Printer Friendly

The market response to environmental incidents in Canada: a theoretical and empirical analysis.

I. Introduction

There is a growing concern that regulations that promote safety (e.g., automobile safety and product safety) may have little impact on the level of risk associated with the utilization of such products |21; 29~. A similar concern has been recently raised with respect to regulations that promote safety in the workplace |12~. A reason often advocated to explain this phenomenon is the lack of adequate enforcement mechanisms. In particular, it is often argued that fines imposed on agents not complying with these regulations are not severe enough to have a deterrence effect |30~. With respect to the enforcement of the Ontario Environmental Protection Act (R.S.O. 1980, c. 141), Saxe writes that "the majority of fines were too low to act as effective deterrents" |23, 104~. However, some authors have challenged this view in showing that the market provides additional monetary incentives for firms to comply with the regulations by punishing non-complying firms through lower stock market prices. For example, some analyses have shown that public announcements of lawsuits against American firms not complying with workplace safety |8~, product safety |31~ and environmental regulations |19~ have caused significant drops of the equity value of these firms. In this last study, it was found that the announcement of lawsuits against firms violating the American Resource Conservation and Recovery Act (RCRA 1976) had a significant negative impact on their equity value on the day of the announcement, while announcements of suit settlements (e.g., fines) had no effect. In most studies, authors argue that the reductions in stock prices have some deterrence effect on firms.

Following Muoghalu, Robison and Glascock |19~ (henceforth MRG), this paper examines the impact of the announcement of environmental incidents on the firm's equity value using a sample of 47 events involving Canadian firms between 1982 and 1991. Like its American counterpart, Canadian environmental regulations (both federal and provincial) are of a "command-and-control" nature.(1) However, it is claimed that enforcement of the regulation has been more severe in the United States. In a comparative qualitative analysis of the regulatory approach in Canada and the United States, Marchant notes that "the United States has one of the most adversarial industry-government relationships in the world, and the hardball attitude of some U.S. industry on some issues may have been the cause of stiffer and less compromising enforcement on the part of the U.S. Environmental Protection Agency" |18, 46~.(2) Our analysis extends MRG's study in two different directions. First, unlike previous papers concerned with similar questions, our analysis is based on a theoretical model that describes how shareholders update their beliefs as to the profitability of the firm when certain environmental incidents are announced. Second, in addition to the announcement of lawsuits and suit settlements,(3) we examine shareholders' reaction to two other categories of events: the announcement of environmental incidents likely to lead to a lawsuit (e.g., levels of pollution above provincial/federal limits, spills) and the announcement of investments in emissions control equipment.

The paper is organized as follows. In section II we present the main features of our theoretical model. In section III our data set is presented and discussed. The event-study methodology is briefly described in section IV and our results presented in section V. We find that the announcement of investments in emissions control equipment are followed by a decrease in the equity value of firms. We also find, contrary to MRG, that Canadian shareholders react negatively to the announcement of suit settlements, not to the announcement of lawsuits. These results yield support to the view that the enforcement of environmental regulations has generally been more severe in the United States than in Canada. Section VI provides concluding remarks.

II. A Theoretical Model of Optimal Timing of Compliance

How is the regulator to decide whether or not there is violation of the emissions standards in a situation where not only the firm's rate of discharge is stochastic, but where also the regulator obtains an imperfect measure of that rate of discharge? Presented this way, the problem is one of statistical quality control to which a number of authors have provided answers |2; 15; 28~. However, these answers are not entirely satisfactory. In all these models, the optimal sample size is chosen before the monitoring process actually takes place. The firm's optimal strategy (choice of pollution level) is also determined at the beginning of the regulatory process. This lack of dynamics prevents the firm's optimal strategy to evolve from a state of no compliance to a state of compliance: in these circumstances, the firm's optimal solution is always either to comply or not comply. There is no room for the firm to change (update) its compliance strategy as the regulator samples its pollution level. Such models cannot explain why a firm would suddenly announce that it will invest in emissions control equipment. How can it be that it was not optimal for the firm to comply with the environmental regulation yesterday, but that it is today?

Moreover, the literature has so far ignored an important characteristic of the enforcement of environmental regulations: even if a violation is correctly detected (i.e., the regulator judges there is a violation while indeed there is one), it does not necessarily follow, as implicitly assumed in previous papers, that the regulator will try to obtain compliance through legal prosecution. There is ample evidence that environmental regulators prefer to obtain "voluntary" compliance from polluters |4; 20~. Legal action is a last resort. Hence, we prefer to model the regulator's monitoring and enforcement problem in such a way as to offer dynamics not only to its decision process, but also to the firm's optimal compliance strategy. Not only shall we observe the regulator updating its belief with respect to the firm's compliance with the emissions standard, but we shall also be able to determine the exact moment for which it becomes optimal for the firm to comply with the emissions standard. Immediate compliance is in general not optimal since the probability of enforcement through penalties is too small. While previous authors have argued that the small number of court cases related to the violation of environmental standards is the prima facie evidence of poor monitoring and enforcement, we show that it may simply be the result of good anticipation by the firm as to the probability of being sued in any given period. This framework of analysis also allows us to provide an explanation for a firm's incentive to announce investments in emissions control equipment. We first present a theoretical model of the regulator's behavior. Then, given this behavior, the firm's compliance strategy is derived.

A Model of Regulator's Behavior

Our main objective is to describe a process by which the regulator updates its belief as to the state of compliance of the firm with the emissions standard. As one would like to observe, we show that the higher the measured level of pollution the higher the regulator's belief that the firm does not comply with the emissions standard.

A firm produces a good |x.sub.t~ and emits a pollutant in quantity |P.sub.t~ in period t. This quantity is stochastic and is a function of whether or not the firm complies with the emissions standard. Let

|P.sub.t~ = |P.sup.c~ + ||Theta~.sub.t~ (1)

be the level of pollution if the firm complies with the regulation, and

|P.sub.t~ = |P.sup.nc~ + ||Theta~.sub.t~ (2)

if it does not comply. ||Theta~.sub.t~ is a pollution shock in period t. We assume that |Mathematical Expression Omitted~. Hence, the firm's mean (expected) level of pollution is |P.sup.c~ if the firm complies with the regulation, and |P.sup.nc~ if it does not comply. If the emissions standard is strictly smaller than the firm's unregulated level of pollution, then |P.sup.c~ |is less than~ |P.sup.nc~.

In each period, the regulator obtains a measure |Mathematical Expression Omitted~ of the firm's stochastic pollution level. In period t, we thus have

|Mathematical Expression Omitted~

where ||Psi~.sub.t~ captures the possibility of a measurement error. We assume that |Mathematical Expression Omitted~. Hence, the measure obtained in period t is unbiased. It follows that |Mathematical Expression Omitted~ if the firm complies with the emissions standard and |Mathematical Expression Omitted~ if it does not comply.

Before measuring the firm's level of emissions, the regulator cannot observe whether the firm complies with the emissions standard. Nonetheless, it holds a prior belief ||Rho~.sub.0~ that the firm does not comply. The value of this prior belief is determined by a number of factors among which is the presence or absence of an emissions control equipment. Indeed, given that emissions standards are technology based,(4) it seems reasonable to postulate that if the regulator observes an emissions control equipment in place (or knows that such an equipment has been installed), then it holds a prior belief of compliance larger than if such an equipment is not installed.(5)

Upon observing |Mathematical Expression Omitted~, the regulator updates his belief on the state of compliance of the firm. Let |Mathematical Expression Omitted~ be the density function of |Mathematical Expression Omitted~ if the firm does not comply with the environmental regulation and |Mathematical Expression Omitted~ if it complies. Hence, the probability of observing any given level of pollution |Mathematical Expression Omitted~ if the firm complies is defined as

|Mathematical Expression Omitted~

and if the firm does not comply, as(6)

|Mathematical Expression Omitted~.

Denote by ||Rho~.sub.t~ the regulator's updated belief, in period t, that the firm does not comply with the emissions standard. Following Bayes's rule,

|Mathematical Expression Omitted~.

Let E(|P.sub.t~) be the regulator's expectation of the firm's true level of pollution in period t,

E(|P.sub.t~) = (1 - ||Rho~.sub.t~)|P.sup.c~ + ||Rho~.sub.t~|P.sup.nc~. (7)

It is then easy to show that the higher the measured level of pollution, the higher is the regulator's expectation of the firm's true level of pollution in period t, i.e. |Mathematical Expression Omitted~. Indeed, note from (6) that

|Mathematical Expression Omitted~.

The monotone likelihood ratio implies that

|Mathematical Expression Omitted~.

Hence, given that the firm violates the emissions standard, with sufficient time the regulator will come to believe correctly that the firm does not comply with the environmental regulation. We think this process captures the dynamics of the regulator's problem at monitoring a firm's level of emissions and its compliance with specific emissions standard. Given these characteristics of the regulator's behavior, we can now analyse the firm's optimal compliance strategy.

Optimal Compliance Strategy

At any time t, the problem of the firm is to calculate its expected present value net of the pollution abatement costs and stochastic penalties for violation of the environmental regulation. The appropriate course of action (install the emissions control equipment or do not install) is obviously indicated by the one that yields the highest expected present value.

The penalty for violation of the emissions standard is stochastic for two different sets of reasons. First, for every period in which the firm does not comply with the emissions standard, it faces a strictly positive probability of being sued in that period. This probability is an increasing function of the regulator's expectation of the firm's level of emissions, which is itself a function of the regulator's belief that the firm does not comply with the emissions standard (||Rho~.sub.t~). Second, conviction is not the necessary outcome of a legal prosecution. Moreover, if convicted, the court has a considerable amount of flexibility as to the type and level of penalty to impose on the firm. As most often described in environmental regulations, upon conviction the court may impose a fine of which only the maximum is determined by law. The court may also require the firm to remedy the problem by installing the appropriate emissions control equipment. In the context of this paper, this means there is a positive probability that the firm will have to incur the total costs of complying with the standard in every period after conviction. Denote by |Omega~ the probability of conviction and by |Gamma~ the probability that the court sentences the firm to remedy the problem. Hence, upon prosecution, the expected penalty in period t, noted |Mathematical Expression Omitted~, is

|Mathematical Expression Omitted~

where T|C.sub.t~ (|P.sup.c~) is the total cost of complying with the emissions standard |P.sup.c~, F is the expected value of the fine and L represents the legal fees incurred by the firm upon prosecution.

Given the present state of environmental awareness, it is increasingly recognised that the most important penalty a firm may incur for being sued for violation of emissions standards lies on losing its ability to maintain any given level of profits. This loss may be a more important threat than simply the legal fees and expected penalty imposed by the court. Let |R.sub.t~ be the firm's profits in period t, not accounting for pollution abatement costs and stochastic penalty. We assume in this paper that |R.sub.t~ is adversely affected if the firm is sued for alleged violation of the environmental standards. Denote by |Mathematical Expression Omitted~ the profits of the firm upon legal prosecution. Obviously, |Mathematical Expression Omitted~. These are otherwise assumed constant. Hence the expected profits of the firm if it does not comply with the emissions standard in period t are

|Mathematical Expression Omitted~

where |Mathematical Expression Omitted~ is the expected probability of being sued in period t.(7) If the firm complies with the emissions standard in period t, its profits are

|Mathematical Expression Omitted~.

Given that |R.sub.t~ and T|C.sub.t~(|P.sup.c~) are time independent and since the firm does not face the possibility of penalties (since it fully complies with the regulation), |Mathematical Expression Omitted~ is constant over time. Note that for any |Gamma~ and |Omega~ strictly positive, a necessary condition to observe a firm complying with the environmental regulation is

|Mathematical Expression Omitted~.

Otherwise, it would not be optimal for the firm to comply with the regulation even if the probability of being sued were to be one. We assume in this paper that (13) holds. Notice that for any given level of probability of being sued, the loss in reputation reduces the fine that is necessary to bring the firm into compliance.

As observed in equation (11), the expected profits of the firm (hence its optimal compliance strategy) are a function of its expectations about the probability of being sued in period t which is itself a function of the firm's expectation of the regulator's belief of non-compliance. Denote the latter by |Mathematical Expression Omitted~, in period t. Given that the firm knows the process of belief updating followed by the regulator, it can form an expectation, in any period t, as to the regulator's belief of non-compliance.

THEOREM 1. Given that |Mathematical Expression Omitted~, then |Mathematical Expression Omitted~, for as long as |Mathematical Expression Omitted~. Moreover, as t |right arrow~ |infinity~, |Mathematical Expression Omitted~.

Proof. From the firm's point of view,

|Mathematical Expression Omitted~.

This latter equation can more generally be written as

|Mathematical Expression Omitted~

where |Mathematical Expression Omitted~ and |Xi~ |is equivalent to~ |f.sup.c~ (|P.sup.nc~)/|f.sup.nc~ (|P.sup.nc~). In particular, |Mathematical Expression Omitted~ and |Mathematical Expression Omitted~. Since |Mathematical Expression Omitted~ it follows that ||Gamma~.sub.1~ |is less than~ ||Gamma~.sub.0~. Hence, |Mathematical Expression Omitted~. Obviously, |Mathematical Expression Omitted~ and, more generally, |Mathematical Expression Omitted~. Moreover, note that the limit of ||Gamma~.sub.t-1~ as t |right arrow~ |infinity~ is zero. Hence the limit of |Mathematical Expression Omitted~ as t |right arrow~ |infinity~ is one. Q.E.D.

COROLLARY. The expected profits of a firm that does not comply with the environmental regulation fall over time.

Let t* = t + n be the time period for which |Mathematical Expression Omitted~. Let |Mathematical Expression Omitted~ be the expected value of the firm in period t given that it complies with the standard in period t. Since |Mathematical Expression Omitted~ is assumed constant, |Mathematical Expression Omitted~ is constant in all period t.

|Mathematical Expression Omitted~

where |Beta~ is the discount rate. Denote by |Mathematical Expression Omitted~ the expected value of the firm in any period |Mathematical Expression Omitted~ given that it complies with the standard in period |Mathematical Expression Omitted~. It is then easy to show that the value of the firm is maximised if |Mathematical Expression Omitted~ i.e., |Mathematical Expression Omitted~. The relationship between those values is illustrated in Figure 1 where |Mathematical Expression Omitted~ denotes the expected value of the firm in period t given that it were not to comply with the emissions standard from period t onward.

Hence, it is optimal for the firm to comply with the standard or announce compliance (e.g., public announcement of investments in emissions control equipment) exactly at period t*. Given that agents are rational and profit maximisers, a clear prediction of this model is that a public announcement of an investment in emissions control equipment should have no effect on the value of the firm unless the firm is forced by the regulator to undertake such an investment at an earlier date than t*. Moreover, for as long as expectations are fulfilled, legal prosecution is never observed. Legal prosecution is observed if the firm, for example, has under-estimated the expected probability of being sued, hence has over-estimated t*.

Upon prosecution, it is easy to calculate the change in the value of the firm. Without any further details, the measured loss in the value of the firm is simply equal to the difference, discounted in period |Mathematical Expression Omitted~, between the expected and realised profits from period |Mathematical Expression Omitted~ to period t* (where |Mathematical Expression Omitted~ is the period where prosecution is announced). Note that even if the firm is not found guilty, we should observe a reduction in its present value. Indeed, not only its reputation is adversely affected, but also the firm may end-up investing in an emissions control equipment at an earlier period than the one that maximizes its present value.

In the above model, we have assumed that the regulator's belief of non-compliance is updated through time using only the measured level of pollution. It would be straightforward to extend the model so as to have this belief also function of environmental incidents in which the firm would be involved. Upon such an incident occurring, one could observe a discrete jump in the regulator's belief of non-compliance and therefore a fall in the value of the firm at the time the incident occurs. This would lead the firm to announce compliance at an earlier date than t*. Finally, it would also be easy to introduce a lag between the date where legal prosecution is announced and the date where suit settlement is announced. Would that be the case, given rational agents, it shall be obvious that there would be no change in the value of the firm upon the announcement of suit settlement unless the market has made an error (upward or downward) in its estimation of the penalty. The next sections are devoted to testing these theoretical predictions. III. Data and Limitations

Our sample consists of 47 events published in Canadian print media (mostly the Financial Post and the Globe and Mail) between 1982 and 1991.(8) These events are divided in four types of announcement: 12 announcements of violation of environmental regulation for which it is likely that the regulator will undertake legal action; 9 announcements of legal action undertaken against firms that violated environmental regulations; 13 announcements of suit settlement; and 13 announcements of investment in emissions control equipment. For four of the 13 suit settlements, the initial announcement of legal action is also available. Note that of the 47 cases, 18 involve firms in the pulp and paper industry, 10 in the mining industry, 6 in the petroleum industry, 6 in the chemical industry and 7 in other industries.(9) These events concern firms operating in Canada and registered at the stock exchange. However, certain subsidiaries of foreign firms are not registered at any Canadian stock exchange. In addition to Canadian-owned firms, we kept in the sample those firms that are totally owned by one American corporation registered at the New York Stock Exchange (14 of the 47 cases). Cases of firms owned by Europeans or jointly owned by more than one corporation were discarded since the impact of the event on the equity of the head firm would likely be too negligible to be detectable.

For each of the four types of announcement, four subsamples are considered. The first subsample contains every case available within a specific type of announcement. The second contains only those cases involving Canadian-owned firms. The third subsample contains all cases (Canadian and American) with the same level of media exposure, i.e., cases that are presented in a feature article(10) of the Financial Post and/or the Globe and Mail. Suret and Pauchant argue convincingly that event studies should be based on events that have the same extent of coverage in the media |26~. Finally, the fourth subsample is composed solely of the Canadian cases with the same level of media exposure. IV. Event-Study Methodology

We use the Capital Assets Pricing Model (CAPM) version of the standard event-study methodology to analyze reactions of a firms' equity value to the announcement of the different events.(11) The event-study methodology is based on the assumption that the market is sufficiently efficient to fully evaluate the impact of different events on future profits of the firms |6~.

The reaction to the announcement of an event is obtained by predicting a normal return for each firm on each day following the announcement and then subtracting this predicted normal return from the actual return. Normal returns are generated by estimating the following CAPM model:

|R.sub.it~ = (1 - ||Beta~.sub.i~)|R.sub.ft~ + ||Beta~.sub.i~|R.sub.mt~ + |e.sub.it~ (15)

where:

|R.sub.it~: the rate of return on security i for day t;

|R.sub.ft~: the rate of return on the risk-free asset |Treasury Bills (90 days) of the Canadian federal government~;

|R.sub.mt~: the rate of return on the Toronto Stock Exchange market (TSE);

||Beta~.sub.i~: estimated parameter;

|e.sub.it~: error term for security i on day t.

In absence of unexpected information, the relationship between the firm's return, the market's return and the risk-free asset should be unchanged. Hence, these returns can be used to forecast the "normal" return for the firm. A prediction error is generated when unexpected information affects the return for the firm without affecting the market's return and the risk-free asset. The prediction error, commonly referred to as the abnormal return (AR) for security i is computed as the following:

|Mathematical Expression Omitted~.

The day the event is announced in the print media is referred to as day 0, and all other days are measured relative to day 0. The CAPM model is estimated for each firm over the 210 day interval before the announcement of the event.(12) The average abnormal return is then computed across firms:

AA|R.sub.t~ = (1/|N.sub.t~) |summation of~ A|R.sub.it~ where i = 1 to |N.sub.t~ (17)

where |N.sub.t~ is the number of securities in a given subsample. A t-test is used to determine the level of significance of abnormal returns for a given subsample. The test uses the estimated standard error of the returns computed for the estimation period:

|Mathematical Expression Omitted~

where |Mathematical Expression Omitted~ is the estimated standard error of abnormal returns during the estimation period (T = 210).(13) This test statistic follows a Student at T - 1 degrees of freedom.

In order to test for the persistence of the impact of the announcement during the period t to t + n, the abnormal returns must be cumulated. The cumulated abnormal return in a period from t to t + n is given by:

|Mathematical Expression Omitted~.

The t test is then defined by(14):

|Mathematical Expression Omitted~

V. Results

Tables I to IV present the average abnormal returns and the cumulated average abnormal returns for the four types of events under study: incidents (potential violations), lawsuits, suit settlements and investments respectively. In each table, results for the four different subsamples are presented: all cases, cases involving Canadian-owned firms, cases with the same media exposure and Canadian cases with the same media exposure.

Table I and II indicate that announcements of incidents and lawsuits are not followed by any significant abnormal returns in any of the different subsamples of firms. That shareholders do not react to the announcement of environmental incidents may indicate that they do not expect the regulator to launch any procedure (including legal action) so as to bring the firm into compliance with the environmental regulation. Furthermore, that shareholders do not significantly react to the announcement of lawsuits may indicate little or no worry as to the outcome of the legal procedure. Indeed, during the last decade, lawsuits in Canada were generally long, and fines, if any, were relatively low |3; 32~. For example, Hetu has calculated that for the period 1984-88, the average penalty under the Quebec Environmental Quality Act has been $667,16 |10~.

Table III indicates that suit settlements with fines imposed on firms result in stockholders experiencing abnormal losses on day 0. Not unexpectedly, this is true only for the two last subsamples of cases (cases with the same media exposure and Canadian cases with the same media exposure) where abnormal losses of respectively 1.65% and 2% of market value are observed. This result is maintained when we consider only the four firms for which we have both the announcement of the lawsuit and the suit settlement: they suffer significant abnormal losses of 2.7% on the day of the announcement of the suit settlement and no loss when the lawsuit is announced.(15) These results suggest that the size of fines, or the fact that there is a fine in itself, is an unexpected surprise for shareholders. This is plausible in a legal context in which, as described above, fines TABULAR DATA OMITTED are the exception rather than the rule. Interestingly, these results contrast with those of MRG who find abnormal losses on day 0 for lawsuits, but not for suit settlements. This suggests that American environmental authorities have been more successful than their Canadian counterpart in designing enforcement mechanisms in which a lawsuit can impose a credible threat on investors |17; 18~.

Whether or not a loss of equity value on the day of announcement of the lawsuit is large enough to have some deterrence effects on firms is debatable |27~. A decline in the equity value of a firm for a few days or a few weeks does not necessarily have a strong wealth effect on shareholders except those who need cashflows in that particular period and have to sell their shares. In fact, there is a transfer of wealth between impatient shareholders and those who are more opportunist, and it is unlikely that this transfer has a strong deterrence effect on firms. Therefore, given our results showing abnormal returns only on day 0, we cannot conclude that the market has the TABULAR DATA OMITTED power to discipline firms not complying with environmental regulation. However, it shall be noted that firms are affected differently by the announcement of environmental incidents. For instance, one firm experienced a loss of 7% of its equity value for a period of 10 days. The deterrence effects of such announcement, if any, would certainly be an increasing function of the extent of the loss and the number of days for which this loss persists. Finally, Table IV indicates that investments in emissions control equipment result in stockholders experiencing abnormal losses on day 0. This is true for two subsamples of cases (cases involving Canadian-owned firms and cases with the same media exposure) where abnormal losses of respectively 1.1% and 1.6% of market value on day 0 and -1 are observed.(16) Recent empirical analysis have found that environmental regulations have a negative impact on industry TABULAR DATA OMITTED productivity. Smith and Sims, for example, have shown that environmental regulations reduced the growth rate of productivity in the Canadian brewing industry |25~. More recently, Barbera and McConnell have analysed the impact of required abatement capital on total factor productivity growth in five U.S. manufacturing industries |1~.(17) They found that the average annual reduction in total factor productivity, for the period 1961-1980, varies between 0.08 and 0.24 percentage points. Jorgenson and Wilcoxen have also found significant impact of environmental regulations on the growth rate of the chemicals, coal mining, motor vehicles, and primary processing industries |11~. At first glance, the reaction of investors as measured in this analysis is consistent with those empirical findings. However, it should be noted that investors react to the announcement TABULAR DATA OMITTED of the investments and not to the investment per se. If investors had no expectation of a firm needing to purchase the equipment, the observed reaction would be an unbiased estimate of the opportunity cost of the expenditure. But this is unlikely to be the case. Would investors have complete information with respect to the required investment in emissions control equipment, as assumed in the previous theoretical model, the value of the firm on the day of announcement shall remain unchanged. We can identify at least two circumstances for which losses would follow the announcement of investments in emissions control equipment. First, as suggested above, investors may have incomplete information as to the need of such investment. Second, the investment may have been imposed on the firm by the regulator at an earlier date than t*. This second factor may explain our results. Indeed, except for one case, it appears that all cases of announcement of investments were forced by the regulator upon the firm.

VI. Conclusion

This paper has examined the impact of the announcement of diverse environmental incidents and investments in emissions control equipment on firms' equity value. First, a theoretical model formalized how shareholders change their expectation about the profitability of firms when different categories of events related to environmental regulation are announced. The model was then tested, using the standard event-study methodology, with a sample of 47 events involving Canadian firms between 1982 and 1991. These events were divided in four categories: announcements of potential violations of environmental regulations, lawsuits, suit settlements and investments in antipollution equipment. Our results showed that the stock value of Canadian-owned firms declined on the day of the announcement of suit settlements resulting in fines (about -2%) and investments (about -1.2%). These results contrasted with those of MRG who showed that American stockholders react, on the day of the announcement, to lawsuits and not to suit settlements. This difference may have been expected given the more conciliatory approach adopted by Canadian authorities responsible of the enforcement of environmental regulation. It supports the view that the enforcement of environmental regulations in the United States is more severe (credible) than in Canada. 1. For a detailed description, see Dewees |4~, Laplante |13~, and Portney |22~. 2. In a personal communication with one of the authors, Professor Don Dewees expressed the same opinion, i.e., that the monitoring and enforcement of environmental regulations is more rigorous in the United States than in Canada. However, it shall be noted that Canadian data on monitoring and enforcement activities are rather dispersed and not readily available. For example, despite his extensive study of the environmental regulation in the Canadian pulp and paper industry, Sinclair writes "the data available on prosecutions are limited" |24, 102~.

3. These are the only events considered in MRG.

4. Emissions standards are generally based on the anticipated performance of the best practicable or best available emissions control technology |7; 13; 16~.

5. For the purpose of this presentation, we assume away operation and maintenance costs. In absence of such costs, there is no need to control the firm's state of compliance with the regulation if the regulator observes that an emissions control equipment has been installed. However, other initial compliance activities may also have been undertaken by the firm (e.g., substitution of inputs, reduction of output, etc.). We assume that the performance of these other activities, in terms of reducing the firm's level of pollution to the required standard, is unknown to the regulator. With operation costs, one needs to distinguish between initial compliance and full compliance activities. Full compliance with the emissions standard is achieved at a later time than initial compliance |14~. Results are qualitatively the same. 6. These are obviously ex post probabilities as the ex ante probability of observing any given level of pollution is zero.

7. Dewees writes: "Three principal forces generate incentives for firms to reduce pollution discharge. The first is the cost associated with violating government regulations. The most obvious such cost is the amount of any fine levied upon the firm for its offence (captured by F; brackets are ours, as are following ones). That these explicit penalties are often small, however, does not mean that enforcement is without effect. Ontario's Ministry of the Environment may order a firm to reduce its discharge when that discharge is unlawful (captured by |Gamma~T|C.sub.t~(|P.sup.c~)). The second factor inducing firms to reduce pollution discharge is potential liability for any harm that might result from those discharges (included in F). The third factor is the polluter's concern about its public image (captured by |Mathematical Expression Omitted~)" |5~.

8. Our initial sample was made of 58 events. However, following Suret, Pauchant and Desnoyers, eleven events were discarded since another event was occurring within 10 days before or after the announcement of the "environmental event" |27~. These other events include an announcement of dividend pay-offs, profits, merger, take-over or new share emissions.

9. Given the dimension of the Canadian economy, the size of our sample is comparable to that of MRG who consider 128 lawsuits and 74 suit settlements in the United States for the period 1977-86.

10. This contrasts with events that are announced in "News-Brief" type columns. 11. A number of alternative tests were made to test the robustness of the results with the single-index market model (SIMM) and the market adjusted returns model. Results were not altered using these two other techniques. For more discussion on these techniques, see Henderson |9~.

12. Daily returns were obtained from the TSE/Western Data Base.

13. Specifically,

|Mathematical Expression Omitted~

where |Mathematical Expression Omitted~.

14. Where |Mathematical Expression Omitted~.

15. The magnitude of these losses is larger than what is reported by MRG.

16. The significant abnormal return on the day prior to the publication date suggests that information about the investment may have been available to shareholders prior to day 0.

17. These are Paper, Chemicals, Stone, Clay and Glass, Iron, and Steel and Non-Ferrous Metals.

References

1. Barbera, Anthony J. and Virginia D. McConnell, "The Impact of Environmental Regulations on Industry Productivity: Direct and Indirect Effects." Journal of Environmental Economics and Management, January 1990, 50-65.

2. Beavis, Brian and Ian Dobbs, "Firm Behavior under Regulatory Control of Stochastic Environmental Wastes by Probabilistic Constraints." Journal of Environmental Economics and Management, June 1987, 112-27.

3. Conway, Tom. "Taking Stock of the Traditional Regulatory Approach," in Getting it Green: Case Studies in Canadian Environmental Regulation, edited by G. Bruce Doern. Toronto: C. D. Howe Institute, 1990, pp. 25-57.

4. Dewees, Don. "Evaluation of Policies for Regulating Environmental Pollution." Ottawa: Economic Council of Canada, Working paper 4, 1980.

5. -----. "The Effect of Environmental Regulation: Mercury and Sulphur Dioxide," in Securing Compliance: Seven Case Studies, edited by M. L. Friedland. Toronto: University of Toronto Press, 1990, pp. 354-91.

6. Fama, Eugene F. Foundations of Finance. New York: Basic Books Inc., 1976.

7. Freeman, A. Myrick, III, "Technology-based Effluent standards: the U.S. Case." Water Resource Research, February 1980, 21-27.

8. Fry, Clifford L. and Insup Lee, "OSHA Sanctions and the Value of the Firm." The Financial Review, November 1989, 599-610.

9. Henderson, Glenn V., Jr., "Problems and Solutions in Conducting Event Studies." Journal of Risk and Insurance, June 1990, 283-306.

10. Hetu, Jean, "Les sanctions penales et la protection de l'environnement." Revue du Barreau, May-June 1989, 494-513.

11. Jorgenson, Dale W. and Peter J. Wilcoxen, "Environmental Regulation and U.S. Economic Growth." Rand Journal of Economics, Summer 1990, 314-40.

12. Lanoie, Paul, "Occupational Safety and Health: a Problem of Double or Single Moral Hazard." Journal of Risk and Insurance, March 1991, 80-100.

13. Laplante, Benoit. "Environmental Regulation: Performance and Design Standards," in Getting it Green: Case Studies in Canadian Environmental Regulation, edited by G. Bruce Doern. Toronto: C. D. Howe Institute, 1990, pp. 59-88.

14. -----. "Optimal Timing of Compliance with Stochastic Penalties." Universite Laval, 1991.

15. Linder, Stephen H. and Mark E. McBride, "Enforcement Costs and Regulatory Reform: The Agency and Firm Response." Journal of Environmental Economics and Management, December 1984, 327-46.

16. Luken, Ralph A. Efficiency in Environmental Regulation, A Benefit-Cost Analysis of Alternative Approaches. Boston: Kluwer Academic Publishers, 1990.

17. Magat, Wesley A. and W. Kip Viscusi, "Effectiveness of the EPA's Regulatory Enforcement: the Case of Industrial Effluent Standards." Journal of Law and Economics, October 1990, 331-60.

18. Marchant, Gary E. "Environmental Protection in North America: a Comparative Analysis of the Regulatory Approach in Canada and the United States." Harvard Law School, 1990.

19. Muoghalu, Michael I., Robison, H. David and John L. Glascock, "Hazardous Waste Lawsuits, Stockholder Returns, and Deterrence." Southern Economic Journal, October 1990, 357-70.

20. Russell, Clifford S. "Monitoring and Enforcement," in Public Policies for Environmental Protection, edited by P. R. Portney, Washington D.C.: Resources for the Future, 1990, pp. 243-74.

21. Peltzman, Sam, "The Effects of Automobile Safety Regulation." Journal of Political Economy, August 1975, 677-725.

22. Portney, Paul R. Public Policies for Environmental Protection. Washington D.C.: Resources for the Future, 1990.

23. Saxe, Diane, "Fines Go Up Dramatically in Environmental Cases." Canadian Environmental Law Reports, 1989, 104-21.

24. Sinclair, William F. "Controlling Effluent Discharges from Canadian Pulp and Paper Manufacturers." Canadian Public Policy, January 1991, 86-105.

25. Smith, J. B. and W. A. Sims, "The Impact of Pollution Charges on Productivity Growth in Canadian Brewing." Rand Journal of Economics, Autumn 1985, 410-23.

26. Suret, Jean-Marc and Thierry Pauchant, "La reaction des prix des titres aux catastrophes majeures." Universite Laval, 1991.

27. -----, -----, and M. Desnoyers, "Prevention et marches financiers." Risques, 1991, 139-52.

28. Vaughan, William J. and Clifford S. Russell, "Monitoring Point Sources of Pollution: Answers and More Questions from Statistical Quality Control." American Statistician, November 1983, 476-89.

29. Viscusi, W. Kip, "Consumer Behavior and the Safety Effects of Product Safety Regulation." Journal of Law and Economics, October 1985, 527-53.

30. -----. "Reforming OSHA Regulation of Workplace Risks," in Regulatory Reform: What Actually Happened, edited by L. Weiss and M. Klass. Boston: Little, Brown and Co., 1986.

31. ----- and Joni Hersch, "The Market Response to Product Safety Litigation." Journal of Regulatory Economics, June 1990, 215-30.

32. Webb, Kernaghan. Pollution Control in Canada: the Regulatory Approach in the 1980s. Ottawa: Law Reform Commission of Canada, 1988.
COPYRIGHT 1994 Southern Economic Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1994, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Lanoie, Paul
Publication:Southern Economic Journal
Date:Jan 1, 1994
Words:6580
Previous Article:Compensating wage differentials and the optimal provision of unemployment insurance.
Next Article:Innovation, imitation, and social welfare.
Topics:


Related Articles
Gambling and Speculation: A Theory, a History and a Future of Some Human Decisions.
Issues in Contemporary Economics, vol. 2, Macroeconomics and Econometrics.
Adverse incentives from improved technology: traffic safety regulation in Norway.
Understanding Consumption.
The Economics of Earnings.
New Perspectives on Business Cycles.
CEOs, information, and decision making: scanning the environment for strategic advantage.
Rami Benbenishty and Ron Avi Astor, School Violence in Context: Culture, Neighborhood, Family, School and Gender.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters