Stock market responses to chemical accidents in Japan: An event study

Despite recent major chemical process accidents in Japan, the top management teams of firms still avoid taking costly risk reduction measures because of their low perceived impact on firm performance. The disclosure of information on accident risks might motivate managers to enhance workplace safety because of the subsequent evaluation of firms by investors in stock markets. If the disclosed risk information is newly available for investors, firms with a high risk of accidents would receive a poor evaluation by stock markets and thus managers would take risk reduction measures to prevent stock prices from declining. In this study, we conduct an event study analysis to examine whether accident risk information is already reflected in stock prices, using data on the Japanese chemical industry. The results of our event study show that the estimated cumulative average abnormal returns of firms' stocks are significantly negative after severe accidents actually occurred. This finding implies that risk information is not already reflected in the stock prices of Japanese chemical firms and that the disclosure of accident risk information has the potential to motivate the top management teams of firms to reduce their firms’ accident risk.     

Vehicle-occupant deaths caused by tornadoes in the United States, 1900–1998

Abstract

Occupants of vehicles encounter an uncertain degree of risk during tornadic storms. The current National Weather Service guidelines suggest we abandon vehicles to lie in a ditch if no sturdy shelter is available. However, these guidelines were developed without the benefit of supporting research. As we are an increasingly vehicle-dependent society, it is important to explore the historical record of vehicle-occupant deaths to see if the National Weather Service guidance is appropriate. The objective of this paper is to investigate the number, distribution, and broad spatial and temporal trends associated with vehicle-occupant deaths that have occurred as a result of tornadoes, and to define the proportion of tornado-induced deaths that have occurred in vehicles.

There were 15,047 deaths caused by tornadoes from 1900 through 1998. Of those, 5685 occurred at a defined site, and 270 were known to have occurred in vehicles. The number of vehicle-occupant deaths, as a proportion of all site-known, tornado-induced deaths, was constant from 1959–1979, but decreased significantly between 1980–1998. The number of deaths was related to several influencing factors. Linear regressions and correlations were employed to determine the degree of relationship between the number of deaths and several explanatory factors. While population, number of vehicle registrations, and number of tornadoes all seemed to influence the number of deaths to some degree, the most significant factor to influence the number of deaths appeared to be vehicle safety features. Most vehicle-occupant deaths occurred during rare F4 tornadoes, when vehicles where thrown from roadways. Regionally, the Great Plains had the highest number of vehicle-occupant deaths from tornadoes.     

Plant/soil concentration ratios for paired field and garden crops, with emphasis on iodine and the role of soil adhesion

In the effort to predict the risks associated with contaminated soils, considerable reliance is placed on plant/soil concentration ratio (CR) values measured at sites other than the contaminated site. This inevitably results in the need to extrapolate among the many soil and plant types. There are few studies that compare CR among plant types that encompass both field and garden crops. Here, CRs for 40 elements were measured for 25 crops from farm and garden sites chosen so the grain crops were in close proximity to the gardens. Special emphasis was placed on iodine (I) because data for this element are sparse. For many elements, there were consistent trends among CRs for the various crop types, with leafy crops > root crops ≥ fruit crops ≈ seed crops. Exceptions included CR values for As, K, Se and Zn which were highest in the seed crops. The correlation of CRs from one plant type to another was evident only when there was a wide range in soil concentrations. In comparing CRs between crop types, it became apparent that the relationships differed for the rare earth elements (REE), which also had very low CR values. The CRs for root and leafy crops of REE converged to a minimum value. This was attributed to soil adhesion, despite the samples being washed, and the average soil adhesion for root crops was 500 mg soil kg⁻¹ dry plant and for leafy crops was 5 g kg⁻¹. Across elements, the log CR was negatively correlated with log Kd (the soil solid/liquid partition coefficient), as expected. Although, this correlation is expected, measures of correlation coefficients suitable for stochastic risk assessment are not frequently reported. The results suggest that r ≈ −0.7 would be appropriate for risk assessment.     

Group utility indifference and resource management decision making

Natural resource managers often rely on the advice of specialists to aid decision making. However, disagreement among these specialists about the relative value of particular management objectives or the risks associated with implementing certain management strategies may complicate the decision effort. Multiattribute utility analysis can facilitate decision making by indicating how attributes of a problem are weighed by individual specialists. This information can then be used to outline bands of potential problem solutions that are acceptable to the advising group and may allow management to further its own objectives (possibly increased efficiency).An example is presented that relates to fire management planning efforts on national forests. Multiattribute utility functions developed from a survey of fire management professionals are used to identify utility-maximizing fire management strategies based on each strategy's level of economic efficiency and risk. Bands of utility-indifferent potential solutions are outlined based on measures of group consensus. It is pointed out that a subset of these would further management's goal (increased efficiency) without significantly altering the value assigned to the risk attribute by the specialists. Finally, the robustness of the technique is discussed with particular reference to environmental management problems and the role that proxy information often plays in decision making.     

Earthquake risk reduction: An examination of local regulatory efforts

We develop an understanding of local governmental efforts to address earthquake risks. Our data for a sample of California and Washington cities' risk-reduction efforts sort into two different clusters of leading jurisdictions as contrasted with a cluster of lagging jurisdictions. We explain differences among the three categories of cities in terms of different political-economic factors. Local governmental willingness and ability to undertake risk-reduction programs have more to do with local political demands and community resources than with objective risk or previous earthquake experience. State mandates have a selective impact on local risk-reduction efforts. The principal policy lessons concern the need to address gaps in local risk-reduction efforts through more careful targeting of federal and state earthquake programs and by including stronger teeth within state mandates.     

Treatment of risk in environmental impact assessment

Risk assessment and environmental impact assessment have developed as separate traditions. While environmental impact assessment is a broad field that includes all activities that attempt to analyze and evaluate the effects of human and related actions on the environment, risk assessment has been concerned with the relatively well-defined regulatory problems and employs formal quantitative analysis of the probability of specific undesired events, such as cancer. Risk analytic approaches, particularly the explicit treatment of uncertainty, can significantly contribute to environmental assessments. This article discusses the type and sources of uncertainty in environmental assessments, techniques for their quantification, and ways to use uncertainty estimates to calculate probabilities of effects or probabilities of exceeding environmental standards and to determine the need for mitigation or additional research.This article was presented at the International Institute of Applied Systems Analysis, Task Force Meeting on Risk and Policy Analysis under Conditions of Uncertainty, Laxenburg, Austria, November 1985.     

Reducing groundwater pollution by toxic substances: Procedures and policies

One major source of water-related health problems is the improper disposal of toxic substances in the environment. Toxic materials leaching from unregulated and unlined pits, ponds, lagoons, and landfills have created a widespread environmental nightmare in the United States and many other parts of the world. At present, there are two major and interrelated components of this problem in the United States. The first is the issue of cleaning up abandoned disposal sites that pose actual or potential threats to water supplies. The second aspect of the problem concerns the necessity of siting proper management, treatment, or disposal facilities in the future. Priorities must be set to allow efficient, effective, and equitable allocation of the scarce resources that are available for accomplishing these tasks.This article examines a number of the issues involved in setting these priorities, and presents the results obtained from a study of risk estimation and evaluation in the context of groundwater contamination by toxic substances. The article introduces a new concept of risk estimation, which is shown to produce more accurate and credible risk analyses. Finally, the relationships between risk credibility and public perceptions of procedural fairness and equity are examined as these factors bear on the institutional aspects of implementing policies for site cleanup and/or facility siting.     

MULTICRITERION ANALYSIS OF FOREST WATERSHED MANAGEMENT ALTERNATIVES1

ABSTRACT: The impacts of alternative forest watershed management practices are examined from a multicriterion viewpoint in order to select the most satisfactory management scheme. The selection process is carried out using two types of multicriterion decision making techniques: the outranking types of ELECTRE I and II, and the distance-based type of compromise programming (CP). The process is illustrated using the U.S. Forest Service Beaver Creek Experimental Watershed in the Salt-Verde River Basin of Arizona as an example. The desired objectives of the experimental study and the alternative forest watershed resources management schemes are transformed into an evaluation matrix of alternatives versus criteria array. Analyses of the matrix using the aforementioned techniques result in a complete preference ordering of the feasible alternatives in the cases of ELECTRE LI and CP and a partial ordering when ELECTRE I is used. In addition, some sensitivity analyses have been performed and showed ELECTRE II and CP to be fairly robust with respect to parameter changes, while ELECTRE I being highly sensitive to changes in threshold levels. Overall the three techniques pointed out that 65 percent vegetation cut is the best management scheme, while the next best is shown to be 50 percent vegetation cut.     

O&M SPENDING LEVELS IN THIRD WORLD IRRIGATION SYSTEMS: EXPLORING ECONOMIC ALTERNATIVES1

ABSTRACT: Third World irrigation and drainage systems have experienced major declines recently in both capability and performance. This is due initially to working them well beyond their design capaci-capacities; scouring, sedimentation, and overtopping result. Chronic O&M underfunding then adds heavily to this worsening malaise. International donors have assisted irrigation departments with rehabilitation projects and programs to improve O&M effectiveness on a grand (billions of dollars) scale. Despite their historical propensity to examine, almost fastidiously, program economic justifications (B/C, IRR, etc), the donors apparently have glossed over two basic analytic elements for (a) more spending on O&M as distinct from an equivalent spending on other means to provide farmers with an m³ of water; and (b) different levels of O&M spending on canals and drains. Surely those different levels provide differing benefits, in total, and at the margin. Which level is most economic? This paper identifies these latter analytical issues, posits methodologies key to an O&M spending level analysis, and discusses the information base. Particular attention is paid to identifying relevant costs and benefits, and to suggesting supplementary criteria for O&M spending level selection. The paper is an exercise in delineation of an imminently practical area of irrigation engineering economics.