Accountancy as Computational Casuistics

(Business & Professional Ethics Journal 17 (4) (Winter, 1998): 21-37; repr. in Australian Accounting Review 9 (2) (July, 1999): 36-43; Eureka Street 9 (1) (Jan/Feb, 1999): 2, 43-6; winner of Eureka Street's Ethics Essay Prize)

Abstract: If a company's share price rises when it sacks workers, or when it makes money from polluting the environment, it would seem that the accounting is not being done correctly. Real costs are not being paid. People's ethical claims, which in a smaller-scale case would be legally enforceable, are not being measured in such circumstances. This results from a mismatch between the applied ethics tradition and the practice of the accounting profession. Applied ethics has mostly avoided quantification of rights, while accounting practice has embraced quantification, but has been excessively conservative about what may be counted. The two traditions can be combined, by using some of the ideas economists have devised to quantify difficult-to-measure costs and benefits in environmental accounting.

When BHP managed to close the Newcastle steelworks, its share price surged, to the benefit especially of the directors who made the decision.1 There has to be a suspicion that capitalism is pulling its usual trick of distributing the profits to itself and the costs to someone else. That means that the accountancy is being done wrongly. That in turn means that there is an obligation to discover how to do it right.

There are plenty of other cases where costs are distributed to people against their will, and where they have no legal or other recourse, because of a combination of difficulty in measuring the loss, and the lack of a legal regime to sheet home losses to those causing them more or less indirectly. US buyers of oil do not pay the cost of sending the USS Enterprise to the Gulf to protect the source of the oil; that cost is not ``internalised'', but paid by the State.2 Financiers insist on their rights to global mobility of capital and the freedom to invest where they like, but those in the third world who lose their livelihood as a result cannot insist on any right to global mobility of labour; nor are they compensated for their not having the right to camp on Rupert Murdoch's lawn.3 Or -- to take an example that may appeal to a different part of the political spectrum -- the benefits accruing to small and isolated nations from the United States' role as global policeman are largely unmeasured and unpaid for.

Now, whose business is it to study such issues? Prima facie it is a problem for applied ethics, or some related discipline. But at present, issues of that kind fall down the cracks between disciplines. Law is not applicable, since there is no legal regime permitting those suffering disadvantage of this kind to sue. Applied ethics itself, which has developed from philosophy with some input from law, has inherited those disciplines' phobia about quantification, and hence is in no position to handle the usually economic nature of the rights and losses involved. Economists study such matters, but typically in a ``big picture'' way, and not from the point of view of inquiring how the individuals affected might be able to do something to recover their losses. Accountants have a good awareness of the nature of small-scale quantified gains and losses, but normally restrict themselves to costs that an entity will have to pay, not ones it merely ought to pay.

The problems can only be attacked by a fusion of the relevant parts of all these disciplines. Applied ethicists must embrace quantification, and create a ``computational casuistics'', while accountants must measure not only the obligations that the present legal situation places on entities, but those that morality requires. They must create a ``moral accountancy'' -- which will be the same thing as computational casuistics. (The name ``casuistics'' is taken from the art of casuistry, the application of moral principles to particular cases, on which there were many books for confessors in the period 1350 to 1650.4

This has been done before, and recalling a few historical points may help suggest how to restart progress. Aristotle's discussion of money and prices occurs in the middle of his treatment of justice, in which he goes into some mathematical detail on the different ways of calculating just outcomes in different cases. In the case of a partnership, for example, it is just to distribute the profits in proportion to the stakes invested, while when there is a matter of compensation for wrong, only the amount of the loss arising is to be calculated. He goes on to say that the advantage of money is that it allows all such things to be compared.5 In the middle ages, the scholastics developed Aristotle's ideas, and caused them to have meaning in the courts. In those days, they thought that everything had a just price, normally the price determined by the market. Selling something to an unsuspecting buyer for substantially more than the just price was wrong, and that moral thought was backed up by legal sanction. In ancient and medieval Roman law, one could ask a court to nullify a contract of sale if the price had been in error ``beyond half the just price''.6 Many subtle and powerful ideas were developed on how to find the just prices of futures contracts, annuities and insurances, leading eventually to the moral-legal problem whose solution created the mathematical theory of probability: what is the just division of the stake in an interrupted game of chance?7 One of the solvers was Pascal, who in his other writings dealt casuistry, the application of morality to particular cases, a blow from which it has only recovered with the rise of applied ethics in the last thirty years.

In the meantime, there have been successive waves of Calvinism, positivism, Marxism, utilitarianism, modernism, postmodernism and so on, all of which have tended to drive apart quantitative and ethical reasoning, and indeed, discouraged reasoning about detailed ethical cases at all. Of the major modern ethical traditions, only utilitarianism approved of calculation, and that was calculation of pleasures rather than of something specifically ethical, such as rights; in any case, the requirement to calculate was generally taken to be one of utilitarianism's weak points. Modern applied ethics has revived casuistry, but not its willingness to get involved with numbers.

There are two main issues, or difficulties, with implementing any project that involves quantifying rights and making the answer have effect. The first concerns the possibility of measuring rights. Does it make sense to quantify rights with sufficient accuracy to make intelligible the demand that they should be recognised by a legally-backed system of accounting? Secondly, is it at all practicable to create a legal regime that would enforce those rights? The natural scepticism one feels about both these matters is, it will be argued, unfounded. The technology to measure rights largely exists already in accountancy and environmental economics, while the international legal regime that is currently developing has the means to enforce any measurable right. It needs some more work, but only more of the same kind of thing that is happening already.

Doubts about the possibility of measurement of certain kinds of rights are perhaps best assuaged by considering the one area where there has been a fairly determined effort to quantify moral rights for which there was at present no supporting legal regime. This is the field of ``environmental accounting'', studied mostly by economists (as it has no legal significance, it has not been of interest to most accountants and lawyers). It began with the recognition that National Accounts, such as the Gross Domestic Product, were measures of economic activity that failed to take into account certain goods, and hence were mismeasures of progress. For example, depletion of a scarce natural resource counts as a positive item in the GDP. Thus the costs to future generations of unrestricted economic growth in the present were not measured. To use GDP as a measure of the Good is thus an instance of ``Great Leap Forward'' pricing, the name taken from the response of the Chinese peasants to the Great Helmsman's demand that they increase the production of iron (they melted down their tools).

Economists attempted to devise a system of national accounts that would take into account environmental goods, as well as, for example, the health of populations, so that there would be a measure of whether economic ``progress'' was actually improving the lot of humanity.8

Accountants have been rather behind in this field, but as legal requirements to disclose costs of cleanup have increased, there has been a certain amount of work.9

This is not to say that the economists have measured fairly. Environmentalists have not been keen to recognise all the benefits of normal economic activity. One needs also to measure the benefits to future generations of unrestricted economic growth in present, such as expensive taxpayer-funded medical research with cures for the future, and the discrediting of Stalinist regimes. A focus on just one kind of good, that of the environment, leads to distortions as bad as those that gave rise to the demand for environmental accounting in the first place. All relevant goods must be quantified, if there is to be any plausibly rational attempt to calculate and balance rights.

Cost-benefit analysis has made a concerted effort at weighing all considerations relevant to such decisions as major infrastructure projects. There have been some reasonable ethical objections to some details of its methododology, which can weight the preferences of the poor less than those of the rich,10 but the vigorous attempts in the field to weigh all considerations are admirable.

These intellectual ferments have largely passed accountancy by. While it may be admitted that the subject matter of accounting is ``equities'', that is, rights or titles,11 the profession has traditionally adopted a conservative attitude to what may be quantified for display on a balance sheet. There has been a strong preference for recording tangible and saleable assets, for example, with a good deal of emphasis on what they cost and what they could now be sold for. But that attitude has tended to erode as modern business has created a demand for information on the true financial position of entities. There is regular criticism of accounting standards whenever a giant corporation goes under shortly after reporting a healthy profit, as happened regularly in the 1980s.12 While the principles of measurement in accountancy are controversial, the recent tendency has been to avoid artificial rules and to look for the expected future economic benefit arising from the assets an entity controls.13 ``Expected'' has approximately the usual meaning it has in mathematical probability theory.

Some of what is now becoming standard comes close to the quantification of a moral obligation. For example, if a company is engaged in open-cut mining, and has an established policy of site restoration to a higher standard than that required by law, it must make provision on its balance sheet for that ``constructive obligation'' as soon as the mining is undertaken. Failure to restore the site to the higher standard ``would cause unacceptable damage to the entity's reputation and its relationship with the community in which it operates''; it is going to have to restore the site, so must disclose its present obligation to do so.14

Further clarifications may be made by replying to various obvious objections to the project.

Objection 1.
Rights are not the kind of thing that admit of quantification. Surely a right to life, or a right of a people to land, are not capable of being weighed in a balance, or subject to more and less?

Answer.
It is not maintained that all kinds of rights admit of quantification. Some obviously do. One's right to be compensated by someone who has stolen one's goods, or to be paid for one's work, is limited to a fixed amount, and that is unaffected by the cases considered in the objection.

But it is also true that there is pressure to quantify somehow even rights to life, when they conflict. When scarce health care resources or investments in safety are to be allocated, someone will lose out, that is, have their rights recognised little or not at all. It is usual to quantify claims somehow, in order to balance them -- for example, when there is a need to allocate scarce health care resources. It may be alarming to hear health economists talking about the dollars needed per ``Quality Adjusted Life Year'',15 or experts on safety putting a money value on life,16 but deviations from the results of those methods are also alarming. It is hard to see any superior way of proceeding.

What rights exactly can be quantified is a further question. Perhaps one cannot improve on Aristotle's opinion that quantification is appropriate in two areas: distribution and compensation. In that case, in a case like allocation of health care resources, it may be preferable not to speak of rights to life as themselves quantifiable, but rather to say that the quantification arises only at the stage when distribution of resources to support those rights is in question. Similarly, it may be that the intrinsic or aesthetic value of endangered species is not quantifiable, but quantification must enter when resources are to be diverted from other uses to save those species.
Objection 2.
The quantification of ``rights'', if it is possible at all, arises from something non-ethical. For example, if I promise you $10, the quantification arises merely from what I said, while the ethics of the matter is founded on a general obligation to fulfil promises, which is not itself quantitative. So there is no need for any kind of ``computational ethics''.

Answer.
That may be so, and in that case pure ethics need not consider quantification. But here we are considering applied ethics, in which we wish to look at the actual obligations that arise from ethical principles in particular cases (and compare those arising from different principles, if necessary). It is not true that because ethical principles are non-quantitative, the obligations arising from them also are. You might as well argue that because probabilities arise from symmetry, which is not quantifiable, probabilities also cannot be measured.
Objection 3.
It is quite unrealistic to expect precise measurements of anything so vaguely specified as one's right to clean air, or one's loss of self-respect through being unemployed.

Answer.
Precise measurement is rarely relevant to action. As is well-known in probability theory, a very rough estimate is all that is needed in most cases. I don't know the precise risk of a crash when I get on a plane: maybe it's one in a million or one in ten million; but if I believe it has got to one in a few thousand, I sweat on takeoff and wish I weren't flying Aeroflot. As this example also illustrates, what is often needed is not so much a precise measurement, as a solid minimum. If I can show that an environmental good is worth at least a certain amount, that is enough to get its claims ``on the table'', even if there is no hope of a precise measurement.
And if precise calculations are needed to decide between two alternatives, it probably doesn't matter much one way or the other.

But again, more entities can be priced exactly than was once thought. Options (the right but not the obligation to buy something for a fixed price at a future date) were once thought too speculative to price, but their exact pricing now keeps many mathematicians in comfort.

Reasonably exact measurement is also possible in the area where quantification is most familiar in law: compensation payments for loss of future earnings, as evaluated by actuaries.17
Objection 4.
Some future outcomes are too speculative to measure even approximately, and if something cannot be measured with some reasonable accuracy, no-one is likely to take it into account when acting in the present.

Answer.
The defence budgets of countries, and the advertising budgets of companies, are not observed to be small. Yet the outcomes of that spending are wildly speculative. The same is true of research and development budgets, which are large in countries other than Australia. Decision makers in such cases are acting to insure against various small risks of large, perhaps catastrophic, loss. The risks and payoffs may be almost unmeasurable, but it is still necessary to distinguish between real or appreciable risks and completely fanciful ones.

As examples of quantities that might seem initially to be too vague, but are in fact priced reasonably, consider the goodwill of a business, and the intellectual capital of a research and development company. The goodwill of a business has a substantial causal effect on its future profits, so it cannot be ignored when the expected future cash flows of the business are to be estimated. It can be measured approximately as the excess that a purchaser of the business is prepared to pay for it over and above the value of its tangible assets (including ``visible intangibles'' like copyright).18 Microsoft has few tangible assets (and not a lot of goodwill, either), but it has ``intellectual capital'' in the sense of a team of researchers with a track record of successful innovations, who can be expected to generate future profits. Microsoft's share price can be taken as an estimate by the market of what that is worth (after tangible assets, licences, etc are subtracted). On one estimate, 94% of Microsoft's worth is represented by such intangibles that do not rate a mention on its balance sheet.19 The size of this figure indicates, if nothing else, the need for accountancy to take some risks over what it regards as measurable. Otherwise, it will be sidelined as an irrelevance.

That completes the consideration of objections based on difficulties over quantification. We now come to those concerning the feasibility of any plans for morally restraining large-scale enterprises.
Objection 5.
Capitalists are not nice people, and are not going to accept more moral rules. They are more than sufficiently challenged working around the few they already have.

Answer.
If capitalists really wanted a deregulated environment, they would migrate to Russia, and take advantage of the free market in judicial decisions, the liquidation of business rivals, and so on. They do not do so, as they prefer to operate in countries where exchange is protected by legally backed moral codes against theft and fraud. A norm of fairness underlies exchange: ``to make exchange possible, we must make theft unfeasible.''20 trust in exchanges through iterated prisoners' dilemma games, which shows the stability of tit-for-tat strategies in a wide range of circumstances.21 As one saying has it, ``When trust is betrayed, stakeholders get even.''22

Corporations worry incessantly about their ``image'' of probity, for the obvious reason that other capitalists don't trust sleazebags. After the crash of the late 1980s, there were fears that Sydney's place as the commercial centre of Australia was at risk through its reputation for corporate fraud.23 Legal pressure on insider trading and similar frauds seems to be increasing (though that is hard to evaluate until years later).24

To return to the middle ages for a moment: when national governments were weak and knew little about business, capitalists created their own international legal regime, the Law Merchant, based on commonly accepted basic moral principles and business practice.25 /
Objection 6.
With all major political parties in the West captured by economic rationalism, there is little prospect for plans that involve an increase in legal control. In any case, the globalisation of business means it can escape the efforts of any nation state to regulate it.

Answer.
This is a problem for an approach like that of Mark Latham's Civilising Global Capital which takes the ways of capitalism more or less for granted, and asks what ``policy'' prescriptions can restrain it from running amok. It is also a problem for the idea that it is the ``State'' whose business it is to coerce business enterprises into internalising external costs.26 The proposals here should act at an earlier stage, and should act through liability at common law, not through action by governments. How capitalism pursues profit depends on what the profits are. And what the profits are, if any, depends on the accounting. It is unclear, for example, if Bond Corp etc ever made a real profit, yet it was able to `buy' anything it liked - including its choice of auditor.27 If it could have afforded to buy only out of its real profits, it would not have been heard of.

Legislative control of business may have been weakening, but it is not so clear that legal control has been. The increases in legal control represented by such decisions as Mabo in Australia and the class actions against tobacco companies in the US have nothing to do with legislation. They depend instead on the common law reaching into the moral underpinning or core of its tradition of decisions. Whatever fads the highest courts of Western countries may be subject to, free market theory is not among them. Robust ethical theories seem to be more popular.

In a remarkable speech on ``Commercial law and morality'', Sir Gerard Brennan, later chief justice of the High Court, said:
Moral values can and manifestly do inform the law ... The stimulus which moral values provide in the development of legal principle is hard to overstate, though the importance of the moral matrix to the development of judge-made law is seldom acknowledged. Sometimes the impact of the moral matrix is obvious, as when notions of unconscionability determine a case. More often the influence of common moral values goes unremarked. But whence does the law derive its concepts of reasonable care, of a duty to speak, of the scope of constructive trusts -- to name but a few examples -- save from moral values translated into legal precepts? 28
The complex maze of rules that make up commercial law may seem an inhospitable domain for moral imperatives, but the opposite is true, according to Brennan. It is for the commercial lawyer to discern the moral purpose behind each abstruse rule, and advise his client's conscience of what is just in the circumstances, not merely what he can legally get away with.

Of course, not all judges of the High Court agree with him. Ian Callinan perhaps does not -- which is one reason why Callinan is in a certain amount of trouble.

The accounting profession cannot set its own standards in splendid isolation from legal opinion, since it must operate in a legal environment not of its own choosing. This became most evident through the legal cases of the early 1990s, when victims of the large accounting firms' view of standards sued for huge sums, sometimes successfully. In one of the largest cases, Deloitte's faced damages of $340m for certifying Adsteam's 1990 accounts, which disclosed a profit of $236m when the true figure was a $244m loss.29 In other cases, the State of Victoria sued KPMG for $1bn over its audit of Tricontinental, and KPMG settled for a reported $136m,30 while Andersens was lucky to avoid trouble over Bond.31 A table as at October 1996 lists twenty-one such cases.32

The ability of law to have an impact on complex financial matters is well illustrated by the move away from the legal support for artificial tax schemes that the Barwick High Court provided. ``Literal'' interpretations of tax laws permitted endless schemes that subverted the intent of the law by appeal to its letter. Then the crucial section 15AA was added to the Acts Interpretation Act in 1981. It provided:
In the interpretation of a provision of an Act, a construction that would promote the purpose or object underlying the Act (whether that purpose or object is expressly stated in the Act or not) shall be preferred to a construction that would not promote that purpose or object.
In 1984, the revolutionary section 15AB was added, allowing recourse to external evidence such as Hansard as evidence of the legislators' purposes.33 Australia's tax avoidance industry, hitherto one of the most cunning in the world, never recovered from the blow.34

There are parallels in accountancy. The standards contain a ``substance over form requirement'' (``Transactions and events should be accounted for and presented in accordance with their financial reality and not merely their legal form''35 and there is a long-standing ``true and fair'' requirement.36 These notions are sometimes thought of by accountants as decorations surviving from the age of the dinosaurs,37 but gross departures from them are subject to legal scrutiny.

All these developments put paid to the notion that can sometimes be gained from both accountants and lawyers, that accountancy and law are very technical fields in which the common person's crude notions of the right, the true and the fair count for little. When the crunch comes, the courts prove to be as scandalised by technical evasions of what is fair as outside observers.

These considerations apply with minor changes in all major jurisdictions: commercial law is remarkably international. It is true that there are some countries where commercial law is a dead letter. But then, there is not much commerce to regulate in Somalia or Afghanistan, for reasons not unrelated to the breakdown of law in those areas.
Objection 7.
The law has a poor track record of actually doing anything to restrain capitalism, at least, restraining it from harming anyone but other capitalists.

Answer.
An interesting historical precedent that may encourage optimism is the gradual success of legal remedies in promoting industrial safety. Accidents and polluted work environments are one of the easiest means for a manufacturer to lessen costs, but the standard of workplace safety is high, and has been rising for at least a hundred and fifty years. Legislation has played a part, but the role of common law damages is far from negligible. Common law has been important both in setting pre-existing standards of liability by employers, and in providing the matrix of interpretation for subsequent legislation.38

More recently, and especially relevantly for a general project of internalising costs through legal means, has been the expansion of legal liability for economic loss. Until recently, the law was generally unwilling to recognise liability for purely economic loss, such as loss of income, suffered as a result of negligence (for example, in oil spills or incompetent audits). There has been strong movement away from this position in the last twenty years, and damages are now often recoverable in such cases.39 The law does not yet recognise a general duty of care in all cases of foreseeable harm, but the tendency to approach that limit is strong.40 In the meantime, the aggressive law firms who specialise in class actions are continually forcing the pace. The firm of Slater and Gordon, for example, have been involved in the Sydney water crisis, pollution from Ok Tedi, asbestos, breast implants, tobacco, peanut butter and the Christian brothers.

One should not forget either the legal regime that is gradually arising internationally from treaties, which has had some success in areas like protecting Antarctica. The speed of development of international e-commerce and Internet gambling means that something dramatic will happen concerning international laws governing finance. It remains to be seen what it will be.
Objection 8.
Attempts to find complex surrogate measures of environmental goods, subtle methods of valuing contingencies and so on are undemocratic. Policy matters about the environment and any other important matters ought to be judged by the citizens in the light of argument, not by alleged experts through measurement. Having experts in back rooms deciding such matters would replace public deliberation about ideas with a scientistic paternalism.41

Answer.
This style of argument will appeal to philosophers and Americans more than non-philosophers and non-Americans. Even American philosophers, though, are grateful to the Food and Drug Administration's experts who test their food and medicines. If rights are quantifiable, no amount of talk about democracy will make them otherwise, and expert opinion as to the quantity of the rights will be part of the information that ought to inform the political (or legal) process.

In any case, the question of what rights exist and can be measured is separate from public policy issues about how to cause them to make a difference.
Objection 9.
No-one is going to be found who is expert in these issues.

Answer.
Australia is crawling with accountants and lawyers: many with glowing testimonials from the best schools and universities in the country. Here is an opportunity for them to do something useful instead of wasting their lives devising tax minimisation schemes.