Suspended sentences in Spain: Decarceration and recidivism

The suspended sentence, along with the fine, is the most commonly used alternative to prison in Spain. This sentence is generally regarded as a good sanction because it avoids the personal and financial costs of prison for occasional offenders which are not in need of rehabilitation. The main aim of this article is to present data and empirical research about the effectiveness of the suspended sentence with respect to its two main goals: reducing the use of prison and preventing offenders from re-offending. The data appear to suggest that this is an effective sentence in these terms, but the article also raises questions about the need to replace suspended sentences with more rehabilitative orders in some special cases.


Introduction
The suspended sentence, a prison sentence that is not implemented as long as the offender refrains from committing another offence, seems to have played a very important role in the limitation of the use of prison in many jurisdictions. 1 However, in the punitive mood apparent in some western countries, Spain among them, the suspended sentence could be submitted to criticism for being a 'soft' sanction. In this context, it seems important to explore whether, in the light of the available evidence, this non-custodial sanction should be supported or replaced by other alternatives more matched to the seriousness of the offence and/or more aimed at the rehabilitation of the offender.
The article begins with a brief discussion about the possible justifications of the suspended sentence in the desert and utilitarian theories of punishment. It seems that the latter theory could support suspended sentence so long as it is true that no other intervention is needed for the rehabilitation of the offender. After presenting the legal framework of the suspended sentence and showing that, in common with other countries, in Spain this sentence has produced a decarceration effect in reducing the use of prison, the article explores the evidence about the level of recidivism of people granted suspended sentences. The data appears to be supportive of this sentence but nevertheless some further debate may be required about the need to replace it by other sentences in some particular cases. a strong correlation between rising severities of punishments and subsequent levels of criminality (Von Hirsch et al., 1999: 48).
The second objection to or 'cost' of the suspended sentence -the possibly lesser capacity to reduce the level of recidivism of offenders compared to other sanctions -becomes the heart of a utilitarian discussion about the effectiveness of this sentence. As stated before, there are at least theoretical grounds for suggesting that the suspended sentence might be effective in preventing recidivism so long as it effectively targeted occasional offenders that might refrain from committing new offences only due to the conviction and the threat of prison in case of a new offence.
This hypothesis is, of course, one that is amenable to research enquiry. It is possible to explore the level of recidivism of people who have received suspended sentences. If the level is very low, then it seems that the 'occasional criminal theory' might be confirmed. But, once we know the level of recidivism of the people granted the suspended sentence we should make appropriate comparisons with other non-custodial sanctions. 3 Only then will we be able to answer the question about the justification of the suspended sentence from a utilitarian point of view.

Legal framework and use of the suspended sentence in Spain
The research about recidivism of offenders presented below has been done with people who were granted suspended sentences in Spain. Given that each legal system can use suspended sentences in a different way, it is worth introducing the main legal features of the suspended sentence in Spain.
The Spanish penal code distinguishes between: minor offences ('faltas') and offences ('delitos'). Minor offences are usually punished with a fine or, in some cases, judges have the discretion to impose a fine or arrest. In practice, the usual sanction for a minor offence is a fine. In the case of offences, the most common punishment is prison and only for some offences does the judge have discretion to impose a fine.
When the mandatory legal punishment for the offence is prison (for example, with offences like theft of more than 400 euros, burglary, drug trafficking and robbery) the law establishes a minimum and maximum sentence length (for example, the legal punishment for a theft is between six and 18 months of prison). In such cases, the judge has to impose the prison sentence and then s/he has to decide whether or not the suspended sentence might be granted.
For a suspended sentence to be granted, 4 the offender has to meet the following criteria: 1. The prison sentence should not be more than two years; 2. The offence should be the first offence by that person (or a relevant time has passed between his or her latest conviction and the present offence). A suspended sentence is possible even when the judge deals with a first offence but the offender has committed another offence subsequently (prior to the court hearing); 3. The person has paid compensation to the victim (except when the person is unable to do so).
If the offender meets the criteria then the judge has discretion to grant a suspended sentence. In cases where the judge decides against imposing a suspended sentence, s/he has to explain the reasons for that decision. According to previous work by Cid et al. (2002), when the legal conditions are fulfilled, the suspended sentence is the most common decision. Table 1 shows the results of this research. As Table 1 shows, judges tend to 'automatically' grant a suspended sentenced when they sentence a first-time offender who has not committed another offence between the current offence and the decision about the suspended sentence. When the offender has an old criminal record or, after the date of the actual offence but prior to disposal, has committed other offences, then a suspended sentence is granted in roughly half of the cases.
The most relevant variable in the decision of judges is the criminal record of the offender; having a clean record is a guarantee of getting a suspended sentence. Other variables that in the study of Cid et al. (2002: 66-70) reached statistical significance in the decision of judges were: public prosecutor agreement with the suspended sentence; compensation to the victim; not being remanded in custody at the time of sentencing; and not being a drug addict. All these make granting the suspended sentence to the offender more probable.
It is relevant to point out that judges tend to make decisions about suspended sentences without the help of a pre-sentence report that might give relevant information about the prediction of recidivism.
The law establishes two kinds of suspended sentence: first, a straight suspended sentence, in which the only obligation of the offender is to refrain from committing another offences during the time of suspension (two to five years); and second, a suspended sentence with probation, in which the offender is placed under the supervision of a probation officer or has to participate in some treatment or has to refrain from some activities.
In practice, according to the research of Cid et al. (2002: 67) judges disregard almost absolutely the possibility of combining the suspended sentence with probation. 5 The main reasons for neglecting probation are, in my opinion, the  Source: Cid et al. (2002: 66-8). The research is a representative sample of the convictions imposed in 1998 by the judges of Barcelona, which deal with offences punishable with a maximum of three years imprisonment.
following: first, the fact that judges do not feel culturally close to the 'rehabilitation ideal' which would support probation because until recently the possibility of imposing probation did not exist in Spain; second, the fact that the introduction of probation in the penal code of 1995 was done without giving additional means to judges for sentencing. For example, judges do not have a probation officer in the court to provide reports or assist with implementing the order, and the judges are dependent on social services, and they are not appropriately equipped to carry out these duties.

Suspended sentence and decarceration
The penal code of 1995 made some important reforms in the system of alternatives to prison. The most important of which were: to set the upper limit for consideration of suspended sentences at two years imprisonment (instead of one year as in the previous penal code of 1973); to allow the replacement of a prison sentence of up to two years imprisonment with a fine (even when the offender has a criminal record) and to abolish the prison sentences of less than six months, making it mandatory for judges to impose a fine or weekend arrest instead. One way of verifying if the new system has been effective in reducing the use of prison sentences consists of comparing the evolution of the number of convictions, with the evolution of prison admissions. If no other reasons are operating, the system would be effective if, in relation to convictions, admissions into prison tend to decrease. Table 2 shows the absolute number of convictions and admissions to prison in the recent years and (taking 1996 as a basis) the relationship between these two variables. Once we know that there is a very significant reduction in the number of admissions to prison in relation to convictions we should explore which is the most relevant factor in operation. From the three main reforms of the system of alternatives to prison of the penal code of 1995, as mentioned above, we know from previous research that the possibility of replacing a prison sentence of up to two years with a fine, for offenders with a recent criminal record, has rarely been used by judges (Cid et al., 2002: 72). Also, although prison sentences up to six months have been abolished, and this has produced an impact in the reduction of prison admissions, the effect has been not very important given that only a few offences are punished with prison sentence less than six months (Cid et al., 2002: 58). For these reasons, it seems that the main factor should be the fact that the penal code of 1995 has increased the upper limit of the suspended sentence up to two years imprisonment.
The confirmation that, as has happened in other countries, the increase in the use of the suspended sentence, following the change of the upper limit from one to two years imprisonment, is the main reason for the reduction in the prison admissions is highlighted in Table 3. It shows the proportion of prison sentences that were suspended between 1996 and 2003. Table 3 demonstrates that in the process of implementation of the penal code of 1995 there has been a progressive proportion of prison sentences which have been suspended and, as it has happened in other countries, it is possible to conclude that also in Spain the suspended sentence has produced a 'decarceration' effect. 6

Suspended sentence and recidivism
Given that the suspended sentence, in conjunction with other reforms, has been shown to be a relevant factor in the reduction of prison admissions, the next question  to consider is the cost of this penal sanction, measured by the recidivism of the people who have benefited from it. The lesser the level of recidivism the greater the confirmation of the 'occasional offender theory', that supports the suspended sentence. Nevertheless, as stated before, the second step of the research, which has not yet been undertaken, would be a comparison with the results of other noncustodial sanctions. The research conducted in this regard analyses the level of recidivism of a sample of people who in 1998-9 were granted a suspended sentence by the judges of Barcelona. The research consists of a follow-up of offenders that in a previous study (Cid et al., 2002) were granted a suspended sentence. This earlier research, aimed at exploring the application of alternatives to prison by Spanish judges, examined the sentences imposed on 1425 offenders. The sample was representative of the convictions imposed by the judges of Barcelona who dealt with those offences that permitted sentences of up to three years' imprisonment in 1998. From these 1425 offenders, 315 got the suspended sentence.
The present research is a modest first step, dealing with one group of the 315 offenders that received a suspended sentence. From these 315 offenders, we have selected 119, who received the suspended sentence from 10 of the 23 judges of Barcelona at this time. 8 These 119 offenders received suspended sentences between April 1998 and December 1999. We followed up these offenders by checking if, up to 31 July 2004 (roughly a five year follow-up), they had been incarcerated in Catalan prisons. There were practical reasons for taking incarceration instead of re-arrest or reconviction as a measure of recidivism. At least in Spain this is the easier data to obtain. Additionally, taking incarceration as a measure of recidivism could be appropriate if the goal is to target persistent offenders, because in Spain it would be very difficult for those with three or more convictions not to spend some time in prison. 9 The recidivism of the offenders is shown in Table 4.
We analysed whether there were statistically significant differences between those incarcerated and those not incarcerated with respect to the following variables: a) age when committing the current offence; b) sex; c) kind of current offence; d) nationality; e) criminal record; e) financial means of the offender. 10 The aim of these comparisons was to get more information in order to do an analysis about the factors most able to predict recidivism. The results are shown in Table 5.
The results of the research could be summarized as follows: 1. When a suspended sentence is granted to offenders with only the current conviction in their criminal record, the recidivism rate, measured by incarceration, is very low (10.6%).
2. When the suspended sentence is granted to offenders that, although not having recent convictions, have either old convictions or convictions for offences committed after the current offence, then the recidivism rate increases significantly, up to 38.1 percent.
3. The only variable statistically correlated with recidivism is the criminal record of the offender.
4. Although, in common with other research (Petersilia, 1997;May, 1999), we found that offenders with some characteristics (for example, being younger than 25 years old when committing the current offence, being male and having been assisted by a state lawyer) show a higher rate of recidivism, none of these differences reached statistical significance. Possibly with a larger sample some of these factors could become significant.

Conclusions
The penal code of 1995 raised the limit of consideration of suspended sentences from one to two years' imprisonment. Spanish judges appear to have continued  the pattern of applying this sentence automatically when offenders do not have any other prior conviction. Consequently, suspended sentences have been a powerful device for reducing admissions into prison in the period 1996-2003. From its origins, the suspended sentence was grounded on the theory that some offenders were occasional and were not in need of rehabilitation. Like other research, the present study confirms that only a small number of the people convicted for the first time become recidivist. 11 Given that in a five-year follow-up, only 17.6 percent of the offenders of the sample which were granted suspended sentences were incarcerated, it seems that the suspended sentence is an inexpensive means of dealing with offenders. However, we do not know to what extent this positive outcome comes from the kind of sentence per se, as opposed to other factors. Only comparative research concerning the outcomes of other non-custodial sanctions can address this question.
Obviously the suspended sentence was not effective with the 17.6 percent of the offenders incarcerated in the follow-up period. It seems that for these offenders it might have been more useful if a non-custodial sanction had been used, like probation or treatment added to the suspended sentence, aimed at confronting their criminogenic needs or the obstacles to desistance (Farrall, 2002: 220).
From this point of view it may be problematic that Spanish judges make decisions about the suspended sentence without the help of a pre-sentence report. Although the problems of prediction are very significant, it seems that taking into consideration some social variables could be helpful in deciding between a straight suspended sentence or a suspended sentence plus probation or treatment. 12 Possibly, with the help of the pre-sentence report, the cases where probation is added to a suspended sentence would increase.
2 In Wasik (1994) there is a more positive view of the suspended sentence, partly for the pragmatic reason that if the suspended sentenced is abolished 'What would replace it?' (Wasik, 1994: 56). 3 Comparisons should be made between people who have been sentenced in a different way but who do not show significant differences in other respects. 4 The present article does not deal with a special form of a suspended sentence for drug-addicted offenders who had committed a drug-related offence. This form of suspended sentence is possible for a prison sentence not more than three-five years according to a recent reform of the Spanish penal code, and it is even possible when the offender has some previous offences. For the offender to be granted this form of suspended sentence s/he has to agree to participate in a programme aimed at confronting his/her drug-addiction. This provision has existed in Spanish law since 1987. 5 The suspended sentence with probation was only imposed in 6 percent of the cases. 6 The 15 percent increase in the Spanish average prison population between 1996 (112 prisoner per 100,000 inhabitants) and 2003 (128 prisoners per 100,000 inhabitants) is basically due to the abolition of remission for good behaviour by the penal code of 1995. For more detailed data see Cid (2005). 7 This data includes also the cases of suspended sentence for drug-addicted offenders (see Note 4). However, according to the research of Cid et al. (2002), this special form of suspended sentence represents only a modest 5 percent of the total amount of suspensions. 8 The selection of these 10 judges was made for practical reasons: with respect of the convictions of these 10 judges we had a copy of the written conviction and therefore we had the name of the offender. With respect to the decisions of the other 13 judges, we only had the number of the decision but not the name of offender. Nevertheless we statistically proved that these 10 judges did not behave in a different way to the other 13 in granting the suspended sentence. The proof confirmed that there wasn't a significant difference between the two groups. If the research should be continued then it would need to take into consideration the suspended sentences granted by the 23 judges. 9 If the new convictions are for offences punishable only with prison the suspended sentence will not be possible and, as we have stated earlier, the power of the judges to replace the prison by a fine is hardly used in practice (Cid et al., 2002: 72). 10 We allowed for the fact that the offender was assisted during the trial by a paid lawyer. 11 Recent research in Spain shows that after a five-year follow-up, only 22.6 percent of the first-time incarcerated have been re-incarcerated (Luque et al., 2004: 101). 12 One of the conclusions of the research conducted by May (1999) was that, for first-time offenders or those with very few conviction, taking into consideration some social factors could improve the prediction of recidivism (pp. ix, 38, 49).