EVALUATION OF LEGISLATION

Proceedings of the Council of Europe’s legal co-operation and assistance activities (2000-2001)

(...)

Estimating the cost of legislation

Professor St. John BATES
University of Strathclyde, United Kingdom

1. Introduction

Let me begin by telling you two allegedly true stories from the United Kingdom – one about benches in parks and the other involving a state financial benefit for those with insufficient resources to rent a home22. These are stories that could be replicated in every jurisdiction.

In the 1990s, there was a restructuring of the state- provided National Health Service. Under the new arrangements, regulations provided that local family doctors would be given budgets and take more direct financial responsibility for their patients, as sub-contractors of the National Health Service.

To operate these financial arrangements, each patient required a postal code (the code that is placed on letters to facilitate their delivery). For most patients this did not create a difficulty; they could provide the postal code of their residence. Unfortunately, there are patients who have no home, and they are commonly in need of medical services. Some of the homeless sleep in hostels in the winter, but on benches in parks in the summer.

To comply with the regulatory regime, it became necessary for some local authorities to assign post codes to park benches so that the homeless could get medical treatment.

The Post Office was not amused. A spokesman said: “This sounds like a way to get around bureaucracy, but we will certainly never deliver letters to a park bench”.

In 1983, the UK Government introduced housing benefit. In March 2001, the National Housing Federation published a very critical report on the operation of the housing benefit scheme. The Federation represents housing associations that build accommodation. This accommodation is commonly then rented at low rates to those with limited financial resources. These people often qualify for state financial assistance in the form of housing benefit. Consequently, a significant proportion of the income of housing associations is derived from housing benefit. The Report of the Federation, Housing Benefit: No Cause to Celebrate, stated that 1.5 million of the 4 million housing benefit claimants in Britain were owed £1.6 billion in delayed payments by local authorities on behalf of the state; and housing associations were owed £84 million of that figure by their tenants. There is a considerable amount of fraud by those who claim housing benefit, and there have been frequent and numerous changes to the regulations governing the benefit, largely to combat fraud. In 1999, for instance, there were 85 changes to the regulations, with local authorities given little notice of the introduction of the changes.

This has led to severe delays in the payment of benefits. The result has been economic hardship for individual claimants, and also for organisations, such as housing associations, which seek to provide accommodation for those in financial need.

2. Preliminary questions
2.1. The cost of what legislation?

The answer is all normative texts. It should apply to both primary legislation and to secondary delegated legislation, such as regulations. However, it should also apply to less formal normative acts, such as codes of practice and guidance notes which have some, perhaps limited, legal effect.

2.2. The cost to whom?

This is a question which is easily answered – the cost is to the private sector and also the public sector – and if there is an increased cost to the public sector, there is inevitably a cost to the private sector in additional taxation.

2.3. The cost of what?

There are many costs of legislation.

There may be cost to the public sector. Increased bureaucracy will have staffing and, therefore, financial implications. There may be other direct public sector costs; for example, if the legislation provides for grants for housing or small businesses.

There may be cost to the private sector. Legislation that imposes taxation, or fees, is an obvious direct cost. Extensive regulatory or compliance provisions is also a cost, because it requires private sector manpower which is ultimately paid for by the customer; if regulatory or compliance provisions impose excessive demands, they will finally be a burden on the taxpayer (because less profit will mean less tax revenue) and the community (because less tax revenue will eventually mean fewer or less extensive public services).

Commonly, there will be a cost to both public and private sectors. So, extensive regulatory provisions will not only be a cost to the private sector, but also to the public sector in additional bureaucracy. The same may be true of legislation that has an environmental impact. It will also be true of inadequately considered, or badly drafted, legislation. Such legislation may not achieve its objective, or may achieve it expensively, or may lead to expensive litigation to resolve textual ambiguities.

2.4. Is the cost justified?

This is often the most difficult question.

For instance, suppose there is a policy decision that banks should be more effectively regulated and licensed. The regulatory regime requires to be extensive to be effective. An extensive regulatory regime will be expensive for the public sector. It will require a large bureaucracy to undertake the investigations necessary; its personnel will have to be highly trained. The banks will have considerable costs is responding to the licensing regime, also needing highly trained staff. The bank costs will be passed to the customers and there may be a loss of business, with a reduction of tax revenue. However, a less rigorous regulatory regime may result increased dangers of money laundering and a lack of international confidence in the banking system. This, in turn, may lead to a reduction in the international economic and political support for the state. The increased cost of an extensive legislative scheme for the regulation and licensing of banks may thus be justified.

On the other hand, the considerations and outcomes might be different were the policy decision taken that doctors or, say, bars, were to be more effectively regulated and licensed.

3. Response; General formal systems

The circumstances which lead to legislative outcomes such as those related to park benches and housing benefit, and the questions they raise, are now addressed by formal systems within Government and Parliament in many jurisdictions. Similar initiatives have been taken by the EU and OECD23.

3.1. United States

Cost-benefit analysis of the exercise of administrative and regulatory functions of US federal agencies can be traced back to the 1900s. In more recent times, US federal agencies in exercising their legislative competence are commonly required by statute to balance competing interests to determine the most efficient manner of achieving their statutory objectives. Where there is such a statutory requirement to conduct a cost-benefit analysis, Executive Order 12291, passed in 1980, laid down general requirements for regulatory impact analysis. The principal requirements are:

(1) a description of the potential benefits of the rule, including any beneficial effects that cannot be quantified in money terms, and the identification of those industries or individuals likely to receive the benefits;
(2) a description of the potential costs of the rule, including any adverse effects that cannot be quantified in monetary terms, and the identification of those likely to bear the costs;
(3) a determination of the potential net benefits of the rule, including an evaluation of effects that cannot be quantified in monetary terms;
(4) a description of alternative approaches that could substantially achieve the same regulatory goal at lower cost, together with an analysis of the potential benefits and costs of the alternatives and a brief explanation of the legal reasons why such alternatives, if proposed could not be adopted; and
(5) unless covered by the description required under paragraph (4) above, an explanation of any legal reasons why the rule cannot be based on the [cost-benefit] requirements…24
3.2. Germany

The German Federal Government published a decision in December 1984 which declared that federal legislation should be examined by government departments to ensure that it was necessary, likely to be effective and would be comprehensible. It requires a series of detailed questions to be posed when drafting federal legislation. These questions are grouped under ten broad questions:

(1) Is action necessary?
(2) What are the alternatives?
(3) Is action required at federal level?
(4) Is a new law needed?
(5) Is immediate action required?
(6) Does the scope of the provision need to be as wide as proposed?
(7) Can the length of the period for which it is to remain in force be limited?
(8) Is the provision non-bureaucratic and understandable?
(9) Is the provision practicable?
(10) Is the cost-benefit analysis of the legislation acceptable?25

3.3. United Kingdom

For many years draft primary legislation in the UK was published with an attached Explanatory and Financial Memorandum. This gave a rather minimalist statement of the financial implications of the legislation, in terms of public expenditure, although like most jurisdictions a rather more sophisticated analysis of the financial effects on the public sector and, to some extent the private sector, was undertaken within Government. The fact that this was not transparent did not encourage a rigorous approach, except where the Treasury (the finance ministry) was active.

When the Labour Government came to power in 1997, it was amenable to the considerable demand to improve parliamentary procedures and scrutiny, and was also keen to maintain a prudent financial policy (which had not been the public perception of many Labour Governments previously). One consequence of these factors, following the Second Report of the Select Committee on the Modernisation of the House of Commons26, was a change in the supporting documentation which is published with draft primary legislation. This is now usually published separately as Explanatory Notes. These Notes contain background material and a commentary on the provisions of the draft legislation (which used to be published separately with a restricted circulation as Notes on Clauses), and also a statement of the financial effects of the Bill, its public service staffing effects, a regulatory impact assessment (or a summary of the assessment published separately), an environmental impact statement where appropriate and a declaration from the parliamentarian (often a Government Minister) promoting the legislation, that it complies with the European Convention on Human Rights. This supporting documentation may be extensive and, therefore, expensive to buy27, but it is now available on the parliamentary website28.

4. Specific responses
4.1. The drafting process

The process of drafting normative texts may affect the cost of legislation.

First, the drafter is subject to the wishes of those who instruct the legislation. Those instructions may not have taken sufficient account of other matters referred to below. They may not actually provide for achieving the objective or they may propose a relatively expensive administrative solution of the issue to be addressed by the legislation. In practice, the drafter may only have limited scope or competence to raise such issues, but should be prepared to do so.

Secondly, the drafter may choose an expensive solution. Let me give you an example from a recent training course which I conducted, where the trainees were simply instructed to draft a statutory offence of making unnecessary noise outside hospitals. One of the issues which they had to address was how to establish what was unnecessary noise. Some chose to determine “unnecessary” noise by reference to the decibel level. This was not a good solution. It would require machines to register decibels at various points in every hospital, operators to manage the machines and prove that the level of decibels recorded related to a particular incident or individual. It would be expensive, bureaucratic, and unlikely to achieve many convictions or contribute significantly to the reduction of noise outside hospitals. As an English High Court judge famously and rightly observed, as well as being unfair law, “”bad law wastes money”29.

Thirdly, the administrative structure of drafting may prove to be expensive in a variety of ways. Historically, most Governments and Parliaments have come to employ lawyers to undertake legislative drafting rather than rely on lawyers in private practice. In the United Kingdom, the Government appointed lawyers specifically for drafting primary legislation from the middle of the nineteenth century30. However, there has often been too few lawyers appointed for the amount of legislation which required to be drafted, although the situation has been significantly ameliorated in the past two years. However, as recently as 1996, there was an experiment of contracting out the drafting of some provisions of the Finance Bill to various lawyers in private practice31. The experiment was expensive and it resulted in a series of provisions drafted technically in different ways which was likely to lead to ambiguity. It has not been repeated.

Unlike many jurisdictions, primary legislation is drafted centrally in the United Kingdom and not in the individual Government Departments promoting the legislation. Many feel that this leads to a consistency of style and thus less risk of ambiguity and expensive litigation. Delegated legislation (for example, regulations) is, however, almost invariably drafted by lawyers employed in individual Departments. Despite stylebooks for drafting delegated legislation, the result is less consistency of style.

By contrast, the UK Parliament has historically not employed lawyers for the primary purpose of assisting parliamentarians to draft legislation or amendments to legislation32, although recently there has been a substantial restructuring of the parliamentary legal service which may conceivably result in drafting services being provided to parliamentarians by lawyers qualified to do so33. Other jurisdictions, such as Canada, at least at the federal level, have distinct separate services within the Government and the Parliament.

4.2. Calculating the direct financial effect of legislation

In many jurisdictions the calculation of the direct financial effect of legislation has become increasingly transparent, and this is a good discipline.

Two examples of the published calculation of the cost of draft primary legislation will serve to illustrate some of the strengths and weaknesses of such calculations.

The Social Security Frauds Bill was introduced into the second chamber of the UK Parliament, the House of Lords, on 18 December 2000. Explanatory Notes on the Bill were published at the same time, and they include an analysis of the financial effects of the Bill34. The principal purpose of the legislation is to reduce fraud and error in the means-tested state benefits system. The Notes indicate that the Bill is expected to result in gross savings of £200-400 million per annum. However, as the legislation empowers the relevant Government Department, the Department of Social Security, to obtain additional information about employees from businesses, there will be a cost to business of £2.5-7.6 million per annum. There will be a cost to the Government of processing the information of £3.65 million (with a set-up cost of £0.75 million). In addition, local authorities have the responsibility of administrating two major benefits (Housing Benefit and Local Authority Benefit) and there will be an estimated additional cost for them of £1.65 million. However, the Notes declare that the legislation is “not expected to have any significant effect on public sector manpower”.

Although there has obviously been a certain degree of work put into providing this analysis for the UK Parliament, there are a number features to note. First, the figures are expressed in price terms for the 1999/2000 financial year, although the legislation is unlikely to be brought into force until the 2001/2002 financial year and possibly later. Secondly, the figures are expressed as approximate figures. Thirdly, the figures that are presented are sometimes in broad bands; so the minimum projected gross savings is half the maximum figure, and the cost to businesses has a 300% variable. Finally, given the figures involved, it could only be suggested that there would not be a significant effect on public service manpower because there would only be a marginal increase in the very large manpower of the Department of Social Security. However, adopting a norm of discounting marginal cost is not an effective strategy for calculating costs of legislation or anything else.

The Football (Offences and Disorder) Act 1999 came into force in September 1999. This was yet more UK legislation seeking to control football hooliganism and related offences in England and Wales, and amongst followers of English and Welsh teams playing abroad. It extended the statutory powers to make exclusion and restriction orders, and it also broadened the scope of certain offences inside and outside football grounds. The Explanatory Notes on the draft legislation considered its financial and manpower effects35. The court determines whether such an order is made when a person is convicted of a relevant football-related offence, so there was “expected to be some increased costs for the courts and the Crown Prosecution Service, these are expected to be minimal”. As the legislation increased the requirements for persons against whom orders are made to report to the police, that was calculated to “result in some additional costs but these are expected to be minimal”. Finally, it was estimated that there would be an additional cost of £100,000 per annum for the Restrictions Order Authority, the administrative body which administers the reporting scheme, and it would have to employ 5 additional staff.

Again, a number of comments may be made about this analysis. First, there is again the danger of discounting marginal cost. Secondly, the analysis is obviously incomplete; it takes account of prosecution costs, but fails to take account of defence costs which would almost certainly by supported by public funds.

Three general points may also be made.

First, this analysis is primarily for Parliament. It presumably tends to minimise costs for political reasons.

Secondly, if the calculations were more sophisticated they would be more expensive. This might be justified in terms of legislation such as the Social Security Bill, but could hardly be justified for legislation of the nature of the Football Bill.

Thirdly, the value of the exercise is perhaps reduced because Parliament has really no resources or procedures to check the accuracy of the figures at the time they are published, or to consider later whether they proved to be an accurate forecast. Again, the economic arguments for establishing such machinery may not always be strong. There is, after all, a danger in investing more in the procedure than in the product.

4.3. Calculating the indirect financial effect of legislation: Regulatory impact assessment

In many jurisdictions, the private sector will complain that it is over-regulated while Government will respond that society is increasingly complex and citizens demand greater legal rights and protection. For both the private and public sectors greater regulation almost invariably means more expense. The difference between the two sectors is that the public sector is in rather greater control of the amount and expense of regulation, whereas the private sector has regulation and its expense largely imposed on it. Even where regulation is to some degree imposed on the public sector, perhaps as a consequence of membership of a super-national body, such as the European Union, or of an international body, such as the World Trade Organisation, the regulation is still likely also to be imposed on the private sector.

This tension between the private and public sectors has led to the development of techniques to measure the financial implications of imposing regulations. In the United Kingdom, this is described as regulatory impact assessment.

In a Guide, the Cabinet Office in the UK Government has defined regulatory impact assessment as “a policy tool which assesses the impact, in terms of costs, benefits and risks of any proposed regulation which could affect businesses, charities or the voluntary sector”36. It is now UK government policy that all departments and agencies where they exercise statutory powers, and make rules with a general effect on others, should produce a regulatory impact assessment.

Regulatory Impact Assessment (RIA) is conducted in three stages: initial RIA; partial RIA (which must be undertaken before there is public consultation on the normative text) and full RIA.

Initial RIA is a general overall analysis based on existing easily available information. At this and subsequent stages it is necessary to consider options that do not involve regulation. So, for instance it may be possible to: (i) rely on consumer choice, competition or innovation; (ii) improve advice or information; (iii) use a code of practice; (iv) use economic instruments, such as user charges, taxes or tax concessions; (v) invite the industry to regulate itself; (vi) simplify or better target existing regulations by a deregulation order (considered further below); (vii) introduce a time limit on the regulation (“sunset clause”) or make it subject to automatic review.

At this stage British civil servants are also required to begin to consider the social impact of the options: will each option impose unfair costs on a specific or vulnerable group such as the elderly? will it create a new vulnerable group?

Partial RIA involves developing work on each of the regulatory and non-regulatory options, with respect to their risks, benefits, costs and compliance. At this stage the policy maker is required to work through a check list involving other areas such as: sustainable development, environmental appraisal, equal treatment appraisal (e.g. with respect to gender, race, age, disability), and health impact assessment, including health and safety at work. Often the assessment of benefits is not satisfactory because there is an assumption that the preferred option has the greatest benefits. Identifying the costs involves making assumptions for each group affected; so, for instance, some costs may have a disproportionate effect for small businesses. Policy and implementation costs will have to be considered separately. Implementation costs are, in general terms, the bureaucratic costs: legal costs; staff time involved in understanding the new regulations; inspection and licence fees and so on. Policy costs are those costs directly attributable to the policy objective: wage costs increase as a result of minimum wage regulations; loss of revenue associated with a ban on certain products (eg. selling beef on the bone). RIA also depends heavily on quantifying all factors, but some are not easily quantified. Some benefits are difficult to quantify in monetary terms, for example reduction in noise pollution or the time saved by a passenger using different modes of transport, but there are technical means of doing so. Finally, compliance should involve a graduated response, with the creation of criminal offences being the last sanction to be incorporated.

Example of Partial RIA37

The Issue: Scrap metal dealers are regulated by statute (Scrap Metal Dealers Act 1964) but the motor salvage industry is not.

There are 2,500-3,000 motor salvage companies (including vehicle dismantlers) in the UK; it is estimated that up to 78,000 stolen vehicles and 12,000 insurance fraud vehicles go through these companies annually.

The Objective: to reduce vehicle theft by making it more difficult to dispose of stolen vehicles, and assist the police and other authorities investigating such offences, and to reduce crime.

The Risks: no identified additional risks for those who will be affected by the legislation. Overall cost to society (including criminal justice costs) for each stolen vehicle is £4,700. So, the total annual cost of vehicles stolen for their parts or to be exchanged for a legitimate written off vehicle is approximately £367 million. Insurance fraud vehicles would bring the total to approximately £424 million annually.

The Options:

(1) rely on present regulatory regime;
(2) apply a voluntary code of practice to the motor salvage industry (and scrap metal dealers);
(3) apply a voluntary code supported by legislation.

Option (1): the motor salvage industry is only regulated by national environmental protection legislation (the Environmental Protection Act 1990) and will be subject in the future to EU legislation with a similar purpose (End of Life Vehicles Directive which will come into force circa 2002).

No perceived benefits of Option (1); and the Option estimated to have no impact on vehicle crime. No additional costs in adopting this Option.

Option (2): Both the insurance industry (in relation primarily to the responsible disposal of vehicles written off by insurers) and the vehicle salvage industry have adopted a voluntary Code of Practice, but have a limited ability to “police” it. Also, the Code of Practice does not apply to about 30% of the salvage industry and up to 90% of the vehicle dismantling industry.

Option (2) would lead to those applying the Code of Practice to complain that significant parts of the industry were benefiting economically by not complying with the Code. The Option would have limited benefit in reducing vehicle crime – possibly 8,000 fewer vehicle thefts per annum, with a saving of £38 million. However, compliance costs would not be high because they would largely involve business records which were already maintained.

Option (3): This would have the benefit of maximising the opportunities to reduce vehicle crime, particularly by making it more difficult to dispose of stolen vehicles and making it easier for the police to investigate such offences. It is estimated reduce vehicle thefts in the first year of operation by 15,000, rising to 39,000 in the third year and thereafter, thus creating an initial economic benefit of £71 million rising to £183 million annually. There would be an estimated additional benefit in the reduction of insurance fraud of £8 million per annum.

Option (3) would have compliance costs of businesses registering under the legislation, making identification checks and assisting with enquiries, estimated to be £302,000 per annum, but these costs would not fall disproportionately on small businesses.

So, recommendation is Option (3).

A full RIA is a more completely argued and formal document which is submitted to the relevant Minister with a clear recommendation for action; if it is approved it is signed off by the Minister and published. It should: (i) identify the objectives; (ii) assess the risks that the proposed regulation is addressing; (iii) compare the benefits and costs for each option; (iv) summarise who or what sectors bear these costs and benefits and any issues of equity or fairness; (v) outline the impact on small firms and any measures for helping them to comply; (vi) set out the arrangements for securing compliance with each of the proposed options; (vii) explain how the proposals would fit with existing requirements; (viii) provide a summary of the results of the consultation exercise, responses received, and subsequent consequential changes in the RIA and (ix) state how the policy would be monitored.

All stages of the RIA process are fully integrated within the Government management structure through the central Cabinet Office and RIA units in individual Government Departments.

4.4. Deregulation

In the United Kingdom, formal procedures for deregulation have been in place since the bringing into force of the Deregulation and Contracting Out Act 1994. The powers have been widened by the Regulatory Reform Act 2001. The fundamental purpose of the deregulation procedure is to reduce the amount of regulation of business within existing legislation. The procedure allows Government Ministers to amend or repeal, by order, primary legislation passed before the end of the parliamentary session 1993-94 which imposes an unnecessary burden on business. It is, essentially intended to be a mechanism to remove excessive regulation whereas regulatory impact assessment is designed to avoid such excessive regulation in future legislation.

The statutory deregulation procedure was viewed with great suspicion when it was proposed. One parliamentary committee described the power of Government Ministers to make such orders to be “unprecedented in time of peace”38. As a result, there are elaborate procedures for the parliamentary scrutiny of deregulation orders. A document containing a proposal is presented to Parliament (“laid before Parliament”) in the form of a draft order, together with explanatory material. Two parliamentary committees, one from each Chamber, then have 60 days to consider and report on the proposal. After that the Government present a draft order to Parliament, either as originally proposed or with amendments recommended by the parliamentary committees. The order must then be formally approved by both Chambers. The procedural rules of the second Chamber require its committee to report again on the Government order before it can be considered for approval by the Chamber39.

Although the power of the Government to make deregulation orders is a significant one, it has to date been used for quite minor purposes. So, for instance, one proposal was to extend the permitted hours that bars could stay open so that they could remain open for a continuous period of 36 hours over New Year’s Eve; as a result of the report of the committee of the second Chamber, this was limited to New Year’s Eve 2000. This is hardly represents a significant reduction on the regulatory burden placed on businesses in the United Kingdom. However, once the Regulatory Reform Act is in force, the deregulation power may be used more widely.

4.5. Other techniques?

There are other procedural techniques which have been given consideration in some jurisdictions and which may serve to reduce the cost of legislation to private sector.

One such technique is the alternative compliance mechanism. The underlying philosophy of this is that the Government may, by a compliance order, approve an alternative method of complying with a regulation affecting a business or industry than that which the regulation provides. So, for example, “a business might propose an inspection schedule for machinery which suits its own maintenance schedule rather than meet the periodic requirements set in regulation”40. The business making such a proposal would have to meet the objectives, although not necessarily the detailed procedural requirements, of the regulation. Draft legislation, the Regulatory Efficiency Bill, providing for alternative compliance mechanisms, was introduced into the Canadian Parliament in 1994. It was heavily criticised by a committee of the Parliament41 and did not progress. A similar proposal was considered by the State of Victoria Parliament, at the request of the Government of the State of Victoria, more recently42.

It may be that procedures such as this which provide greater legislative flexibility will find favour if they are not seen as a threat to the power of parliaments as compared to that of Governments. They are perhaps best presented as procedures which provide a hybrid between a normative text and an individual discretionary decision of the Executive.

5. Conclusion

These are then some of the means of estimating the cost of legislation. I am conscious that there are major and increasingly significant considerations that I have mentioned but not fully explored – such as the environmental and social costs of legislation. However, I trust that this analysis is sufficient to demonstrate that rigorous evaluation of the cost of legislation, while determining policy and before the legislation is enacted, is likely not only to be to the economic benefit of the state but also to improve the quality of legislation.

(...)


22 These stories are taken from a website (www.sillyregs.com) which is supported by various UK business organisations.

23 See the Report from the Commission to the Council and the European Parliament: Results on the Third Phase of SLIM and follow-ups of the First and Second Phases COM (1999) 88 final (the Report is on the EU website; see www.europa.eu.int).

24 See C M Heimann et al., “Project: The Impact of Cost-Benefit Analysis on Federal Administrative Law” (1990) 42 Admin L Rev 545.

25 Vol. 13 No. 2 Statute Law Review (1992), editorial.

26 HC 389 (1997/98).

27 Vol. 21 No. 2 Statute Law Review (2000), editorial.

28 www.parliament.uk

29 Hon Mr. Justice Brooke, chairman of the Law Commission, in presenting the 1994 annual report of the Commission; Law Com. No. 232, HC 244 (1994-5) (14 March 1995).

30 A.G. Donaldson, “The High Priests of Mystery: A Note on Two Centuries of Parliamentary Draftsman” in Finnie, Himsworth and Walker (eds) Essays in Public Law (Edinburgh U.P.: 1991).

31 T. St.J. N. Bates, “Privatised drafting and the Finance Bill” 1996 Scots Law Times 153; T. St.J. N. Bates, “Contracting out drafting: a British experience” 17 Statute Law Review (1996) 152.

32 For the different experience in the U.S.A. see M. D. Bellis, “Drafting in the Congress of the United States” (paper presented to the conference on “Legislative Drafting: Emerging Trends”, Dublin, Ireland, 6-7 October 2000).

33 The House of Commons now has a Legal Services Office; for its staff, see Vacher’s Parliamentary Companion (December 2000) p.143.

34 HL Bill 10 - EN, paragraphs 176-186.

35 HL Bill 68-EN, paras. 24-27.

36 Good Policy Making: A Guide to Regulatory Impact Assessment (Cabinet Office:2000) available at www.cabinet-office.gov.uk.

37 ibid., Annex 2(i).

38 Report of the House of Lords Committee on the Scrutiny of Delegated Powers, HL 60 (1993-94) para. 1; this Committee, re-styled the Delegated Powers and Deregulation Committee, now plays a central role in the scrutiny of deregulation orders made under the 1994 Act: see P Tudor, “ Secondary Legislation: Second Class or Crucial?” 21 Statute Law Review (2000) 149.

39 House of Lords Standing Order 70(1)(b).

40 Liberal Party of Australia (Victorian Division) Small Business Policy, p.8.

41 Standing Joint Committee for the Scrutiny of Regulations, Report on Bill-62 (16 February 1995).

42 Parliament of Victoria Law Reform Committee Discussion Paper, Regulatory Efficiency Legislation (May 1997).