Albert Klijn

1. reassessment of the scope of entitlement to state subsidy for legal services;
2. guaranteeing an adequate supply of legal services delivery;
3. improvement of the budgetary control of legal aid expenditures;
4. modernizing the governance structure of the legal aid provision.

In our paper we focused on the central question of the probability of actual effects of the law.

According to our opinion those promises would come closer to reality to the extent that the measures proposed by the government as well as their implementation could be thought to be adequate stimuli changing behavioral conditions for those who are seeking for legal aid and the providers as well. We argued that such would be the case if three preconditions would be fulfilled. Legislation would be as effective as intended only when:

• research has provided reliable knowledge of the behavior of both users and providers acting in order to reach their respective goals in the most efficient way ;
• policy makers can rely on devices derived from that knowledge that can be used to change the conditions of those who are in need of legal services and those who are able to deliver those services;
• those devices are accordingly applied by policy makers in their legislation.

Today, five years from the moment our new legislation came into effect, it might be worthwhile to look back on both our arguing and the actual findings of all the research that has been carried out, meanwhile probing the effects of that legislation. Doing so, I first will present some figures showing you global trends in the use of legal services as well as the public expenditures on legal aid. Secondly, I will go much more into depth with respect to the question to which extent two important goals the legislation aimed at, in effect are fulfilled and, to what extent our sociolegal research has contributed to that. As I will show, this contribution will be greater to the extent that researchers have been able to rely on the existing body of accumulative knowledge.

The 1994 Legal Aid Act did not come out of the blue! Actually it took quite a long time to get all relevant actors to agree with the bill which intentionally would bring a fundamental change in the structure of the legal provision. To place actual findings in perspective, one might have to keep in mind that the legislation as approved by Parliament was - as usual - the end of a long and difficult bargaining process. Great differences of opinion between the government and the legal services providers as well as between those providers themselves had to be bridged. What was the difference the legislation made, seen in that perspective?

For an overall picture of the importance of the Dutch legal aid provision over time as well as the eventual effects caused by the legislation under study, we can rely on two indicators: the volume in use (sheer number of granted certificates) as well as the public expenditures on legal aid. In effect, both are closely interconnected presenting only two sides of the same coin.

* number of certificates per 100,000 inhabitants

With respect to the volume, table 1 shows that the number of granted legal aid certificates rose steadily from 1968 to 1993 both in civil and in criminal law. While in criminal matters the use of legal aid increased nearly four times (from 123 certificates per 100,000 inhabitants in 1968 to 469 in 1993), the level of granted civil certificates in 1993 proved to be seven times as high as in 1968 (from 244 to 1,739). Although different in size, both areas show a break in the trend break starting in 1994, with legal aid use returning to the level of the late Eighties, i.e. nearly 2,000 certificates per 100,000. These figures clearly suggest that the legislation had some effects.

* Expenditures in 1993 prices

The same goes for the trend in spending as table 2 shows. Both in civil and in criminal law, we see continuous growth from 1968 to 1993, then decreases of respectively 14% to 16% in the year 1994 which lowers the expenditures close to the level of 1988.

Two remarks must be made with regard to two important factors with direct relevance to the figures presented. First, one has to take into account the actual use of legal aid which shows a continued growth even during the same period of the eligibility rate - i.e. the percentage of the population entitled to state subsidy based on their net monthly income - was declining from 80% of the population in the Seventies, 66% in the late Eighties to 54% in 1993. The new legislation proved to have lowered that percentage much more to 45% in 1994. That, of course, indicates that most of the provision is used by the low income class people not being affected by this reduction.

Secondly, when interpreting figures on actual use of legal aid, one has to take into account the development of the potential use which for a variety of reasons, of course, can vary over time showing increases or decreases. Assuming that the number of situations people define as ‘troublesome’ could be a useful indicator for such a potential demand, we can estimate the developments relying on two population survey data from Central Bureau of Statistics (CBS). Comparing the answers given by representative samples of the population (from 16 years and older) covering the years 1992 and 1993 versus the years 1995 and 1996 - so before and after the LAA came into effect - there proved to be no significant changes in the numbers of problems respondents had experienced with respect to their labor situation, their housing, social welfare payments and other contractual relationships. In addition to this, one might have a look into figures on the recourse to lawyers by the Dutch population within the same period of time. Population survey data from the Social Cultural Planning Office (SCP) show a statistically significant decline in seeing a lawyer by 6.5% of the population in 1991 to 5.9% of the population in 1995. This decline was especially prominent among middle income people, with a decrease in use from 6.9% in 1991 to 4.7% in 1995.

These findings suggest that the 1994 Legal Aid Act had significant effects on the volume of state subsidy, provided to those with limited financial means. That leaves us with the central question whether or not those effects are designed by the Government. Answering that question is the heart of the matter of empirical evaluation research.

Empirical evaluation research is meant to render judgment about the effects of an adopted policy through comparison of the actual consequences of the legislation with the explicitly stated goals. Such research presupposes at least two conditions to be fulfilled.

The first condition is that these goals in effect are explicitly formulated by the Government. More often, however, that is not the case. Thus, in order to reach conclusive judgment, researchers often have to make crucial decisions themselves on the actual empirical content of goals as stated. This, of course, makes the research findings vulnerable to criticism afterward by the policy makers, who fault the degree of subjectivity brought in by the researchers’ operational decisions. However, such a criticism seems more properly to be re-addressed to the critics themselves.

Secondly, evaluation research assumes that the officials in charge of the policy have made deliberate choices with respect to the measures for achieving the goals they have in mind. This is too often not reality. But for this deficiency, we as social researchers might bear some responsibility having not always provided our policy makers with the relevant knowledge that they have to make those decisions. I want to stress the latter because it forces us to structure our research in a way that our findings can contribute to the situation at hand. This we will achieve into the degree our research draws on solid social theory and its cumulative body of tested general principles of social behavior which can be applied to the actual circumstances policy makers ask us to look at.

From this, it follows that we should be very persistent in conceptualizing the actual policy questions at hand in a theoretical framework that allows us to profit from the accumulative knowledge. Such a frame one might find embodied in the so called ‘boat-metaphor’ proposed by James Coleman (1990). In this theory, social changes on macro societal level are linked to changes on the micro level, i.e. the individual behavior as guided by cost-benefit considerations. It means as he points out: " where an intervention is at the level of the system, such as policy change by a nation’s government, its implementation must ordinarily occur at lower levels, and that implementation is what determines the consequences for the system". (Coleman, 1990, p.3). So we have to analyze the changes in the conditions to which individuals are responding through their (changed) behavior based on their rational cost-benefit considerations.

Following Coleman’s guidelines in analyzing the effects of our Legal Aid Act, one might start to identify the different ways in which a specific measure, as put forward by the legislation, effectively influences the conditions of the specific actor whose behavior is considered to be changed. Two different types of effect can be distinguished. There could be direct effects, induced by changed cost-benefit considerations of the relevant actor due to circumstances altered by these measures. There also can be indirect effects caused by the behavior of other actors reacting to changes brought in their behavioral conditions. Let me illustrate these different effects and their eventual relationship.

It is hard to say to what extent each of these eventual effects in fact will occur and to unravel whether they are in the end reinforcing or contradicting each other. That is in effect a matter for empirical research.

In the remainder of this paper I will focus on two research projects, both of them being a part of the ministry’s evaluation research program. The first project was related to the ‘supply-goal’ (Van Leeuwen, Klijn & Paulides, 1996), the second project was aimed at the ‘access-goal’ (Klijn, van der Schaaf & Paulides, 1999). They differ in scope. The first project, being an ex-ante evaluation, was intended to explore the conditions for successful implementation of the policy as adopted. The second study was designed as an ex-post evaluation to estimate the degree actual policy implementation had succeeded in achieving the stated objectives. As I mentioned before, both research projects also differ in their policy relevance; the first project being ‘instrumental’ to a lesser degree than the second one. As I see it, this difference is far from accidental but closely related to the degree in which the research could build upon knowledge at hand.

As I have argued extensively, Dutch sociolegal research on the patterns of the distribution of legal services over the last two decennia have seen a remarkable continuity in studies of the accessibility of legal services (Klijn, 1996). Special attention has been given to the importance of the financial costs of legal services. The common denominator of all these projects has been to find the answer to this question: To what extent does the price people have to pay for legal services effectively influence their decision whether or not to see a lawyer? The body of knowledge gained from this research allowed us - as shown in the 1994 conference paper I just mentioned - to ‘forecast’ effects from increasing price measures the Minister of Justice was about to take in the new legislation. Policy makers could, in other words, profit from social research and prevent or minimize unwanted consequences (Klijn & Van Tulder, 1994).

Much less theoretical, cumulative, and, by consequence much less ‘instrumental’ to social policy, has been sociolegal research on the supply structure of legal services. There has been little research that has tried to state specific questions in a more general theoretical frame that would allow us to foresee the probable behavioral reactions of lawyers subjected to legislative measures that in way or another change the conditions of their actual daily legal practice. By consequence, it is fair to say that policy decisions aimed at specific changes in the Dutch supply structure must be characterized to a great extent as ‘trial and error’.

For these reasons, I will present both projects in reverse chronological order.

The official wording of the access goal - as stated in the explanatory memorandum to the bill - states that the legislation’s is to ‘offer entitlements’ to those who have insufficient financial means to pay the costs of legal services themselves (cfr. Ministry of Justice, 1998, par. 4.1). When the authors of the evaluation report describe the foreseen effects of this legislation, they put it in a somewhat different way, stating that goal to be a "reassessment of access" (par 5.1.1.). The latter is, of course, more accurate, considering the measures put forward by the legislation with respect to this goal: Income limits for entitlement were lowered, substantive criteria for legal problems covered by legal aid were tightened, the fees to be paid by applicants were raised, and the applications submitted were checked in a stricter way. These measures all point into one and the same direction: increasing the prices people have to pay for legal services. Stated in more theoretical way: the new legislation deliberately changes people’s cost/benefit considerations by increasing their costs.

However, different from previous legislative interventions which were justified by budgetary constraints only, this legislation was aimed at ‘broader’ goals. First, this legislation was assumed to stimulate people’s responsibility and accountability as well force them to a more scrupulous reassessment of their ‘need’ to rely on public money for solving their private problems. According to this reasoning, people may decide either to look more for other strategies than using a lawyer, or they may continue to look for lawyers’ services but at their own expense. However, as the Minister of Justice stated explicitly in the explanatory memorandum, access to law (i.e. to lawyers) is still to be considered as a fundamental right implying that access should not be primarily hindered by financial thresholds. Secondly, changing the actual fee-tariff would bring more equality under the law between single applicants at the one hand and those who were married or cohabiting at the other. Under the previous scheme, the former were evidently ‘favored’ by comparatively lower fees compared to the latter. Changing the tariff would mean a contribute to the government’s maxim: ‘taxation according to his/her financial abilities’. Analyzing the actual effects of the respective measures, we referred to both of these goals as an increase in the selectiveness, respectively an increase in equality.

Analyses of the legal aid registration data over a three year period of time (1994-1996) shows that the numbers of granted certificates in civil & administrative law have decreased with 38%. . Part of this effect results from the decrease in entitlement to subsidized legal services which do not necessarily imply abstaining from lawyers’ services at all because people can still decide to seek a lawyer at their own expenses. Based on population data analysis we estimated an 24% decrease in people’s recourse on lawyers. As mentioned above (and earlier elsewhere) we also were able to show that the decrease in use of lawyers’ services varied among income classes, showing substantial lower figures among those income classes which were confronted with severe increase of prices leaving both the lowest and the highest income classes unchanged. All these findings corroborating our price-based assumption - everything else kept constant, decrease in use will be seen in according to the degree of increase of financial costs - are to be considered as direct effects of the legislation.

However, before we could answer the crucial question to what extent these findings underline the Ministers’ claim, the legal aid act having contributed to people’s selectiveness, two additional questions need still to be answered. First, one wants to know whether or not the decision not to see a lawyer can be thought to have been a reasonable one. That means that the decision was taken under such circumstances that left the people involved, at least any room to decide. Because when problems can hardly be considered to be resolved without recourse on law, abstaining from lawyers’ help might indicate the opposite. Skipping a detailed account of our empirical analysis - while already presented earlier (Klijn, 1997 - we concluded that there proved to be some indications that the legislation, having caused some unwanted effects by creating financial thresholds, especially in social security law and to a lesser degree in labor law.

The second question that still has to be answered regards the indirect effects from measures taken by the new legislation. Because we took it for granted that the higher prices people have to pay are, in effect, charged to them by those who are due to do so: the legal services providers as well as the administrative professionals at the Legal Aid Boards. All of them are considered to be relevant actors in applying the ‘law in the book’. But do they or would they have reason - based on their own cost-benefit calculations - to divert from the book bringing ‘law into action’?

With regard to the decisions to be made by the Boards’ administrators, given differences in managerial styles allows them less or more own discretion in deciding on not standardized and computerized issues in the application form. There might be differences in the pattern granting certificates causing differences in prices as charged to litigants. Our analysis proved that not to be the case. Neither were we able to trace such effects caused by the alleged ‘exit’ from lawyers no longer handling legal aid clients because of the level of remuneration which they held still falling far below decent standard. We argued that if such an exit would be substantial in numbers, it could in effect cause litigants to pay higher prices forcing them to overcome greater distances. By consequence the above reported decrease would be partly ‘too’ high because of this unforeseen as well as unwanted behavior. But, as far as we could see, there has not been such an exit.

Although not very substantial in volume, we found some remarkable effects causing some price evading by the Legal Advice Centers’ lawyers. They seemed to exceed the time limits to their cheaper services which, if they did, would force them to charge clients in the same way as private lawyers have to do. Legal Advice Centers’ lawyers behaved in that way partly because they - as the private practitioners - hate the increased paperwork that comes with it, partly because they prefer the ‘one touch, one play’ principle instead of referring clients to private practitioners. Both these considerations increase their own ‘benefits’ as well as those of their clients, bringing them a lower price to pay. In addition some Legal Advice Centers, fearing their constituency would be overcharged by the new legislation, started delivering full services charging clients a moderate hourly-based as compared with the actual fee rates in Dutch private practice especially middle income class clients profit from this new type of supply while the prevailing legislation would confront them with considerably higher prices. Even, not substantial in volume, these effects seem important to be mentioned pointing to the interdependencies in behavioral responses from the relevant actors to the law as ‘black dots on white paper’.

So, we have to conclude that the new legislation holds its promise on the increase of selectiveness, what about the equality promise? Has this come true? For reasons of space available I will summarise this. Referring to the predictions Van Tulder and the present author made in the paper mentioned before, one would expect still inequality to have left caused by the unbalance in the fee tariff at the expense of the middle class applicants. From the analyses we made, it follows that this inequality in essence seems to be still there. So, there still seems to be a shortfall on the direct effects, compared to the intended effects.

However, as far as the legislation has changed the fee tariff by putting up the price to pay by single applicants, the given promise proved to be held. In contrast with the situation in former days showing a much higher recourse on lawyers by single persons, the new legislation leaves us with a comparable level of recourse on lawyers by single as well as non-single applicants. All in all: mixed results as far it comes to improving equality.

The second mentioned goal of the Legal Aid Act focuses on the supply side of the provision. At first glance, the phrasing of this goal seems rather unspecific, stating only that ‘sufficient supply’ had to be guaranteed. However, considering the measures related to this goal, one would be tempted to argue that the legislation deliberately seems to have focused on the quality issue. Two different measures are relevant for consideration in relation to this goal.

First, the one most explicitly related to this issue is the introduction of the so called ‘quality requirements’. By law, lawyers who wish to provide their services under the legal aid provision are required to register at one of the five Legal Aid Boards. These Boards were empowered to set some requirements lawyers have to fulfill. According to the law such requirements relate to the minimum and the maximum number of legal aid cases a lawyer must/may handle, the lawyer’s professional expertise, the organization of his/her office, and the way records of practice is kept (cfr. art. 15 LAA). Although the explanatory memorandum argues, these requirements are expected to improve efficiency as well, one may wonder whether these requirements would do, given the measures already put forward under that heading. We safely may consider them as primarily focused to ‘make things better’.

Secondly, related to this goal is the increase in lawyers’ remuneration. Here one might have to recall the austerity policy adopted by the Dutch government in the recent past. Notwithstanding severe opposition from the Dutch Bar Association, remuneration levels were lowered in the mid eighties. The new legislation increased that remuneration level by 25% across the board. This was done to implement a promise made to the Bar in 1989 to create an ‘adequate climate’ for loyal co-operation.

What about the effects of both of these measures? Actually, there is no empirical research which provides us with conclusive findings we could discuss. However, that has not prevented all the parties from being engaged in a debate which seems to be tuned in minor. To put it in a somewhat provocative way: The legislation has caused a drain in the numbers as well as the quality of lawyers willing to take legal aid clients. What to think about?

As the Ministry’s evaluation reports states that the years prior to the year 1994 have not signaled a shortage in lawyers willing to handle legal aid cases, this statement may be misleading to some extent; unintentionally I suppose. Partly it suggests that we can rely on figures on that topic, which we can’t. Partly also because such a statement seems to neglect the persistent claims about the ‘silent dying’ of adequate supply in numbers as well in quality. There was a widely spread ‘feeling’ that the legal aid bar - especially the ‘social advocacy’ i.e. lawyers whose practice substantively relies on legal aid - was not any longer in the same healthy shape as it was in the late Seventies. The findings of our 1993 survey under the Dutch Bar showing the actual participation rate to be about 80% - compared to 95% in 1979 - seems to underline those impressions. But actually, as we argued explicitly, the sheer number of lawyers available for legal aid clients had doubled because of the immense growth of the total bar. With regard to the ‘social advocacy’, although their share relatively spoken has decreased over the years - from 16% in 1979 to 8% in 1993 - their sheer numbers seems to have been stabilized. And, in addition to this, it also would be hard to argue in favor of a causal link if one considers the meaning of the increase of lawyers’ remuneration in terms of real income as table 3 does.

* Hfl, in 1993 prices excluding client fees

As shown, the legislation actually seems to have brought the lawyers ‘back on track’, i.e. recovering the loss they suffered from the austerity policy during the eighties. Even if one admits that there is no real increase in lawyers’ incomes at all - the case the Bar Association rightly and for understandable reasons still holds - it is difficult to argue the exit-thesis. If there was any trend downwards in lawyers’ participation - it could easily be that the 1993 participation rate presents the unknown rate already there in the late eighties - this legislation should eventually have caused a halt to that trend, or having had no impact at all, stabilizing the status quo. As far as we could rely on actual registration figures - including all lawyers who in effect applied for certificates - the ‘stabilization’-thesis seems most likely: 1994: 6563; 1997: 6491; 1998: 6602 .

The Legal Aid Act empowers the Legal Aids Boards to conceive their own requirements’ policy, applying them after they have been approved by the Ministry of Justice. However, with regard to the first year that the legislation would be in effect, the Minister of Justice would formulate a ‘provisional ruling’ that it would be applied nationwide by each of the Boards. Having consulted both the Dutch Bar Association and the Legal Aid Boards themselves, such a ruling was put forward (in the autumn of 1993). It stated that those lawyers would be allowed to register, who - based upon the previous year - were supposed to handle at least 10 certificates whereas at the same time they would not be allowed to handle more then 250 in the coming year. Only for some specific areas of law some substantive criteria - supposedly probing lawyers’ expertise on this areas of law - were set. Each of the Legal Aid Boards, however, had some discretion to divert from them if local conditions would force them to do so. All this was mailed to the lawyers around October 1993 informing them about the ‘law in the books’.

Given these requirements, essentially one question started our research project: What kind of consequences would these requirements have on the supply? Would they improve the quality of the services delivery, and, if not, what kind of criteria would be more preferable? According to Coleman’s theory, we would expect requirements to be adequate to the extent that they would be in agreement with the ways lawyers think about ‘quality’ and their ways to assure the quality of their services in their daily practice. Because, if the requirements were not, any criteria proposed by the government (or the Legal Aid Board) would bear great risk not to match the organization of lawyers’ practice and consequently being perceived by the lawyers as only an increase of external control. Stated in our theoretical way: to the extent that measures would only increase lawyers’ costs, they most probably would not respond in the way as assumed by the legislation. Consequently, we focused our research on lawyers’ actual practice respective, their opinion on criteria which would adequately probe ‘quality’ and their perception of actual criteria as well. This was done by a postal survey among a random sample of the members of the Dutch Bar (excluding only those who had started their practice less then 1 year before the time the survey was conducted - i.e. first months in 1994).

In order to understand lawyers’ perception of criteria which they perceived to probe ‘quality’ in an adequate way, we asked them to rate several possible options eg a minimum number of certificates per year, a minimal percentage of time spent a year on a specific area of law, a minimum number of cases on a specific area of law, a restriction on the number of areas of law and a lawyers’ investment in upgrading their actual abilities. From their answers, we were able to conclude that especially lawyers’ time investment in specific area’s of law - i.e. specialization - as well as his/her investment in upgrading - i.e. permanent education - were perceived as being far the most adequate. We then proceeded in constructing 6 hypothetical standards that might eventually be applied by the Legal Aid Boards. These standards respectively referred to:

1. a minimal percentage of time a lawyer actually invests in a specific area of law;
2. a minimum number of certificates in a specific area of law a lawyer is allowed to handle;
3. a maximum number of areas of law a lawyer actually focuses his/her attention in her/his legal practice;
4. a maximum number of areas of law a lawyer is allowed to handle certificates;
5. a lawyers’ investment in permanent education;
6. lawyers’ most recent investment in professional training.

As you might recognize, the first four of these are focused on ‘specialization’, whereas the other two standards tend to grasp with the concept of ‘permanent education’. Based on statistical simulations we then estimated these standards’ probable effects on a number of issues among others: their support among the Bar, the numbers of lawyers who actually would meet the standard as well as their professional abilities, and, the implementation costs the Legal Aid Boards would have to face applying them. Specialization standards based on lawyers’ total practice (1 & 3) appeared to contribute most strongly to the quality of the supply. The latter we estimated by computing the numbers of lawyers who, according to their own assessment, did lack the required competence on specific areas of law. We also estimated the degree of ‘specialism’ that would be lost by applying a standard. To the degree that a standard to be applied would decrease the number of ‘incompetent’ lawyers on the one hand and would not affect the number of ‘specialists’, that standard seems to be adequate to the purpose at hand. At the same time, however, their implementation costs seem to be higher too. This is because their feasibility strongly depends on lawyers’ willingness to provide the information to the Legal Aid Boards which would be unable to get the information they need otherwise. The same goes for the standard which refers to professional training. From this we concluded that those requirements which seem to be most adequate were at the same time the most ‘unrealistic’ given the required co-operation from the local Bars in order to make their application possible. Such a co-operation was highly unlikely since the relations between the Boards and the Bar were rather tense at that time.

What did the Legal Aid Boards do? How they applied the rules? Effectively they ‘neglected’ them to quite a large extent, especially as far as it concerns the minimum number of certificates lawyers were supposed to do. Their registration figures proved that about one third of all lawyers registered in 1994 were supposed to handle less then the required minimum 10. In effect, one could conclude that compared to the situation at hand, nothing had changed in 1994. This statement in essence still stands today.

How to explain the behavior of the Boards? There is a simple answer: the Boards disliked the consequences to be expected from enforcing these rules. We tried to compute these effects, based on the assumption that lawyers would not change their mind overnight. This means that we could consider their behavior in the recent past as the best guess about their behavior in the near future. Concerning their numbers, even in areas of law where state subsidy is very preeminent, such a minimum standard would leave the Boards with 50% to 70% of the actual lawyers as ‘not qualified’. With regard to an eventual ‘quality’ drain, the effects seem pretty good, but still on some areas of law there might be some unwanted decrease of lawyers. So, why should the Boards prevent lawyers from being registered if they themselves want to be, knowing that such a decision would not only contribute to the ‘tense’ relationships with the local Bars, but also could end up in the loss of lawyers whose abilities sometimes may be needed? This was only for reasons of a rule on paper, from which no-one was assured that it would help. Better to wait and see!

However, there might be some change in the very near future. Some Legal Aid Boards have decided to divert from the individual requirement- model and to contract specific law firms enabling them to handle the number of certificates agreed upon ‘on their own terms’. This means that these law firms themselves are responsible for assigning the cases to the lawyer which they consider to be capable to handle specific cases. Instead of the ‘control in advance’ - the type of control the law is based upon - the Boards will be allowed some kind of ‘control afterwards’. In essence, such a policy means a fundamental change in the type of governance: from steering along the way of external, bureaucratic control to steering based on an internal, professional control. This at least seems to be in full agreement with the findings of our research regarding lawyers’ acceptance of such a quality control.

Both of our research projects have shown the importance to take account of the ‘indirect effects’ from policy in order to understand the ‘social working’ of law. Rules enacted, have to become real incentives, factors that influence daily life decisions. They always work indirectly: by the way of behavior of actors who have to adjust these factors in the setting that already exist enstructures their behavior. Instead of asking ‘how the law works’ it may be more accurate to ask ‘ how the actors involved work with the law’.

This is not merely some kind of ‘playing with words’; it helps us to understand how ‘law behaves’. Let us look back to the behavior of the Legal Advice Centers’ lawyers in their ‘refusal’ to stick to the letter of the law when ‘time was up’, i.e. when they had spent two hours on a case but not yet able to finish. What to do? They did what they did most of the time before the legislation came into effect. Relying on their (although not very conclusive) statistics, in previous days they dealt with most of their cases in about three hours. However, for one or another reason the Legal Advice Centers were not able to convince our policy makers that such a time span was an adequate one. In the end, they lost as the rules in the books were stated otherwise. We know what happened: they continued their behavior. And what happened next? Well, as the evaluation report of the Ministry of Justice remarks: the law in the book was changed according to the behavioral wishes of the Legal Aid Centers’ Lawyers (Ministry of Justice, 1998, par. 5.1.3.4). And remember the change in the requirement policy some Legal Aid Boards seem to consider. This change is in essence a neglect of the rules in order to get closer to the goal the original rules were supposed to achieve. They might be right, and if so, who would care about the rules?