E , N V I R O ~ N T & SPEdwDpNG AND SHARE PREE PERFORWUVCE: THE PETROLEUM .INDUSTRY

Jenifer Piesse' Centre for Business and Management Studies Brunel University

Introduction

The current increase in public concern for the environment has resulted in some companies announcing major environmental spending programmes. The petroleum sector is particularly interesting and receives high levels of media exposure, partly because the results of an oil spill are so disastrous and also because clean-up operations are frequently undertaken at an i n t em~t io~ l level. Therefore, this p r e h b r y study concentrates on environmental spending in this industry.

However, while there is a large literature about the evaluation of the costs and benefits of environmental protection, and the implementation of policy on behalf of the public sector, (see for example, Barde and Pearce (1991)), less research has been done ae the corporate level to measure the response of fums to this issue. There are a couple of possible reasons why the private Sector response has been largely ignored First, until recently, companies were not required to disclose in their corporate accounts, details of research and development expenditure, either regarding the nature of the research or the areas of operations affected, and a fair amount of R&D is based around safety and reducing pollution levels. Only the aggregate emission needed co be reported, and this gave a minimal amount of information, both to existing or potential shareholders, and to researchers.

Secondly, the established view in finance theory is that the objective of managers is to maximise the wealth of shareholders. Therefore decisions about investment expenditure, including that which specifically protects the environment, are undertaken using traditional capital budgcting procedures. Any evaluation of the costs and benefits of this expenditure is difficult myway since many of the resources arc variably affected and are not assets Lhat are subject to clear private ownership. They are elements such as clean air and water, or the continued existence of certain species of flora and fauna, the value of which cannot be easily measured in monetary terms. This point is further compounded by the fact that valuations of this sort arc normally based on individual preferences, and the principle of maximising individual expected utility does not easily extend to a community aggregate.

But despite these difficulties, the impact of private sector spending on environmental protection is becoming increasingly important, as is evidenced by the volume of expenditures incurred, and also by the effect on shareholders of what can be seen as a social or ethical investment policy, or a lack of it.

This paper first discusses the theoretical aspects of market failure, and the traditional correction mechanisms. Then a number of approaches taken by corporations in the petroleum industry are identified and compared. Case studies are used to examine market reaction following policy implementation by the companies involved

I. Market Failure

Before discussing the causes and consequences of market failure, we should briefly consider the sense in which markets are efficient - since the best way to understand market failure is to first understand market success. This requires that a collection of perfectly competitive markets should achieve an equilibrium allocation of resources which is Pareto optimal. This is more formally defmed as the First Fundamental Theorem of welfare economics, which states the following: (1) if there are enough markets, (2) if all consumers and producers behave competitively, and (3) if an equilibrium exists, then the allocation of resources in that equilibrium will be Pareto opha l . (Ledyard, 1990).

45

Any departure from the conditions above results in market failure, and the traditional literature identified two main causes. These were the existence of elements of monopoly power (as in (2) above) and the occumnce of externalities or spillovets', which correlates with point (I), in the modem approach used above. The modem approach thus tends to make "market failure" tautological; markets fail because they don't exist, are incomplete, or have too few actors. It also makes it obvious that the two traditional causes are not really very Werent; the distinction between not having a buyer and a seller at aU, for a particular contingent claim for a p i b l e state of the world (see footnote 2) and not having enough buyers and sellers for the market to be competitive is merely a question of degree. Thus, the modem approach unifies market failure, but we will find that both of the causes of market failure have direct application to the petroleum industry.

i) Not Enough Competition

As a general rule, markt structures fail to conform to the perfectly cmpetitive ideal. There are often an insufficient number of players, whether buyers or sellers, i.e. the market is too thin. In the case of the petroleum industry, the OPEC cartel is a particular form of market structure which distorts prices and restricts output. It has been shown that OPEC has worked very successfully both because it controls a substantial share of the reserves and the production of oil (about 66%), and the non-OPEC players cannot respond very competitively to a cutback in supply (pindyke, 1978). The organisation of fums into a cartel denies the market the most common form of correction for non- competitive behaviour, that is to ensure that each firm is small, in terms of its control over resolllces and access to information.

Apart from the cartel aspect, the industry is dominated by a handful of multinational companies, with considerable market power and a tendency to collude. The particular aspect of collusion that is of m a t interest is the communal arrangements for bearing the costs of cleaning up in the event of a disaster; this raises the issue of moral hazard.

U) Not Enough b5arket.s

Market failure due to a lack of a complete set of markets is the rule, rather than the exception, once time and risk are taken into account, since fiutures and insurance markets are not complete; they do not cover all goods and seMces, or all states of the world2. There are i ~ ~ s w c e and futures markets in which dated commodities are traded, but both of these are limited; for instance, insurance is not available to cover every outcome, and it is not possible to buy or sell a barrel of oil at a given price for every date in the future.

However, the lack of markets in the area of ~ tu ra l and environmental murces is more fundamental, since the common property element prevents there being even a spot market. In the absence of welldefmed private property rights, the probability is that the market will not exist at all, or it will at best be incomplete or fragmented. This allows some transactions to have non-market spillover effects, or extedit ie3.

ili) Externalities

J.-J. M o n t (1990, p.112) defines technological externalities as the indirect effect of a consumption activity or a production activity on the consumption set of a c o m e r , the utility function of a consumer or the production function of a producer. By indirect, we mean that the effect concerns an agent other than the one enacting this economic activity and that this effect does not work through the price system. The most frequent examples of externalities are those involving negative pollution activities and unfortunately there is no shortage of evidence that they exist in the petroleum industry. Indeed it is reasonable to suggest that the theory of technological externalities forms the basis of environmental economics.

XI. Some Solutions

Several possible solutions are offered, which can be implemented with varying degrees of success. These are detailed below:

46

i) Establish a AUarket for the Externality by DeZintng Property Rights

It may be pssible to create a market for the externality, under two scenarios. If the initial condition is that the operators can be given the right to pollute, then consumem c,an purchase the right to a clean environment. In the extreme case, the consumers can buy all these rights, and the fm will cease to operate, and the externality has disappeared. Alternatively, if the starting condition is a clean envircmment, then potential polluter^ have to pay to purchase the right to pollute. This is the 'polluter must pay' state, which is one facing the companies in this analysis. h the event of an oil spill, propem rights lie with those whos environment is damaged, and the producer is liable for clean-up costs, damages and fines. Thus, thase at risk are protected, at least to an extent, by legislation.

U) Technological Solutions

Technical measures can sometimes be installed to solve externality problems; catalytic converters on cars are a good example. Legislation is passed to regulate auto emissions, and the industry must respond by producing vehicles that pollute less. In this case, the vehicle that pollutes less is a safer tanker, and there are minimum safety regulations for tankers.

iii) Other Remedies

If it is possible to correctly measure the extent of the externality, the government is able to tax a negative externality or subsidies a positive one. The payment will replace the value of the unpriced item in the market, so that an equilibrium is achieved. The structure of the tax or subsidy can be used to set constraints to output, like a zero tax to a certain level of pollution, and then a pasitive tax beyond. The imposition of clean-up costs etc. could be viewed as a pollution tax.

For any solutions to be implemented, an external legislator is required with both information about, and the ability to measure the market imperfections, and the power to set regulations and enforce compliance. This may not correct the market failure totally, but will keep the externality within reasonable bounds. If property rights have been accurately assigned, and a market in the exchange of these rights functions properly, this will enable the correct allocation of responsibility when environmental damage or conservation occurs.

Details of operating conditions are communicated to managers and minimum standards are known to the organisations concerned. The result is that, in a regulated industry, management must operate within the constraints of the standards set, while continuing to meet their commitment to their shareholders. What this paper suggests is that this constrained maximisation problem is further complicated by the inclusion in shareholders' own utility functions not only of wealth maximisation, but also of pollution minimisation.

iv) Summary

To summarise, in the oil industry, there is not a complete set of futures and insurance markets, so firms will have to bear some risks. There is a distinct lack of markets for the environmental resources that are threatened by oil spills. Finally, there is, or at least was, a particularly effective cartel controlling prices and outputs in the oil industry.

All of these aspects of the industry should be taken into account, but this paper will concentrate on the environmental externality problem. Suppose that there are two possible states of the world. In the first, oil is transported successfully, and there is no environmental externality'. In the second state, a major spill OCCUTS, and environmental damage results. If the offending operator could avoid the costs of the environmental damage, then there is a considerable negative externality. The authorities have reacted in two main ways to this problem. First, the oil company is liable for the costs of the clean-up operation and will be fmed. Secondly, oil companies are not free to take any risks they like; the potential danger is regulated by requiring that tankers meet particular standards.

Thus, the public sector has reacted to the potential externality with liability legislation and safety regulations. But, as was argued in the introduction, private companies are also in a position to

respond, both to the initial potential externality problem, and to present and expected fume legislative and regulatory controls.

m. Corporate Response to Regulation

One issue which recue~ throughout the literature on mgulation is the choice of the most efficient policy for neutralisig externalities. One option is to implement ex unre policies, such as a tax or subsidy, or a safety standard, which take effect before the externaliw is generated. The other is to i m p an ex post correction if there is proven liability following the creation of an externality. Various authofi have questioned both policies, generally on punch of inefficiency. Ex ante measures are criticised because of informational deficiencies on the part of regulators about the value of the potential damage (Baumol and Oaks (1971), Rose-Ackerman (1973) and Shavell (1984)). Ex posr controls are challenged because of the possibility tbat liability will not be proven, or if it is, bankruptcy denies the imposition of a financial penalty (Shavell (1984, 1987)). These alternatives were generally viewed as substitutes, although Shavell(l987) has suggested that ex unre and ex post regulation can complement each other. Kolstad, Ulen and Johnson (1990) show that where there is uncertainty, such a complementary use of policies is optimaL

But although public policy may include both espects of regulation, expenditure decisions at the level of the f m are more likely to reflect the individual corporate view of the environment. Aldersey- Williams (1990) surveyed several companies to elicit Views from managers about the affect of environmental spending on share prices, and their perception of its importance as a source of competitive advantage, but the resulb were inconclusive. It is suggested m this paper that firms manage their environmental strategy by following one of two Merent approaches. One of these results in a higher level of environmental protection than the other, but both are subject to the chance of a negative response from the market. These appraaches can be distinguished as follows.

i) Proactive Environmental Expenditure

Proactive expenditure is defined as when a company voluntarily spends more than it is legally required to do, in order to improve its environmental performance, and reduce the risk of an accident This may result in the fm reaching a higher level of safety than that required by the regulators.

ii) Reactive Expenditure

Reactive expenditure is characterised by fines or clean-up expenses which are imposed following an environmental violation. This has increased as a result of stricter legislative pressure and tighter standards leading to a greater number of companies being required to undertake reactive expenditure.

lV. Three Case Studtes

The remainder of this paper compares rhe performance of three companies. One undertakes voluntary ex anre preventative expenditures, where this can include both research and development costs and implementation of new technology, while the other two respond ex po~r to the costs necessitated by an accident Movements in the share price following the announcement of either increased environmental spending or the imposition of a fine, are used to measure the extent of investor reaction.

Three ohsewable events were used in this preliminary analysis; Conow’s proactive spending programme; Exxon’s Exxon Valdez accident cleanvp costs (and h e ) ; and Shell UK’s fl million fine for polluting the Mersey as a result of a pipehe failure.

i) Conoco: An Environmental Spending ProgTamme

In April 1990, Conoco announced that it had adopted a nine point plan to reduce toxic emissions and prevent oil spills. The plan was expected to cost $50 million per year, and it was to include the building of two new double hull tankers, to be completed by 1992. The tankers would cost $70 to $90 million - 15% more than conventional tankers.

48

ElWIROKHEX'lYLL SPEXDIRGAXD SEARE mCE PERFORMANCE

DuPont, which owns Conoco, has 685 million shares in issue. The $SOmn extra expenditure, if deducted from the 1989 dividend, discounts the share price by the present value of 7 cents annually and over the life of the tanker, approximately 50 cents. However an effect of this size is not discemable in the share price.

Figure 1 shows the share price of DuPont compared to the FT All Oils index at around the time of the announcement. DuPont is not a typical oil company; the bulk of its activities are industrial and chemical, with petroleum yielding around one third of its sales. The share price performance does not closely track the oil sector, as the divergence at the beginning of the Smt m d thud quarter of 1990 shows.

~

Figure 1: Share P r l a of DuPont Compared with the FT All 011s Index

DuPont v s FT All Oils 2 6

25

24

p 2 3 .- - k 22

2 1 e

cn C .- $ 2 0

k 1 9 .-

1 8

2500

2400

2300

2 2 0 0

2100

2000

1900

1800

16 1700 1Q89 2089 3089 4089 1090 2090 3090

- DuPont (sterling) - FT All Oi!s Index

X a, U C -

Both of these departures were due to oil price changes. The second, monger divergence was related to the oil price increase at the end of June, which was due to increased tension in the Gulf, prior to the invasion of Kuwait by Iraq, as well as other seasonal pressures on the oil price. Increases in the oil price improve the share price of oil companies, and lead to increased raw material costs for industrial companies, thus driving their share prices down. DuPont behaves like the industrially based company it is, and mova in the opposite direction to oil companies when the oil price changes.

It is clear that DuPont matched or outperformed the market at around the time of the announcement of their environmental spending programme, but no significant change is seen in their share price at that time. It can reasonably be inferred from the share price performance that the announcement of the environmental spending programme did not significantly discount the share price, and may have increased it relative to the rest of the oil sector.

ii) Exxon: The Exxon Valdez Disaster

In March 1989, the Exxon tanker "Exxon Valdez" grounded in Prince William Sound in Alaska, spilliig a significant quantity of crude oil into an environmentally very sensitive area. Figure 2 shows the share price of Exxon compared with a basket of its rival companies from the third quarter of 1988 to the end of 1989.

49

prior to the spill, Exxon shares were trading at a premium to the average of the peer group of companies of $2 to $4. After the disaster, on the 23rd March 1989, the shares were. trading at a discount of $2 to $8. This fall of $4 to $12 per share au ld imply that the market expected the disaster to cost Exxon between $5 and $15 billion. This figure is in excess of the actual estimated costs of cleanup which are around $2.5 billion, and the fine which is $15, although the opportunity cost of the incident has been estimated by one source at over $10 bdlioa(3) (Exxon's profits (not assets) 1989 = $5 billion.)

Figure 2: Exxon Share P r l a Compared to a Basket of Peer Group Cornpanles

Exxon vs P e e r s 3 Q88 -4Q89

5 8 56 54 5 2

0

Q

- 5 5 0

c 48 p: 46

dr 44 42

40

1-

.-

3 8 88 3 Q Q4 1989 02 Q3 Q4

- Exxon - Peer Group

It is apparent from the share price history of Exxon that the effect on the sbare price of a company involved in a major oil spill can far exceed the actual costs involved

iii) ShtU UK: Oil Spill in the M e r ~ e y

In August 1989, Shell were responsible for a spill of oil into the Mersey River. In February 1990, they were fined f 1 million, after the fvst Crown Court action taken by the newly formed National Rivers Authority. ?he fine was the largest imposed for environmental damage in the UK at that time, but the judge stated that he had taken Shell's previously unblemished environmental record into account when deciding on the fme. The clear implication is that further pollution would be less leniently dealt with.

Shell also had to cover the costs of cleaning up the spill, which were estimated at f15 million. Theoretically, the market capitalisation of Shell Transport should only fall by .€25 million in total, or less than one tenth of one penny for each of the 3.3 billion shares.

Figure 3 shows the comparison of the Shell Tka~sprt (Shell UK) share price with that of the Royal Dutch group (RDS), which is another part of the Shell corporation. In general, the two prices track each other very closely. At the time of the spill, the two share prices continued to track closely, but when the fine was announced, the price of Shell Transport began to trade at a discount of around 20p per share to that of RDS.

50

Flgum 3: Sham P r l a P e d o n n ~ c e Of Shell Transport Compwtd With Royal Dutch Group

500

4a o 46 0

L

vl 420

.r 400 m 0

a ‘;I 380

360

3 40

320

Shell v s Royal Dutch Group

2Q89 3Q89 4Q89 lQ90 2090 3Q90

-.A- Shell Transport - RDS (sterling)

It is apparent that, at least temporarily, the market felt that the expected value of Shell Transport’s earnings stream was lower than that of the RDS group. The discrepancy in values was 20p per share, or 1670 million. This a m o u t is far greater than the actual cost of the cleanup and the fine, and demonstrates that even small environmental accidents can detrimentally affect a company’s share price.

V. Share Price Behaviour and Market Valuation

Tbe theory of share valuation states that the market value of any share is the sum of expected cash flows, either based on past earnings or dividends, or future revenues, which are then discounted by a factor which reflects the level of risk associated with thcse cash flows. Cash flows are net of the costs of production and therefore have to take into account any predicted capital expenditures which are included in the strategic plan, and by unexpected demands on company funds such as losses, bad debts or fines. All of these elements are included in the valuation of the equity. This is consistent with the notion that the securities market is eficienc and that prices fully reflect all publicly available information. All of these elements are reflected in the share price. However, there may be an additional effect (a response to good or bad news) arising from the nature of the expenditure, which can be explained by some degree of market sentiment. The next stction speculates on what these influences may be.

i) Proactive Environmental Spending

Where environmental spending is undertaken, which is in addition to current investment, there will be an immediate effect of lower earnings due to increased capital outflows. If the investment is made out of retained earnings there is the possibility of a lower dividend. Thus in the short-term any dividend-based or earnings-based valuation should result in a lower sham price. Supporting this is the fact that 8s the volume of oil transported in a reinforced tanker is the same as in the original ones, there will be no change in revenues, the capital market will judge this an U M ~ C ~ S S ~ ~ Y expense which should not be made unless absolutely necessary.

However, in the case of Conoco, the move to double hulled tankers preempts likely legislation, legislation which will obviously affects all companies in the industry equally. If their forecasts of

51

the change in standards is correct, they will have benefitted from the opportunity to undertake the increased expenditures at a time of their choice, and probably at a more agreeable cost of capital. Furthermore, the fm which installs pollution controls now will have less risk of an interruption in operations in the future tban a competitor which delays such expenditures until it is forced by legislation to undertake them et an inconvenient time. In addition, the capital expense can be depreciated over the life of the tanker (unlike the payment of a fine which is a one period loss). The increase in asset value will appear on the balance sheer, leading to a lower risk of bankruptcy, adding to the already low risk of an accident Finally, these expenditures act an a tax shield, and therefore do not affect the distributable profit in the Same away as fines, which are not tax deductible.

Therefore any negative effects of excessive pollution avoidance expenditures on earnings may be compensated by the positive effect of increasing the value of the fm and reducing the risk of damage liability and polluting the environment 'Ihese aspects will appeal to both the rational, wealth- maximising and the envhmentally conscious investment clientele. So on balance, proactive environmental spending is unlikely to affect the share price by its fully discounted cost

The overall net effect of proactive spending is difficult to determine. Belkaoui (1976) examined several companies who had disclosed their pollution control expenditures. He concludes that managers may be advised to increase their expenditures on environmental safely, and importantly, to fully report these expenditures. Belkaoui also verities the existence of an 'ethical' investor, which emphasises the public relations aspects of environmental protection, although later work by Barrett (199 1) claims that investors may think that fums who actively protect the environment are especially attractive, but do not have the specialist knowledge to make correct judgements. This author would claim that information about the effectiveness of environmental protection is a commodity l i e any other, and the increasing number of company environmental audits will help to disseminate knowledge successful in the future.

U) Reactive Environmental Spending

Conversely, all retroactive environmental spending, including clean-up costs, civil liability suits and fines, are likely to negatively, and, most importantly, randomly affect the firms cash flows. The magnitude and timing of the costs are not anticipated, are not included in the budget, and are not built into expected capital outflows (unless there is some provision made for losses such as these). Insurance is not an adequate solution, because of its limited contract Clean-up costs will be reflected in both earnings and dividends, and therefore the share price. They are also written off in one year through the profit and loss account, directly affecting the profit figure of the fm in the year in which they are incurred. The announcement of an oil spill will therefore downgrade the expected value of earnings and future dividend stream, and the share price should fall by the expected discounted c a t of any cleanups and fines as soon as the infomation about an environmental accident is received by the market.

But in addition, there may be negative welfare effects which bias the views of investors. Evidence of shareholder reaction is always too complex to unravel accurately, but some work has been done on the share price of companies involved in Crises caused by hazardous products. Marcus, Bromiley and Goodman (1987) analysed the market reaction to Union Carbide following the of toxic chemical in Bhopal, India, resulting in the loss of life of two thousand people, and severe injuries for several thousand more, Immediately after the accident, the shares suffered a steep decline, mching an almost 50% loss 75 days after the event. However, there followed a remarkable turnaround as rumours of a takeover bid for this now undervalued company were received by the market. 'Ihase shareholders who has retained their equity were rewarded with a return on their investment which was higher than it was before the accident. Apparently, sentiment is a short-run effect, although there may be long-term consequences.

Conclusion

From this initial inspection of company performance, it appeas that the share price of oil companies behaves broadly as would be predicted in terms of direction of change following news of positive or negative environment effects, but by an amount in excess of expectations.

52

-RMEXTAL SPEXDKNG AND SEfARR PJ3RFOR116ARCE

In the f m t case, Conoco's decision to implement a mejor environmental spending programme was not strongly reflected in the share price. 'Ibis is probably because the market saw the items in the spending programme as unavoidable, and they were already factored into the market's view of the shares' value. However, the price of DuPont as a whole me sbove the FT oils, BS other companies in the sector did not undertake expenditure to increase safety until forced to by legislation.

In the c8se of Exxon, where large unforeseen c a b were incurred, the market price for the shares was heavily discounted relative to the rest of the oil sector. In fact the fall accounted for more than the actual mts of the ckinup, and may have taken account of the likely fall in profits resulting from a consumer shift away from &on. In this example, market sentiment affected the share price of Exxon more strongly than is explained by current theory.

In the case of Shell UK, the fines and costs of cleanup should have had a negligible impact on the financial performance and the expected value of the future earnings streams, since the size of the payment was fairly small. However, the market temporarily discounted the share price of Shell Transport quite substantially, relative to the Royal Dutch Group, even though their financial performance should be very similar. This indicates that companies which have had their share prices discounted for these reasons are undervalued in the short tern.

The p m p i t i o n that the major objective of firms is to maximise shareholders wealth is at the Centre of fmce theory. However, this strategy may be complicated in some areas of social policy.

* I would like to thank John Akfersey-Mlliams, (Schmder Wragg) for early discussions on this papec and for providing the case study m e r i a l .

Footnotes

1.

2.

3.

4.

We could add public goods, which tend to be regarded as strong case of externalities. This is inexact, but makes sense, since the two defining characteristics of public goods are that exclusion is costly and that consumption is non-rival. The first of these is the defining characteristic of an externality and if the second feature is also present, the externality can be called non-depletable (Baumol and Oa t s , 1971). More formally, Arrow (1971) showed that if spot markets are complete and prices perfectly forecasc then a complete set of markets requires that there be one contingent claim (security) for each possible state of the world. State contingent outcomes are obviously directly relevant to the oil spill problem. There has been same debate in the literature Centred around the difference between technological and pecuniary externalities. One attraction of the modern approach is that it rehabilitates pecuniary externalities, showing that the cause is a lack of futures markets (Ledyard, 1990, p.186). However, the analysis which follows concentrates on technological externalities only. This abstracts h m reality, in that tankers habitually foul the ocean, but the extent is within limits that society fmds tolerable.

References

1.

2. 3.

4.

5.

6. 7.

Aldersey-Williams, J., (1990) 'Environmental Policy', Unpublished MBA dissertation, Kingston Business School. Artow, KJ., (1971) h a y s in the Theory of Risk-Bearing, North-Holland Barde, J-P and Pearce, D.W. (eds), (1991) Valuing the Environment, Earthscan Publications Ltd, London. Barrett, S., (1991) 'Environmental Regulation for Competitive Advantage', Business Straegy Review, Spring, p.1-15. Belkaoiu, A., (1976) 'The Impact of the Disclosure of the Environmental Effecls of Organisational Behaviour on the Market', Financial Management, Wmter, p.26-30. Daily Telegraph, 'Shell Fined f 1 Million For Oil Spill', via Textline, February 24th 1990. Datastream, (1990) Sham pnce information 1990.

53

8. DuPont, (1989), DuPont Annual Report. 9. Fisher, A.C., (1983) Resource and Envimnmental Economics, Cambridge University 10. Inwsor Relutions, Sun Company, Radnor, Pennsylvania, (1990). 11. Kolstad, C., Wen, T. and Johnson, G., (1990) ‘Ex Post Liability for Harm vs Ex Ante Safety

Regulation: Substitutes or Complements?’, Amencan Economic Review, Vol.80, No.4, p.888- 901.

12. Wont , J.-J., (1991) ‘Externalities’, Allocuzion, Znjbmafion and Markets, Eatwell, Milgate and Newman, (eds), Micmillan, London.

13. Ledyard, J.O., (1991) ‘Market Failure’, Allocation, Injbrmufion and Markets, Eatwell, Milgate and Newman, (eds), Macxnillan, London.

14. Leresche, R., (1989) Executive Director, Alaska Energy Authority, State Oil Spill Coordinator, Conference Dinner Remarks, Fourth International Energy Conference, London.

15. Marcus, A., Bmmiley, P. and Goodman, R., (1987) ‘Preventing Cotprate Crises: StockMarket Losses as a Deterrent to the Production of Hazardous Products’, Columbia Journal of World Business, Spring, p.3342.

16. Pindyke, R.S. (1978), ‘Gains to Producers from the Cartelisation of Exhaustible Resources’, Review of Economics and Starkria, V01.60, p238-51.

17. Rose-Ackerman, S., (1973) ‘Effluent Charges: A Critique’, C a d a n Journal of Economics, V01.6, ppJ12-28.

18. Reuters, ‘Conoco’s Green Plan to Cost $50 Million Per Year’, via Textline, April loth 1990. 19. Shavell, S., (1984) ‘A Model of the Optimal Use of Liability and Safety Regulation’, Rand

20. Shavell, S., (1987), Economic Analysis of Accident Law, Harvard University Press. Journal of Economics, Vo1.15, pp.271-80.

54 BusnVEfm S m m Y AND TEE ERvraoRlaERT voune 1. PARr 1.8pnnsa 1993