Emeritus Professor, University of Manchester
Paper to be presented at the Symposium on Law Reform and Regulation, University of Warwick, 14 September 2011
1. What is “law-and-economics”?
Law-and-economics involves the application of economic theory and methodology to legal principles and institutions, in order to predict the behavioural response of individuals and firms to different legal forms – the positive dimension - and to evaluate the capacity of different legal forms to generate economically preferable (generally referred to as “allocatively efficient") outcomes – the normative dimension. The focus of most of the early law-and-economics literature was on private law. Nevertheless the University of Chicago, where the origin of law-and-economics is to be located, was also a prime mover in the critical evaluation of public law, in particular regulation. Economists there had remained hostile to Keynesian arguments for state intervention. They sought to show that regulatory structures were not conducive to efficient outcomes. The private interest theory of regulation which emerged seeks to explain how politicians and bureaucrats may be motivated to meet the demands of interest groups for regulation of a particular form. The basic idea is that regulation is a commodity made available in the political “market-place” and “supplied” by politicians and bureaucrats by reference to the demand of those who will benefit from its promulgation.
Such an approach proved to be valuable for lawyers who, perhaps naively, had tended to assume that government was predominantly well-intentioned in its approach to regulation, attempting to generate outcomes consistent with the public interest. In their view, if regulation failed, in general this was because it was insufficiently stringent or inadequately enforced. Private interest analysis was an alternative and tool for understanding regulatory failure. But there was a danger of oversimplifying the debate in the light of such analysis. The policy solution could not always be a crude abolition of interventionist measures since in many areas there existed a significant degree of market failure. The question then became one of investigating whether the measures could be adapted to meet the failures at lower cost - in short, whether a more sophisticated analysis could lead to more efficient solutions.
To meet this challenge, mainstream law-and-economics, which hitherto had applied standard price theory, transaction-cost analysis and organisation theory primarily to private law institutions and principles, developed a new branch of public interest analysis, sometimes referred to as “progressive law and economics” The new approach involved a three-stage inquiry into the market for particular products or services: first, there was a need to identify and explain instances of market failure; secondly, alternative methods of correcting the failure had to be investigated; thirdly, the predicted response of actors to the different methods had to be assessed, with particular attention to the minimisation of administrative costs, notably information and enforcement costs. Such analysis can lead to the selection of optimal regulatory forms, that is regulatory principles and institutions which can meet the goals of regulation at lowest cost; it is the focus of this paper.
2. Justifying Regulation
Regulation deals with a huge variety of industrial and non-industrial activities and involves a multitude of different legal forms. The choice of an appropriate regulatory instrument must, in the first instance, depend on the justification for the intervention. Once the justification/s has/have been identified, the policy-maker must then consider the advantages and disadvantages, primarily in terms of costs, of different legal instruments. In the light of those considerations, the task is to select the instrument which can meet the regulatory objective resulting from the justification(s) at lowest cost.
Logically regulation should only exist where the unregulated market will fail to reach the desired outcomes. Let us begin with the classic instances of market-failure. The first category comprises monopolies or other significant impediments to a competitive market and regulation deemed to necessary to correct this kind of failure is generally referred to as “economic regulation”. The other categories which give rise to “social regulation” centre on (1) inadequate or asymmetrical information affecting the relationship between suppliers and consumers or (2) externalities (spillover effects) whereby activities affect third parties in ways not reflected in the prices set by producers. To these can be added (3) co-ordination problems where, though desired outcomes can in principle be achieved by private transactions, the costs of co-ordination are so high that it is cheaper for the law to prescribe appropriate conduct.
If these are the key instances of market failure, phenomena which hinder wealth-maximisation within a society, we should be aware that governments may wish to regulate for other, non-economic, reasons. These can be very varied but two stand out as being particularly important. First, the unregulated market may lead to outcomes which do not accord with what is perceived by government to be a just or fair distribution of resources. They may wish, then, for regulation to intervene on grounds of distributional justice. Secondly, governments may believe that, in particular markets or instances. individuals are assumed not to be good judges of, or are not trusted to act in accordance with, what is in their own best interest. Such regulation may be referred to as paternalist. Note that whereas the classic instances of market failure can, in principle and on the assumption of adequate data, be assessed objectively, the non-economic justifications for regulation are crucially linked to ideological and political values.
3. Selecting Regulatory Forms: Traditional Instruments
a) Efficiency and costs
There are many regulatory instruments and no consensus on how they should be classified. To illustrate the law-and-economics approach, we can take a brief look at some of the main forms used for social regulation and see how their appropriateness can be judged by reference to the goal of allocative efficiency. More specifically, the aim is to minimise three sets of costs. In relation to a particular market for goods or services, these are: the costs arising from the unregulated market; the costs to industry of complying with the regulation; and what may be called “tertiary costs”, mainly the administrative costs arising from the regulatory intervention. Minimising the first two of these sets involves balancing the benefits arising from the intervention, in terms of reduction of social losses, against the primary cost, that to industry in preventing or abating the harm, as required. That gives us the notion of optimal harm reduction.
For regulatory policy-making purposes, and from a law-and-economics perspective, minimising the third set of costs is the most challenging and perhaps the most interesting task. What are to be included in tertiary or administrative costs? Clearly account should be taken of the cost of formulating and implementing a particular instrument, and that includes also the acquisition of information necessary to do this effectively. Then there is the expenditure incurred in monitoring behaviour to ensure compliance and, if necessary, engaging in formal enforcement procedures. There are also some less obvious, but no less important, consequences which may be relevant to the selection process. First, different instruments have different incentives for technological development, for example for industry to evolve new and cheaper techniques for risk-abatement. Second, as private interest theory has been so quick to reveal, some instruments impede competition and thereby create significant welfare losses. Finally, instruments vary in the extent to which they can be well targeted on particular phenomena and regulatory goals, thereby potentially creating what may be described as error costs. And so to a survey of the traditional forms of social regulation.
The most interventionist approach involves a system of prior approval. This prescribes that firms, before lawfully engaging in an activity or supplying a product or service, must first obtain a licence or permit from an authorising agency; and for such approval they have to satisfy the agency that certain conditions of quality meet the regulatory goal. The administrative costs of scrutinising all applications is very high and to these must be added the opportunity costs arising from any delay before the licence is granted. Moreover, significant welfare losses arise if the system is used, within the industry, for the anti-competitive purpose of creating barriers to entry. The benefits from prior scrutiny must therefore be very large to justify, on public interest grounds, these substantial costs. Such benefits may include the weeding out ex ante of notoriously unfit suppliers and, as we shall see, when discussing the licensing of taxi services, the ability to enforce regulatory compliance by the severe penalties of suspension or abrogation of licences.
The mandatory standards technique allows the activity to take place without any ex ante control but the supplier who fails to meet certain standards of quality will be subject to penal or administrative sanctions. Standards can be subdivided into: performance (or output) standards requiring certain conditions of quality to be met at the point of supply, but leaving the supplier free to choose how to meet those conditions; and specification (or input) standards compelling the supplier to employ certain production methods or materials; or prohibiting the use of certain production methods of materials. The most important economic variables in choosing between these types of standards are the costs of being informed on the technological means of achieving the regulatory goals, and the administrative costs of formulating appropriate standards and monitoring compliance. In principle, firms should be given choice as to how to meet the goals, since that encourages innovation in risk abatement. Hence there is a presumption in favour of less interventionist measures. However, the benefits of such measures might be outweighed by the costs of administering them and/or the costs to firms of acquiring information on the appropriate technology.
d) Information disclosure
Rather than impose standards on suppliers, forcing them to adopt optimal loss abatement, legislation may simply require that they disclose to purchasers and others information regarding harms or risks which may arise from the activity or product. If regulation forces suppliers to reveal adequate information as to quality/safety, on the basis of which consumers can exercise choice, market transactions will ensure that preferences are met; and there will be no welfare losses from consumers being deprived of choice, as can occur under a standards regime. Moreover, mandatory disclosure will reduce costs where the consumer is the least cost-abater, notably by responding to published warnings. The administrative costs of formulating and enforcing disclosure rules are also relatively low, given in particular that policy-makers do not themselves have to determine optimal levels of loss abatement.
On the other hand the potential application of this technique is limited since not all those affected by the product or activity will receive the information and be able thereby to adapt their behaviour. Moreover, even within the narrower group of purchasers who can use the information there may be problems. It may be impossible to summarise the necessary information in a form which the great majority will read and understand. The “bounded rationality” of individuals may constitute a further obstacle: there is evidence that individuals tend to overestimate risks associated with low-probability events and underestimate those arising from higher-probability events. Given the often significant costs to purchasers of assimilating information and making decisions, it may be cheaper to force suppliers to adjust the product or service to what purchasers would presumptively have chosen if those intellectual processes had been completed. This solution may be particularly apposite where the costs arising from consumer error are high, for example where death or serious personal injury may result.
4. Selecting Regulatory Forms: Newer Instruments
In the last two decades, there has been much discussion, and some exploration of alternatives to traditional social regulation. Economic analysis has played a significant role in the debate. Technological advance has perhaps led to the most important criticism of traditional regulatory forms which attach directly to the activity of firms. If such activity-based intervention is to be successful, the agency must be able to understand and evaluate latest developments. One consequence of a heavily prescriptive regime is that the regulatees have little or no incentive to develop or discover cheaper means of meeting the regulatory goal. Another is that, given the diversity of activities and techniques which must be controlled, traditional “rule-books” become very detailed and bulky; and hence give rise to heavy administrative costs.
Economic reasoning suggests that regulation is likely to be improved if decisions relating to activities and products can be taken closer to their production than is usually possible with command-and-control regulation and if regulatory mechanisms can be harnessed to market-based incentive structures. Here we examine two regulatory forms designed to meet these goals: fiscal instruments and co-regulation.
a) Fiscal instruments
The idea of correcting or internalising negative externalities by the imposition of financial charges or taxes on the responsible individual or firm has gained much currency in the recent debate. Advocates of regulatory taxes claim for them the principal advantages that they reduce information and administrative costs because, to secure an optimal level of abatement, the regulator need have no knowledge of a firm’s compliance costs. Also taxes create incentives for technological innovation, since the firm is free to explore different ways of minimising the financial charges.
On the other hand, the idealised tax systems envisaged in the economic models give rise to a number of difficulties. First, to set the optimal tax, regulators must have adequate information on the total social costs generated by the activity. While they may proceed by estimates, adapting tax rates on an iterative basis as the effect of the instrument becomes known, this generates uncertainty and thus increases planning costs. Second, the collection of financial payments involves substantial administrative costs. Third, while the regulatory objective will be met if the activity of the firm is price-elastic and thus behaviour changes in response to the charge, inelasticity will hinder this outcome, because the firm’s preference will be for paying the charge. Since governments benefit from greater revenue, officials may prefer the latter, thus distorting the regulatory objective.
For the purposes of traditional regulation a public agency formulates the rules and enforces them against industry. Under a system of self-regulation, the state delegates part or all of the rule-formulating and enforcement functions to agencies representing the regulated industry. In terms of private interest theory, this might seem to imply “capture” in its most complete form, but there are, nevertheless, some powerful public interest arguments for self-regulation. In particular: self-regulating agencies normally have greater technical expertise than public agencies - the costs of formulating and interpreting standards tend therefore to be lower; interaction between self-regulating agencies and the firms should involve mutual trust - thus there should also be a lowering of monitoring and enforcement costs; and self-regulating systems should involve less formalism, thus involving savings in administrative costs.
The challenge is then to devise institutional arrangements which retain the advantages of low-cost rule formulation and enforcement but involve some degree of public accountability, to constrain private interest influence. In some areas “co-regulation” has emerged: self-regulatory agencies, or firms themselves, issue, and sometimes enforce, rules but with a degree of oversight from, or participation by, public agencies.
5. The Regulation of Taxis and Private Hire Vehicles
“The taxi trade should be a model of textbook economics. There are lots of sellers (drivers), lots of buyers (passengers) and low barriers to entry (the price of care). It isn’t. Throughout the world the trade is distorted - by government rules, monopoly, political lobbies, mafias, racial exclusiveness and every other sin in the free marketeer’s book.” (The Economist, 22 December 1990)
a) Justifications for regulation
Taxicab regulation provides an excellent example of the law-and-economics approach to regulation. There are significant variations between regimes in different jurisdictions and allegations that many of them are ill-targeted and/or excessively restrictive on the supply of services (OFT, 2003). The relevant market failures justifying regulation are, however, not difficult to identify.
The main problem is that of information. Customers hailing a vehicle in the street almost certainly have insufficient information about the quality and reliability of service being offered, and – in the absence of regulated fare tariffs – the same will be true of prices. Note that these failures are less likely in relation to services which are the subject of pre-booking, since then the customer has the opportunity to acquire some of the relevant information and, indeed, to compare what is being offered by different suppliers. Thus we would expect different regulatory regimes as between taxis and private hire vehicles which cannot be hailed in the streets.
Externalities are also likely to be present. The poor quality of services supplied can affect third parties, that is individuals other than those hiring and paying for the service. Also taxi services can significantly add to problems of congestion and give rise to other adverse environmental effects. Of course, these arguments apply to all road users, but regulation designed to deal with other problems may take account of such external costs. Note too that potentially there may be external benefits, as well as external costs, since taxi services may contribute to, or be a partial substitute for, public transport, and hence play a part in the local infrastructure. Reference is also sometimes made to the fact that taxi services are insufficiently competitive. However, this is invariably a consequence of regulation rather than of market conditions. Let us now examine the regulatory forms typically used to correct the acknowledged market failures.
Regulatory systems invariably require that both taxicab vehicles and their drivers are licensed. It follows that many of the standards involved in vehicle and driver quality controls – see below - have to be verified and satisfied prior to any lawful supply of the service. The administrative costs of such an ex ante scrutiny system, combined with those of monitoring ex post, are very high. As regulation theorists have been telling us for some time, there is also the risk that standards imposed under such a system are used to limit the number of suppliers, even though these standards are ostensibly designed only to control quality (Ogus and Zhang, 2005).
What then are the justifications for the use of this method of quality regulation? There is, of course, the familiar assertion, considered above, that ex ante measures serve to prevent harm whereas ex post measures can only hope to deter it; and notoriously unsuitable vehicles and drivers can be weeded out at an early stage. But the most powerful argument for the licensing technique is probably that based on enforcement considerations (Gallick and Sisk,1987). The capital invested and acquired in licence plates by licence-holders operates as a bond which the latter will forfeit to the authority should they, or the vehicle, fail to comply with the quality standards imposed. Reliance on the ex post infliction of financial penalties in an unlicensed regime may generate insufficient incentives for compliance, especially where the probability of apprehension is relatively small and the resources available to pay the penalties are limited.
c) Quantity controls
Limiting the number of taxicabs permitted to ply for trade is the most direct form of entry control. It has been a feature of most regimes and still operates in many jurisdictions. Given the adverse effects generally attributed to quantitative controls, limiting the availability and variety of services, and enabling supra-competitive profits to be earned, strong public interest arguments are necessary to justify them. “Excessive competition”, by itself, has little meaning and if the concern is rather with possible deteriorations in the quality of the service, these can be addressed by means other than quantitative controls (OFT, 2003). The same applies to important externalities such as road congestion and pollution.
Where regulation has imposed quantitative limits, it has often conferred considerable discretion on the authority responsible for issuing licences. And such power is obviously open to abuse, encouraging corruption or at least capture by rent-seeking individuals.
There are some quantitative measures which seem to proceed on the assumption that the supply of taxis in a free market will not match demand, although it is difficult to find support from economists for the proposition. In any event, why should regulators be able to assess supply and demand better than those exercising the particular trade? Take, for example, peak demand periods. Regulating the quantity of supply to meet this problem is almost impossible, while in an unregulated market vehicles which have other functions during off-peak periods can enter to meet the demand, the result being reduced costs and prices (Australian Trade Commission, 1999).
d) Price controls
Where road space is very limited and congestion is a major problem, price controls can be used to regulate the demand for taxis relative to that for other forms of transport (Yang et al, 2000). Indirectly this amounts to a pricing of road use, justifiable by reference to externalities. But there are several more traditional economic arguments justifying price controls (Frankena and Pautler, 1986). First, demand for the services is inelastic since a customer hailing a cruising cab will not be able, at low cost, to compare the price offered with that of an alternative supplier; and the same applies at a taxi stand insofar as there is in operation there a “first-in; first-out” allocation scheme. In addition, the costs of bargaining a fare may be unduly high, or such as to enable the supplier to exploit the customer.
Of course, these arguments do not apply where cabs are hired by telephone or otherwise in advance, hence the separate regime for private hire vehicles. Further, it may be possible to dismantle allocation practices at stands enabling customers to exercise freedom of choice in that context (Seibert, 2006). Deregulation of price controls may then be possible, provide that customers have sufficient information as to prevailing tariffs before entering a particular vehicle, and this can be achieved by a “price-posting” regime.
For vehicles hired by hailing in the street, price controls are generally desirable although, to encourage some degree of competition, these can be in the form of price-caps, rather than a mandatory tariff (OFT, 2003).
e) Quality controls
Customers hiring a taxi will normally be insufficiently informed on the safety and quality of particular vehicles and their drivers. To some extent, the problem may be alleviated where firms supplying in the market are able to develop a reputation for the quality of their service. But since it is only consumers able to identify and select cabs operated by the firm who will be able to rely on this reputation, the argument does not apply to large areas of the market. It is therefore widely accepted that safety and quality regulation is necessary. The more difficult questions are how extensive the regulation should be and what forms it should take.
Take first obligations on drivers. Uncontroversially they must have the relevant driving skills, competence in the relevant language and specialist knowledge of the area where they seek custom. But some regimes go clearly beyond what can be justified by information asymmetry, for example the driver’s dress. Also if the conditions to be fulfilled by the licence-holder are vague, such as a record of “good conduct”, there is always the risk that they can be used to restrict entry on arbitrary grounds.
Quality conditions applying to the taxicab are subject to similar considerations. The consumer information problem can certainly justify regulations concerning the installation and maintenance of meters as well as the means of identifying the cab and its driver. So obviously when they govern the safety of the vehicle. However, over-rigid specification standards can give rise to unnecessary costs which must then be passed on to the consumer (OFT, 2003); and what are we to make of regulations which, as in London and some other British cities, require that taxis conform to a certain design and appearance? No doubt they may be more easily identified and many customers may be reassured to be conveyed in the traditional format, but if cheaper designs meet their needs just as well, why should the customers pay for the increased cost?
Australian Trade Commission.1999, “Regulation of the Taxi Industry”, Commission Research Paper.
Frankena, M.W. and Pautler, P.A. 1986, “Taxicab Regulation: An Economic Analysis”, Research in Law and Economics, 9: 129-165.
Gallick, E.G. and Sisk, D.E. 1987, “A Reconsideration of Taxi Regulation”, Journal of Law, Economics and Organization, 3: 117-128.
Office of Fair Trading. 2003. “The Regulation of Licensed Taxi and PHV Services in the UK”, OFT 676.
Ogus, A and Zhang, Q. 2005. “Licensing East and West”, International Review of Law and Economics, 25: 124-142.
Seibert, C. 2006. “Taxi Deregulation and Transactions Costs”, Economic Affairs, 26: 71-73.
Yang, H., Lau, Y.W., Wong, S.C. and Lo, H.K. 2000, “A Macroscopic Taxi Model for Passenger Demand, Taxi Utilization and Level of Services”, Transportation, 27: 317-340.