Introduction
When should prosecutors in England and Wales start or continue a prosecution? According to the Code for Crown Prosecutors (the Code), they should do so only in cases that pass both stages of the Full Code Test.Footnote 1 The first stage is the evidential stage: there must be ‘sufficient evidence to provide a realistic prospect of conviction’.Footnote 2 The second is the public interest stage: even if there is sufficient evidence, ‘prosecutors must go on to consider whether a prosecution is required in the public interest’.Footnote 3 Exactly when the public interest requires prosecution is, however, left unspecified in the Code. The key paragraph instead reads:
It has never been the rule that a prosecution will automatically take place once the evidential stage is met. A prosecution will usually take place unless the prosecutor is satisfied that there are public interest factors tending against prosecution which outweigh those tending in favour. In some cases the prosecutor may be satisfied that the public interest can be properly served by offering the offender the opportunity to have the matter dealt with by an out-of-court disposal rather than bringing a prosecution.Footnote 4
The Code then goes on to provide a non-exhaustive list of public interest factors.Footnote 5 This list is supplemented by guidance on specific offences and legal issues published by the Crown Prosecution Service (CPS),Footnote 6 which identifies additional public interest factors that prosecutors must consider.Footnote 7
In this paper, I draw on these sources to construct an account of when prosecution is in the public interest. I catalogue the reasons that they recognise in favour of and against prosecution and ask what principles, if any, might explain their requirements. I conclude that, in most cases, CPS guidance imposes on prosecutors an obligation to seek an appropriate penalty for those against whom there is sufficient evidence. In more serious cases, this obligation effectively renders prosecution mandatory. In less serious cases, it may be outweighed by countervailing factors, and out-of-court disposals may constitute an appropriate penalty. However, only a limited range of factors are clearly recognised as counting against prosecution. CPS guidance also lacks clear principles regarding when prosecutors may exceptionally pursue non-punitive aims.
The public interest test embodies a long-standing principle in English law: that the justifiability of prosecution depends not only on whether there is sufficient evidence but also on various non-evidential considerations.Footnote 8 This principle is significant given how criminal cases proceed after charge. At that point, non-evidential considerations become largely irrelevant: either defendants will plead guilty to the charges against them or adjudication of their guilt will be the focus of the trial. Although these considerations play an important role in determining who should be convicted and punished, they thus do so outside of open court and in a relatively non-visible way. Examination of the public interest test therefore promises to reveal normative commitments underlying the criminal process that might otherwise remain hidden.
Despite its significance, however, the public interest test has rarely been explored in previous research. Early editions of the Code attracted some relevant commentary.Footnote 9 Empirical research on prosecution decision-making has also sometimes examined the role played by non-evidential considerations in practice.Footnote 10 However, only one previous analysis has considered the test itself in any depth, and this was concerned mainly with potential reforms to the Code.Footnote 11 The present analysis, by contrast, is based on a study of all published CPS policies and guidance. As of August 2024, more than 200 such documents were published on the CPS website, of which 101 included additional guidance on decisions whether to prosecute. This sizeable body of sources has gone virtually unexplored in previous work. As we shall see, however, the understanding of the public interest test that emerges from them differs in important respects from that which the Code suggests.
Before proceeding further, I should clarify the object of my enquiry. I am concerned with how prosecutors should decide whether to prosecute, according to the guidance that they are given. I am not primarily concerned with how prosecutors make these decisions in practice. Nor am I concerned with how CPS guidance should structure these decisions in an ideal sense. Rather, my aims are to analyse what the public interest test currently requires of prosecutors and, where possible, to identify principles that might explain these requirements. Whether these principles are ultimately defensible, and the extent to which they provide a sound basis for prosecution guidance, are questions that I leave for discussion elsewhere.
The paper is structured as follows. Section 1 provides a more detailed introduction to the public interest test, its nature and origins. Section 2 then examines the most important set of public interest factors: those relating to the seriousness of the alleged offending and what would constitute a proportionate response to it. Section 3 covers other reasons to prosecute. Section 4 addresses reasons against prosecution, including cost, legal obligations not to prosecute, and various harms that prosecution can cause. Finally, section 5 examines alternatives to prosecution and when and why they are deemed appropriate.
1. The public interest test: nature and origins
The CPS was established in 1986. Before then, there was no national prosecution service in England and Wales. Prosecutions in a small number of serious and sensitive cases were conducted by a Director of Public Prosecutions (DPP). Most prosecutions, however, were conducted by solicitors instructed by local police forces. This unusual arrangement had its roots in the nineteenth century, when the new public police assumed responsibility for prosecution from private victims and associations.Footnote 12 Only in 1981 was there decisive impetus for change, with the report of the Royal Commission on Criminal Procedure. The Royal Commission endorsed the principle that decisions to prosecute should be taken by legal experts independently of the police. It therefore recommended the establishment of a new prosecution service that would decide whether to prosecute in all police-initiated cases.Footnote 13
The Royal Commission considered the criteria for prosecution that this new service should apply. It concluded ‘that it is not necessary or desirable for all offences to be prosecuted’Footnote 14 and recommended the promulgation of non-evidential criteria for prosecution.Footnote 15 These conclusions reflected the DPP’s long-held views. As the Attorney General had confirmed in 1951, in a much-quoted statement to Parliament: ‘It has never been the rule in this country – I hope it never will be – that suspected criminal offences must automatically be the subject of prosecution’. Rather, the ‘dominant consideration’ for the DPP was whether ‘prosecution … is required in the public interest’.Footnote 16 This consideration also informed the DPP’s and Attorney-General’s criteria for prosecution, which prescribed something like the two-stage Full Code Test in the years preceding the creation of the CPS.Footnote 17
Upon the CPS’s creation, the DPP was required to issue a code setting out general principles governing decisions whether to start or continue a prosecution.Footnote 18 There have since been eight editions of the Code for Crown Prosecutors, all of which have retained the Full Code Test. All have also declined to define the public interest explicitly and have instead listed factors that prosecutors must consider at this stage. While there have been numerous changes to this list, the most significant came with the third edition of the Code in 1994. Since then, the Code has been oriented more towards explaining CPS decision-making to the public in easily understandable terms.Footnote 19 One aspect of this change was a shorter and explicitly non-exhaustive list of public interest factors.Footnote 20 In the current Code, this list comprises seven questions that prosecutors are required to consider.Footnote 21
The list of public interest factors in the Code has always been supplemented, however, by guidance to prosecutors on specific offences and issues. Much of this guidance was previously contained in confidential manuals, but the CPS has increasingly made it public over the past two decades or so.Footnote 22 As noted above, the present study encompassed 101 documents containing guidance on prosecution decisions. Of these, around three quarters include additional public interest factors for prosecutors to consider. It is unclear whether these factors exhaust the public interest when taken together with those in the Code. CPS guidance likewise sometimes states that it is non-exhaustive, and it almost never excludes factors from prosecutors’ consideration.Footnote 23 The Code says, however, that it and the guidance ‘should enable prosecutors to form an overall assessment of the public interest’.Footnote 24 This statement may imply that considerations are irrelevant if they are not recognised in either source.
Several preliminary questions arise from this introduction to the Code and CPS guidance. First, what purposes do these documents serve? The obvious answer is that they guide prosecutors in their decision-making. At least some prosecutors apparently consult these documents in making their decisions, while others refrain from doing so only because they feel they have absorbed their principles.Footnote 25 Research suggests, however, that prosecutors also use these documents for other purposes. In particular, they see them as a vehicle for explaining their decisions to others and for justifying and defending those decisions when they are challenged.Footnote 26 Although we will focus below on the decision-guiding role of these documents, we should thus also bear in mind these ‘rationalising’ roles and how they might influence their content.
Secondly, what style of decision-making are prosecutors guided to adopt at the public interest stage? At first glance, the Code may seem to require a balancing of competing factors. If a case passes the evidential test, then a prosecution ‘will usually take place’ unless ‘there are public interest factors tending against prosecution which outweigh those tending in favour’.Footnote 27 One problem with such approaches is that the factors to be balanced are often incommensurable. There is not clearly any rational basis for weighing, say, the seriousness of the alleged offence against the harm that prosecution may cause the victim.Footnote 28 The Code’s only comment on this issue is that the weight of such factors will ‘vary according to the facts and merits of each case’.Footnote 29 As previous critics have argued, this statement recognises the problems of weighing and prioritising competing factors without attempting a solution.Footnote 30
On closer inspection, however, the public interest test is not merely a balancing exercise. For one thing, CPS guidance often contains explicit presumptions in favour of and against prosecution. Although it is almost never conclusive regarding when prosecution is in the public interest, it often identifies cases in which prosecution is ‘normally’, ‘usually’, or ‘almost always’ required. Less often, it even identifies cases in which prosecution would be ‘unusual’, ‘unlikely’, or ‘exceptional’. Moreover, we shall see that, in most cases, the Code and the guidance effectively create a presumptive obligation to prosecute. Prosecutors, that is, have inherent reason to prosecute where the evidential test is passed, which only a limited range of countervailing reasons can defeat. Thus, the Code does not merely state a tie-breaking rule when it says that a prosecution ‘will usually’ take place unless the reasons against ‘clearly outweigh’ those in favour. This wording instead reflects an inherent public interest ‘in upholding the criminal law and placing the defendant before the courts’.Footnote 31
Thirdly, what remedies are available if prosecutors fail to apply this guidance correctly? Such failure is not, in itself, a ground for any private law remedy.Footnote 32 However, it is a ground on which victims might seek judicial review of a decision not to prosecuteFootnote 33 and on which defendants might argue that prosecution is an abuse of process.Footnote 34 In both cases, however, courts consistently emphasise that intervention in prosecution decisions should be exceptional. One reason is that these decisions require a balancing of multiple considerations that is properly a matter for prosecutors’ professional judgement. Hence, challenges will likely succeed only where prosecutors have failed properly to inform themselves or to ask the correct questions. Prosecution decisions should not be judged unreasonable simply because the court might disagree with them.Footnote 35
The result of the above is that the public interest in prosecution will sometimes turn on a balancing of incommensurable factors with which courts will not interfere. Particularly given prosecutors’ use of guidance as a vehicle for ex post rationalisation of their decisions, this result creates concerns. One is that certain factors, such as previous convictions, assume an overriding weight that is not evident from the guidance.Footnote 36 The more serious concern, however, is that the guidance becomes entirely presentational: it can be used to defend decisions that were taken on notionally irrelevant grounds. Cases may be treated differently, for instance, according to the type of suspect and/or victim involved.Footnote 37 As McConville et al highlight, these discrepancies often reflect police priorities and ‘working rules’ that prosecutors may be unable or reluctant to challenge.Footnote 38
Two further preliminary points are worth addressing at this stage. The first is the out-of-court disposals that might constitute acceptable alternatives to prosecution. Several such alternatives are available to the police, including warnings, simple cautions, and fixed penalty notices. For CPS prosecutors, however, the disposals most likely to be relevant are conditional cautions.Footnote 39 Conditional cautions require the suspect to comply with specified conditions, whose purpose may be rehabilitative, reparative, and/or punitive.Footnote 40 They may be offered only if the suspect admits the offence and there is sufficient evidence to charge them, and the suspect must consent to their imposition.Footnote 41 If the suspect fails to comply, then they may be prosecuted for the original offence.Footnote 42 The Code says little about when these disposals are appropriate,Footnote 43 although there is further guidance on their use that is examined below.
Finally, prosecutors must often decide not only whether to prosecute but also for which offence(s). Since criminal offences often overlap, a suspect’s conduct might constitute several at once. Alternatively, a suspect might have committed multiple offences during an incident or series of incidents. In such cases, prosecutors must choose which of the possible offences to charge. Additionally, they might choose not to continue with some charges if the defendant is prepared to plead guilty to others. These decisions – on charge selection and the acceptance of guilty pleas – are dealt with separately in the Code.Footnote 44 However, since they likewise concern when to proceed with cases that pass the evidential test, they are also in effect decisions about when prosecution is in the public interest. Guidance on these decisions is therefore also examined in what follows.
2. Seriousness and proportionality
By far the most important public interest factors concern the seriousness of the alleged offending. These factors relate closely to another: what would be a proportionate response to that offending, given its seriousness? In the Code, five of the seven public interest questions mentioned above relate wholly or partly to seriousness. In the policies and guidance documents that list further public interest factors, factors relating to seriousness feature in almost all. Indeed, in most of these documents, discussion of the public interest consists mainly in lists of context-specific aggravating and mitigating factors. Only a minority of these documents mentions any of the other factors discussed below.
Seriousness itself is the first factor listed in the Code: ‘The more serious the offence, the more likely it is that a prosecution is required.’Footnote 45 It also explicitly encompasses the next two factors: the suspect’s culpability and the circumstances of and harm caused to the victim. Culpability factors include the suspect’s level of involvement, their prior record, any planning or premeditation, and any benefit from the alleged offending. They also include mitigating factors, such as mental ill health and the suspect’s having been coerced or exploited.Footnote 46 Victim-related factors include the impact of the offence on the victim; their vulnerability; and motivation by, or demonstration of, hostility towards their protected characteristics.Footnote 47 The next two factors listed in the Code also bear on seriousness: the suspect’s age and maturity;Footnote 48 and the impact of the offence on the wider community.Footnote 49
Many of the additional factors identified in the guidance are simply specific instances of those listed in the Code. Others are not listed in the Code but relate clearly to the scale of the harm caused or risked by the offending or the level of the suspect’s culpability. The same is true even of offences that need not have specific victims or that are more regulatory in character. The public interest in prosecuting these offences is often dictated mainly by seriousness, adjusted for the mischief that they are assumed to target. For just one example, consider the offence of running an unregistered independent school. Relevant factors here include the breadth of the curriculum taught and whether it disadvantages students, as well as any failure to observe applicable standards or to comply with warnings from regulators.Footnote 50
Why might factors relating to seriousness play such a significant role in determining the public interest? An obvious explanation is that they also determine the sentence that the alleged offending will likely attract. Sentencing decisions likewise begin with an assessment of seriousness, harm, and culpability. The harm and culpability factors identified above also overlap to a great extent with those listed in the Sentencing Council’s general guideline.Footnote 51 Although this connection is only hinted at in the Code,Footnote 52 research evidence suggests that it is understood by prosecutors themselves.Footnote 53 The more severe the likely sentence, it seems, the stronger the public interest in proceeding against the suspect in court.
The most important demonstrations of this point are the policies on out-of-court disposals such as conditional cautions. These policies require prosecutors to ‘assess the seriousness of the case to ensure that this out of court disposal provides an appropriate and proportionate response to the offending behaviour’.Footnote 54 If a conditional caution would provide adequate reparation, rehabilitation, or punishment, then prosecutors should offer one, unless the case is too serious for such an outcome to be proportionate.Footnote 55 What counts as ‘too serious’, however, is primarily a question of the appropriate sentence. If imprisonment or a high level community order is appropriate, then a conditional caution should not be offered.Footnote 56 In effect, therefore, the public interest requires prosecution where the offence concerned would attract such a sentence. Only in less serious cases – by implication, those where the likely sentence is a low- or mid-level community order or a fine – can out-of-court disposals be considered proportionate.
Neither the Code nor the guidance explicitly says why, under these conditions, out-of-court disposals are an appropriate response to offending. Presumably, though, the answer is that they facilitate outcomes similar to those of prosecution, often more quickly and at a lower cost.Footnote 57 Conditional cautions can include requirements to pay a small fine or undertake a small amount of unpaid work.Footnote 58 They therefore resemble the penalties that suspects might receive in cases where they may be offered. Moreover, since they can include reparative requirements, such as compensating the victim, they can substitute for reparative orders made on conviction.Footnote 59 Prosecution is required, by contrast, where the likely penalty significantly exceeds that which conditional conditions can provide.
Similar reasoning motivates other presumptions in favour of prosecution: other cases in which, according to CPS guidance, prosecution is ‘usually’ or ‘almost always’ required. Most such presumptions are rationalised in terms of the seriousness of the offending concerned, the likely sentence, or both. Some relate to types of offending that are inherently serious and likely to attract custodial sentences, such as homicide,Footnote 60 firearms offences,Footnote 61 and non-consensual sexual offences.Footnote 62 Others relate to particular examples of offending that are likely to attract such sentencesFootnote 63 or that involve certain aggravating features.Footnote 64 The guidance implies here that seriousness not only strengthens the case for prosecution but also requires it above a certain threshold. Research evidence again suggests that prosecutors themselves often share the same view.Footnote 65
Perhaps unsurprisingly, no deeper explanation of this view emerges from CPS guidance. An explanation might begin, however, with the Code’s instruction that prosecutors ‘must always act in the interests of justice’.Footnote 66 Justice, some would argue, demands that offenders be held accountable for their wrongdoing. Prosecutors should therefore call people to account where there is sufficient evidence that they have committed an offence.Footnote 67 This obligation is not necessarily absolute: it might be outweighed by at least some countervailing factors, and for the reasons already noted, it might sometimes be met through out-of-court disposals. Nevertheless, it entails that prosecutors have inherent reason to seek an appropriate penalty in cases that pass the evidential test. Such an underlying view would explain why the primary question at the public interest stage is apparently what the appropriate penalty is, rather than whether any penalty should be sought at all.
Seriousness and appropriate sentence are also the dominant considerations in guidance on charge selection. According to the Code, prosecutors should select charges that ‘reflect the seriousness and extent of the offending’ and ‘give the court adequate powers to sentence’.Footnote 68 In the guidance, most instructions on charge selection reflect these principles, albeit in differing ways. Notably, these instructions prioritise ensuring adequate sentencing powers over other ways of reflecting seriousness. Where prosecutors have a choice between overlapping charges, they are often told to choose the one that is more specificFootnote 69 or that better describes the overall offending.Footnote 70 However, these principles tend to be overridden where a more generic offence carries a higher sentence that matches the seriousness of the alleged conduct.Footnote 71 Again, prosecutors’ primary responsibility is seemingly to enable a sentence that reflects the seriousness of the offending concerned.
The same responsibility underlies guidance on the acceptance of guilty pleas. According to the Code, prosecutors should accept only those pleas that will enable the court ‘to pass a sentence that matches the seriousness of the offending, particularly where there are aggravating features’.Footnote 72 In the guidance, this principle again informs most of the specific instructions on the acceptance of pleas to lesser charges. Prosecutors must consider whether such pleas would likely result in a lower sentence than the evidence might support,Footnote 73 or in some cases, should simply refuse to accept them for the same reason.Footnote 74
In summary, the prominence of seriousness in guidance on the public interest test reflects two related principles. First, prosecutors should ensure that offending of which there is sufficient evidence can be met with an appropriate penalty. This principle underlies limits on the use of out-of-court disposals, as well as guidance on which charges prosecutors should select and continue with. Secondly, the strength of the case for prosecution depends largely on the seriousness of the alleged offending. In relation to less serious offences, that case is weaker. Above a certain threshold of seriousness, however, it seems to be overwhelmingly strong. Demands of justice, it is perhaps implied, require prosecution in order that an appropriate penalty can be imposed.
3. Other reasons to prosecute
The case for prosecuting those against whom there is sufficient evidence thus depends primarily on the appropriate penalty for their alleged offending. The next question is whether the appropriate penalty is all that matters. If an offence is likely to attract only a nominal penalty, does it follow that there is no public interest in prosecution? And if a charge is unlikely to add to a defendant’s sentence, does it follow that there is no public interest in continuing with it? In practice, the answer to these questions may be ‘yes’. By far the most frequently given reasons for discontinuance on public interest grounds are that the penalty would be small, nominal, or not significantly increased by continuing with the charge.Footnote 75 At least in theory, however, CPS guidance is now clear that other factors can justify prosecution besides impact on the likely sentence. These factors are the focus of this section.
The clearest recognition of other reasons to prosecute can be found in a specific policy on cases likely to end in nominal penalties. According to this policy, prosecution in such cases can still be in the public interest, partly because of the value of conviction itself. Conviction provides formal recognition of the offender’s wrongdoing. It ensures that justice is seen to be done and that any harm to the victim is acknowledged. It is also held on the offender’s criminal record, which may provide information relevant to other decisions, such as decisions about safeguarding.Footnote 76 We might add that conviction allows courts to make ancillary orders, such as disqualifications, preventive orders, and orders for confiscation.
Policies on specific offences rarely list the formal value of conviction as a pro-prosecution factor. However, they sometimes list the availability of particular ancillary orders as such.Footnote 77 The potential uses of criminal records also provide reasons to prosecute some offences. A conviction may be important to decisions on, for example, paroleFootnote 78 or licensing applicationsFootnote 79 even where the likely penalty is minor. Similar benefits of conviction affect which charges prosecutors should select and which guilty pleas they should accept. For both types of decision, the Code requires prosecutors to ensure that appropriate ancillary orders can be made.Footnote 80 In some places, the guidance likewise lists the availability of specific orders as a reason for selecting or continuing with particular charges.Footnote 81 Very occasionally, even where continuing with a charge would not affect the likely sentence, other benefits of conviction can tell against accepting a guilty plea.Footnote 82
Prosecution can even be in the public interest when an outcome other than conviction is likely. This can occur in two scenarios. The first is where the court is likely to find the suspect not guilty by reason of insanity. The second is where the court is likely to find the suspect unfit to plead but, following a fact-finding hearing, to find that they did the act alleged. The Code stipulates that these cases pass the evidential test.Footnote 83 And guidance confirms that prosecution in these cases can still be in the public interest. This is because these outcomes have values analogous to those of conviction: they may provide a measure of justice in themselves, and they empower the court to make hospital or guardianship orders. They also provide information that is relevant to further decisions, such as risk assessments and decisions about treatment.Footnote 84
The public interest in prosecution, then, can consist in formal outcomes of criminal proceedings besides punishment. It is less clear how far it can consist in other positive consequences associated with these outcomes. Sometimes, it is counted in favour of prosecution that it would disrupt ongoing offending or lead to the incapacitation of dangerous offenders.Footnote 85 Occasionally, more general benefits, such as deterrence and maintaining community confidence, are also counted in its favour.Footnote 86 However, such benefits are rarely mentioned in relation to specific offences, and there is no suggestion that they can justify prosecution by themselves. Rather, they seem to add weight to the case for seeking a conviction or other formal outcome, which becomes salient mainly when a significant penalty is unlikely.
Can we go further and say that the public interest in prosecution consists only in the value of such formal outcomes? Although this position is nowhere stated explicitly, it is implicit in some aspects of CPS guidance. For example, the potential benefits of the prosecution process, as distinct from its outcomes, are almost never recognised as valid reasons to prosecute. There is little suggestion that, say, the value of fact-finding in open court can justify proceeding with a chargeFootnote 87 and none that prosecution itself may be used as a deterrent or informal punishment.Footnote 88 Moreover, there is seemingly no public interest in prosecution where no formal disposal whatsoever is available. This can occur in one exceptional type of case: non-imprisonable offences in the magistrates’ court where the suspect is unfit to plead. Hospital and guardianship orders are unavailable in these cases,Footnote 89 and magistrates have no other disposal powers. Prosecution is therefore unlikely to be in the public interest,Footnote 90 which suggests that some potential disposal is required for prosecution to be justified.
To summarise, CPS guidance recognises inherent reasons to prosecute in almost all cases where there is sufficient evidence to do so. These reasons derive primarily, and perhaps entirely, from possible formal outcomes of criminal proceedings. As we saw in the previous section, the most important such outcome is the appropriate sentence for the alleged offending. The outcomes discussed in this section tend to become salient only when a charge is unlikely to lead to a significant (increase in) sentence. Nevertheless, the value of conviction and its consequences clearly supply reasons to prosecute even where the likely penalty is nominal. Since conviction is possible in almost all cases that pass the evidential test, prosecutors almost always have some reason to prosecute at the public interest stage.
There is, however, one significant exception to this general rule: a group of presumptions against prosecution that seem to embody a de minimis principle. In these cases, the conduct concerned technically constitutes an offence; however, prosecution is unlikely to be in the public interest, as the mischief targeted by the offence is deemed absent. The most important examples concern sexual offences that do not require non-consent: for example, offences of sexual activity with children or adult relatives. The CPS has long had a policy of not prosecuting these offences where the activity is consensual, the parties are similar in age, and there is no evidence of coercion or exploitation.Footnote 91 Similar thinking now underlies presumptions against prosecution in several other contexts.Footnote 92 The implication is that there is not automatically any public interest in prosecution simply because the conduct concerned breaches the letter of the criminal law.
The principle underlying these presumptions is nowhere made explicit. Mostly, though, it seems to be limited in its scope. It is not merely that the conduct concerned is non-serious and that the public interest in prosecution is therefore minimal.Footnote 93 Rather, it is that the conduct does not involve the kind of harm or wrong that the offence concerned is assumed to target.Footnote 94 What is less clear is how prosecutors should determine what that harm or wrong is. Some of the policies cited in the previous paragraph refer explicitly to legislative intention, but others do not. This point is significant partly because there seem to exist further presumptions against prosecution that are not acknowledged in CPS guidance.Footnote 95 Although such presumptions can seemingly be legitimate, no general principle is articulated that explains when or why.Footnote 96
4. Reasons against prosecution
With limited exceptions, then, prosecutors have inherent reason to prosecute at the public interest stage. As the Code acknowledges, however, there may be factors tending against prosecution at this stage that outweigh those tending in favour. This section examines the most important such factors. Although it is not completely exhaustive, almost all reasons against prosecution identified in CPS guidance can be grouped under the headings below.Footnote 97 As we shall see, only a limited range of such reasons are identified, and some of these are relevant only conditionally. The combined result is that CPS guidance effectively imposes a presumptive obligation to prosecute in cases that pass the evidential test.
(a) Cost
Contemporary criminal justice systems operate with high caseloads and limited resources. Under such conditions, some charges that are supported by sufficient evidence will inevitably escape prosecution, such that prosecutors must judge which cases they have strongest reason to pursue. This concern underlies the sixth of the seven public interest questions in the Code. Prosecutors must consider the cost of prosecution ‘to the CPS and the wider criminal justice system, especially where it could be regarded as excessive when weighed against any likely penalty’.Footnote 98 Cost is rarely mentioned, by contrast, in guidance on specific offences. This likely reflects the fact that all prosecutions consume time, money, and resources. Hence, this provision in the Code is notable mainly for framing the relevance of cost as a question of proportionality to the likely outcome.Footnote 99 Once again, the likely penalty is the primary outcome that prosecutors are required to consider.
Cost could also be relevant to decisions on charge selection and the acceptance of guilty pleas. Prosecutors might, for example, seek to reduce the cost of a trial by selecting charges that are less complex or easier to prove. They might also eliminate such costs altogether by accepting a guilty plea to fewer or lesser charges and thus avoiding a trial. In relation to these decisions, however, both the Code and CPS guidance are reluctant to portray cost as a relevant consideration. Prosecutors ‘must never accept a guilty plea just because it is convenient’,Footnote 100 and some policies provide that they should not consider convenience or ease of proof in choosing between charges.Footnote 101 On the other hand, the Code encourages prosecutors to select charges that ‘enable the case to be presented in a clear and simple way’.Footnote 102 And in the guidance, clear presentation occasionally shades into ease of proof: at least where two charges carry a similar sentence, the one that is easier to prove may sometimes be chosen.Footnote 103
This ambivalence about the proper role of cost in prosecution decisions is perhaps revealing. In reality, its role is probably more significant than the Code and guidance allow. Plea bargaining in particular has been characterised as a process akin to factory farming, in which the efficient production of guilty pleas is prioritised over other considerations, including the justice of the outcome.Footnote 104 For our purposes, however, the point is that such an efficiency-driven approach is contrary to the principles that CPS guidance apparently endorses. Prosecutors should prioritise cases based on the likely penalty; and when they choose prosecution, the charges pursued should allow for a penalty that is at least not disproportionate to the offending concerned.Footnote 105
(b) Legal obligations not to prosecute
Prosecutors have numerous legal obligations that can require non-prosecution of an offence. The most important are obligations arising from international conventions – especially the European Convention on Human Rights (ECHR), but also those relating to, for example, refugeesFootnote 106 and victims of human trafficking.Footnote 107 Such obligations are arguably not best understood as public interest factors.Footnote 108 Indeed, both the Code and the general guidance on the ECHR classify them as considerations separate from the Full Code Test.Footnote 109 Other CPS policies, however, classify compatibility of a prosecution with international obligations as a public interest matter. These obligations should therefore be noted briefly here.
The most frequently discussed such obligation derives from the right to freedom of expression under Article 10 of the ECHR. CPS policies sometimes take Article 10 to limit the kinds of conduct that may be prosecuted.Footnote 110 More often, however, they require freedom of expression to be balanced against other public interest factors. The most striking examples are the policies on cases involving the media. These policies examine the relevant human rights jurisprudence in great detail. Ultimately, however, they require prosecutors to balance the seriousness of the alleged offending against the public interest that it serves: for example, in disclosing wrongdoing or contributing to public debate.Footnote 111 Such a requirement to balance incommensurable factors again leaves room for concern about the considerations that will in fact prove decisive in these cases.Footnote 112
(c) Harms of prosecution: victims, witnesses, and other third parties
Prosecution can cause various harms to innocent third parties. We might distinguish initially between harms resulting from conviction and punishment and harms of prosecution in a stricter sense. The latter are the harms caused by the prosecution process itself, even when it does not end in conviction. Rogers argues that prosecutors are responsible only for these latter harms. They therefore need not account for the harms of punishment, including its ‘incidental effects’ on persons other than the offender.Footnote 113 As a rough generalisation, CPS guidance complies with Rogers’ principle. While it rarely addresses harms of punishment, it treats harms of prosecution to third parties such as victims and witnesses as reasons against it.
Harms of punishment are recognised as public interest factors tending against prosecution in only a few cases. These include both harms to specific individualsFootnote 114 and general harmful effects of enforcing the criminal law.Footnote 115 In cases of solicitation, for example, prosecutors should consider the impact of enforcement on sex workers’ safety: ‘it may be better that [solicitation] takes place in well-lit and safe places rather than being displaced’.Footnote 116 Such instructions are exceptional, however. Punishment often harms persons other than offenders, and law enforcement can be harmful or counter-productive in contexts other than sex work. Yet there is no suggestion that this consideration counts against the prosecution of, say, drug possession.Footnote 117 Overall, CPS guidance does not embody a general principle requiring prosecutors to consider such harms.
Harms of prosecution, by contrast, are often recognised as factors tending against it. The most important examples are harms to victims and witnesses. Besides the stress and inconvenience of attending court, these groups may be forced to re-live the offence when giving evidence. They may also face damaging questions about their credibility during cross-examination. Prosecution may thus be likely to damage victims’ and witnesses’ mental health – especially, but not exclusively, if they are children or otherwise vulnerable.
The Code alludes to this possibility in relation to victims. Prosecutors must ask whether ‘prosecution is likely to have an adverse effect on the victim’s physical or mental health’. However, this possibility need not be decisive against prosecution: prosecutors must also consider ‘the availability of special measures and the possibility of a prosecution without the participation of the victim’.Footnote 118 Other witnesses are not mentioned in the Code but are considered in CPS guidance. In some places, the same principle is suggested: likely harms to witnesses count against prosecution, but prosecutors must also consider special measures and proceeding without their participation.Footnote 119 In other places, the focus is entirely on support measures that might assist witnesses.Footnote 120 Prosecutors’ first resort, in short, should be to minimise these harms where possible.
In some cases, the likely effects of prosecution may be so adverse that victims and witnesses are unwilling to give evidence. Besides the fears already mentioned, they may be dependent on the suspect or face retaliation from them. CPS guidance addresses this issue in several contexts and again suggests similar principles for dealing with it. Ideally, prosecutors should implement protective measures for the victim or witness and thus (re)gain their support. Alternatively, they should find ways of proceeding without them: for example, by applying to adduce their evidence as hearsay. As a last resort, they may consider a summons requiring them to give evidence rather than discontinuing the prosecution. This choice requires a balancing of the public interest in the prosecution continuing against the impact of this course on the victim’s or witness’s safety.Footnote 121
A related issue is the views of the victim on whether the suspect should be prosecuted. Does it count in favour of prosecution that the victim would prefer this option or against it that they would not? In the former case, prosecutors must consider the victim’s views in deciding whether to accept a pleaFootnote 122 or offer an out-of-court disposal.Footnote 123 The reference here is deliberately to victims’ views, not merely their interests; however, prosecutors should treat these views only as weighing factors and not as decisive against outcomes of these kinds.Footnote 124 In the latter case, by contrast, the guidance strongly implies that victims’ views are not relevant in themselves. As just noted, non-prosecution should be a last resort where the victim does not support a prosecution. Even then, it is the likely impact of prosecution on the victim, not their views, that prosecutors should consider. Similarly, the Code does not mention victims’ views on prosecution and emphasises that ‘the CPS does not act for victims … in the same way as solicitors act for their clients’.Footnote 125
What about harms of prosecution to third parties other than victims and witnesses? One such group is the object of the seventh and final public interest question in the Code: harms to sources of information, such as police informants and undercover operatives. In some cases, prosecutors can protect the identity of these sources through an application for public interest immunity. In others, however, ‘details may need to be made public that could harm sources of information, ongoing investigations, international relations or national security’.Footnote 126 In these cases, disclosure can be avoided only if a prosecution does not take place or is discontinued. Harms to sources of information must therefore be treated as a public interest factor tending against prosecution.Footnote 127
In short, in relation to third parties, CPS guidance roughly reflects Rogers’ distinction between harms of prosecution and harms of punishment. While the former supply reasons against prosecution, the latter generally do not. Even harms of prosecution, however, are relevant only conditionally. Prosecutors should prefer ways of avoiding them that allow the prosecution to continue. Even when there is no such alternative, prosecutors must balance these harms against the public interest in prosecution. Overall, harms to victims, witnesses, and informants are treated as regrettable side-effects of prosecution that prosecutors should minimise if possible, rather than as decisive reasons against it.
(d) Harms of prosecution: suspects and defendants
Prosecution also adversely affects suspects and defendants. It can lead, of course, to conviction and punishment, as well as the collateral consequences that they entail. But defendants can likewise suffer harms of prosecution even when they are not convicted. Prosecution itself will often cause stress, stigma, and fear of prospective punishment. It can affect the defendant’s employment or finances. It can also lead to restriction of their liberties: a court may impose bail conditions or remand them in custody pending their trial.Footnote 128 Generally, such harms are not recognised as factors tending against prosecution in either the Code or CPS guidance. They are treated as such only in relation to two groups: children and those suffering from mental ill health.
Before examining these exceptions, we should ask why harms of prosecution to defendants might be irrelevant outside these two cases. Perhaps the obvious answer is that offenders in some sense deserve to endure such harms. Just as they cannot legitimately complain about being convicted and punished, so they cannot legitimately complain about enduring the process that leads to these outcomes. This answer has explanatory power when coupled with the widespread assumption that defendants are in fact guilty.Footnote 129 One function of the evidential test, it might be argued, is to justify this assumption, and thus to help confine the harms of prosecution to those who deserve them.Footnote 130 My intention here is not to defend this principle: perhaps there are compelling arguments against the idea that harms of prosecution can be deserved.Footnote 131 The point is that some such defence would be necessary to justify accounting for harms of prosecution in relation to third parties while disregarding them in relation to suspects and defendants.
Turning now to the exceptions, consider first child suspects. As noted earlier, the suspect’s age and maturity are among the public interest factors identified in the Code. Youth is relevant partly because it speaks to the suspect’s culpability. However, the CPS has emphasised that it also generates special concerns.Footnote 132 These include ‘whether a prosecution is likely to have an adverse impact on [the suspect’s] future prospects that is disproportionate to the seriousness of the offending’.Footnote 133 Prosecutors must also ‘have regard to the principal aim of the youth justice system, which is to prevent offending by children and young people’.Footnote 134 Although these considerations do not obligate prosecutors to choose diversion for child suspects, it counts against prosecution that it would frustrate these welfarist and rehabilitative aims.Footnote 135
Consider next suspects with mental health issues. According to the Code, it can count against prosecution that ‘the suspect is, or was at the time of the offence, affected by any significant mental or physical ill health or disability’.Footnote 136 Although this factor is listed as one affecting the suspect’s culpability, it encompasses suspects who are currently ill, as well as those who were ill at the time of the alleged offence. The implication, again, is that ill health is also relevant to the impact of prosecution on the suspect.Footnote 137 The guidance confirms this impression at least in relation to mental ill health. Prosecutors must consider the impact of prosecution on the suspect’s mental health, particularly ‘where there is evidence that it can be exacerbated, or it is degenerating’.Footnote 138 Again, the CPS has emphasised that this consideration is special to those with pre-existing conditions and should be distinguished from stress brought on by prosecution itself.Footnote 139
What justifies these exceptions to the apparent rule that harms of prosecution to suspects and defendants are irrelevant? In relation to children, legal obligations might again supply an answer. Prosecutors have duties to consider children’s best interests,Footnote 140 and how best to prevent them from offending,Footnote 141 that they lack in relation to adults. In relation to mental ill health, however, a justification is harder to see. The decisive factor here is seemingly whether the mental health impact of prosecution involves the exacerbation of a pre-existing condition. Yet it is not obvious why the relevance of such harms to prosecution decisions should depend on their precise aetiology. It is therefore unclear whether a principled rationale for CPS guidance in this area can be advanced.
(e) Enabling prosecution of accomplices
In the context of crimes committed by groups or organisations, participants are often both suspects and potential witnesses. Prosecutors might thus offer non-prosecution to some participants to encourage them to give evidence against others. As we have seen, CPS guidance generally requires prosecutors to seek an appropriate penalty where there is sufficient evidence against a suspect. On that principle, the non-prosecution of participant witnesses is obviously problematic. At best, it represents a sacrifice of justice in one case in the name of doing justice in others. We might therefore expect any permission not to prosecute in these cases to be correspondingly limited: for instance, to cases in which non-prosecution is truly necessary to secure evidence that will enable a more serious conviction and penalty.
Such a principle indeed seems to underlie CPS guidance on this issue. The most detailed treatment is in a dedicated policy on ‘assisting offenders’.Footnote 142 The starting point of this policy is that the public interest normally requires prosecution of participant witnesses. Prosecutors should prefer alternative ways of encouraging co-operation, such as sentence reductions and undertakings not to use suspects’ evidence against them. In exceptional cases, however, non-prosecution may be appropriate. This will depend partly on the relative seriousness of the offence(s) that the participant may have committed. It will also depend on the importance of their evidence to the case against their accomplices. Non-prosecution may be unnecessary if the case is likely to succeed anyway or the evidence could be obtained from elsewhere.
In this section, we have seen that CPS guidance recognises only a limited range of reasons against prosecution. The limitations reflect several underlying principles. Cost is relevant mainly because prosecutors must prioritise among cases. And some harms resulting from prosecution are not relevant at all – perhaps because they are thought to be deserved, or perhaps because prosecutors are thought not to be responsible for them. Other factors tending against prosecution are relevant only conditionally. In these cases, enabling a prosecution should be the priority, and non-prosecution should be chosen only if necessary to avert a greater harm or injustice. CPS guidance thus imposes only a limited obligation to weigh the inherent reasons in favour of prosecution at the public interest stage against its negative effects.
5. Alternatives to prosecution
A final set of public interest factors concerns the availability of alternatives to prosecution. As we saw earlier, these alternatives are often considered suitable because they allow for an appropriate penalty for offending to be imposed at a lower cost. Sometimes, however, CPS guidance instructs prosecutors to prefer such alternatives for other reasons. Some cases are deemed properly dealt with outside the criminal process, while in others, rehabilitative or welfarist aims should supplant punitive ones. As we shall see, no explanation is apparent of why the usual requirement to seek an appropriate penalty is suspended in these cases, or why only certain suspects should benefit from the alternatives on offer. It is therefore difficult to identify any principle underlying CPS guidance on this issue.
Consider first cases that are deemed properly dealt with through mechanisms outside the criminal process. A prominent example is disciplinary action taken by other bodies. Prison disciplinary procedures, for instance, might be a better way of dealing with offences committed in prisons.Footnote 143 Professional discipline must also be considered in relation to some offences, most notably those committed by police officers,Footnote 144 but also some committed by other public servants.Footnote 145 The logic here is presumably that disciplinary sanctions – including, potentially, losing one’s job – might represent adequate ‘punishment’ for an offence. It is unclear, however, why this logic should apply only to police officers and some other public servants when it could be extended to anyone whose employment is threatened by prosecution.
Other examples are cases where the guidance deems civil rather than criminal proceedings to be appropriate. The most notable examples are from the policy on fraud offences.Footnote 146 Prosecutors must ensure that the criminal law is not ‘used as a debt collection agency’ in fraud cases or to protect commercial interests or ‘private confidences’. Where the case involves a dispute about ownership of property, it should be settled in the civil rather than the criminal courts. Civil proceedings may also be the appropriate way of dealing with ‘advertising puffs’ and with cases where a person gains by taking sensitive commercial information from a former employer. Again, it is unclear why civil proceedings should be preferable in these cases but apparently not in most others.
Consider next cases in which prosecutors are instructed to pursue non-punitive aims. In some cases, CPS guidance instructs prosecutors to prefer conditional cautions to prosecution in order to promote desistance from offending. One example is the presumption against prosecuting sex workers for solicitation and related offences. Prosecutors’ aim in these cases should be ‘to encourage [sex workers] to engage with support services and to find routes out of prostitution’.Footnote 147 In other cases, welfarist interventions are considered potentially appropriate – whether from agencies such as social services or even informal intervention by parents.Footnote 148 A notable example is public justice offences involving false allegations of sexual offences and/or domestic abuse. Here, the avoidance of a criminal justice response is considered justified by the fact that suspects in these cases are often vulnerable.Footnote 149 Why other vulnerable suspects should not benefit from the same consideration is again left unaddressed.
Analogous considerations arise in relation to corporate prosecutions. In recent years, prosecutors have acquired a power to negotiate deferred prosecution agreements (DPAs) with corporate suspects. Corporations agree to the terms of DPAs in exchange for the suspension of criminal proceedings against them. Terms may include the payment of fines and compensation, but also (among other things) the institution of compliance programmes.Footnote 150 Correspondingly, the promotion of future compliance supplies reasons against prosecuting corporations and in favour of offering DPAs. Relevant considerations include the corporation’s co-operation with investigators, its introduction of compliance measures, and whether there has been cultural or structural change within the company.Footnote 151 Such considerations also typically motivate the judgments of courts, who, unusually, are required to approve this alternative to prosecution. Significantly, offence seriousness seems not to override these considerations as it generally does in relation to individual suspects.Footnote 152
These examples show that an obligation to seek an appropriate penalty for offending is not the inevitable starting point for prosecution policy. Prosecutors could instead ask what course of action is most likely to secure future compliance with the law – a question to which the suspect’s welfare and future prospects are likely to be relevant. Such an approach is infamously already taken by most regulatory agencies. Many have doubted that there is any justification for the discrepancy between their approach and the generally pro-prosecution stance of the CPS.Footnote 153 For our purposes, however, the notable point is that this approach occasionally takes over even in CPS policy. We have seen how welfarist and rehabilitative aims become relevant in relation to child suspects, as well as in the cases mentioned in this section. What remains unclear is why these aims should supplant punitive ones only in this limited set of cases. This lack of clarity is particularly unfortunate given that alternatives to prosecution are mentioned only briefly in the Code.Footnote 154
Conclusion
In this paper, I have sought to provide an account of when prosecution is in the public interest. I have argued that, in most cases, CPS guidance recognises inherent reasons to prosecute those against whom there is sufficient evidence. These reasons derive primarily, although not exclusively, from a requirement to seek an appropriate penalty for offending behaviour. The more serious the offending, the more likely it is that prosecution is required to enable an appropriate penalty. CPS guidance, moreover, recognises only some adverse effects of prosecution as factors weighing against this requirement. Even when such effects are considered relevant, prosecutors should prefer options that minimise them and that allow a prosecution to continue. Not only, therefore, can crime control imperatives override prosecution guidance in practice;Footnote 155 CPS guidance itself steers prosecutors towards prosecution to a greater extent than a casual reading of the Code might suggest.
As emphasised at the outset, I have not tried in this paper to assess whether the requirements and principles that I have discussed are normatively defensible. To conclude, however, it is worth noting two implications of my analysis. First, we noted that the Code for Crown Prosecutors is intended to explain CPS decision-making to the public in easily understandable terms. In that light, it is notable that many of the principles highlighted above are not stated clearly in the Code.Footnote 156 Secondly, it has proved difficult to find principled rationales for some aspects of CPS guidance. In relation to some topics (such as de minimis cases and the harms of prosecution), the apparent principles are at best implicit and observed inconsistently. In relation to others (particularly the relevance of aims other than seeking an appropriate penalty), no principles whatsoever are apparent. This paper therefore highlights at least a need for greater clarity and internal coherence in guidance on the public interest test.
Open access
