Grantee Research Project Results
Final Report: Environmental Policy and Endogenous Technical Change: A Theoretical & Empirical Analysis
EPA Grant Number: R826610Title: Environmental Policy and Endogenous Technical Change: A Theoretical & Empirical Analysis
Investigators: Opaluch, James J. , Grigalunas, Thomas A. , Jin, Di
Institution: University of Rhode Island , Woods Hole Oceanographic Institution
EPA Project Officer: Chung, Serena
Project Period: October 1, 1998 through September 30, 2001
Project Amount: $325,000
RFA: Decision-Making and Valuation for Environmental Policy (1998) RFA Text | Recipients Lists
Research Category: Environmental Justice
Objective:
The objectives of this research project were to: (1) develop a deeper understanding of the relationship between technical change and alternative environmental policies that accounts for environmental inputs and depletion of natural capital stocks; (2) use a case study to measure historic rates of technical change, accounting for environmental inputs; (3) compare ex ante, engineering estimates of the costs of complying with environmental regulations to actual, ex post performance that includes innovative means of achieving standards, similar to process change; (4) estimate benefits of environmental policies that provide increased flexibility and encourage innovation; and (5) simulate the long-term relationship between productivity change and environmental protection.
Summary/Accomplishments (Outputs/Outcomes):
In this report, we provide a more fundamental understanding of productivity change, where productivity is broadly defined to include both market and environmental outputs. This allows one to identify potential gains from flexible environmental regulations within a dynamic context, where technology changes over time, and where environmental regulations can contribute to (or inhibit) development of new technologies. This extends previous economic studies, which consider the adverse economic impacts of regulations only in the context of a fixed set of production technologies.
Our approach also allows one to extend productivity measures to environmental outputs as well as market outputs. In contrast, traditional studies have measured the effects of environmental regulations on productivity by considering only the effects on market inputs and outputs. From a broader social perspective, the goal of environmental regulations is to change the mix of outputs by reducing undesirable outputs (pollution emissions) at the expense of reducing market outputs and/or increasing market inputs. These traditional studies of productivity focus only on measuring the effective cost of regulations, and not the associated environmental benefits that are achieved. An appropriate measure of the impact of regulations on productivity must consider both the costs associated with reductions in market outputs (and/or increase in inputs) and the associated environmental benefits that are achieved. We also had hoped to compare ex ante estimates of costs of complying with U.S. Environmental Protection Agency (EPA) regulations with ex post costs, but we were unable to obtain proprietary ex post cost measures.
The first steps in the project were to develop timelines for environmental regulations and for important new technologies. The list of technologies is augmented by carrying out an extensive analysis of technology announcements in industry publications, thereby extending and refining the work of Moss (1993). We further refined this index of new technologies using survey data developed by the National Petroleum Council (National Petroleum Council, 1995) on the importance of specific technologies. Together, these were used to construct an index of identifiable new technologies, which was used to decompose productivity change, as discussed below.
Simultaneously, we developed a conceptual model for endogenous technological change as a random process, whereby research and development expenditures increased the probability of making a discovery, and where technologies are complementary (Opaluch, 2000). Complementarity of new technologies is a common phenomenon, as new discoveries often make existing technologies more valuable. For example, there are important synergistic relationships between technologies such as improved computer processing power and advances in materials science. New materials improve computer processing power, which contributes to further advances in materials. Similarly, complementary relationships exist for computer components; faster hard drives and faster memory chips make faster computer processors even more productive. However, the existing literature models technologically advance as either pure substitutes, as in the Agion and Howitt (1992) vintage model, or as additively separable, as in the Romer (1990) model.
Next, we collected extensive data on inputs and outputs for outer continental shelf oil and gas production in the Gulf of Mexico. We constructed a unique micro-level data set, including the following: (1) production data, including monthly oil, gas, and produced water outputs from every well in the Gulf of Mexico from 1947 to 1998. The data include a total of 5,064,843 observations for 28,946 production wells; (2) borehole data describing drilling activity for each of 37,075 wells drilled from 1947 to 1998; (3) platform data with information on each of 5,997 platforms, including substructures from 1947 to 1998; (4) field reserve data, including oil and gas reserve sizes and discovery year of each of 957 fields from 1947 to 1997; and (5) reservoir-level porosity information from 1974 to 2000. These data include a total of 15,939 porosity measurements from 390 fields.
These data were used to measure productivity change and to decompose productivity change into various constituents to provide a more fundamental understanding of the process within the context of our case study. We recast the traditional issue of technological change by recognizing that production of market goods implicitly embodies joint production of market outputs and environmental commodities, so that our measures of productivity include both market and non-market outputs. In comparison, traditional measures of productivity change consider market outputs only.
Recent literature has suggested that it may be possible to develop environmental regulations that encourage innovation, such that regulations increase productivity of market outputs, while also providing environmental benefits. This is the so-called Porter hypothesis (e.g., Porter, 1991; Porter and van der Linde, 1995). Recent theoretical literature has confirmed that the Porter hypothesis is not necessarily inconsistent with economic theory of rational behavior by firms (e.g., Xepapadeas and Zeeuw 1999; Mohr, 2002). As part of our research, we provide the first true empirical test of the Porter hypothesis discussed below.
We apply Data Envelopment Analysis ([DEA]; e.g., Charnes, et al., 1978; Färe, et al., 1985) to measure Malmquist indices (e.g., Malmquist, 1953; Caves, et al., 1982a, 1982b) of productivity change in the offshore oil and gas production in the Gulf of Mexico. This is an important industry because energy resources are central to sustaining our economy and because petroleum products currently are the key energy resources. Furthermore, offshore oil production is a technology-intensive industry that embodies important tradeoffs between production of market outputs and associated adverse impacts on the environment.
The first product of our research (Managi, et al., 2002a) focuses on the relative sizes of depletion effects and technological progress for offshore oil production in the Gulf of Mexico using our unique field-level data set from 1947 to 1998. We update and adapt previous innovation measures (Cuddington and Moss, 2001). The Cuddington and Moss index is based on a simple count of innovations. We use an industry survey of technology needs by the National Petroleum Council (NPC) (NPC, 1995) to create an importance-weighted index.
The study supports the hypothesis that technological progress has mitigated depletion effects over the study period, but the pattern differs from the conventional wisdom for nonrenewable resource industries. Contrary to the usual assumptions of monotonic changes in productivity or an inverted "U" shaped pattern, we found that productivity declined for the first 30 years of our study period. More recently, the rapid pace of technological change has outpaced depletion and productivity has increased rapidly, particularly in the most recent 5 years of our study period. This is consistent with reports of accelerating technological progress in recent years (e.g., Bohi, 1998). The central role of technological change in maintaining economic viability of this industry underscores the importance of designing environmental policy so as to encourage (or at least not unduly inhibit) development and implementation of new technologies.
We also provide a more fundamental understanding of different components of technological change and depletion. We find that both diffusion and learning-by-doing play more important roles in productivity change than innovations based on specific new discoveries. We find that, early on in our time series, field size is a more important factor in limiting productivity, but that water depth is more important later. This is consistent with findings of very large fields in increasingly deep waters in recent years.
We work with an aggregate model of offshore production to study technological change in new discoveries in the Gulf of Mexico (Managi, et al., 2002b). We employ a yield per unit effort (YPE) model that relates discoveries to the aggregate "effort" placed in exploration. We again find an increasing pace of technological change over time, so that net productivity has been increasing in recent years. We also find that recent productivity increases have overcome depletion-related productivity declines of the earlier years (late 1940s through the 1970s).
Next, we rethink the Porter Hypothesis and test a revised version (Managi, et al., 2002c). The Porter Hypothesis states that strict environmental regulations can induce firms to innovate and become more efficient, ultimately contributing to productive efficiency and profitability in the long run. Thus, the Porter Hypothesis implies that well-designed environmental regulations can potentially provide a win-win solution, with increased profitability and reduced pollution. Recent theoretical literature has demonstrated that, because of market imperfections in technical innovation, the Porter Hypothesis is not necessarily inconsistent with rational behavior by firms.
We recast the Porter Hypothesis to measure efficiency with respect to joint production of market and nonmarket outputs (e.g., Repetto, 1996). Thus, productive efficiency is measured, taking into account not only market outputs, but also pollution levels that affect the supply of environmental commodities. Our results show a long-run upward trend of productivity in the environmental sector, despite increasing environmental stringency. Our results reject the standard Porter Hypothesis, which considers productivity of market outputs only. Our results support a recast version of the Porter Hypothesis, where efficiency is measured with respect to joint products comprised of vectors of market and environmental commodities.
The causality between innovation and stringency of environmental regulations could go in either direction. More stringent environmental regulations could result in innovation to meet new challenges. Alternatively, cost-reducing technological innovations could result in the implementation of more stringent technology-based environmental regulations. We use Granger causality tests to test the direction of causality between innovation and environmental regulation. Although we cannot come to any firm conclusions of the causal relationship between environmental stringency and technological innovation (new inventions), we find a clear causal direction from environmental stringency to less structural aspects of innovation, such as so-called "learning by doing." This implies that tougher environmental standards do not necessarily lead to new technological innovations, but they clearly lead to increased efficiency of operations as a result of learning from experience.
We also compare alternative indices of technological innovation (i.e., identifiable new technologies) to assess whether some proxies perform better than others in the offshore oil and gas industry (Managi, et al., 2002d). To do so, we compare use of patent counts, weighted patent counts, innovation counts, and weighted innovation counts to explain the Malmquist index calculated using DEA. Our initial analysis finds that all of proxies fit relatively well as an approximation in cumulative case, but none of the alternative innovation indices appear to be clearly superior to the others. Cumulative innovation indices appear much more promising than incremental innovation indices as a proxy measure of innovation. This suggests that there may be a problem of using proxy measurement for short-term analysis.
We also use stochastic production frontier analysis (SPF) (Aigner, Lovell, and Schmidt, 1977; Meeusen and van den Broeck, 1977) to assess technological change in the offshore oil and gas industry (Managi, et al., 2002e). Being a statistical technique, SPF can be used to construct confidence intervals and can validate the DEA results. Results of our SPF model suggest that the effect of technological change on the offshore oil and gas industry at the field level was substantial over the study period from 1947 to 1995. Because of technological progress, the negative effect of resource depletion on field-level production frontier has been declining over time. Similarly, the negative impact of water depth on the production frontier has been falling. The results reveal that environmental regulation had a significantly negative impact on offshore production, although such impact has been diminishing over time because of technological change and improvement in management.
We then develop a simulation model based on the disaggregated, field-level data discussed above (Managi, et al., 2002f). The model is used to forecast the future production and pollution of the offshore oil industry through 2050 under alternative assumptions regarding: (1) the rate of new resource discoveries; (2) the rate of technological change; (3) the stringency of environmental regulations; and (4) the form of environmental regulations (command and control versus flexible environmental regulations). Given the inherent uncertainties involved, the emphasis is on developing a reasonable range of benefits that might result in identifying critical parameters through the use of sensitivity analyses.
Historic data are used to simulate evolution of the industry to date. The model based on disaggregated field-level data is used to forecast production and pollution through the year 2050 under different scenarios regarding technical change, future resource discoveries, and alternative environmental policies. We address various policy questions, such as identifying potential benefits from innovative pollution control policies and the associated benefits that can be derived from flexible approaches, such as market-based approaches for pollution control. This improved understanding of the potential role of technology and environmental policy can provide policy-relevant information for designing and implementing sound environmental policies.
We use different scenarios to explore the significance of various factors in determining forecasts. The baseline scenario uses historic rates for technological change, the number of new field discoveries, and the change in the stringency of environmental regulations. The baseline scenario also assumes that environmental regulations will continue to be based on command-and-control.
In our baseline scenario, oil and gas production increases by approximately 1.5 percent per year until 2020, when a declining trend sets in. Pollution levels remain relatively constant until 2014, and start to gradually decrease thereafter.
Various scenarios are used to explore how results change with alternative assumptions regarding: (1) R&D expenditures and associated levels of technological change; (2) new reserve discoveries; (3) environmental regulations; and (4) flexible regulations in the Gulf of Mexico. The alternative scenarios are defined as follows:
1. R&D Expenditures and Technological Change. The high scenario for R&D assumes expenditures increase at 1 percent per year. The R&D low scenario assumes expenditures decrease at 1 percent per year. The levels of R&D expenditures then were used to forecast future rates of technological change, which in turn affect the future levels of production.
2. New Discoveries. The high scenario for new resource discovery follows the Energy Information Administration high price scenario for new discoveries. The low scenarios assume discoveries decrease linearly over time, ceasing altogether at dates ranging from 2015 to 2045.
3. Environmental Stringency. The high scenario for growth in the stringency environmental regulations assumes that stringency grows at the same rate as the historic decade with the highest rate of growth (1980s). The low scenario for environmental stringency assumes an equivalent decline in the rate of change in environmental stringency. That is, regulations continue to become more stringent, but at a slower rate.
4. Flexibility of Environmental Regulations. The high scenario for flexibility of environmental regulations assumes reductions in compliance costs following the estimates of Popp (2001). Two different assumptions are compared. The first assumes the Popp results are applicable to all fields, and the second assumes that the Popp results are applicable only to new fields, while old fields are assumed to be locked into historic technologies associated with past command-and-control regulations.
Technological change, as influenced by R&D expenditures (Alternative Scenario 1), had the greatest effect on production. The high scenario for technological change increased production (and pollution) by 189 percent compared to the baseline scenario. The stringency of environmental regulations (Alternative Scenario 3) had the smallest impact on production. The number of new discoveries also has significant impact in maintaining the long-term production. Flexible regulations (Alternative Scenario 4) applied to all fields, results in production and pollution of about 85 percent higher than baseline scenario in 2050. If flexible regulation is applied only to new fields, production and pollution increase around 20 percent higher than the baseline scenario.
Conclusions:
Technological change is an important determinant of future standards of living, particularly in a society facing natural resource depletion coupled with increasingly stringent environmental regulations. Therefore, the impacts of regulations on technological change should be a central concern in the design of environmental policy. However, regulatory requirements for assessing economic effects of environmental policies tend to be limited to static notions based on available technologies. A notable exception to this are so-called "technology-forcing" regulations, which set standards beyond current state of the art, to force development of new technologies (e.g., CAFÉ regulations).
Traditional arguments regarding the cost effectiveness of flexible environmental regulations are reinforced when the implications for technological change are considered. Indeed, our simulation results suggest implications of flexibility of regulations for technological change could be more important than the level of stringency. This means that "win-win" solutions can be attained by providing more stringent but more flexible regulations.
Benefits associated with technological change are primarily long term in nature, and short-term economic impacts can be very important to industry. This argues for long-term regulations with phased implementation, consistent with historic practices. An additional potential impediment is industry acceptability. Companies that benefit from new approaches to achieving environmental regulations may not be the incumbent market leaders at the time of the regulations. This may suggest that flexible regulations could provide opportunities for new entrants, increased competition, and changes in market leadership. This could imply an additional source of resistance by industry, as objections to regulations could be amplified by the fact that private costs to powerful incumbent firms exceed the social costs to industry as a whole. New research is necessary to test these hypotheses and to explore their implications for optimal regulatory design, which may need to consider the likelihood of obtaining political support for new regulations.
Our results, with respect to the Porter Hypothesis, are somewhat sobering. We find no support for the hypothesis that more strict environmental regulations spur innovation that leads to an increase in productivity with respect to market outputs. However, we do find evidence for a weaker version of the Porter Hypothesis, which indicates that more strict environmental regulations increase productivity of the joint production function for market and nonmarket outputs. The source of increased productivity is a goal for future research. Our results suggest that productivity increases are associated with learning obtained from experience with new environmental technologies, rather than with development of new technologies. This is as expected for an industry faced with command-and-control regulations.
An open empirical question for future research is whether the form of the environmental regulations in our particular case study has impaired industry’s ability to achieve potential productivity gains. That is, the Porter Hypothesis states that well-designed environmental regulations could spur innovation leading to increases in productivity of market goods. However, regulations in the offshore oil industry historically have been command-and-control. Therefore, an interesting hypothesis is whether it is the form of regulations that have impeded potential improvements in productivity. A cross-industry study might be useful in exploring this hypothesis, where the data include sectors with varying degrees of flexibility in achieving environmental regulations.
References:
Agion P, Howitt P. A model of growth through creative destruction. Econometrica 1992;60(2):323-351.
Aigner DJ, Lovell CAK, Schmidt P. Formulation and estimation of stochastic frontier production function models. Journal of Econometrics 1977;6:21-37.
Bohi DR. Changing productivity in U.S. petroleum exploration and development. Resources for the future, discussion paper 98-38, 1998.
Caves DW, Laurits RC, Diewert WE. Multilateral comparisons of output, input and productivity using superlative index numbers. Economic Journal 1982a;92(365):73-86.
Caves DW, Laurits RC, Diewert WE. The economic theory of index numbers and the measurement of input, output and productivity. Econometrica 1982b;50(6):1393-1414.
Charnes A, Cooper WW, Rhodes E. Measuring the efficiency of decision making units. European Journal of Operational Research 1978;2(6):429-444.
Cuddington JT, Moss DL. Technical change, depletion and the U.S. petroleum industry: a new approach to measurment and estimation. American Economic Review 2000;90(5).
Färe R, Grosskopf S, Lovell CAK The measurement of efficiency of production. Boston: Kluwer-Nijhoff, 1985.
Malmquist S. Index numbers and indifference curves. Trabajos de Estatistica, 1953;4(1):209-42.
Meeusen W, van den BJ. Efficiency estimation from cobb-douglas production functions with composed error. International Economic Review 1977;18:435-444.
Mohr R. Technical change, external economies, and the porter hypothesis. Journal of Environmental Economics and Management 2002;43(1):158-168.
Moss DL. Measuring technical change in the petroleum industry: a new approach to assessing its effect on exploration and development. Presented at the National Economic Research Associations, Working Paper #20, 1993.
Opaluch JJ. Research development and demonstration needs of the oil and gas industry. Presented at the National Petroleum Council, Washington, DC, 1995.
Popp D. Pollution control innovations and the clean air act of 1990. Presented at the National Bureau of Economic Research, Working Paper Number 8593, 2001.
Porter ME. America’s greening strategy. Scientific American 1991;264:168.
Porter ME, van der Linde C. Toward a new conception of the environment-competitiveness relationship. Journal of Economic Perspectives 1995;9(4):97-118.
Repetto R. Has environmental protection really reduced productivity growth: we need unbiased measures. Presented at the World Resources Institute, Washington DC, 1996.
Romer PM. Endogenous technological change. Journal of Political Economy 1990;98(5):S71-S102.
Xepapadeas A, de Zeeuw A. Environmental policy and competitiveness: the Porter Hypothesis and the composition of capital. Journal of Environmental Economics and Management 1999;37(2):165-182.
Aigner DJ, Lovell CAK, Schmidt P. Formulation and estimation of stochastic frontier production function models. Journal of Econometrics 1977;6:21-37.
Cuddington JT, Moss DL. Technological change, depletion, and the U.S. petroleum industry. American Economic Review 2001;91(4):1135-1148.
Moss DL. Measuring technical change in the petroleum industry: a new approach to assessing its effect on exploration and development. National Economic Research Associations 1993.
Journal Articles:
No journal articles submitted with this report: View all 8 publications for this projectSupplemental Keywords:
petroleum, productivity, innovative technology, economics, technical change, offshore oil, Porter Hypothesis., RFA, Economic, Social, & Behavioral Science Research Program, Scientific Discipline, Economics, decision-making, Engineering, Economics & Decision Making, ecosystem valuation, technical innovation, decision making, economic benefits, endogenous technical change, offshore oil, cost benefit, economic incentives, empirical analysis, environmental policy, theoretical analysis, compliance costs, innovative pollution control, benefits assessmentProgress 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.