In July 1989, The Wall Street Journal estimated the cleanup cost to exceed $1.25 billion, and less than one-third was covered by insurance. Exxon said its net income for the June 30, 1989, quarter, after the spill-related expense, amounted to $160 million, or 13 cents a share, a pittance compared with the previous year's second-quarter net of $1.2 billion, or 90 cents a share. In fact, that's the lowest Exxon profit for any quarter in 20 years. Earnings had consistently topped $1 billion a quarter, the only exceptions being two periods in 1986 when net dipped to $745 million. Exxon's accounting staff must have racked their brains- and their disclosures files-to find some precedent for spreading the cleanup costs, either forward or backward.

Although in several respects this spill was unique, environmentally sensitive transactions are not rare. The matter of settlements for large-scale pollution had already arisen in Japanese fishing villages hit by mercury poisoning and, more recently, in the Union Carbide chemical accident in Bhopal, India. Most industries in the petrochemical field and other sectors already face escalating public expectations about environmental protection. Exxon must have known that to retain respect at the gas pumps of North America, it had no alternative but to respond immediately and comprehensively.

But how does our profession prescribe the accounting for something like the Valdez oil spill? Are there accounting alternatives for these environment-destroying transactions? it is unlikely that conventional insurance can be obtained to hedge against costs of this magnitude.

Most businesses today spend more than ever before on environmental protection. But these costs are not segregated in their financial statements. That's why the public has no way of judging the degree to which companies are responding to environmental concerns.

Why are companies spending more? Partly because of the changing social conscience, and partly because of increasingly stringent and comprehensive environmental legislation at all levels of government. No North American manufacturer or transportation company can make or move anything without running into tough environmental regulation. In Canada, as an example, most movements of dangerous goods are governed by the Transportation of Dangerous Goods Act. An oil spill would come under the Canada Shipping Act. The polluter is liable for the cleanup costs. As a result, a tightening net of contingent liabilities now encircles every environmentally sensitive transaction, a net that can squeeze the economic viability of any sized entity in the future.

Along with the legislation on marine pollution, tougher laws concerning ground waste have greatly increased company liabilities for site cleanups. North American actuarial estimates for contingent liabilities of this kind now run in the billions of dollars. Everything from credit ratings to potential corporate takeovers in the U.S. is affected by such legislation. Interested investors want to know the potential cleanup bill they may be inheriting.

The point is, companies today are spending more money on the environment, as well as more time in the board room discussing what they're doing to the environment and what the environment may be doing to the bottom line. Will the;, drill for oil in the Arctic, given the environmental sensitivity? How much will it cost to comply with new emission standards and the push for zero effluents?

Governments on many levels are also addressing the public's concern with more comprehensive and stringent anti-pollution laws. Nonetheless, environmental disasters such as the Exxon Valdez oil spill will continue. Technological failure or human error is always a risk; every industry involved in environmentally sensitive transactions faces potential liabilities.

SHOULD THERE HAVE BEEN AN ACCRUAL?

After the spill, Exxon's accountants might have wondered whether they'd overlooked an unrecorded liability for all the years oil had been shipped from Valdez, Alaska. It's unlikely the "implicit" social contract between Exxon and the residents of Prince William Sound was based on the assumption that a major environmental accident was probable. At most it was assumed Exxon would take responsibility for cleaning up if, not when, this happened. otherwise a recurring accounting accrual would have been made. Such accruals are common in the mining industry, when a company knows it will have to restore the land when a mine is shut down.

Exxon's accountants must have known that every shipment of North Shore crude carried with it an associated risk and a potential liability. In 1989, the risk became reality and the potential liability a real one-a net cost of $850 million.

For the 1989 financial statements, they found themselves stuck with a solution that some might argue allowed for no real matching of past profits and deferred costs. The incident is important in that, because of legislation and public expectation, the company assumed financial responsibility for environmental damage-a social cost that probably would not have been acknowledged 50 years ago. As I followed the story, it became clear to me that the accounting confession, for all its experience, still lacks models for handling transactions with implied social costs.

Accounting models have not dealt with social costs, despite such costs having been associated with profit making. Environmental damage is only one kind of social cost. Early industrialization depended, in part, on child labor; mining went hand-in-hand with catastrophic hazards and long-term illnesses like black-lung disease; and ventures like building railroads, cost many lives. Now we see a single organization-Exxon- assuming financial responsibility for a billion-dollar social cost. With companies now booking these kinds of costs, the accounting profession must analyze the adequacy of existing practices and consider models.

REVENUE RECOGNITION AND SOCIAL COSTS

When I say "There's no accounting for the Exxon Valdez," I mean there's no accounting for social costs. Assigning the entire net cost of $850 million to one quarter, a cost that almost offset that period's earnings, is poor accounting; it fails to match revenues with related social costs. How can we measure costs that aren't matched with revenues in the traditional sense, costs incurred not for the good of an individual company, but for the good of society?

Is there a connection between financial accounting, accounting for natural resources, and changing public attitudes about who should pay for the social costs spawned by private enterprise?

Adam Smith told us the invisible hand of self-interest would guide resources to their best use in the creation of wealth. With hindsight, we can point to the Industrial Revolution as the source of the gap between the ordered world of double-entry bookkeeping and the modern reality of large-scale pollution and its attendant costs. That period brought with it the first evidence of unplanned social costs; double- entry, accounting could not handle pollution from tall smokestacks, the dislocation of human life and the consequences of child labor.

CURRENT PRACTICES ARE FLAWED

Current practices suffer from two fatal flaws when it comes to the problem of environmental accountability. First, we cannot account for the full costs of production, including the costs of "consuming" essential natural resources such as air, water, and fertile land; often they have no assigned monetary costs. our double-entry accounting efforts are geared to measure financial transactions rather than resource consumption. We cannot account for social costs and the related benefits.

Second, our accounting rules penalize, rather than encourage, the environmentally responsible enterprise. The more a company spends on prevention and cleanup, the less its earnings. We lack a vehicle for recording "green assets" and monitoring their use, for distinguishing between costs of renewable and non-renewable resources and for providing accounting incentives to improve environmental protection. Let's look at how these flaws might contribute to the occurrence of another huge oil spill-how they might affect corporate decision making.

Companies engaged in environmentally sensitive transactions like drilling, shipping, and refining oil have to strike a balance between cutting costs and cutting margins of safety against pollution. At present, companies have considerable discretion in deciding on the level of investment in pollution prevention: how much to spend on double- hulled ships, staff training, and enforcement of company policies to promote safe operating practices, like those prohibiting alcohol consumption on duty. Standards that result in other major costs include: tanker design and construction, operating equipment such as advanced radar, oil ballast monitors, inert gas systems, plus operating and manning requirements. Although some of these are incurred complying with international conventions, many are discretionary.

All of these prevention activities carry costs. Today they are buried in operating and capital accounts. The more a company spends, the worse its short-term bottom line. Business would say that operating unsafe tankers in an unsafe way is bad business and bad economics. whatever a company spends on prevention will, if a major spill is successfully averted, reap rich benefits for society. But there's still no way of accounting for this.

Underlying a company's investment decisions is its reading of the changing mood of public expectations, the network of legislation, and an increasing awareness of the economic impact of environmental damage. Anybody in the pollution prevention business will tell you that big gaps remain between the expectations for response capability and what those involved with cleanup operations can usually deliver.

The public expects that an oil spill can be contained, controlled, or cleaned up before it hits the shoreline. But so many variables affect a cleanup: its accessibility, sensitivity of the local environment, the size of the waves, the volume spilled, weather conditions, the experience of those involved. The best response capability and the best network can only deal with spills of a certain size. If the spill is large enough, then some contamination along the beaches becomes likely. All these factors underlie the importance of a company's frank disclosure of its response capability.

At a minimum, all oil companies should have a risk analysis that identifies high-risk areas. They should also consider the costs and benefits of alternative transportation strategies, such as pipelines. But, when a company wants to determine its level of investment in environmental protection, it also requires some method of accounting for social costs.

STANDARDS OF ENVIRONMENTAL ACCOUNTABILITY

How do we account for environmentally sensitive transactions so that we:

* Alert investors to the limits of a company's insurance coverage for cleanup costs and related claims?

* Encourage responsible corporate behavior and an investment in prevention, rather than cleanup?

* Better match the cost of prevention and cleanup with the annual flow of oil and related earnings?

* Record the cost of consuming resources, as opposed to the costs of purchasing them? and

* Reflect the social contract between buyers and sellers and "invisible stakeholders"-all those affected by an oil spill?

Perhaps the first accounting lesson is that investors need more information on the potential impact of environmental protection costs on their investment. Specifically, disclosure about contingent liabilities in statements might include:

* The obligation for environmental protection imposed by existing legislation;

* The limits of a company's insurance coverage for environmental disasters including major oil spills; and

* Specific, identifiable, contingent liabilities for the cleanup of existing toxic wastes.

This would be a good first step, given Exxon's precedent in assuming responsibility for cleanup costs. Such a change would also be consistent with the trend toward increased disclosure and would not change the traditional purposes of financial statements. Perhaps the next lesson is that investors, and other stakeholders, such as concerned consumers, want more information on how much a company spends on prevention efforts. In my review of published annual reports, I found virtually no financial information on how much, for example, a major oil company spends on safe ships, operating procedures, or specialized pollution- prevention equipment.

There must be an accounting of these costs, both how much and how little. Again, separate disclosures detailing a company's expenses for environmental protection would be desirable. While this might pose some definition problems, these issues can be resolved. Again, this step could be taken within the context of traditional financial accounting.

THE NEED FOR NEW FINANCIAL MODELS

Now my efforts to explore ways of meeting the ambitious accounting objectives begin to falter. Each alternative I considered required an accounting model that was beyond the stated purpose of traditional Financial statements. According to tradition (and current practices), financial statements are primarily to help investors and creditors to make decisions. Perhaps we need to develop other models.

Start With the Normative

One might be a "normative" accounting model that starts with this premise: We're going to pay for pollution either way-through better prevention or eventual cleanup-but we're going to pay. The cost accrues from a smokestack's first emission, from the first shipment of oil. We have to start tracking the potential costs of resource consumption, through pollution, at the production phase.

This might mean that the cost of prevention (as measured by state-of- the-art equipment) has to be recognized at the first point of production, even if such costs are not actually incurred by the specific company. That is, we would book a social cost, on an industry-wide basis, regardless of whether the specific entity has actually spent money on pollution-prevention equipment. The rationale here is that once a company has drilled or shipped its first oil, it has incurred a potential social cost. Hence, an accrual should be set up for every ton shipped. Companies should also improve their disclosure of potential environmental liabilities and the limits of their insurance coverage for environmental disasters.

There are obvious problems with this model. Who will set the norms for normative accounting? One of the keys to accounting's credibility is its objectivity. Once accounting addresses "What should be," it moves onto subjective ground. In societal issues-and the environment is only one of them-an elected government is primarily responsible for setting minimum standards. The norm may be set at a higher level, but it cannot be lower. Were accountants to take on this responsibility it would set a new, and perhaps dangerous, precedent.

Proceed to the Radical

Let's turn to another model. One way to encourage investment in the environment might be to devise an accounting regime in which the successful prevention of problems brings about a social-benefit revenue.

This model is intriguing because it addresses the reality that relatively limited efforts by one company could reap benefits for society in terms of what is prevented. it deals with the dual notions that the monetary costs of one company's actions may have to be borne by others, and that since those costs are not borne by the company, they are not recorded in its accounts. Again, the problem here is one of objectivity: How could these benefit revenues be determined?

Or how about a "green" model, based on the notion that industrial activity involves costs to future generations? Some "soft" resources, such as human talent and labor, are renewable. They do not have the same environmental effects as the consumption of nonrenewable resources like oil and gas-resources that take millions of years to develop but are consumed in seconds. The green model might also take into account the concept of resource use, or borrowing, versus resource consumption.

When a company siphons river water through a plant for cooling and returns it uncontaminated to the stream it was drawn from, that resource has been borrowed but not consumed. When a company uses air in a chemical process, but filters out impurities before returning it to the atmosphere, that resource has been borrowed but not consumed. Clearly, sulphur-dioxide emissions consume the atmosphere. But how do we account for the use of clean air, when air has no assigned price?

We could set the price of air at the same level as for state-of-the- art pollution equipment. After all, some pollution may be inevitable, but there's a big difference between high and low pollution counts, one that could cause repercussions for future generations.

But to pursue the green model would require nothing less than a redefinition of traditional accounting. The new "product" might include a statement of resources noting the quantities of air, water and land a company had purchased, borrowed and consumed.

Move Back Toward the Traditional

Perhaps what is really needed is an environmental incentives model that would revolve around the concept that traditional accounting principles (such as matching, conservatism and comparability) are secondary to promoting investment in environmental protection. It would allow the capitalization of prevention and cleanup costs. What's more, it would end the penalizing of companies for spending on environmental protection and cleanup.

The problem, however, is that the asset being built up is the result of accumulated spending on environmental items. Initially, such expenditures would pertain to current business activities-a relationship that would lessen over time. The business location on which the expenditures were made, for example, might no longer be owned by the same company. Similarly, the process that caused the environmental concern might become outmoded. Furthermore, the items on which expenditures are made today might bear no relationship to tomorrow's business. What, then, is the asset in this sort of circumstance?

Normally, assets are amortized unless, like land, their value never diminishes. This practice would resolve the problem I've just described, but would conflict with the objective of not penalizing the earnings of companies investing in the environment.

No doubt other alternatives exist, too, but these were a few of the choices that came to mind as I puzzled over environmentally sensitive transactions. Although I recognized the drawbacks of each from the perspective of traditional accounting concepts, I couldn't resist the temptation of using parts of those models to try a non-traditional approach to the Exxon Valdez spill.

I carried out this exercise because of my personal belief that it's time society moved from entrepreneurship to "eco-preneurship." Adam Smith's invisible hand may allocate scarce resources to the most efficient and, consequently, most profitable enterprises. But the companies that have accounted for the full costs of consuming natural resources, plus other social costs and related benefits, do merit investment funds.

Settle on Results Based Accounting

I experimented with a model that represents a departure from traditional double-entry systems in several ways based on normative accounting-accounting for what should be, rather than what is-and booking the related social costs and benefits it moves us to results- based accounting. if we can prevent pollution, the resulting social benefit should be accounted for and recognized in income.

The starting point is establishing government standards for investing in prevention and cleanup that can be translated into industry standards. The chemical industry, for example, might set standards for emissions, based on legislative requirements. Likewise, the oil industry could set standards for response time and capability for spills up to a certain magnitude. Industry standards could be set for the types of vessels used as well as their size.

Industry could then say to producers, "You have to meet these prevention standards and possess a stated response capability if you're shipping that much oil." The standards would be modified to each company's scale of operations. Benchmarks could be based on investment per ton of oil shipped.

Objectives of the Model

This model's objective is threefold: 1) to better match potential cleanup costs with earnings; 2) to avoid penalizing a company for its cleanup expenses once an environmental disaster has occurred; and 3) to record a social benefit if a disaster is successfully averted. Achieving this is possible if an industry-wide cleanup liability is accrued per ton shipped. Operations would be billed for an annual charge and a corresponding liability would be set up. In the event of a spill, the liability would be drawn down. Once the accrued liability was exceeded, cleanup costs would be capitalized to prevent inhibiting cleanup efforts and those costs would then be amortized subsequently.

And if there's no major accident or disaster? That's where results- based accounting comes in. If desired results are achieved, society reaps a significant benefit that should be recognized. A portion of the accrued cleanup liability should be credited to income in recognition of this benefit.

We must also account for restoration costs, which may extend well into the future. Studies of past spills have shown that environmental effects carry ramifications for decades. Oil eventually solidifies into an asphalt that affects shell life. Those costs, too, must be accounted for, even though difficult to measure. The accounting treatment would be disclosed in notes to the financial statements, indicating the duration of a future liability and the estimated annual charges during this restoration period. Each year, an estimated notional expense for restoration, calculated by environmental experts, would be charged to income, and a corresponding credit made to the environmental liability account.

Companies that invested heavily and effectively in prevention would show the best earnings and balance sheets. The longer the duration between spills, and the smaller their environmental impact, the better the financial results.

Still Not the Answer

On reflection, I soon realized this experiment with non-traditional accounting approaches for the Exxon Valdez suffered from its own conceptual problems. For example, the liability presents difficulties, given that environmental costs are often borne by parties other than those who cause them. In this case, a liability may be recorded by a company that will never pay it; indeed, this liability would remain even after other parties hid paid theirs in full. To disclose a liability in this case would only mislead investors. Even if the company bore the costs, no clear relationship could be drawn between the liability recorded, and the expected amount the company would have to pay to make good on specific environmental damage.

This new accounting treatment, based on whether a company has been "good" or "bad" in its environmental spending, would cause its own problems. While this treatment would allow for incentives, it could reduce the comparability of different companies' financial statements and those of an individual company over time. Such a departure from traditional accounting is a good example of the risks to financial reporting if we shift our focus from objective reporting to an active role in shaping social policy. Acknowledging these problems, I reexamined the current state of affairs and the alternatives.

I decided that the actions taken by Exxon had set an important precedent. Major companies are now prepared to accept wider responsibility for social costs; in short, they are more accountable. Yet, traditional financial accounting remains severely limited in its ability to account for these increased responsibilities.

FINAL WORD-RESOURCE BASED STATEMENTS

The bottom line is the acceptance by industry of a wider sphere of responsibility for environmental protection; yet it also has to have some way of reporting (to a wide range of stakeholders) on how this accountability, is managed. Certainly, traditional GAAP statements could be modified to respond to this increased accountability, but only at risk to their credibility. This suggests the need for a resource-based statement, something beyond the rather short, bland and non-specific annual coverage of health and environmental issues. investors and other stakeholders need more information on:

* The quantities of resources "consumed," rather than borrowed;"

* A company's perceptions of the major risks to the environment and its risk management plans to deal with them;

* its level of investment in environmental protection;

* Its response capability, including the degree of reliance on an industry network;

* Potential liabilities exceeding insurance coverage; and

* A frank discussion of the full consequences of a major oil spill, and, in the absence of a major spill, a discussion of the social benefits realized.

The issue facing the accounting profession is whether we wish to respond to the challenge of trying to develop standards and practices to provide this level of reporting. As the Exxon Valdez spill indicates, the time to make this choice is now.