Published in F.Regan, T. Goriely, A. Paterson, D. Flemming (eds) The Transofrmation of Legal Aid: Historical and Comparative Studies (Oxford University Press) ISBN: 0-19-826589-1
The Construction of ‘Comparison’ in Legal Aid Spending:
The Promise and Perils of a Jurisdiction-Centred Approach to (International) Legal Aid Research[]
CYRUS TATA[]
Public expenditure has become the key battleground where legal aid policy is contested (e.g.: Goriely 1997; Moorhead 1998). Research examining current and projected costs of policy (both within and between jurisdictions) inevitably commands attention in government circles. Although the promise of comparing costs within and between jurisdictions seems immense, care needs to be taken in interpreting what the comparisons actually mean; and, it is imperative that attention is paid to issues underlying comparative methodology. This chapter reflects on some of the key issues in comparing criminal legal aid expenditure. It arises from a research study which compared criminal legal aid expenditure in three European jurisdictions. This paper first describes the origins, background, and objectives of the study before reporting some of the main findings. The chapter then proceeds to examine critically key issues in comparative research of this nature. It uses the inter-jurisdictional study and its findings as a means of discussing the methodological issues underlying ‘comparative’ research.
Although the chapter reports recent comparative criminal legal aid research, it necessarily raises questions for comparative civil legal aid research, as well as ‘comparative’ methodology more generally. It would be tempting to assume that inter-jurisdictional empirical research is in principle necessarily far more problematic than single-jurisdictional research (e.g.:Bierne and Nelken 1997). This chapter questions the view that inter-jurisdictional research is necessarily more difficult than single- or ‘intra’-jurisdiction research. Rather, it will argue that ‘jurisdiction’ (the legal limits of the State) and legality (i.e.: legal doctrine and institutional legal obligations) are not the primary sources of the major challenges facing research on legal aid spending. The chapter instead suggests that the key intellectual difficulties are located in the attempt to make any kinds of comparisons whether this over time; between organisations; between different populations; geography; cultures etc. I will suggest that labelling only inter-jurisdictional research as ‘comparative’ is misleading and that instead we need to recognise that the essence of empirical legal aid research is necessarily comparative. First, however, the paper discusses the conduct and findings of recent research on the spending patterns of three European countries.
Criminal Legal Aid Expenditure and Constitutional Requirements
Criminal legal aid expenditure is not solely a matter of political choice. In its 1995 Green Paper, the Lord Chancellor's Department of England and Wales acknowledged that ‘criminal legal aid is different from family and other civil legal aid’ (para 4.38). Thus:
Unlike most other forms of legal aid, demand is determined by the state. Defendants in criminal cases have no choice but to defend themselves against the power of the state ranged against them in the form of prosecuting authorities. (para 4.40)
Furthermore, the right of a criminal defendant to legal assistance is guaranteed under Article 6 of the European Convention on Human Rights. Article 6.1 guarantees the right to ‘a fair and public hearing within a reasonable time by an independent and impartial tribunal established by law’. In the case of criminal trials, Article 6.3 sets out the following minimum rights:
(a) to be informed promptly, in a language which he understands and in detail, of the nature and cause of the accusation against him;
(b) to have adequate time and facilities for the preparation of his defence;
(c) to defend himself in person or through legal assistance of his own choosing or, if he has not sufficient means to pay for legal assistance, to be given it free when the interests of justice so require;
(d) to examine or have examined witnesses against him and to obtain attendance and examination of witnesses on his behalf under the same conditions as witnesses against him;
(e) to have the free assistance of an interpreter if he cannot understand or speak the language used in court.
However, it is not possible to say what level of expenditure would be necessary to comply with the Convention and the amount actually spent differs considerably over time; between, and, within jurisdictions. So why might identical constitutional requirements give rise to differential levels of expenditure? For these purposes, the Convention’s key provision is Article 6.3 (c) which requires that defendants with insufficient means must be given free legal assistance ‘when the interests of justice so require.’ This, however, must be read in conjunction with the other provisions, which demand, for example, that there is adequate time to prepare a defence and the chance to examine and cross-examine witnesses.
There is now an emerging body of case law interpreting Article 6.3 (c) (Ashworth 1996). It has been held that when legal assistance is provided for free the accused does not necessarily have the right to choose a lawyer,[1] but his/her wishes should be considered. Where the relationship between a lawyer and client has broken down, schemes should make provision for another lawyer to be assigned.[2] The key question is when and in what circumstances ‘the interests of justice’ require legal assistance. The determining factors appear to be: the complexity of the case; the seriousness of the charge; and, the severity of the possible sentence. In Quaranta v Switzerland,[3] the accused received a custodial sentence and the court found that ‘free legal assistance should have been afforded by reason of the mere fact that so much was at stake.’ Pham Hoang v France[4]suggests that the interests of justice may also require representation for a serious charge when the proceedings are ‘...clearly fraught with consequences...’, even if the outcome was a large fine.(Ashworth 1996).
Expenditure levels
At one level, therefore, all the signatories of the Convention are bound by the same rules. It is clear, however, that these rules result in different levels of expenditure. Although there are no official figures on criminal legal aid expenditure within Europe, the 1994 International Conference on Legal Aid made the following estimates of 1992 per capita expenditure within the most developed legal aid schemes. They cover three signatories to the European Convention - the UK, the Netherlands and Sweden.
Table 1: Estimate of expenditure per capita on criminal legal aid 1992
US $England and Wales / 15
The Netherlands / 3
Scotland / 21
Sweden / 9
Source: Klijn and Huls (1994)
As the authors of that paper acknowledge, these figures are only estimates. They relate to different time periods; do not extract tax; and, do not necessarily extract the criminal elements of combined expenditure (on, for example, children or advice). However, the magnitude of these differences inevitably raises questions as to the different ways in which Scotland, England and Wales, the Netherlands, and Sweden interpret their obligations to comply with the European Convention in terms of per capita spending.
A recent pilot study investigated some of the reasons for these apparent differences (Goriely, Tata, and Paterson 1997). In particular, it considered why Scotland appears to spend so much more on criminal legal aid than other signatories to the convention. The pilot study did not aim to provide complete answers to these questions, but to provide some reasonably robust hypotheses which might be pursued by subsequent research. The study, commissioned by the Lord Chancellor’s Department of England and Wales and by the Scottish Office, compared the spending patterns of England and Wales, Scotland and a third country: the Netherlands. The Netherlands was included because of its seemingly very low level of spending; the extent of information which is available; and the high-quality reputation of its scheme. The first Director of the American Legal Services Corporation, Clinton Bamberger, described the Dutch legal aid scheme as “one of the best in the world: probably the best” (Bamberger 1989). Like the UK schemes, it relies on private practitioners to provide criminal services: there are no salaried public defenders providing assistance in criminal matters. Furthermore the Netherlands has a relatively high level of compliance with the Convention - at least equal to that of the UK (Swart and Young 1995).
‘Adversarial’ and ‘Inquisitorial’ Models of Criminal Justice
It might be expected that differences in criminal legal aid expenditure between Britain and other European jurisdictions can be explained by differences between ‘adversarial’ and ‘inquisitorial’ models of criminal justice. In an ‘adversarial’ model, where the judge is a neutral umpire between prosecution and defence, it is clearly vital for both sides to be represented. Thus, it might be supposed, the interests of justice require more people to be represented. Furthermore, the defence lawyers must carry the burden of presenting evidence: they must therefore do more, and be paid more for it. In the typical ‘inquisitorial’ case, however, the main burden of finding evidence relies on the judge. Defence lawyers may be needed in fewer cases, and may be required to do less.
However, empirical examination suggests that the varying assumptions behind the ‘adversarial’/ ‘inquisitorial’ divide cannot provide the primary answer to the question of why international differences arise in criminal legal aid expenditure. There are four reasons for this. First, it cannot account for the differences between ‘adversarial’ systems. If anything, Scotland's system, with the important historical role given to the Procurator Fiscal in criminal trials, is more inclined towards the ‘inquisitorial’ system than that of England and Wales. Yet, as this chapter explains, its criminal legal expenditure is significantly greater. Differences within Canada and the USA also show that even within the same legal family, variations in culture and political preference can result in substantial differences in legal aid expenditure (Goriely 1997; Wall 1996). Secondly, it is too easy for comparisons between ‘adversarial’ and ‘inquisitorial’ systems to concentrate on trials, especially the symbolic high-point of the British system - trial by jury (Darbyshire 1997; Duff and Findlay 1997). Although British jury trials are relatively slow and expensive, it should be remembered that in Scotland at least, the bulk of legal aid expenditure does not involve jury trials (or even its possibility) but rather Summary procedures prosecuted in the lower courts. In addition, in most cases the accused pleads guilty, at least at the point the case comes to trial (Samuel and Adler 1994; Samuel 1996), and, (in common with research findings in England and Wales), proceedings tend to be abrupt (McBarnett 1979; Mcconville, Sanders, and Leng 1991). Therefore, although the assumptions underlying the Dutch and Scots approach may differ, it is by no means self-evident why the Scots’ system should be so much more expensive in the way that these cases are handled. Thirdly, the easy invocation of ‘adversarial’ and ‘inquisitorial’ models of truth-finding underestimates the extent to which the Dutch scheme is increasingly relying on defence lawyers to cut investigative corners (Jörg, Field, and Brants 1995). The criteria governing when the court may refuse such requests have been relaxed by statute, resulting in a great increase in litigation on the issue. There is also more pre-trial contact between prosecution and defence in the Netherlands. Jörg, Field, and Brants (1995: 48) comment that ‘the incidence of negotiations between defence and prosecution is on the rise: while not called plea bargaining yet, the terminology has emerged in official documents.’. Finally, there are some aspects of ‘inquisitorial’ justice which may actually increase legal aid costs. For example, in an ‘inquisitorial’ system an uncorroborated confession does not necessarily dispose of a case in the same way as a plea of guilty does in the ‘adversarial’ system (Jörg, Field, and Brants 1995: 50)
Therefore, in the study discussed here, it was recognised that the primary explanation for the marked differences in per capita spending does not lie in the differences between ‘adversarial’ and ‘inquisitorial’ models. To understand the dynamics of spending we must consider explanations other than simple constitutional requirements. Below I focus on the key findings of the comparative study.[5]
Criminal Legal Aid Expenditure in England and Wales, Scotland, and the Netherlands : Analysis of Spending by Population
The initial task was to calculate expenditure on criminal legal aid by head of population in each of the three jurisdictions under discussion. Table 2 below shows total expenditure by per head of population, broken down into its constituent categories.
Table 2: Expenditure on criminal legal aid, per head of population (£ sterling)[6] 1994-5
England and Wales / Scotland / The NetherlandsPopulation (1991) / 51,099,000 / 5,107,000 / 15,010,000
Representation in jury courts / £5.76 / £4.79 / nil[7]
Representation in non-jury courts / £3.16 / £10.32 / £1.41
Court duty solicitor schemes / £0.24 / £0.18 / n/a
Assistance at Police stations / £1.34 / n/a / n/a
Advice / £0.47 / £1.36 / £0.16
Appeals / £0.25 / £0.29 / £0.20
TOTAL / £11.22 / £16.94 / £1.77
less contributions / £0.08 / nil / nil
TOTAL NET COSTS[8] / £11.14 / £16.94 / £1.77
The Table distinguishes between the cost of representation in cases dealt with in jury courts and those which are not. Thus, the category ‘Representation in jury courts’ relates to cases which go to the Crown Court in England and Wales, and those dealt with under solemn procedure within the Sheriff and High Court in Scotland. These latter cases do not necessarily involve a trial, but include guilty pleas and cases where the Procurator Fiscal takes no proceedings, as well as trials. The Dutch do not have jury trials. The second category includes cases dealt with under summary procedure in the Sheriff and District Court in Scotland; the Magistrates' Court in England and Wales; and, the Dutch District Court.
How does Scotland compare?
In the year 1994/5 expenditure in Scotland was clearly the highest, with spending per head of population 52% greater than in England and Wales, and more than nine times that of the Netherlands. The differences lie primarily in payment for representation before summary courts, Scotland’s per capita expenditure is three times that of England and Wales and seven times as great as the Netherlands.
At first sight, the difference between England and Wales and Scotland seems surprising given that Scottish criminal procedure contains several features which have been proposed in England and Wales as money saving devices. First, in contrast with England and Wales, accused persons in Scotland do not have the right to choose jury trial: the decision is that of the prosecutor alone. Decisions over who should be granted summary legal aid are made by the Scottish Legal Aid Board, rather than by the courts as in England and Wales. Secondly, Scottish solicitors also have more extensive ‘rights of audience’ than their counterparts in England and Wales, and, in Scotland, solicitors can appear in all Sheriff Court proceedings. Thirdly, the Scottish prosecutor has more powers to divert cases from formal court procedures, and, fourthly, the Scottish criminal legal aid system appears to be less generous to those pleading guilty.[9] Finally, the Scottish system has not had to bear the high costs of city-related fraud trials which can prove to be so expensive for the English and Welsh scheme. The finding that the Scottish system costs more than that of England and Wales is unexpected for another reason. Research by Samuel and Adler (1994) found a widespread perception among Scottish solicitors that, compared with England and Wales, the Scottish system offered good value for money.
So far I have identified differences in per capita spending. In an attempt to provide some further analysis and explanation, and given space constraints, the remainder of this chapter concentrates exclusively on the element which appears to account for the vast bulk of spending: namely, representation before non-jury courts by examining first the number of cases and then the costs per case.
Representation before Summary Courts
A key question in understanding the differences identified by Table 2 in representation before non-jury courts is whether they relate to volume or price of cases. Do the Scots fund more cases, or does each case cost more? Do the Dutch fund fewer cases, or is each case cheaper?
The number of summary bills
Table 3: Bills for Representation before Summary Courts in England and Wales and Scotland
1993-94 / 1994-95N / Rate per
1,000 pop / N / Rate per
1,000 pop
Scotland / 68,480 / 13.41 / 65,839 / 12.89
England and Wales / 355,787 / 6.96 / 392,577 / 7.68
The Netherlands / 48,341 / 3.22 / 51,055 / 3.40
Source: England and Wales: Legal Aid Board Annual Reports; Scotland: unpublished information supplied by the Scottish Legal Aid Board; Netherlands: information supplied by the Ministry of Justice.
The differences in volume in the three jurisdictions are striking. Per head of population, England and Wales pays double the number of summary bills as the Dutch, while in 1993-94, the Scots paid almost double the number paid by the English. Finally, it must be remembered that, as the Dutch do not have jury trials, the summary figures include all the serious cases which in Britain would have been dealt with before the higher courts. All these factors make the high number of Scottish bills particularly surprising.