Search

Environmental degradation is one of the key challenges of our times. Economists see the problem of environmental degradation as one in which economic agents impose negative environmental externalities upon society in the form of pollution. Market-based environmental policy instruments, such as taxes or tradable permits intend to provide incentives for polluters to reduce negative environmental externalities. They seek to incorporate the external cost of production through taxes or by creating property rights and facilitating the establishment of a market for the use of environmental services. This Thesis studies the design and enforcement of market-based environmental policies from a theoretical and empirical perspective. The special focus is thereby on the often overlooked strategic interactions and the information asymmetry between the designer of an environmental regulation and the regulated agents. For instance, when designing an emissions tax system, regulatory authorities may require polluters to self-report on their level of pollution, because especially in the case of non-point emissions, it is not observable to the regulator directly which firm polluted how much. Furthermore, when designing an emissions trading scheme, regulatory authorities may require information about how many tradable permits should be allocated free of charge during the initial allocation to offset potential losses to the industry. In both examples, the incentives of the designer of the regulation and the regulated agents are not aligned. In case of an emission tax, agents might underreport their emissions to save on tax payments. Of course, it is common practice to audit some fraction of reports and to impose penalties when underreporting is discovered. However, penalties are limited and auditing is costly for the regulator and therefore the detection probability is determined by its operating budget, which may be tight. In case of an emissions trading system, firms may demand higher amounts of free permits than actually necessary to offset any losses, which might not be easily verifiable by the regulator either. The first chapter is entitled "Firms’ Emissions and Self-reporting under Competitive Audit Mechanisms''. It focuses on the restricted audit capacity of environmental regulatory authorities to verify whether or not firms are in compliance with environmental policies. I introduce a novel type of audit mechanism that induces more truthful reporting and lower emission levels by firms in comparison to the commonly applied audit mechanism. The second chapter is entitled "On Optimal Audit Mechanisms'' and builds on the analysis in the first chapter. I derive an audit mechanism that leads to socially optimal emissions while the commonly used audit mechanism in the literature fails to achieve this first-best emissions levels given the same audit resources of the regulator. The third chapter is entitled "Carbon Emissions and Stock Returns – Evidence from the EU Emissions Trading Scheme''. I jointly worked on this chapter with Professor Ilias Tsiakas. We investigate the effect of the European Union Emissions Trading Scheme (EU ETS) on the profits of firms as reflected by their stock returns. Despite facing the burden of a new regulation, portfolios of firms which were directly affected by the EU ETS significantly outperformed portfolios of firms which were not directly affected. We provide evidence that this is due to the subsidy-like effect of substantial free carbon certificate allocations from European Governments to the affected firms. Our new result informs the environmental policy debate in North America and elsewhere, as the EU ETS is seen as the prototype for a potential global climate policy regime that would be based on emissions trading. I acknowledge the financial support from the University of Guelph, my advisors, the Ontario Graduate Scholarship program and Sustainable Prosperity.

This thesis investigates the cost-effectiveness of different environmental policies in the automobile market. The focus of the second and third chapters is on the cost-effectiveness of "Mandatory Vehicle Inspection and Maintenance" programs. The results predict that the marginal abatement cost for a major representative program (the Ontario Drive Clean program) is so high that even a small reduction in the abatement target leads to substantial social cost savings. Furthermore, even for relatively high levels of the abatement target, the optimal minimum testing age is substantially higher and the frequency of testing is much lower than what is common in many jurisdictions. The fourth and fifth chapters look at the cost-effectiveness of market-based incentives in automobile market. The results suggest that a higher price of gasoline shifts vehicle holdings towards more fuel efficient vehicles and also reduces annual demand for miles traveled, whereas changes in vehicle prices have little to no impact on annual demand for miles traveled and only shift vehicle holdings. Furthermore, achieving any abatement target through a wide range of fees and/or tax on miles driven is more expensive than a tax on gasoline. The only exceptions are fees on new fuel inefficient vehicles and the Feebate program, where vehicles with low fuel efficiency are charged fees and vehicles with high fuel efficiency receive rebates. However, the maximum amount of abatement that can be achieved by these is relatively small.

The dissertation consists of three essays whose common theme is the allocation of natural resource use over time in the presence of environmental amenities. The 'first' essay consists of a modification of a classic problem in forestry economics, that of determining the optimal rotation period for a forest managed for timber production, to include the benefits of carbon uptake in wood, and the costs of subsequent carbon oxidation from wood products. When carbon values are included the private rotation age, determined by timber value only, is generally less than the socially optimal rotation age. The essay gives conditions under which a forest is harvested sooner, later, or not at all, and proposes appropriate societal intervention to insure that the private landowner harvests at the socially desired time. Despite the large literature that has evolved since Hotelling wrote his seminal paper on the optimal depletion of exhaustible resources, there remains considerable confusion about the time path of scarcity rent and price when costs tend to rise with depletion. The 'second' essay shows that Hotelling's fundamental results of rising scarcity rent and price paths are sustained and that the path of scarcity rent converges on the r-percent rule provided the objective function is concave. Nonmonotonic or declining scarcity rent paths found in the literature are shown to be due to assumptions that lead to nonconcave objective functions typically associated with models of economic, rather than physical, depletion. The 'third' essay discusses the timing of resource use given trade liberalization in raw materials and resource based commodities when there are environmental benefits that flow from the resource stock. The results indicate that trade usually leads to an increase in the aggregate amount of resource use but that the initial rate of resource use under trade is uncertain when compared to autarky. Results strongly depend on the class of preferences considered. An example is provided where the shadow price of the resource falls with unanticipated trade liberalization and provides a possible explanation for the empirically observed failure of exhaustible resource prices to satisfy Hotelling's rule.

This thesis is an investigation of three topics in environmental regulatory economics. The first essay reconsiders Winston Harrington's paper "Enforcement leverage when penalties are restricted". We show that Harrington's results are not necessarily robust to a more general specification of information and compliance cost structures. The research done in the first topic reconsiders Harrington's original work and exposes the critical effect that uncertainty and asymmetric information can have on a state-dependent enforcement regime. The second essay examines the use of citizen complaints as a potential environmental monitor that can be used by regulatory agencies. Empirical evidence suggests that although citizen complaints may be useful in deploying resources to clean up potentially harmful spills, the general lack of significance between specific types of complaints and industry spills types seems to imply citizens know something has gone wrong, but that they cannot determine exactly who is responsible for the spill. The third essay examines the role that litigation plays in regulation. We find that when the litigation process is not certain and is costly, the enforcement agency may find it optimal, under plausible conditions, to lower the fine for non-compliance to in order to raise the industry compliance rate.

The thesis examines the problem of determining the optimal selectivity and age-specific catch for multiple cohort fisheries with an empirical application to the Northern cod fishery (NAFO division 2J3KL). Two of the main assumptions of the previous literature are examined: exogenously fixed age selectivity and the use of a Schaefer production function (which is Cobb-Douglas with constant marginal productivities of effort and stock numbers). The latter assumption's validity is empirically tested for the harvest of Northern cod and age selectivity is endogenously determined in a general analytical model and in numerical simulations for the Northern cod fishery. Analytical results about age selectivity have been derived in a multiple cohort model by Clark (1973, 1990) under the restrictive assumptions of zero harvesting costs and by Stollery (1984) for a steady state age distribution with nonzero harvesting costs. The thesis derives optimal age selectivity and the optimal harvest path for a two-cohort model and therefore provides the first analytical extension of Clark's single cohort model without assuming a steady state solution. The optimal harvest starting age for each cohort is determined and it is shown that knife-edge selectivity always dominates perfect selectivity when the harvest of the younger cohorts commences. Clark's single cohort model is re-evaluated by relaxing the assumption of a Schaefer production function and significant changes to the optimal harvest extraction path are discovered. For the empirical application to the Northern cod fishery, the production function is estimated by using scientific population data and alternatively by developing an estimation technique that avoids the use of any population estimates. Production function estimation results do not support the assumption of a Schaefer production function and stock predictions that are based on production function estimation results differ significantly from the Department of Fisheries and Oceans (DFO)' figures. Production function estimations as well as other estimation results serve as inputs into a simulation of the optimal harvest rule for the Northern cod fishery. Policy implications from the dissertation suggest a methodological change in stock assessment and an alteration of the age selection of fishing gear that consists of allowing more young fish to escape.

The Hotelling rule, which states that the shadow price of an exhaustible resource must grow at a rate equal to the interest rate, has been recognized as the fundamental principle of the economics of exhaustible resources. Recently, many economists began to focus attention on testing whether the data support the Hotelling rule. These tests, however, have provided only limited empirical support for the Hotelling rule. This thesis derives a modified version of the Hotelling rule for the case of the "exhaustible" resource of old-growth forest and tests its empirical validity using data from old-growth timber sales in the U.S. Pacific Northwest. Since the choice of old-growth timber obviates many of the problems faced by previous researchers working with minerals in measuring shadow price, the Hotelling rule of old-growth is tested directly using the data. Three models of old-growth harvest are set up corresponding to the three types of harvesting cost functions. For the basic case in which the unit harvesting costs are assumed to be constant, the modified Hotelling rule for old-growth forestry reveals that the shadow price of standing old-growth timber must increase at a rate equal to the interest rate plus a term reflecting the opportunity cost of capital adjusted for the carrying cost of holding a unit of old-growth stands. When the harvesting cost function is either generalized to be nonlinear or to allow for stock effects the derivation of the modified Hotelling rule is made more complex. Despite the complexities, a relatively simple Hotelling rule for each case has also been obtained. Each one predicts the determinants of the time path of the stumpage price for the optimal harvest of old-growth in its corresponding case. Empirical models are constructed to test the modified Hotelling rule directly. In addition, two variations of alternative empirical models are developed to incorporate a stock-based model and an inflation-based model with the purpose of providing further interpretation and corroboration for the findings from the direct models. The empirical result demonstrates that the set of the Hotelling hypotheses is not rejected as an empirically valid description of the harvest path for the old-growth forests. The predictive success of the Hotelling rule of old-growth significantly improves the empirical performance of Hotelling's theory, which is a major contribution to the literature in natural resource economics. It is also shown to be useful to identify and direct appropriate resource conservation and management practices for the old-growth forestry.