative coefficient. Results also suggest that the higher the property tax in a town, the more likely a town may be to adopt a PAYT program. This result suggests some sort of “fiscal stress” category of influence.
POLICY IMPLICATIONS
This study illustrates the dynamics of managing the State of New Hampshire’s solid waste stream. The analysis found that an increase in marginal price for solid waste disposal above and beyond the normal price of store bought garbage bags, and a pro-active approach to community planning, reduce annual per capita waste generation. It was also apparent that certain community characteristics, such as average size of households, can affect solid waste generation.
Although results show that an increase in marginal price reduces per capita MSW generation rates, PAYT may not be right for every town. This study presents results gleaned from analyses using mean annual data taken from all towns in New Hampshire. Each town must consider specific characteristics and variables that may influence solid waste policy and make a determination based on that specific data. The design and implementation of a PAYT program is not without administrative costs, which range from staff time to record keeping to education and outreach.
Although this study shows that implementing a PAYT program lowers solid waste generation rates, the overall savings will only be beneficial to the community if they outweigh the cost of program implementation. These cost analysis studies should look at all aspects of solid waste management options, such as recycling programs and local level composting.
The question of why or how disposal rates decreased was not explored. Further research is needed to explore where the decrease in MSW generation originates. Are households in towns where a PAYT program exists more environmentally conscious with tendencies to buy in bulk or make purchasing choices of items with less packaging or re-use value? Or do these communities engage in illicit dumping or burning because of the increase in cost?
Shanna Hallas-Burt is City Planner for Laconia, New Hampshire. John M. Halstead is Professor and Chair of the Department of Resource Economics and Development at the University of New Hampshire. Additional details of the study, including full references, are available from Professor Halstead. Thanks to Liz Bedard, Ju-Chin Huang and Mark Morgan for assistance in this research project, as well as the New Hampshire Agricultural Experiment Station for funding support.