Is the Model Broken? Regulation and Assessment of Legal Costs in New South Wales
Mark Brabazon SC
I am honoured to be asked by Blackstone Legal Costing to give their first annual lecture in Sydney. Blackstones started in Melbourne. They do things differently there. My subject this evening is not any Melbourne-Sydney rivalry, but the autochthonous experiment of costs assessment in New South Wales. Should it now be declared a failure, or is there yet hope?
The assessment of legal costs is a dry topic, but a necessary one. It is a fine conversation-stopper unless you are talking to a lawyer with a costs problem. The subject is large. I shall not attempt to cover it all, but shall focus on a few key themes which have caught my attention and which, I hope, may point to a useful answer.
1. The New South Wales Model
The New South Wales model of costs assessment with which we are all familiar has a set of fundamental features: the use of deregulated market-based criteria for party/party and practitioner/client assessment, a set of disclosure-based consumer protection rules, the establishment of a single extra-judicial structure for quantification of party/party and practitioner client costs, what might be described as formal informality of process, outsourced decision making, and a user-pays system of funding.
The model was first enacted as a bold and unprecedented statutory experiment in 1993. It took the form of an entirely rewritten Part 11 of the Legal Profession Act 1987. It is now Part 3.2 of the Legal Profession Act 2004. The 2004 rewrite significantly increased the disclosure requirements and forbad fee uplifts conditional on success in damages. There has been some other tinkering at the edges but, for the most part, Parliament has left the original model untouched. What we have today largely reflects the legislative vision of 1993 with the addition of a regulatory overlay from 2004.
Nearly two decades after its introduction, Bathurst CJ commissioned a review of the costs assessment system. The legal status of the review was unusual. There was no statutory provision for it. It was not commissioned by Parliament or the Executive. It was simply an initiative of the Chief Justice. The Court published terms of reference and invited submissions from the public. The review was conducted by a judge of the Court, Brereton J, assisted by an advisory panel drawn from the judiciary, from government agencies and from professional bodies. His Honour presented his report to the Chief Justice, who published it and invited further public comment in March last year. The Report and subsequent comments are now in the hands of Government.
The peculiar constitution of the review has some important consequences. First, its terms of reference focused on the assessment system, not on the general regulation of legal costs or the disclosure rules. Secondly, reflecting its status as an initiative of the judicial branch, the review was conducted on the basis that ‘the deregulation of practitioner costs, the ‘fair and reasonable’ criteria for party/party costs, and the ‘user pays’ funding model are not open to general reconsideration’. The general approach was to look for ways to improve the performance of the system within its basic parameters. Thirdly, the review had no resources beyond the personal and official resources of its immediate participants. There was no funding for research. The review did not have the resources to compare New South Wales with interstate or international models in any depth, or to undertake statistical research into the market cost of legal services.
The interstate dimension is nevertheless important when we are considering possible futures for costs assessment and regulation in this State. New South Wales is not an island. Its costs reforms have influenced other Australian jurisdictions, and developments elsewhere may influence the future of costs assessment in New South Wales. And from an economic perspective, Australia increasingly operates as a single market for legal services, or a set of inter-penetrating markets.
Other States and Territories have looked at the New South Wales experiment and found it to be like the curate’s egg – good in parts. The deregulation of practitioner/client costs, the adoption of market-based criteria for party/party and practitioner/client assessment, and the system of disclosure-based consumer protection rules have been widely copied. Many provisions have been copied verbatim. There are reservations, of course: Victoria has adopted fee deregulation, but has retained scales as the measure of party/party costs and as a default measure of practitioner/client costs. But what the other jurisdictions have really balked at is the system of outsourced and extra-judicial decision making. Those features remain peculiar to New South Wales.
More importantly for the future, those aspects of the New South Wales system which already found favour interstate are set to be replicated, with some genuine improvements in policy and drafting, in Part 4.3 of the new Legal Profession Uniform Law. The Uniform Law is a product of the COAG process. Its main function is to provide uniform regulation of the legal profession across Australia and to resolve any remaining choice of law or jurisdiction issues. It has been formally enacted by Victoria as host jurisdiction, with application legislation to follow in New South Wales and elsewhere. The Uniform Law leaves most aspects of the costs assessment process to be determined by local legislation. Victoria, for example, will continue to have its Costs Court, a structure within the Supreme Court headed by an Associate Justice in which the effective decision makers are either judicial officers or registrars. Like its New South Wales counterpart, the Costs Court is conceived as a State-wide one-stop shop for the quantification of costs, but its process is centralised and judicially controlled.
That brings me back to the fundamental features of the New South Wales experiment. I have found it convenient to divide them into two groups. The first group is made up of the deregulation and consumer protection elements. These provide legal criteria and a regulatory framework for assessment. I shall talk about them first. The second group comprises all the rest – the one-stop shop, outsourcing, informality and user-pays. These are as integral to the process and mechanism of assessment as practised in New South Wales, and I shall talk about them second.
2. Legal Criteria and Regulation
The starting point in deregulation is the choice of a single verbal formula – ‘fair and reasonable’ – as the criterion for all costs assessment. In the practitioner/client context, this replaced an amalgam of scales, discretionary allowances and qualified contractual freedoms. In the party/party context, it replaced the use of scales and the conceptual test of what is ‘necessary and proper’. Brereton J said of this in his Report:
The evident goal of adopting common ‘fair and reasonable’ criteria for party/party as well as for practitioner costs was that a successful litigant who conducts litigation in the manner of a hypothetical reasonably prudent person and who has the expectation of meeting their own costs from their own resources and has adequate but not extravagant means should be fully compensated for costs by an ordinary party/party costs order.
In a practitioner/client context, the replacement of scales with freedom of contract and the single ‘fair and reasonable’ benchmark was balanced against the second fundamental element, a set of disclosure-based consumer protection rules. It represents an unmistakable act of faith in the mechanisms of an informed free market to deliver rational and fair outcomes. Disclosure requirements apply at the point of retainer, during the course of retainer, and at the point of billing. Failure of disclosure has various adverse consequences for the practitioner. The 2004 rewrite increased the nature and scope of disclosure obligations and made more severe the consequences of non-compliance. This aspect of the rewrite was heavy-handed. It reflected no real understanding of the practicalities of legal practice, of litigation, or of the real needs of clients. The Uniform Law strikes a more realistic balance. It retains the obligation to disclose the basis of charging, as well as the obligation to give estimates covering the entire retainer and, where a significant change occurs, updated estimates. It also retains pre-settlement estimate obligations and disclosure obligations at the point of billing. The estimate obligations will, without doubt, be the most contentious aspect in actual practice.
What is Really Needed?
So what is really needed for the assessment of deregulated costs and for the disclosure of information about costs to clients?
The deregulated market-based approach to costs assessment in New South Wales and elsewhere requires decision-makers to identify what is a ‘fair and reasonable’ cost for particular legal services. Under a time-billing system, this usually resolves to a number of more specific questions: Was the hourly rate for the particular practitioner fair and reasonable? Was it reasonable to do the particular work that is charged for? Was it reasonable that this practitioner do it, and not another? Was the time claimed for particular work by a particular practitioner reasonable? And so on. Under a billing system defined by milestones or events, or where an agreed lump sum is charged for an entire piece of work, the question of what is fair and reasonable is necessarily broader. It cannot just be determined in retrospect by reference to the actual work done. The client and the lawyer each buy certainty of price, and take the risk that the amount of actual work may turn out to be less or more than they have guessed in their bargain.
The New South Wales model makes an assumption which may or may not be made elsewhere: that it should be possible to recognise what is fair and reasonable for an entire piece of work just by reference to certain high-level characteristics of the work. This focuses on the services themselves. A one-day personal injury action in the Supreme Court should usually cost about this much; a two-day criminal trial in the Local Court should cost about that much; and so on. To put the matter in those bald terms is overly simplistic. Any number of factors can reduce or increase (it is usually increase) the appropriate extent or expertise of professional input. The assumption is that experienced legal professionals have the ability to recognise and quantify those factors in a fashion that is both reliable and relatively broad-brushed.
If the assumption is true, it has not been demonstrated. Its implementation has not even been attempted. If the recommendations arising from the review are adopted, that attempt may yet be made. With what outcome remains to be seen.
It has long seemed to me that market-based assessment of legal costs needs a scientific foundation. Professional experience gives practitioners a sense of what hourly rates are commonly charged in a particular field. It should give them a sense of what is prudent or sensible and what is silly or extravagant, particularly in their own field of practice. Whether and to what extent it gives them a sense of how long particular tasks should take to perform, particularly amorphous tasks such as gathering evidence or getting-up a case in which they are not personally involved, is more debatable. An experienced practitioner’s opinions about these things are based on individual experience and anecdotal observation. They are not worthless, but they are not particularly scientific.
Science assumes a body of organised knowledge based on objectively verifiable data. The market price of legal services is the professional domain not of the lawyer, but of the statistician. The relevant data set would correlate relevant classes of legal services with what people pay for them by reference to the services themselves. I am not aware of any body of data that answers this description. Not being a statistician, I cannot say how difficult or practicable its assembly would be. Data aggregating the outcomes of costs assessments clearly would not do: that would be a measure of the behaviour of costs assessors in a non-representative sample of cases. Data aggregating charge-out rates would not do: that is just a measure of the price of lawyers’ time. The closest thing we have at present is the aggregated anecdotal experience of a relatively small number of specialist costs consultants, whose personal experience may serve as a proxy for an objective data set. The courts have recognised and relied on expert evidence of that kind in the context of lump sum party/party costs orders.
The need for a scientific foundation is not limited to costs assessment. If information is the client’s panacea, it must be reliable. The present model assumes that lawyers know, or should know, how to identify at the outset of a retainer a monetary range within which the total cost of the retainer should fall, subject to allowance for contingencies. It assumes expertise in budgeting for litigation, or whatever the subject matter may be, both at the outset and as the matter progresses. In order to acquire the assumed skills, it is reasonable to think that lawyers would receive training. They do not. They develop their budgeting skills ad hoc, in the course of legal practice. Some of them get good at it.
The legislature requires lawyers to give estimates. By implication, it expects them to give estimates which clients can treat as reliable, albeit subject to revision if circumstances change or assumptions are falsified. Given those obligations and expectations, the training of lawyers in private practice should include legal budgeting. Legal budgeting is a cross-disciplinary skill: commerce informed by law. It should be taught at the College of Law. The development of a training programme should also facilitate the development of a body of organised knowledge concerning the objective market cost of legal services, particularly in litigation.
We should also ask just what it is that clients need in the way of information. They do not all need the same thing. There are some carve-outs for sophisticated clients from the full rigour of disclosure and assessment, but for the most part the disclosure system is still a one-size-fits-all regime. If we must have a one-size-fits-all regime, it should fit as many cases as possible. So what does the hypothetical template client really need in the way of information about costs? The answer, I suggest, is a meaningful, objectively reliable and continuing conversation about legal budgeting. What the client gets at present, if the lawyer is professionally competent and compliant with the legislation, is a pile of information that protects the lawyer’s back. The two things are not the same. The present heavy-handed regulatory requirements are not the solution, they are part of the problem.
The Uniform Law is a move in the right direction. It reduces the excesses of the 2004 Act in relation to disclosure. The next step should address the practice and culture of lawyers by giving them the training and the confidence to have that budgeting conversation with the client.
3. The Assessment Scheme
I want to return now to the other key elements of the New South Wales model. These are the elements which I have described as integral to the process and mechanism of assessment. They are generally not addressed by the Uniform Law. They comprise the costs assessment scheme proper.
The scheme provides a one-stop shop for all kinds of costs assessment. It is not constituted as a court or even a tribunal, but as an administrative facility within the Attorney-General’s Department. Its decisions are made by publicly appointed but otherwise private solicitors and barristers functioning as quasi-arbitrators. These costs assessors have information-gathering powers, a duty of procedural fairness, and a duty to give reasons, but otherwise their process is practically unregulated and formless. There is no provision for hearings; all decisions are made on written evidence and submissions. There is a right of full merits review by a panel of two assessors, but the only judicial oversight of the scheme is by way of appeal (as of right for error of law, otherwise only by leave) or judicial review in the general jurisdiction of the Supreme Court. The assessors are paid by the Government. In party/party assessments, this cost is fully covered by inter-partes costs awards made by the assessors. In practitioner/client assessments and merit reviews, the scope for costs-of-assessment awards is more limited: the shortfall is made up from filing fees in all kinds of assessment applications.
In short, these are the bits that the other States don’t like.
If deregulation of costs, the unified test and information-based consumer protection amounted to an act of faith in the mechanisms of an informed market, the design of the assessment scheme was an act of faith in the privatisation of public services and in a combination of ‘informality’ and a nominally inquisitorial process as a means to reduce cost and increase efficiency. This faith included an underlying assumption, as Brereton J put it in his Report,
that experienced practitioners would be able to make a global determination of what a particular piece of legal work should reasonably cost, having regard to its characteristics and history, without the need for the kind of detail and itemisation that had been customary in the taxation of hills of costs, and therefore more quickly and efficiently than under the former regime.
The assumption has two aspects. There is a substantive assumption that global assessment is possible – I have already mentioned this in the context of deregulation. And there is a procedural assumption that experienced practitioners already have the skill, knowledge and ability to do it. At least as the scheme has developed to date, that assumption has not been realised.
The procedural aspect goes further. There is an assumption that experienced practitioners appointed as assessors can make reliable assessments with nothing more than written information from the lawyers concerned, written submissions from each side, and access to the relevant lawyer’s file. Taken as a general proposition, I think the assumption has been exposed as naïve. I do not want to overstate this: I am not saying that outsourcing is proven to have failed. I do not think the experiment has been given its best chance.
In considering what changes may now be necessary or wise, it is also important to remember that any changes will not be made in a vacuum. The costs assessment scheme has given rise to a substantial industry. This was probably inevitable. First, those barristers and solicitors who also practise as costs assessors rely on the scheme for part of their income. Secondly, the paper-driven assessment process demands vast quantities of written material in a form that costs assessors can use and digest. This is usually provided by costs consultants, occasionally by non-specialised solicitors.
Thirdly, it is the anecdotal but widespread observation of practitioners in the field that legal costs disputes, particularly on the practitioner/client side, have become far more litigious than ever they were before 1994. It is difficult to say for certain whether this reflects a cause-and-effect relationship with deregulation, with the costs assessment scheme, or perhaps with both. Some statistics from the five years to 2012 may be of interest. Of 8,108 original applications for assessment, party/party assessments outnumbered practitioner/client ones by only 4:3. Of 3,442 practitioner/client assessments, those initiated by practitioners outnumbered those initiated by clients by 7:3. And the proportion of determinations by a single assessor that were taken to review by a review panel was about one in nine. This does not count appeals from assessors or review panels.
The Chief Justice’s review focused particularly on the process and mechanism of assessment. Brereton J’s Report made a total of 56 recommendations. They are strongly focused on process with a view to efficiency. An implicit conclusion of the Report is that informality has not produced efficiency, and that process will be improved by imposing structure and a degree of procedural control. Formlessness tends to chaos, which is inefficient. That has been a hallmark of the assessment scheme to date. Formality serves a purpose. Mindfully targeted, it orders human activity and enhances efficiency. It remains to be seen which of the Report’s recommendations will be implemented, and what effect they will have on the performance of the assessment scheme.
At this point in the history of costs assessment, I think it is useful to stand back and reflect on a few broad themes – not improvements of process as such, but the life force of the outsourcing model itself.
Can Outsourcing Survive?
The big question is, can the outsourcing model survive? To put it another way, more consistent with the approach of the review, what would be necessary for it to survive? What conditions would be likely to produce an optimal outcome from the outsourcing experiment?
The first and most obvious necessity is a tectonic improvement in consistency of decision making. Practitioners in the area have long identified inconsistency between individual assessors as a major problem. One experienced costs consultant no longer gives her clients predictions of outcome: she finds the assessors too unpredictable. Inconsistency was a major theme in submissions to the Chief Justice’s review.
The probable causes of inconsistency are not difficult to find. The assessors are numerous and geographically isolated from one another. They are not members of an institution that could provide an institutional culture. There is little institutional supervision, beyond the exercise of review and appeal rights by dissatisfied parties to assessment. Their mandatory professional development is relatively limited. There is no scientific or otherwise organised body of knowledge by reference to which they exercise their functions. Their assessment work is part-time. The criteria for appointment as an assessor are generic, and there is no specialist qualification that could be required. And the remuneration of assessors is relatively poor by professional standards.
Short of rejecting the outsourcing model, a cure for inconsistency must involve a range of strategies. A reduction in the number of assessors, at least in the short term, would be useful. But I think it is possible to draw a larger and perhaps less obvious conclusion. I have come to the clear view that it is essential to bring costs assessment under direct and centralised judicial supervision at an institutional level. That is not to say that the assessors should sit in a building in Sydney. Some of the best assessors now are in regional New South Wales. The point is that consistency requires a degree of cohesion which can only be achieved by leadership, which given the nature of the task can only come from the judiciary.
Centralisation under judicial supervision is also necessary if the procedural recommendations arising from the review are to be successfully implemented.
Brereton J’s Report recommends reconstituting of the Manager, Costs Assessment as a Registrar of the Supreme Court and instituting judicial leadership of the Costs Assessors Rules Committee. Those steps, or something equivalent, are I think indispensable.
So far I have focused on consistency of decision making and organisational structure. Now I want to focus on the role of the costs assessor.
What is a costs assessor? An assessor is not a judicial officer; not an officer of a court; not an administrator. An assessor has inquisitorial powers and uses his or her personal expert knowledge, but makes quasi-judicial decisions which result in legally binding determinations of rights, and has corresponding public duties and statutory immunities. An assessor acts only on referral from an administrative official and is paid by the hour by the Government, but the assessor’s fees in many cases are recovered by Government from the parties whose dispute the assessor has determined. It seems that the assessor is best described as a public referee/arbitrator.
Costs assessors are poorly paid by professional standards – the hourly fee of $175 was modest but respectable in 1994, but the only increase since has been a 10% uplift for GST, and assessors must still meet all their own professional overheads. Assessing may suit a part-time employee or a retired practitioner with negligible overheads, or a lawyer practising from home. A barrister or solicitor in active professional practice with Sydney overheads cannot justify practising as a costs assessor, other than as a public service.
You get what you pay for. No wonder there is criticism of the quality and consistency of decision making. The wonder is rather that any and indeed many assessors still produce decisions of high professional quality. Reform of the costs assessment scheme will inevitably put further pressure on the assessors. The outsourcing model assumes that assessors will be ‘practitioners well versed in the running of a legal practice’. We need to keep the best, not the worst; and we need to attract new talent year by year. If we really want to improve costs assessment, we should not expect charity from the assessors.
Brereton J’s Report recommends that the remuneration for assessors be reviewed and increased to at least $250 per hour. The words ‘at least’ deserve emphasis. Allowing for GST, a payment of $100 today equates to $53 in 1994; for a taxable supply, it equates to $59. To replicate a payment of $175 in 1994, you would have to pay $328 today for a taxable supply or $298 for a non-taxable one.
And here is something to ponder. Parity of remuneration could be restored with little or no real impact on the State budget. If legal costs go up (as they have), the filing fees go up. If assessors’ fees go up, the amount recovered under costs-of-assessment awards goes up too.
So is the experiment a failure? I think it is too early to say.
I have argued that deregulated costs assessment should be given a genuinely scientific foundation, preferably by developing a body of statistically sound knowledge. I have argued that lawyers in private practice should not just be told to give estimates, they should be trained in legal budgeting. I have argued that clients need a reliable conversation about legal budgeting, not the present heavy-handed disclosure. Hopefully we can make some progress down that path when the Legal Profession Uniform Law is enacted here.
I have argued that the outsourcing model is not dead yet. It is very sick and we hope it might get better. To give it the best chance, it needs to become more centralised. It needs direct judicial supervision. Among other things, we need a great improvement in the consistency of decision making. And we need to pay our costs assessors properly.
If we achieve all these things, there is hope.
 The 1993 reforms had permitted conditional fee uplifts generally: contrast Legal Profession Act 1987 (NSW) (‘LPA 1987’) s 187, Legal Profession Act 2004 (NSW) (‘LPA 2004’) s 324.
 Hon Justice Paul Brereton, Report of the Chief Justice’s Review of the Costs Assessment Scheme (Supreme Court of New South Wales, 12 March 2013) <http://www.lawlink.nsw.gov.au/practice_notes/nswsc_pc.nsf/pages/607> (‘Report’).
 Report, idem [1.4.4].
 See e.g. Legal Profession Act 2004 (Vic) Pt 3.4; Legal Profession Act 2006 (ACT) Pt 3.2; Legal Profession Act 2006 (NT) Pt 3.3Legal Profession Act 2007 (Qld) Pt 3.4; Legal Profession Act 2007 (Tas) Pt 3.3; ; Legal Profession Act 2008 (WA) Pt 10.
 Victoria has followed New South Wales in adopting the ‘fair and reasonable’ criterion for party/party as well as practitioner/client costs. In a party/party context applying the standard basis of taxation, ‘all costs reasonably incurred and of reasonable amount shall be allowed’ (Supreme Court (General Civil Procedure) Rules 2005 (Vic) r 63.30) The practitioner/client provisions apply a ‘fair and reasonable’ criterion (Legal Profession Act 2004 (Vic) ss 3.4.19, 3.4.44). The difference between the two jurisdictions is more in the process than in the level of permitted or recoverable fees because the present Victorian scales are generally set at commercially realistic levels. Practitioner remuneration orders are prescribed under the Legal Profession Act 2004 (Vic) s 3.4.22 (cf s 3.4.19).
 The Legal Profession Uniform Law (‘Uniform Law’) is Schedule 1 to the Legal Profession Uniform Law Application Act 2014 (Vic).
 The constitution and jurisdiction of the Costs Court are prescribed by the Supreme Court Act 1986 (Vic) Pt 2 Div 2B.
 Note 2 [1.1.4], citing the Attorney-General’s second reading speech (Hon J.P. Hannaford MLC, NSW Legislative Council Parliamentary Debates, Hansard, 16 September 1993, at 3278).
 See M L Brabazon, ‘Costs Disclosure: New Regime’ (2005) 43(5) Law Society Journal 59.
 The central provisions are ss 174 and 175 of the Uniform Law (compare LPA 2004 ss 309, 310, 316); the main consequences of inadequate disclosure are set out in ss 178 and 204(2) (compare LPA 2004 ss 317 and 369(3)(a)).
 Uniform Law s 177.
 Uniform Law ss 192, 194(1) (boilerplate notice of rights). The client’s right to require an itemised bill is set out in ss 187 and 194(2). The Uniform Law recognises lump sum bills and itemised bills (s 186) but does not define either term.
 See e.g. Idoport Pty Ltd v National Australia Bank Ltd  NSWSC 23, which resulted in a ‘gross sum’ $50 million quantification of costs under the Civil Procedure Act 2005 s 98(4). The principal expert evidence for the Bank as receiving party was given by Michelle Castle, then of D G Thompson & Co. Her evidence is described in detail in the judgment of Einstein J (see  ff).
 See e.g. LPA 2004 ss 312, 314, 323, 333, 395A.
 Some of the provisions governing access to the assessment mechanism and appeals from it are covered by the Uniform Law and the other interstate statutes (n 4) in terms which reflect corresponding New South Wales provisions, but the focus here is on the mechanism of assessment itself. The various labels attached to the process of quantification – assessment, taxation, review – are unimportant. The characteristics of the process are what matters.
 Note 2 [1.4.2].
 See n 2 [1.3].
 It must be recognised that the pool of practitioner/client matters includes non-litigious matters and litigation in non-costs jurisdictions and in Courts and tribunals whose costs orders are not subject to assessment under the Legal Profession Act.
 The statutory criteria do not have that character.
 See Report, n 2, recommendations 49 to 56 and Chapter 6 passim.
 Idem, recommendations 49 and 50.
 Attorney-General’s second reading speech (Hon J.P. Hannaford MLC, NSW Legislative Council Parliamentary Debates, Hansard, 16 September 1993, 3269 ff), quoted in Report, n 2 [1.1.2].
 Comparison based on cost price index, weighted average of capital cities, June quarter 1994 to March quarter 2014 (www.abs.gov.au). GST allowed at 10% on the basis that the pre-GST wholesale tax component inherent in legal services was negligible.