Trash Talk: Pay-As-You-Throw Systems

Should communities adopt "pay-as-you-throw systems" for garbage collection to reduce the amount of garbage flowing into landfills or incinerators?:

Kicking the Cans, by Robert Tomsho, WSJ: Plymouth, Mass. -- In this historic community ... garbage has become ... a touchy subject...

During months of debate, Mr. Quintal, chairman of the town's governing board of selectmen, argued that people who throw out more trash should pay higher disposal bills. "I got emails from people saying they thought I was right," he says. "But there were just as many from those who thought I was an idiot."

Like Plymouth, more and more communities are grappling with whether to abandon traditional garbage service and adopt so-called pay-as-you-throw systems. With PAYT, residents are charged based on how much garbage they generate, often by being required to buy special bags, tags or cans for their trash. Separated recyclables like glass and cardboard are usually hauled away free or at minimal cost. ...

PAYT represents an effort to curb garbage's impact on the ecosystem by pressuring consumers to create less of it. But the effort to make people change their habits has often stirred tension...

While Americans are accustomed to paying for utilities like water and electric based on use, that's not true about garbage in most places. ...

Tampering with that notion can be tricky in communities that switch to PAYT. Illegal dumping has cropped up in about 20% of such communities, according to a 2006 U.S. Environmental Protection Agency report. Local officials also complain about variations of the so-called Seattle stomp (named after one of the first PAYT cities), where homeowners try to beat the system by compacting huge amounts of trash into a single can or bag.

There has also been a recent backlash in some locales over costs and inconvenience...

Supporters of PAYT say it gives residents a direct economic incentive to recycle. Skumatz Economic Research Associates, a waste-consulting concern in Superior, Colo., estimates that PAYT programs lead to a 17% reduction in the flow of residential waste to incinerators and landfills... "Every analysis shows that this is a very cost-effective thing to do," says Lisa Skumatz, the firm's principal.

I should stop here and comment, every analysis doesn't show that, but we'll come back to the cost-effectiveness in a moment. Continuing:

...The system was supposed to go into effect July 1, but the grumbling began almost as soon as details were announced this year. While few Plymouth residents spoke out against the concept of recycling, there was criticism aplenty for the plan designed to pressure them to do more of it in a hurry.

At public meetings, senior citizens complained that, even with the reduced annual fee, their personal disposal costs would go up. Some critics said the new trash plan would be unfair to big families. ... [There are] also fears that PAYT will prompt residents to burn trash in their fireplaces and dump it illegally. ...

[The] board voted last month to hold off on overhauling the town's garbage system for at least another year. Mr. Hammond, the public works director, says the garbage debate has been an eye opener. "A lot of people don't want to recycle," he says. "They just want to throw everything in the bin."

I remember seeing a seminar on this awhile back. I think this was the paper:

Household Responses to Pricing Garbage by the Bag, by Don Fullerton and Thomas C. Kinnaman, The American Economic Review, September 1996: Introduction The average tipping fee paid by garbage collectors to landfills has tripled over a six-year period, largely due to rising land prices and new EPA regulations... Several communities and private firms have responded ... by implementing volume-based pricing programs that require households to pay for each bag or can of garbage presented for collection. These towns employ unit pricing not only for additional revenue, but to reduce their direct costs and external costs from using landfills and incinerators. Households might recycle more, compost more, and demand less packaging at stores. Unfortunately, they might also bum garbage or dump it along deserted roads. The attractiveness of unit pricing depends crucially on the extent of each such method of garbage reduction.

The price per bag might also induce households to compact garbage into fewer bags. This practice, known as the "Seattle Stomp," was noticed first when Seattle started an early unit-pricing program. It is not helpful, since collectors compact the garbage anyway.

This paper employs individual household data to estimate the effect of such a program on the weight of garbage, the number of containers, the weight per can, and the amount of recycling. We also provide two indirect measures of illegal dumping. The data are based on a natural experiment... On July 1, 1992. Charlottesville, Virginia, implemented a program to charge $0.80 per 32-gallon bag or can of residential garbage collected at the curb. Before and after the implementation of this program, we counted and weighed the bags or cans of garbage and recyclable materials of 75 households. In response to this new price, the average person living in these households reduced the weight of garbage by 14 percent, reduced the volume of garbage (number of containers) by 37 percent, and increased the weight of recycling by 16 percent. Our indirect measures suggest that additional illegal dumping may account for 28 percent to 43 percent of the reduction in garbage.

Based on these data, the change in weight of garbage is statistically significant, but small. The implied arc-price elasticity is only -0.076. We ... conclude that this pricing program has little effect on the weight of garbage. ...[H]owever, we find more substantial effects on volume, density, recycling, and illegal dumping. ...

We ... calculate the effect of introducing a minimum of one bag per week, and we conduct a simple cost-benefit comparison. Welfare benefits from unit pricing range from $0.08 to $0.15 per $0.80 bag of garbage, but administrative costs are likely to exceed $0.19 per bag.

...

Conclusion This paper has used original data gathered from individual households to estimate responses to the implementation of a price per bag of garbage. We find that households reduced the number of bags, but not necessarily the actual weight of their garbage. Thus households stomped on their garbage to reduce their costs. They also increased the weight of recycling, and they might have increased illegal dumping. The reduction in weight of garbage at the curb is 14 percent. If we account for the amount of illegal dumping, using our lower estimate, then the true reduction in garbage is only 10 percent. Recycling increased by 16 percent. Many in Charlottesville were already participating in the voluntary recycling program before unit pricing began. Thus the incremental benefit of unit pricing is small. In our simple comparison, this social benefit does not cover the administrative cost.

Here's something more recent from a Resources for the Future commentary. Skipping over the statistics on landfill usage and recycling, here's the part on the PAYT programs. It's from one of the same authors as the paper above, Don Fullerton, and its conclusions are similar but a bit more generous toward the program:

Trash Talk, by Don Fullerton and Margaret Walls, RFF, December, 2007: ...The economist's typical solution to an externality problem is a Pigouvian tax: charge a tax or fee per pound of trash exactly equal to the social damages imposed by that trash. That would reduce waste in landfills, but it raises two questions. The first is whether the social damages can actually be estimated. Even if policymakers know what to charge, however, the second question is whether any such fee can feasibly be administered and enforced.

Some communities charge for each can or bag of trash, under a system commonly called "Pay as You Throw" (PAYT). Households might be charged one monthly amount for one can a week, or a higher monthly amount for a larger can or two cans a week. But not every can gets filled every week, leaving households with no incentive at the margin to reduce waste. A better system, closer to true marginal cost pricing, requires households to buy a special bag at the grocery store, or a special tag to use on a bag of garbage of a particular size.

EPA estimates that approximately 7,100 communities in the United States use some kind of PAYT, making it available to approximately 25 percent of the country's population. The number of communities has risen over time and, in some areas of the country, is quite high. Some states (Wisconsin, Oregon and Minnesota) even have a law requiring that communities use PAYT.

Does it work? Results from the economics literature suggest that demand for garbage collection is relatively unresponsive to prices, but PAYT towns have experienced some reductions. And it is important to keep in mind that even if reductions are small, charging the right price may result in the right amount of garbage disposal. Fixed monthly charges - the norm in many places - set a zero price for an additional bag or can and thus provide no incentive for households to conserve.

The big question for PAYT communities, though, is what households are doing with the garbage they no longer place at the curb. To avoid paying the fee, households can reduce their waste by recycling, composting, consuming less in the first place, or disposing illegally - burning, finding a commercial dumpster, or throwing it by the side of the road. Recycling does increase with PAYT but not enough to account for all of the reduction in trash. Clearly, municipalities can help themselves by providing free curbside collection of a wide variety of materials for recycling and yard waste collection for central composting. Towns also must choose how much to spend on enforcement, and how to set penalties.

PAYT is most effective in small cities and suburban areas but has not worked so well in densely populated urban areas where apartment dwellers use chutes and dumpsters for their normal disposal (and might easily use vacant lots for everything else). PAYT is also not as well-suited to very rural areas where illicit dump sites are similarly easy to find. In general, it is most feasible where we can measure and monitor individual households' weekly trash and recycling.

Even in towns where a PAYT fee works well to reduce waste amounts without increased dumping, it does nothing special for separate handling of hazardous and other troublesome items like batteries, tires, or used electronic equipment. These products, especially, are candidates for some kind of deposit refund system (DRS). Experience has shown great success with a DRS applied to certain products: beverage containers in "bottle bill" states have recycling rates that range from 60-95 percent, significantly higher than in states without such a program; 96 percent of lead-acid batteries are recycled; and tires in states with a DRS are recycled at a 72 percent rate. But the idea can be generalized, in a "two-part instrument," a general sales tax on everything at the store - all of which eventually becomes waste - along with a subsidy per ton of waste handled at the recycling center. Products like computer monitors could still be specifically targeted with a special fee, but most items could be treated in bulk, without time-consuming transactions to count or weigh individual items.

Thus the "best" policy is not any single policy. PAYT can successfully be employed in at least some communities, and probably in more than are currently doing so. Other towns, however, need a two-part instrument - a general sales tax on new items at the store, plus a subsidy for recycling. And products that pose special problems may need targeted deposits or refunds. Different circumstances therefore call for different policies, PAYTs, DRSs, or two-part instruments. All of these policies have a key feature in common and one that economists invariably seek in all of their policy prescriptions: they provide the proper incentives to consumers and others to generate a socially desirable outcome.