From Industrial Relations to Personal Relations:
The Coercion of Society

Social Engineering in the Market Place

Professor Ross Parish

An American sociologist, Peter Rossi, has enunciated what he calls "metallic laws" of the evaluation of social programmes; and his iron law of evaluation is that "the expected value of any net impact assessment of any large scale social programme is zero". In other words, no effect. The stainless steel law of evaluation is that "the better designed the impact assessment of a social programme, the more likely is the resulting estimated net impact to be zero"; and the brass law of evaluation is that "the more social programmes that are designed to change individuals, the more likely net impact of the programme will be zero".

(I might say that Rossi is not deterred by his negative conclusions, and remains a dedicated interventionist. He says 'the double message of this paper is an argument for further development of policy relevant to basic social science and the establishment of the new profession of social engineer!')

I think Rossi is a bit too pessimistic. I think social programmes do have effects; our two previous speakers certainly testified to that, and what I am going to say suggests that they often have effects which are different from what was intended. But, more fundamentally, it is often very difficult to determine what the effects of social policies are.

Evaluation is made difficult by the complexity and flux of the social world, and its failure to perform the sort of experiments needed. Journals are full of econometric studies of these things---for example, how many lives have been saved by some safety programme---but nobody really takes much notice of these any more. People are quite properly rather sceptical of econometrics.

The problem of evaluating programmes is compounded by the tendency of governments and their agencies to attack any problem on a broad front using several policies so it is difficult to disentangle the effects of any one of them from those of the others. I think that may very well be a calculated policy because it is a form of insurance. If none of the policies work, and it is obvious that none of them work, then that is a problem; but if one of them works they all get a bit of the credit. Therefore, no programme can be eliminated on the basis that it doesn't work. A good example is road safety where we have better roads, booze buses, red light and speed cameras, seat belts (and soon airbags), and frightening TAC advertisements; and nobody knows how much the lower road toll is due to any one of them and how much is simply due to the recession. There is good evidence in the United States that there is a very close correlation between the economic activity and road fatalities. In fact someone---an econometrician---has estimated that the economic factors account for 30-40% of the decline in the last three years in Victoria.

As a consequence of the difficulty of evaluating programmes ex post they tend to be judged by their intentions rather than their consequences. This links up with what Colin Howard said: if the intention makes you feel good, then it is a good programme. People seem to think that the effects of policies and programmes will reveal themselves over time. I remember years ago when Britain was considering entering the Common Market "The Times" editorialised to the effect that there are arguments both ways but we will only know the effects of entry five years after we enter. Of course we did not know then. We did not know what it would have been like if Britain hadn't joined the market. Evaluative, ex post, studies involve the comparison of factual and counter-factual situations, just as predictive studies do.

I think we need to distinguish between the ostensible and latent purposes of policies. For example, sex education in schools has presumably the ostensible purpose of reducing teenage pregnancies, but I would suggest has the latent purpose of breaking down sexual taboos among young people .

It would be interesting to develop a typology of social engineering interventions but as I haven't been able to do that I am going to deal with only one type: that is, interventions which seek to alter the rules for transacting so as to favour one of the parties in the transaction. Now the sort of transactions I have in mind are between producers and consumers, lenders and borrowers, landlords and tenants, employers and employees, husbands and wives---whatever you like. With the exception of the last, all of these transacting pairs consist of a buyer and a seller---and some may say that applies to husbands and wives as well. Interventions typically favour the buyers who, it is alleged, are in an inferior bargaining position vis-a-vis the seller, and are exploited or are victims and so on. The exception is the employer/employee relationship where intervention usually favours the employee (the seller of labour) rather than the buyer. Actually, labour contracts are more complicated than that---the employee sells his labour and the employer sells working conditions and promotion prospects, etc. Note that in all cases intervention is in favour of the more numerous group, something that one would expect in a democracy.

Intervention is usually based on a bargaining or a power model of the transaction relationship. Buyers are at a bargaining disadvantage vis-a-vis sellers---or so it is said---and hence the rules should be tilted in their favour. Standard contracts are criticised by radical lawyers as being unconscionable because they seem to favour the seller. I think there is a lot of confusion on the part of lawyers---market exchange doesn't take place in a sanitised legal environment. It is done in a cultural milieu where certain rules of the game are tacit, but understood and is based to a considerable extent on trust and on custom rather than on strict law.

Although there are areas of possible conflict between buyers and sellers, the over-whelmingly important thing about their relationship is its complementarity. Buyers needs sellers and vice versa. Sellers compete with one another. They don't compete with buyers. Buyers compete with one another. Voluntary exchange is mutually beneficial. Complementarity is often ignored, suppressed, or denied by the so-called reformers who see only the element of conflict in the relationship. For example, militant feminists deny the complementarity of men and women and Marxists see the interests of labour and capital as being antithetical.

Now let me discuss to some of the concrete examples of the sort of things I am talking about.

1.One sort of reform which has been popular in recent times is the denial to sellers, vendors, landlords of so-called self-help remedies against contractual default by buyers. These include the legality of a utility cutting off service to a customer who is not paying his bills; the repossession of vehicles under hire purchase agreements when the purchaser fails to keep up the payments, and the retention by landlords of security deposits to cover unpaid rent and property damage by tenants. Now the last two of those have certainly been done away with through hire purchase no longer being legal and the Residential Tenancy Legislation of some years ago has put the security deposits under the control of a government agency.
2. Another example is the inclusion of mandatory cooling-off periods in certain types of contract---used cars, houses at non-auction sales, and door-to-door sellers.
3. Mandatory warranties for used cars.
4. The substitution of caveat venditor (or seller beware) for the caveat emptor which used to be the dominant rule in liability law.

Now all of these reforms raise the sellers' costs. It is fairly cheap and easy to send somebody out to repossess a car but it is more costly if you have to do it through the courts, which is the only way it can be done today. If the producer or the seller is likely to be held liable for product-related accident costs, he will take out liability insurance or possibly self-insure and that adds to his costs. He will attempt to pass these costs on in the price of the product or service, and to the extent they are passed on buyers will pay (in whole or in part) for their new rights. That is the first major point and one which is often neglected by the advocates of this sort of thing because they have some notion that these costs can easily be absorbed and will be absorbed for some reason by the sellers.

You will notice that these policies are expressed in terms of a dichotomous or dualistic view of the world---buyer/seller, employer/employee and so on. But in undertaking further analysing this simple dichotomy becomes inadequate and we have to distinguish between different sorts of consumer/borrower, etc., and different types of producer, lender and so on. So let me now speak of "good" and "bad" members of each group, where "good" and "bad" describe the perception of the person or economic entity by the opposite side of the market. "Bad" customers are those that don't pay or are slow in paying their bills, or their credit repayments; "bad" tenants are those who wreck the property and don't pay the rent. They may be "bad" customers because they are dishonest, feckless, or are unfortunate. This category would also include the litigious and politically motivated or alienated persons who behave in a manner that most would describe as unreasonable. "Bad" sellers are the Arthur Daleys of the world, the shonky sellers who behave unethically and unlawfully; loan sharks and rack-renters who prey upon minorities who are unaware of their rights, or who are unable because of their own criminality to seek the protection of the law.

Of the four groups under discussion, only one, that is the "bad" buyers, will benefit unambiguously from the legal enhancement of buyers' rights or the curtailment of sellers' rights. "Good" sellers will be harmed to the extent that they are unable to pass on their increased costs to buyers and even if they can, they are likely to suffer from reduced sales.

"Good" buyers will be harmed by the rise in the price of the goods or service. Some will gain, admittedly, from the curtailment of the activities of "bad" sellers. But many will pay for a right which is of no use to them. It is a right which is only useful to the "bad" buyers.

"Bad" sellers will be harmed by being denied the use of questionable practices. However a change in the law will have little effect on those already acting extra-legally. Loan sharks' propensity to knee-cap under-performing borrowers is unlikely to be reduced by a law forbidding less drastic self-help remedies.

So the effects of legal innovation of this sort increasing the rights of consumers is likely to be perverse in that good consumers are harmed, bad ones benefited. Even the one presumably positive outcome, that is the deterrence of "bad" selling practices may in practice be of little importance since shady businessmen are least likely to be inhibited by legal considerations.

That is the first round effect. There are some further effects that are also likely to occur.

If sellers are denied certain remedies for contractual default, they may seek means of avoiding that problem. For example, they may be more selective in whom they accept as customers. They may demand references, check credit ratings, or they may simply discriminate on the basis of various stereotypes. To the extent that such screening is successful, the costs of sellers will be reduced, but to the extent that it is misdirected potentially satisfactory customers will be rejected. This sort of response is characteristic of rent control. In Canberra they had rent control briefly in the 1970s. There was a practice in Canberra for groups of three or four students to rent a suburban house and the landlord, knowing students' propensities, would charge them somewhat higher rent, but the students were happy as it was cheaper than living in colleges. It seemed to be a satisfactory arrangement all around. When rent control came in of course the landlords were forbidden to charge higher rents, discriminating against students, and the result was no more houses for students. Students protested about the lack of student housing but, of course, at the same time favoured rent control. Of course what the landlords wanted was a nice married couple with no children and both of them working.

Another consequence is that customers certainly have more incentive now to be "bad" customers, since they are more likely to get away with that sort of behaviour. If the gas company can't cut off the gas because you haven't paid the bill, they are likely to have more slow-paying customers and more bad debts.

One good effect is that sellers will have a reduced incentive to behave badly.

There will be less adherence to the customary commercial morality, that is the tacit rules of the game by customers, and that form of behaviour being replaced by the legalistic, even a litigious approach to market relations.

It would be a mistake to think that all of these interventions were the result of legislated changes. Most of the examples I have been talking about have been of that sort. But the Anglo-Saxon attachment to judge-made law means that changes in terms of contracts can and have been brought about by judicial interpretation. The shift from the doctrine of caveat emptor to caveat venditor in the United States has been, I believe, accomplished largely by judicial innovation. The shift in liability together with the award of huge damages by both juries and judges, has resulted in large increases in costs and prices of various products and services. A most perverse consequence of this has been that some products have simply disappeared. For example, I believe, that the manufacture of single engine aircraft in the United States has ceased because the liability insurance is so high that it is not worthwhile to produce them and, again in my talking to doctors, I have been given two examples of highly skilled doctors in the United States (one a brain surgeon and one a spinal expert) who took on hard cases, a lot of which were unsuccessful. As a result they were refused insurance and had to give up practice and restrict themselves to research.

The very high damages frequently awarded in American cases are often blamed on the ignorance and soft-heartedness of juries. However, I understand that in Victoria, doctors prefer to have their cases heard by juries because judges tend to give the most extravagant awards. It was suggested to me that judges want to do good, to feel good about helping poor people whereas juries are a bit more circumspect in their deliberations.

To sum up. In social policy, as in anything else, good intentions are not enough, but owing to the difficulty of prediction and evaluation, policies are often judged by their intentions alone. Evaluation is difficult because of the complexity of the social world, and by the propensity of governments to attack any "problem" using a variety of instruments, thus ensuring that their effects are confounded. A major deficiency of policy makers and advocates is neglect of the economic effects of their policies, that is, the way markets respond to and interact with particular intervention. Economic analysis can usually predict the nature of the effects of a policy, but is less successful in predicting their magnitudes.


Ross Parish, "Industrial Censorship". Quadrant, XXII:7 (July 1978) 12-17.
Ross Parish, "The Economic Effects of Victoria's Residential Tenancies Bill" in Robert Albon, (ed). Rent Control: Costs & Consequences (St Leonards: Centre for Independent Studies, 1980) 191-221
Ross Parish, "Consumer Protection & the Ideology of Consumer Protectionists", in AJ Duggan & LW Darvall (eds) Consumer Protection Law & Theory, (Sydney: Law Book Company, 1980) 229-243.
Peter Rossi, "The Iron Law of Evaluation and Other Metallic Rules", Research in Social Problems & Public Policy Vol 4 (1987) 3-20.

Why HR Nicholls?