The Scottish Government

« Previous | Contents | Next »

Listen

Chapter 4: Statistical Analysis

4.1 This chapter of the report presents a statistical analysis of data relating to case trajectories in the High Court for the periods prior to and following the introduction of the reforms. The way in which the data was collected and analysed was described in detail in chapter 3.

4.2 The analysis presented in the chapter is generally at the level of the individual indictment. The existence of cases in which there are multiple accused always raises the question of whether each individual accused should be treated as a whole case or only a fraction of a case. The main problem with the former approach is the distortion of results: where a case involving seven accused is adjourned, for example, this would result in seven separate adjournments being recorded and the risk that the incidence of adjournments in the sample is over-reported. Thus the approach taken here was (unless otherwise stated) the latter approach. This presents a fresh problem of how to record outcomes in multiple accused cases where the fate of each accused differs ( e.g. in a case involving two accused where one accused pleads guilty at the first preliminary hearing, whereas the other is found guilty at a subsequent trial diet). The approach taken to this problem differed according to the context and, where the issue arises, the approach taken is described and explained in the relevant section of this chapter.

4.3 The percentage totals in the tables that follow do not always total 100 due to rounding.

The Sample Profile

4.4 As chapter 3 has already outlined, two samples were collected: a 'pre-reform' sample, comprising all High Court cases where the indictment was registered and the case concluded between 1 April 2004 and 31 March 2005 ³¹ and a 'post-reform' sample, comprising all High Court cases where the indictment was registered and the case concluded between 1 April 2005 and 31 March 2006.

4.5 An initial comparison that can be made is to examine the cases in our samples as a proportion of the total number of indictments registered in the 12 month period in question. Table 4.1 presents the results of this exercise.

Table 4.1: Registered and concluded indictments

4.6 A number of points can be noted from table 4.1. First, it shows that the number of indictments registered during the first 12 months following the reforms was lower than the number registered in the 12 month period immediately prior to the reforms (1104 compared to 1234). This is almost certainly due in part to the increased sentencing powers of the sheriff courts and a consequent shift of at least some business from the High Court to the sheriff courts. This issue is examined further in chapter 10.

4.7 Second, it shows that the proportion of indictments registered during the 12 month period that also concluded during that 12 month period is roughly similar: 69.4% in the pre-reform sample compared to 64.2% in the post-reform sample. ³² This is an important point for our analysis as it shows that in drawing comparisons between the two samples (as we go on to do in the remainder of this chapter) we are comparing like with like.

4.8 Tables 4.2 and 4.3 present the profile of the two samples in terms of the most serious charge contained in the indictment and the proportion of cases involving multiple accused respectively. In table 4.2 below, which presents the sample profile in terms of the main charge in the indictment, where the case involved multiple accused, and the accused faced different charges, the most serious of the charges was selected on the basis that this would serve as an indication of the complexity of the case.

Table 4.2: Main charge on indictment

Table 4.3: Number of accused per indictment

4.9 Suffice to say, tables 4.2 and 4.3 indicate that the profile of the pre-reform and post-reform samples, while not identical, are sufficiently similar such that comparisons between the two datasets can validly be made.

Case Disposal Point

4.10 This first analysis examines the point at which each case in the two samples was disposed of. Case disposal here (and indeed for the remainder of this chapter) means one of the following outcomes: the accused's substantive guilt or innocence was determined (either by a plea that was accepted by the Crown or by the case resulting in a guilty, not guilty or not proven verdict); the case was deserted; or a warrant was granted for the apprehension of the accused. ³⁴ No account is taken here of any subsequent sentencing proceedings. Where the case involved multiple accused, the disposal point was taken to be the point at which the last remaining accused was dealt with ( e.g. in a case involving two accused, with one pleading guilty at the first preliminary hearing and the other being found guilty at the second trial diet, the disposal point was taken to be the second trial diet).

4.11 Table 4.4 displays the point at which the case was disposed of for both the pre-reform and post-reform samples.

Table 4.4: Case disposal point

4.12 The first point to note from table 4.4 is the increased proportion of cases which were disposed of by way of a section 76 hearing in the post-reform sample (6% in the pre-reform sample, compared to 19.2% in the post-reform sample). While this may be attributable in part to Du Plooy v HM Advocate, ³⁵ where the appeal court endorsed explicit sentence discounts for guilty pleas (and especially those made at an early stage in proceedings), ³⁶ this cannot be a complete explanation. Du Plooy was decided on 3 October 2003, well before the reforms came into effect, and if this alone was the influential factor in encouraging guilty pleas under section 76, one would have expected to have seen a greater number of such pleas in the pre-reform sample.

4.13 What we may be seeing here is the influence of a combination of factors. First, it might simply have taken a while for the implications of Du Plooy to filter down to participants in the criminal justice process, meaning that its effects only started to become apparent in the post-reform sample. Second, it may be that the wider range of judges who now preside over section 76 hearings is a factor. In the past, section 76 hearings were always presided over by the Lord Justice-Clerk and one particular occupant of this position was perceived as a heavy sentencer, which perhaps acted as a disincentive to accused who might otherwise have pled guilty under section 76. ³⁷ Third, it may be that the flat fee of £1,250 paid to counsel for an accelerated guilty plea under section 76 is having some influence in encouraging the use of section 76. ³⁸ Fourth, and finally, it may be that improved disclosure from the Crown to the defence means that the accused is more likely to be aware of the strength of the case against him or her and therefore is able to take a more informed decision about the benefits of pleading guilty.

4.14 A second point to note from table 4.4 is that the proportion of cases disposed of at a trial diet has massively decreased in the post-reform sample. Of the 857 cases in the pre-reform sample, 806 proceeded to a trial diet, whereas of the 709 cases in the post-reform sample, only 236 proceeded to a trial diet. Expressed in percentage terms, this represents a change from 94% to 33.3%. This is hardly surprising, as it is now possible to dispose of cases at a preliminary hearing which, of course, it was not possible to do prior to the reforms. It is also some indication, however, that preliminary hearings are proving effective in avoiding the need for a trial diet.

Eventual Case Outcome

4.15 Table 4.5 below displays the nature of the outcome for all of the cases in the two samples. In cases where there were multiple accused, the eventual case outcome was sometimes different for each accused ( e.g. one pled guilty but the other was found guilty following a trial). For the purposes of this analysis, each accused has been treated as a fraction of the case as a whole. For example, in a case involving four accused, each accused was treated as 0.25 of the case; in a case involving two accused, each accused was treated as 0.5 of a case. The reason for this is to avoid over-representing the outcomes of cases involving multiple accused. If, in a case involving nine accused, all of them were found guilty, to count this as nine separate outcomes would distort the results.

Table 4.5: Case outcomes

4.16 Table 4.5 indicates that the nature of case outcomes has changed very little between the two samples. It is perhaps slightly more common in the post-reform sample for a case to conclude with the accused pleading guilty as charged (26.4% of cases in the post-reform sample compared to 22.7% of cases in the pre-reform sample) - this is consistent with the increased proportion of cases being settled at a section 76 hearing in the post-reform sample (see table 4.4 above). By and large, however, the nature of case outcomes is very similar between the two samples. Likewise, one thing that can be drawn from table 4.5 is that the proportion of cases in which a trial actually took place (as opposed to the case being settled by a guilty plea or deserted in some form) has changed very little. In the pre-reform sample it was 27.1%, whereas in the post-reform sample it was 25.3%. What has changed (and this is the important point), is that in the post-reform sample it was unusual for a guilty plea to be tendered at the trial diet and this issue is examined below.

Guilty Pleas: When Were They Tendered?

4.17 Table 4.6 below is concerned only with those cases that were settled by a guilty plea and it shows the point at which the guilty plea was tendered. In multiple accused cases, where guilty pleas were tendered at different times by the accused involved, the latest point at which a guilty plea was tendered was recorded ( e.g. in a case involving two accused, where one pled guilty at a section 76 hearing and the other pled guilty at the first trial diet, the disposal point is recorded as the first trial diet).

Table 4.6: Disposal point (for cases that concluded with a guilty plea only)

4.18 The main point to be drawn from table 4.6 is that it is unusual in the post-reform sample for a guilty plea to be tendered at a trial diet. Only 6.9% of guilty pleas were tendered at a trial diet in the post-reform sample. This provides a strong indication that preliminary hearings are operating effectively in encouraging the tendering of guilty pleas at a stage earlier than the trial diet, saving court time and all the attendant inconvenience and distress this may cause to any witnesses who have attended unnecessarily.

4.19 Table 4.6 also shows very clearly that it was far more common to tender a guilty plea at a section 76 hearing in the post-reform sample (where 30.7% of guilty pleas were tendered at a section 76 hearing) compared to the pre-reform sample (where only 10.3% of guilty pleas were tendered at a section 76 hearing.

4.20 One further point might be drawn from table 4.6. If one looks at the proportion of guilty pleas in the pre-reform sample that were tendered at either a section 76 hearing or at the first trial diet, this is almost identical to the proportion of guilty pleas in the post-reform sample that were tendered at either a section 76 hearing or the first preliminary hearing (77.9%, compared to 77.5%). All of these individuals might be termed 'early pleaders' as they pled guilty either at the first scheduled diet (the first trial diet or the first preliminary hearing) or they chose to make arrangements to plead even earlier than this, by asking for a special section 76 hearing to be arranged. What is clear, then, is that the proportion of 'early pleaders' has not changed, which suggests that there is a relatively stable proportion of accused who are pre-disposed to plead guilty at the earliest possible opportunity (or conversely, a relatively stable proportion of accused who will always wait until the last possible opportunity to plead, no matter what incentives are on offer for early guilty pleas). What has changed, as was noted above, is that more of these 'early pleaders' are choosing to plead guilty at a section 76 hearing, rather than waiting until the first scheduled diet. This seems to be a further indication that, where it is possible to influence behaviour, incentives designed to encourage the early tendering of guilty pleas are proving effective.

Number of Court Diets/Adjournments

4.21 Prior to the reforms, a case could be subject to three main types of event: a procedural hearing (either an application for a postponement of a trial diet under section 80, a section 275 hearing to deal with applications relating to sexual history evidence, or a hearing under section 72 to deal with any other procedural issues); an accelerated hearing to tender a guilty plea under section 76; or a trial diet. Following the reforms, matters which were previously dealt with at a section 72 hearing or a section 275 hearing are now heard at a preliminary hearing. Thus the three types of court diet that could take place subsequent to the reforms are preliminary hearings, section 76 hearings and trial diets. ³⁹ Applications for postponement of court diets which were previously made under section 80 and required the parties to attend court are now dealt with under section 75A, which does not require a court appearance where all the parties join in the application.

4.22 The aim of this section is to establish the impact the reforms have had on the average number and type of court hearings per case. To this end, table 4.7 below displays the average (mean) number of various types of court diet in both the pre-reform and post-reform samples. It also displays the average (mean) number of trial diet adjournments per case.

4.23 A number of points can be drawn from table 4.7. The most important relates to the row highlighted in bold. This calculates the average number of court hearings per case (including all types of court hearing that could possibly happen). What this shows is that the average number of court hearings per case has hardly changed at all following the reforms (2.06 compared to 2.03). ⁴⁰ What has changed is that in the post-reform sample the predominant type of hearing is the preliminary hearing whereas in the pre-reform sample the predominant type of hearing was the trial diet.

4.24 A number of further points are worth noting from the table. First, in the pre-reform sample, the average number of trial diets per indictment was 1.5, whereas in the post-reform sample it was 0.4, confirming the point made in relation to table 4.4 earlier that cases are far less likely to proceed to a trial diet in the post-reform sample.

Table 4.7: Average (mean) number of court hearings per case

Key to table: TD=trial diet; PH = preliminary hearing

4.25 Second, the table displays the average number of adjournments of trial diet per case and this has decreased enormously, from 0.56 in the pre-reform sample to 0.06 in the post-reform sample. This can also be seen in table 4.8 below, which displays the total number of trial diets in each case in the pre-reform sample and the post-reform sample. As table 4.8 shows, in the pre-reform sample 32.6% of cases had two or more trial diets ( i.e. the trial diet was adjourned at least once), whereas for the post-reform sample, this figure was 4.5%.

Table 4.8: Total number of trial diets per case

4.26 On the other hand, table 4.8 does not display the full picture for the post-reform sample because it does not indicate how many preliminary hearings were necessary before the trial diet. Table 4.9 below provides a more complete view. It focuses on the 28.8% of cases in the post-reform sample (204 cases) that were settled at first trial diet and shows the number of preliminary hearings that occurred before that trial diet.

Table 4.9: Total number of preliminary hearings (cases disposed of at first trial diet)

n=204

4.27 As table 4.9 shows, the majority of cases that settled at the first trial diet required at least two preliminary hearings before that trial diet (57.9% of cases). This provides further support for the finding, discussed in relation to table 4.7 above, that what has changed as a result of the reforms is not so much the total number of court diets held, but their nature: from trial diets to preliminary hearings.

Outcome of all Callings

4.28 Further analysis of outcomes has been undertaken at the level of each individual calling of the case. Only substantive hearings were included in this analysis, that is, trial diets, preliminary hearings and section 76 hearings. Purely procedural hearings (in the pre-reform sample, section 80, section 275 and section 72 hearings) were not included (although these were analysed above in table 4.7).

4.29 Two analyses of the pre-reform sample were carried out. The first was undertaken on all substantive hearings - trial diets and section 76 hearings. The second was undertaken on trial diets only. In the first analysis, which includes trial diets and section 76 hearings, the 857 indictments in the pre-reform sample called 1331 times in total. Table 4.10 displays the outcome of these callings.

Table 4.10: Outcome of all callings of indictments (pre-reform sample, including section 76 hearings)

n=1331

4.30 In the second analysis, which is an analysis of trial diets only, the 857 indictments in the pre-reform sample generated a total of 1280 trial diets. Table 4.11 displays the outcome of these trial diets.

Table 4.11: Outcome of all trial diets (pre-reform sample)

n=1280

4.31 As table 4.11 shows, the most common outcome of a trial diet, prior to the reforms, was for that trial diet to be adjourned (this was the outcome of 37.1% of trial diets). The second most common outcome was for a guilty plea to be tendered (34.8% of trial diets). Neither of these outcomes is particularly desirable when it occurs at a trial diet. An adjourned trial diet is a waste of court resources and, especially if cited witnesses have not been countermanded prior to the day of the trial, a cause of witness inconvenience. Even if witnesses were countermanded in advance of the trial diet, they may still have been inconvenienced by having made arrangements to be present at the trial. A guilty plea tendered at a trial diet likewise causes witness inconvenience that could have been avoided. It also means that, unless additional trials have been scheduled for that slot, or other business can be found to fill the gap, court time that could have been utilised is wasted.

4.32 An analysis of the outcomes of all court diets was also undertaken for the post-reform sample. Here, the analysis was undertaken separately for all preliminary hearings and for all trial diets, as it makes little sense to try and combine these different types of diet, given that their outcomes can - and should - be of an entirely different nature.

4.33 Taking the trial diets first, the 709 indictments in the pre-reform sample generated a total of 284 trial diets. Table 4.12 displays the outcome of these trial diets.

Table 4.12: Outcome of all trial diets (post-reform sample)

n=284

4.34 Comparing tables 4.11 and 4.12, it can immediately be seen that there have been considerable changes in the outcome of trial diets between the pre-reform and post-reform samples. In the pre-reform sample, only 18.1% of trial diets resulted in a concluded trial, compared to 63% of the post-reform trial diets. Adjournments were less common in the post-reform sample (16.2% of trial diets resulted in an adjournment, compared to 37.1% of the pre-reform trial diets) and guilty pleas at the trial diet were also less common (11.3% of trial diets in the post-reform sample concluded with a guilty plea, compared to 34.8% of the pre-reform trial diets). This is a strong indication that preliminary hearings are working effectively.

4.35 A separate analysis was undertaken of the outcomes of all preliminary hearings in the post-reform sample. The 709 indictments in the post-reform sample generated a total of 1020 preliminary hearings. Table 4.13 displays the outcome of all of these preliminary hearings.

Table 4.13: Outcome of all preliminary hearings (post-reform sample)

n=1020

4.36 As table 4.13 shows, the most common outcome of a preliminary hearing was the continuation of that preliminary hearing to a further preliminary hearing (42.8% of preliminary hearings). A further 26.7% of preliminary hearings concluded with a guilty plea, with 23.4% concluding with the trial date being set. The issue of continued preliminary hearings is analysed further below. ⁴⁶

Length of Trial

4.37 In 232 of the 857 cases in the pre-reform sample and in 179 of the 709 cases in the post-reform sample a trial concluded. Table 4.14 indicates the length of the trial in these cases. ⁴⁷

Table 4.14: Length of trial

4.38 The findings of this analysis can also be presented in a slightly different way. In the pre-reform sample, the shortest trial in the sample lasted for 2 days and the longest for 48 days. The average (mean) length of trial was 5.85 days; the median length was 4 days. In the post-reform sample, the shortest trial in the sample lasted for one day and the longest for 44 days. Here, however, the average (mean) length of trial was 4.88 days; the median length was, as in the pre-reform sample, 4 days.

4.39 What can be concluded from these figures is that in those cases where a trial did take place, trials in the post-reform sample tended to be slightly shorter. As well as the average length of trial being a little shorter in the post-reform sample, it was also less common in the post-reform sample for a trial to last for more than 20 days (1.1% of trials, compared to 2.6% of trials in the pre-reform sample).

4.40 While these differences are not particularly striking, it is worth pointing out that they cannot be accounted for by the cases in the post-reform sample being less complex; in fact if anything, the cases in the post-reform sample that went to trial were more complex than those in the pre-reform sample. ⁴⁸ For example, in the pre-reform sample, 19.9% of cases that ended with a concluded trial involved multiple accused whereas in the post-reform sample, this figure was 22.8%. Likewise, 30.6% of cases that ended in a concluded trial in the pre-reform sample were murder or attempted murder cases, whereas the figure was 36.9% for the post-reform sample. The slightly lower average trial length in the post-reform sample might be accounted for at least in part by an increase in the use of the agreement of evidence provisions of the 1995 Act, thus lessening the number of witnesses who are required to give evidence at trial, although this is not something that can be concluded with any certainty. ⁴⁹

Reasons for Adjournments

4.41 As tables 4.11 and 4.12 indicated earlier, 475 (37.1%) of trial diets in the pre-reform sample and 46 (16.2%) of trial diets in the post-reform sample resulted in an adjournment on the day of the trial. Table 4.15 shows the reason for the adjournment of these callings. ⁵⁰ The analysis is not a perfect one by any means. Information about the reason for the adjournment was available for only 330 of the 475 trial diets that were adjourned in the pre-reform sample and 40 of the 46 trial diets that were adjourned in the post-reform sample. Thus table 4.15 is an analysis only of the 330 trial diet adjournments in the pre-reform sample and the 40 trial diet adjournments in the post-reform sample for which this data was available.

4.42 These figures should be interpreted with some caution due to the very low number of cases in the post-reform sample, where trial diet adjournments were relatively rare (and where information on the reasons for these was not always available). That said, the table does indicate a shift in the reasons for trial diet adjournments between the two samples. The most common reason for a trial diet to be adjourned in the pre-reform sample was that one of the parties had inadequate time to prepare (63% of trial diet adjournments for which information was available). This was the reason for only 15% of trial diet adjournments in the post-reform sample, where the most common reason for an adjournment was that an essential Crown witness was absent (40% of trial diet adjournments for which information was available). This is not surprising. The earlier provision of information to the defence from the Crown and the introduction of mandatory preliminary hearings post-reform should indeed have led to a reduction in trials adjourned because of a lack of time to prepare. It must also be stressed again that because adjournments of trial diets were rare in the post-reform sample, the numbers are very small - the 40% of adjournments that were due to a Crown witness being absent equates to only 16 actual adjournments.

Table 4.15: Reasons for adjournments on the day of the trial

4.43 All of this would seem to suggest that the reforms have been very effective in cutting down on the number of what might be seen as 'preventable' adjournments. It is near impossible to avoid trial diet adjournments entirely. Some types of adjournment are very difficult to prevent - the accused who suddenly becomes unwell or the essential Crown witness who, despite giving assurances that he or she is available for the trial, simply fails to turn up. Others are more easily preventable - the most obvious example being adjournments because the defence or Crown had inadequate time to prepare their case. It is the 'preventable' adjournments that are very rare in the post-reform sample, as table 4.15 shows. In this sense, preliminary hearings are doing their job - especially in ensuring that cases are not adjourned at trial (with the attendant problems of witness inconvenience) simply because one or both of the parties has not yet prepared their case. The higher proportion of 'lack of court time' trial adjournments post-reform may be a consequence of fixed trials being assigned initially. Any trial over-running could cause the adjournment of another. As the post-reform year progressed, it was more common for floating trials to be assigned.

Continued Preliminary HearingsS

4.44 It was established in table 4.13 that the most common outcome of a preliminary hearing was for it to be continued to a further preliminary hearing. This section of the chapter explores further the issue of continued preliminary hearings in the post-reform sample.

4.45 Table 4.16 shows the total number of preliminary hearings per indictment for the post-reform sample. For the purposes of this analysis, cases where the accused pled guilty at an accelerated diet under section 76 ( i.e. cases where no preliminary hearing was held) have been excluded.

4.46 As table 4.16 shows, 38.7% of cases in the post-reform sample (excluding those where the accused pled guilty at a section 76 hearing) had two or more preliminary hearings, including 5.2% of cases in the sample that had five or more preliminary hearings. It is likely, furthermore, that this is an under-estimate of the incidence of continued preliminary hearings, especially in cases indicted towards the end of the sample period. The majority of those interviewed were of the view that continued preliminary hearings had become more prevalent over time. If this is an accurate perception, it would not necessarily be reflected in our data, which only covers cases indicted and concluded before the end of March 2006. Thus our analysis may under-represent the incidence of continued preliminary hearings towards the end of that period, because cases with multiple preliminary hearings are, by definition, less likely to have concluded. ⁵¹

Table 4.16: Total number of preliminary hearings per indictment

n=573

4.47 Table 4.17 examines the average (mean) number of preliminary hearings according to the main charge on the indictment. Once again, cases where the accused pled guilty at an accelerated hearing under section 76 have been excluded from the analysis. As the table shows, there is some variation, with rape/attempted rape indictments and indictments involving other sexual offences having the greatest number of preliminary hearings (2.23 per case and 2.0 per case respectively). The cases that had the least number of preliminary hearings were indictments involving drugs offences and robbery indictments.

Table 4.17: Average (mean) number of preliminary hearings by main charge

4.48 Another way of examining the issue of continued preliminary hearings is to look separately at the outcome of initial preliminary hearings and continued preliminary hearings. As such, table 4.18 displays the outcome of first, second, third and fourth preliminary hearings respectively. It was not thought useful to focus on preliminary hearings 5 and above, given the small numbers involved.

4.49 As table 4.18 shows, the most common outcome of an initial preliminary hearing was for a guilty plea to be tendered (38.2% of first preliminary hearings). A further 35.8% of initial preliminary hearings were continued to a further preliminary hearing. These 35.8% are what might be termed 'unproductive preliminary hearings': outcomes other than continuation all lead to a conclusion of sorts (a guilty plea tendered and accepted by the Crown, a trial date set, the case deserted, or a warrant granted for the non-appearance of the accused). Table 4.18 shows that 52.4% of second preliminary hearings were productive, with 47.6% being further continued. It also shows, however, that a majority of third preliminary hearings were unproductive, with 52.2% resulting in a further continuation. The proportion of unproductive hearings lessened somewhat in relation to fourth preliminary hearings, but still 47.5% of fourth preliminary hearings were further continued. Thus, each time a preliminary hearing is continued, there is, approximately speaking, a 50% likelihood of it being continued again.

Table 4.18: Outcome of preliminary hearings 1-4

4.50 The number and productiveness of preliminary hearings alone, however, does not present a complete picture. Table 4.19 goes on to show the proportion of cases with one, two and three or more preliminary hearings respectively that eventually proceeded to trial compared to the number that were settled at the preliminary hearing itself.

Table 4.19: Proportion of cases proceeding to trial with one, two and three or more preliminary hearings

4.51 As table 4.19 shows, those cases requiring multiple preliminary hearings are more likely to proceed to trial than to conclude at the final preliminary hearing. In 64% of cases that had three or more preliminary hearings, the case proceeded to a trial diet. The proportion of cases with only one preliminary hearing that proceeded to a trial diet was 28%. This does suggest that cases involving continued preliminary hearings tend to be complex ones that cannot be settled at the preliminary hearing itself, although that is not to say that all of the multiple preliminary hearings were necessary.

4.52 Perhaps a better measure of the effectiveness of continued preliminary hearings is to look at the cases that did proceed to trial and examine the outcome of that trial diet. As such, table 4.20 is concerned only with those 237 cases in which a trial diet took place and breaks down the eventual outcome of the case according to the number of preliminary hearings that preceded the trial diet. If cases in which multiple preliminary hearings take place are still resulting in guilty pleas being tendered at the trial diet, for example, that would be an indication that the multiple preliminary hearings were ineffective.

Table 4.20: Trial diet outcome of cases with one, two and three or more preliminary hearings

n=237

4.53 What is clear from table 4.20 is that cases in which three or more preliminary hearings took place were the least likely to result in a guilty plea being tendered at the trial diet and the most likely to proceed to trial. This is an indication that holding those three or more continued preliminary hearings was reasonably effective in those cases. It also makes one wonder how many of the 16.9% of cases in which two preliminary hearings occurred before a guilty plea was tendered at trial might have been resolved at a further preliminary hearing if they had not proceeded to a trial diet when they did.

4.54 One thing that is clear is that where preliminary hearings are continued, they are not generally heard by the same judge. This is shown in tables 4.21 and 4.22 below. Table 4.21 examines all continued preliminary hearings and calculates the proportion of cases where the presiding judge was the same between the first and second preliminary hearing. This is not a perfect analysis, as information on the identity of the judge presiding over the preliminary hearing was not always available. ⁵³ Thus table 4.21 focuses only on the 172 cases with two or more preliminary hearings where the identity of the judge was known for both the first and second preliminary hearing.

Table 4.21: Did the same judge preside over the first and second preliminary hearing?

n=172

4.55 Table 4.22 examines separately cases where there were two, three, four and five or more preliminary hearings. It displays the proportion of cases in which the identity of the judge stayed the same throughout all of the preliminary hearings in the case. As before, those cases where information on the identity of the judge was not available for all of the preliminary hearings in the case are excluded from the analysis.

Table 4.22: Percentage of cases where the same judge presided over all of the preliminary hearings

n=172

4.56 What both tables show is that it is relatively rare for the judge to stay constant throughout the preliminary hearing stage of cases where there are multiple preliminary hearings. Ideally, a case would be heard by the same judge all the way though as this means that judges are hearing cases with which they are familiar and thus where they are more likely to recall what the parties undertook to do and to question whether or not this has been done. If the same judge does not hear the case, this is not necessarily problematic in itself. It means, however, that the importance of making detailed records is heightened, so that the judge picking up the case can familiarise him or herself with it relatively easily. ⁵⁴

Witness Numbers

4.57 This section of the chapter examines the impact of the reforms on Crown witnesses. ⁵⁵ There are three measures that might be relevant here: the number of witnesses listed on the indictment; ⁵⁶ the number who are then actually cited to give evidence (which is likely to be lower, as it might be decided that not all of those listed on the indictment are necessary to the case and witnesses whose evidence has been agreed would not be cited); and the number who actually do give evidence, assuming the case proceeds to trial. As chapter 3 outlined, data was available only on the first and third of these measures: the number of witnesses listed on the indictment and the number that were actually called to give evidence at trial.

4.58 The first thing to note is that the average number of witnesses listed on the indictment per case does not appear to have changed between the pre and post-reform samples. Taking the 806 cases in the pre-reform sample where some witnesses were listed ( i.e. excluding cases where a guilty plea was tendered at a section 76 hearing), the average number of witnesses listed was 26.86. Looking at the 570 cases in the post-reform sample where some witnesses were listed (again excluding those cases that were disposed of at a section 76 hearing), the average number of witnesses listed was 26.68, although if one exceptional case in which 389 witnesses appear to have been listed is removed from the calculation, this figure falls to 26.0.

4.59 What does seem to have changed as a result of the reforms, however, is the extent of witness inconvenience. There are a number of ways in which this can be measured.

4.60 One way of measuring witness inconvenience is to examine the average number of witnesses called in each concluded trial. In the pre-reform sample, there were 232 completed trials at which, on average, 12.92 witnesses per case were called. In the post-reform sample, there were 179 completed trials at which, on average, 11.03 witnesses were called.

4.61 An even better indication can be found by examining all cases in which a trial took place and calculating the number of witnesses who actually gave evidence as a proportion of those who were originally listed on the indictment. This shows a slight difference between the two samples. In the pre-reform sample, on average, 50.6% of witnesses listed on the indictment actually gave evidence at trial, whereas for the post-reform sample, this figure was 47.8%. Table 4.23 below displays these results in categories and shows a similar picture. For example, in the pre-reform sample the proportion of cases in which less than 40% of the listed witnesses gave evidence was 28.1% whereas in the post-reform sample it was 35.5%. Perhaps more strikingly, in the pre-reform sample, the proportion of cases in which less than 60% of the listed witnesses gave evidence was 65.4% whereas in the post-reform sample it was 80.3%.

Table 4.23: Percentage of listed witnesses who gave evidence

4.62 The reasons for these differences can only be speculated upon, but it is likely to be attributable at least in part to an increased use of agreement of evidence and/or to earlier case preparation meaning that prosecutors are better able to identify those witnesses who are vital to the case and to countermand those that are not absolutely necessary (or not cite them in the first place). ⁵⁷ The differences are certainly not due to the cases that went to trial in the post-reform sample being less complex (in terms of the charges and number of accused involved) than those that went to trial in the pre-reform sample. If anything, the reverse is true. ⁵⁸

4.63 A second way in which witness inconvenience can be measured is to focus solely on adjourned trial diets. If a trial diet is adjourned on the day of the trial, witnesses will be inconvenienced by turning up at court only for the trial not to go ahead. Even if they are countermanded in advance of the trial diet, they may still have been inconvenienced by having made arrangements to be present at the trial. Table 4.24 displays the number of adjourned trial diets in each sample alongside the total number of witnesses listed on the indictment in those cases.

Table 4.24: Adjourned trials and Crown witness numbers

4.64 As table 4.24 shows, the number of witnesses possibly inconvenienced by adjourned trial diets has decreased massively between the two samples, from 16,795 to 1,295. It should be noted, however, that this is a somewhat crude measure of witness inconvenience. First, it assumes that (a) all of the witnesses listed on the indictment were cited (which is unlikely to have been the case) and (b) if they were, that adjourning the trial did actually inconvenience them.

4.65 Second, the two samples consist only of cases concluded and indicted in the 12-month time period in question. The pre-reform sample consists of all High Court cases indicted and concluded between 1 April 2004 and 31 March 2005. As chapter 3 noted, 377 cases indicted during this period remained unconcluded as of 31 March 2005 and thus were not included in the dataset. The post-reform sample consists of all High Court cases indicted and concluded between 1 April and 31 March 2006. A total of 395 cases indicted during this period were not included in the dataset because they had not concluded. Thus the numbers shown in table 4.24 are an under-estimate of total witness inconvenience during the 12-month period particularly in the pre-reform sample, where over a third of trials were adjourned (see table 4.11) compared to less than a fifth of trials in the post-reform sample (see table 4.12).

4.66 Third, as has already been noted, the analysis focuses only on Crown witnesses and does not include any witnesses for the defence. This is unlikely to have had a major impact on the figures, however, as it is rare for large numbers of defence witnesses to give evidence, even in High Court cases.

Time Bar Extensions

4.67 One of the reforms introduced by the 2004 Act was to extend the time bar in custody cases from 110 days to 140 days. ⁵⁹ Table 4.25 displays data on the number of extensions to the 110-day and 140-day custody time bars in the pre-reform and post-reform samples respectively. Table 4.25 focuses only on custody cases, defined as cases where at least one of the accused appeared from custody during the period covered by the data. It should be noted that in none of the cases in either sample was an application recorded as having been refused.

4.68 As table 4.25 shows, in the pre-reform sample, there were 336 custody cases and in 84 of these cases (25%) at least one extension to the 110-day time bar was granted. In the post-reform sample, there were 313 custody cases and in 67 of these cases (21.5%) at least one extension to the 140-day time bar was granted.

4.69 It should be borne in mind in interpreting these figures that they are not an absolute measure of the total number of time bar extensions in the years April 2004-March 2005 and April 2005-March 2006. As chapter 3 outlined, our samples contained only cases that were indicted and concluded in this period. There were, for example, 395 cases indicted during the period April 2005-March 2006 that had not concluded by the end of March 2006 and thus were not included in the post-reform sample. At least some of these cases are likely to have involved extensions of the 140-day time bar. What is important, therefore, is not the actual number of time bar extensions, but the comparison between the pre-reform and post-reform samples.

Table 4.25: 110-day and 140-day time bar extensions in custody cases

4.70 Bearing in mind that slightly fewer cases indicted between March 2005 and April 2006 (the post-reform time period) had concluded in time to be included in our analysis compared to the equivalent pre-reform time period, ⁶⁰ and assuming that cases that have not concluded are more likely to involve at least one time bar extension, it can safely be concluded that the proportion of custody cases in which there is at least one extension of the time bar has changed very little following the reforms. The new 140-day time bar is being extended almost as frequently as the old 110-day time bar.

4.71 In the post-reform sample, in the 67 cases where the 140-day time bar was extended at least once, the average (mean) length of the extension was 46 days. ⁶¹ In the pre-reform sample, in the 84 cases where the 110-day time bar was extended at least once, the equivalent average (mean) figure was slightly longer, at 57 days. Table 4.26 displays this information in categories.

Table 4.26: Total length of extensions to the 110-day and 140-day time bars

Summary of Main Findings from Statistical Analysis

4.72 An increased number of cases were disposed of by way of a section 76 hearing in the post-reform sample: 19.2% compared to 6.0% in the pre-reform sample.

4.73 In the pre-reform sample, 94.0% of cases proceeded to a trial diet compared to 33.3% in the post-reform sample. This is an indication that preliminary hearings are working effectively in avoiding the need for a trial diet.

4.74 The proportion of cases in which a trial actually took place was similar in the two samples: 27.1% of cases in the pre-reform sample compared to 25.3% of cases in the post-reform sample. What has changed is that it was very unusual for a guilty plea to be tendered at a trial diet in the post-reform sample: 6.9% of guilty pleas were tendered at a trial diet in the post-reform sample compared to 89.7% in the pre-reform sample. This is a further indication that preliminary hearings are working effectively in encouraging the tendering of guilty pleas at an earlier stage than the trial diet.

4.75 The average number of court hearings per case has changed very little - 2.06 in the pre-reform sample compared to 2.03 in the post-reform sample. What has changed is that in the post-reform sample the predominant type of hearing was the preliminary hearing whereas in the pre-reform sample it was the trial diet.

4.76 In the pre-reform sample, 32.6% of cases had at least one adjournment of a trial diet, compared to 4.5% in the post-reform sample.

4.77 In the pre-reform sample, the most common outcome of a trial diet was for that trial diet to be adjourned (37.1% of trial diets). The second most common outcome was for a guilty plea to be tendered (34.8% of trial diets). Neither of these outcomes is desirable (in the sense that if a case is to conclude with a plea of guilty, this would ideally be tendered at an earlier stage than the trial diet). Only 18.1% of trial diets in the pre-reform sample resulted in a concluded trial.

4.78 By contrast, in the post-reform sample, the most common outcome of a trial diet was for a trial to conclude (63.0% of trial diets). Only 16.2% of trial diets were adjourned and only 11.3% resulted in a guilty plea. This is a strong indication that preliminary hearings are working effectively.

4.79 The most common reason for a trial diet to be adjourned in the pre-reform sample was that one of the parties had inadequate time to prepare (63.0% of trial diet adjournments). The most common reason in the post-reform sample (where only a very small proportion of trial diets were adjourned) was that an essential Crown witness was absent (40.0% of trial diet adjournments), although because adjournments were rare in the post-reform sample this equates to only 16 actual adjournments.

4.80 The average length of a concluded trial in the pre-reform sample was 5.85 days. In the post-reform sample it was 4.88 days. This cannot be accounted for by the cases in the post-reform sample being less complex. It may be an indication that there has been an increase in the use of agreement of evidence provisions or that more time is now available for case preparation.

4.81 In the post-reform sample, the most common outcome of a preliminary hearing was for that preliminary hearing to be continued to a further preliminary hearing (42.8% of preliminary hearings). A total of 38.7% of cases had two or more preliminary hearings (if cases where the accused pled guilty under section 76 are excluded) and 5.2% of cases had five or more preliminary hearings. Each time a preliminary hearing is continued, there is an approximately 1 in 2 likelihood of it being continued again.

4.82 In cases not settled by way of a guilty plea at a section 76 hearing, continued preliminary hearings were most common in rape cases (which had, on average, 2.23 preliminary hearings); cases involving other sexual offences (which had, on average, 2.0 preliminary hearings); and murder cases (which had, on average, 1.89 preliminary hearings).

4.83 When a preliminary hearing was continued, it was unusual for the same judge to preside over both the initial and the continued preliminary hearing. Where there were three or more preliminary hearings in a case, it was extremely rare for them all to be presided over by the same judge.

4.84 In the pre-reform sample there were 232 completed trials at which, on average, 12.92 Crown witnesses gave evidence. In the post-reform sample, there were 179 completed trials at which, on average, 11.03 Crown witnesses gave evidence. Further, in the pre-reform sample an average of 50.6% of witnesses listed on the indictment actually gave evidence, whereas this figure declined to 47.8% for the post-reform sample. This reduction in witness numbers cannot be accounted for by the cases in the post-reform sample being less complex. It may be an indication that there has been an increase in the use of agreement of evidence provisions or that more time is now available for case preparation.

4.85 The number of Crown witnesses possibly inconvenienced by adjourned trial diets has decreased massively between the two samples, from 16,795 (although some of these may have been countermanded where it was clear in advance that the trial was going to be adjourned) to 1,295.

4.86 The new 140-day time bar in custody cases was extended almost as frequently as the old 110-day time bar. There was at least one extension of the 110-day time bar in 25% of cases in the pre-reform sample. The equivalent figure for the 140-day time bar in the post-reform sample was 21.5%. While valid for comparative purposes, these figures are likely to be under-estimates as they refer only to cases which concluded within the sample periods.

« Previous | Contents | Next »