Grantee Research Project Results
2002 Progress Report: Looking Inside the Black Box: Microlevel Empirical Analyses of the Impact of State and Federal Policy Instruments on Hazardous Waste Generation and Management
EPA Grant Number: R828952Title: Looking Inside the Black Box: Microlevel Empirical Analyses of the Impact of State and Federal Policy Instruments on Hazardous Waste Generation and Management
Investigators: Good, David , Richards, Kenneth
Institution: Indiana University - Bloomington
EPA Project Officer: Aja, Hayley
Project Period: April 1, 2001 through March 31, 2003 (Extended to May 20, 2005)
Project Period Covered by this Report: April 1, 2001 through March 31, 2002
Project Amount: $180,917
RFA: Market Mechanisms and Incentives for Environmental Management (2000) RFA Text | Recipients Lists
Research Category: Environmental Justice
Objective:
Although many have developed analytic ideas about how federal- and state-level regulations affect the production and treatment of hazardous wastes, there is little empirical evidence about how these regulations work in practice. The objective of this research project is to empirically evaluate the effects of the several alternative policy instruments that states have at their disposal, and under what circumstances they are most effective. To achieve this objective, we are developing databases of policy instruments that have sufficient temporal and cross-sectional variation that allow us to model decisions and identify impacts on hazardous waste generation and management.
Progress Summary:
Progress has been made in three main areas. The first is in the explanation and comparison of state-level hazardous waste policy instruments. The second involves characterization of hazardous waste generation in each state. The third area involves developing the models and methodology to be used in data analysis. The summary below is divided into corresponding areas.
Inventory of State-Level Hazardous Waste Policy Instruments
One of the key tasks in this research project is the development of a database of measures that states have adopted under the authority granted them by the Resource Conservation and Recovery Act (RCRA). We used the Bureau of National Affairs (BNA) Environment and Safety Laboratory Digests of State Programs, which outlines how state programs exceed the national RCRA requirements. We also looked at policies according to the targeted entity: generators, transporters, or treatment, storage, and disposal (TSD) facilities. The results showed that states have largely focused on four types of instruments: (1) voluntary programs; (2) increased tracking and reporting requirements; (3) liability and audit privilege laws; and (4) fees for generators, transporters, and TSD facilities. Relatively few states use command and control regulation or informational programs.
Voluntary programs involving technical assistance and public recognition have been adopted by many states. These programs often include assistance in legal interpretation and incentives for environmental beneficial activities such as the redevelopment of brownfields, pollution reduction equipment purchases, and recycling. Also, the vast majority of states have developed tracking and reporting requirements that exceed RCRA requirements in either scope of materials covered, detail of reports, or frequency of filing, and sometimes all three.
Liability laws vary across states. Often, parties are required by the state to demonstrate predetermined levels of preparation to assume liability by posting bonds or obtaining environmental insurance at a mandated level. This provides quite different incentives with respect to the precautions that firms will adopt. Periodic studies by the Environmental Law Institute regarding hazardous materials liability laws are useful, yet subject to the limitation that the judgments used may change from year to year with the change in individuals conducting the study, although there is no real change in the rules. These studies have been re-evaluated and adjusted for use in our analysis. Our revised data, completed for all states from 1988 through 2001, now more consistently reflect the laws on the state books and are less reliant on the changing judgments of individuals compiling the data, or on those of state regulators answering the survey regarding the nature of implementation. We have constructed a similar database for audit privilege laws.
Fees and taxes have been used widely to cover administrative costs and, in some cases, to encourage pollution prevention. Of the various instruments reviewed, both fees and liability rules showed distinct promise as explanatory variables. Fees were particularly favorable because they appeared to vary considerably across states and over time, could be categorized into relatively few distinct types, and provide a quantitative rather than qualitative variable. The research team decided to develop a complete, detailed database of state hazardous waste fees, which will allow us to begin the econometric testing with fees as an explanatory variable. As the econometric analysis proceeds, we hope to expand the policy instrument database to include a detailed characterization of state liability laws.
The project also is developing the database of fees required, generating a taxonomy of fees and taxes that would completely characterize the many types of fees currently and previously in existence in all states. We examined our preliminary taxonomy by applying it to the first 14 states to be analyzed. It was discovered that the original taxonomy was inadequate to allow us to completely categorize the range of fees that states employ. We then developed a new taxonomy with 56 distinct categories that should allow us to systematically characterize any of the state fee structures with acceptable accuracy. We have begun classification of the fees employed by all of the states using this revised taxonomy.
Some difficulty has arisen surrounding the large number of possibilities for fee and tax structures. Because several of the states employ fees based on a step function that is not a strictly quantity-based fee or an annual fee, we will have to generate some form of roughly equivalent quantity-based fee for use as an independent variable. Progress on state-level taxing information has been slow, due to the complexity of the systems and lengthy correspondence often required to clarify how they are implemented. To date, fee data for 35 states are complete.
Characterizing Hazardous Waste Generation
We build models to describe a quantity of produced undesirable output associated with each type of waste stream. Each Biennial Reporting System (BRS) participant completes the survey for every hazardous waste stream produced or managed onsite or offsite using a series of waste codes, source codes, and form codes. Given the large number of possible waste stream codes, form codes, and source codes, facilities may report essentially identical waste streams quite differently. This leads to a model, which is parametrically complicated because each of these waste streams, as well as substitutions among them, must be identified.
Our models rely on an unbalanced panel data set, in the sense that firms do not always appear in the data set in all years, but typically, firms will be in the data for more than 1 year. This allows us to focus on changes in behavior for the same firm, and how these correspond to changes in laws and regulations over time, rather than making inferences. To maintain a reasonable level of homogeneity for our models, because of our estimation of production technologies, we have focused our efforts on six main industries: Printed Circuits (Standard Industrial Classification [SIC] 3672), Plating and Polishing (SIC 3471), Wood Furniture (SIC 2511), Commercial Printing (SIC 2752), Paint (SIC 2851), and Automobile Parts (SIC 3714). The choice of these industries was dominated by four factors: (1) the production of reportable levels of hazardous wastes by a large number of facilities, (2) the availability of summary information about hazardous waste generation technologies from either the U.S. Environmental Protection Agency's (EPA) industry notebooks or similar documentation, (3) relatively homogenous type of product produced, and (4) special attention in hazardous waste laws and policy.
We accommodate the variety of reporting and the complexity of the waste stream by aggregating them using four different approaches. The first is based on the type of waste (as defined by the 40 Code of Federal Regulations [CFR] § 261) and is mapped into streams that are acutely toxic, toxic, corrosive, ignitable, and reactive. The second approach is based on the U.S. EPA's industry notebooks (or similar trade association sources) and maps waste streams into three to five industry-specific aggregations based on the generation process. The third and fourth approaches examine the form of wastes and are more appropriate aggregations for describing the waste management approach. The third aggregation method uses generic groupings, while the fourth approach is industry specific.
Models and Methodology
Our investigation of state-level hazardous waste laws has prompted us to rethink our original model along two different dimensions. Most states use some form of nonlinear or multipart pricing, which must be converted to a usable format. In the extreme, maximum likelihood techniques would lead to different functional forms for a state’s fee structure at each point in time. Because BRS allows facilities considerable latitude in how they choose to report waste streams, the waste stream data are incompatible from one facility to the next. We find many instances where a facility eliminated one waste stream, but simply started another, which appears to be functionally equivalent under a different source code.
We begin by viewing the generation of hazardous wastes as a single quantity without weighting for toxicity or any other characteristic. This approach is most amenable to the fee structures, because the majority of states do not distinguish by type of waste. In some of our models, these features are added by disaggregating the waste streams using their codes. In other approaches, these are added as quality attributes, much like what would be done in a hedonic model. Our methodology uses quality adjustment as part of the efficiency estimation process, rather than adjusting the efficiency scores after the fact and doing so potentially inconsistently. We call it Semi-Parametric Envelopment Analysis (SPEA), because it is a mix of parametric and nonparametric techniques.
Our efforts in estimation have been complicated by the Census change in rules that no longer permit the use of compiled languages. In addition, our heavy reliance on new models, which are not part of standard programs, many involving likelihood functions that do not exist in close algebraic form, rely on either multivariate Monte Carlo integration (truncation models) or layered optimization (data envelopment models). Our models are being transformed to use a mix of SAS, Systat, Limdep, and Matlab.
Future Activities:
Our project has three types of studies. The first is based solely on the analysis of state-level hazardous waste laws. The second involves state-level analysis of hazardous waste generation. The third type of study involves developing the models and methodology of data analysis.
The compilation of state taxes and fees related to hazardous waste alone is beneficial. We will publish several law journal articles using the results of the database. The initial set of articles will be targeted at state or regional law journals providing a comprehensive description of how states have used hazardous waste fees and what effect the fees will have on various types of businesses. This will allow state regulators and legislators to meaningfully compare the implications of their choices for fees and taxes relative to their neighbors. We already have data sufficient for articles on the Midwest and Northeast. As we complete additional regions, we will use a similar approach to generate analysis for the Mid-Atlantic, the Southeast, the Gulf states, the Southwest, and the West. We soon will have the 48 contiguous states completed, at which point we will develop a national comparison. Once the national database is completed, we will write an article considering the rationality of the fee and tax systems used by the states. Given the tremendous amount of variation among states’ fees, both in form and level, it is hard to imagine that they are all based on a normatively rational approach. Which ones make sense based on regulatory theory, and which ones could likely be improved? Are there political implications of various designs that might explain why states adopted them? Could variation among the states with respect to their generation levels, composition of industries and types of wastes, and population density explain why a design that is optimal in one state is not optimal for another? The article will explore these issues, while drawing on the empirical observations provided by the database.
Our aggregate-level studies will use the state-level aggregated data on broad categories of hazardous waste generation and treatment, and relate them to liability and audit privilege laws. Given the unrepresentative nature of BRS data, these studies are based on our model of imputation of the quantity of wastes generated at the state level. The advantage of these studies is that they are relatively simple econometrically, though we must address potential selectivity biases. The time series cross-section approach of these studies allows us to separate differences in waste generation due to laws from effects more related to geographic differences of corporate culture.
A large portion of our efforts has been focused on micro-level studies. Similar to the methodologies described above, they are focused on examining production relationships (including the onsite production of hazardous waste treatment). For many of the hypotheses we wish to test, studies focus on the production technology. For others, they focus on the relationship between productive efficiency and the production of hazardous wastes, their treatment, and laws.
Journal Articles:
No journal articles submitted with this report: View all 7 publications for this projectSupplemental Keywords:
policy instruments, policy incentives, risk management, hazardous waste production, corner solutions, nonlinear pricing, data envelopment analysis, behavior change, corporate environmental behavior, effects of policy instruments, environmental behavior, environmental decision-making, environmental policy, environmental quality, hazardous waste generation, hazardous waste management, impact of federal policy instruments, impact of state policy instruments, incentives, market incentives, market-based mechanisms, policy making, pollution prevention, pollution reduction, public policy., RFA, Economic, Social, & Behavioral Science Research Program, Scientific Discipline, Waste, Hazardous Waste, Corporate Performance, decision-making, Hazardous, Market mechanisms, Environmental History, Economics & Decision Making, Social Science, Environmental Law, hazardous waste management, environmental quality, market incentives, market-based mechanisms, policy instruments, effects of policy instruments, impact of state policy instruments, impact of federal policy instruments, policy making, policy incentives, technology-based regulation, incentives, decision making, environmental decision making, risk management, hazardous waste generation, environmental policy, pollution reduction, public policy, environmental behavior, behavior change, pollution preventionProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.