October 24, 2025
By Lize Glas
In Hora v. the United Kingdom, published on 23 September 2025, the European Court of Human Rights (the Court) had to decide whether the applicant’s disenfranchisement was compatible with Article 3 of Protocol 1 (P1), which protects the right to free elections. However, the ongoing dialogue for more than twenty years between the UK, the Court and the Committee of Ministers (CM) regarding the disenfranchisement of convicted prisoners required the Court to do much more. It had to respond to execution measures approved by the CM in the meantime, and explain why it did not find a violation in Hora’s case despite having found one in previous cases involving similar facts.
Hora is serving an indeterminate prison sentence for two counts of rape and one count of sexual assault. When the UK held a general parliamentary election in 2019, he was ineligible to vote based on Section 3 of the 1983 Representation of the People Act (the 1983 Act), which states that person sentenced to prison cannot vote while serving their sentence.
Hora is not the first convicted prisoner to bring a complaint against the UK under Article 3 P1 because of his disenfranchisement. In Hirst v. the UK (No 2) (2005), the Court described the blanket ban in the 1983 Act as a ‘blunt instrument’ that disenfranchises ‘a significant category of persons’. This ‘general, automatic and indiscriminate restriction’ fell outside ‘any acceptable margin of appreciation’ and violated Article 3 P1 (para 82). In a subsequent pilot judgment, Greens and M.T. v. the UK (2010), the Court found a similar violation because the 1983 Act had not been amended, but rather extended to cover European Parliament elections. The Court, under Article 46 of the Convention (binding force and execution of judgments), ordered the UK to ‘bring forward, within six months of the date upon which the present judgment becomes final, legislative proposals intended to amend the 1983 Act […] in a manner which is Convention-compliant’ (para 6(a) operative provisions). The required legislation had to be enacted within a period determined by the CM, which supervises the implementation of Strasbourg Court judgments.
These judgments caused quite a stir in the UK, with Parliament refuting any proposed legislation to change the 1983 Act. The domestic authorities explained this position in Strasbourg. In a final resolution in 2018, the CM accepted the impossibility of legislative amendment and closed its supervision of the implementation of the Hirst group of cases, despite having previously called for this execution measure. In the final resolution, the CM recalled ‘the wide margin of appreciation in this area’ and ‘noted the administrative measures taken and in particular the changes to policy and guidance to make it clear that two categories of prisoners who were previously effectively disenfranchised (prisoners released on temporary licence and on home detention curfew) are now able to vote’. As only prisoners released on temporary licence or home detention curfew are permitted to vote, i.e. mainly those serving short sentences, probably only around 100 prisoners were enfranchised through this measure, while the UK’s total prison population was over 80,000 people in 2018.
As others have also noted in blog posts on this issue and elsewhere, it was doubtful whether the execution measures were fully compliant with the previous judgments. This was due not only to their form (administrative rather than legislative amendments), but also to their content. After all, the administrative measures do not apply to convicted prisoners who are permanently in prison. These prisoners are still subject to a blanket ban on voting which, in the Court’s words in Hirst, ‘applies automatically […], irrespective of the length of their sentence and irrespective of the nature or gravity of their offence and their individual circumstances’ (para 82).
Prior to dealing with the applicant’s complaint, the Court discusses three preliminary matters. First, it clarifies its Article 46 indication in Greens and M.T., stating that the supervisory process is not restricted to such indications. If this were the case, the CM would not be able to supervise flexibly ‘on the basis of the information provided by the respondent State and with due regard to the evolving situation, the adoption of measures that are feasible, timely, adequate and sufficient’ (para 121). Furthermore, the Article 46 indication did not require legislative amendment, ‘but rather proceeded on the basis that the Government and the [CM] had already identified such amendment’ as the required execution measure (para 121).
Second, the Court highlights three relevant developments that have occurred since Hirst. It summarises its new case law on the disenfranchisement of prisoners and explains certain developments in the UK, including that the government and parliament have considered amending the 1983 Act, but decided against doing so. Lastly, the Court refers to a UK Supreme Court judgment in which the UK judges, based on post-Hirst Strasbourg case law, refused to declare Section 3 of the 1983 Act incompatible with the Convention.
Third, the Court summarises the content of the CM’s final resolution of 2018 and the process leading up to its adoption. It then concludes that it will not conduct an abstract review, but will instead examine whether applying the 1983 Act to the applicant was compatible with Article 3 P1.
The Court finds no violation of Article 3 P1. The interference, it concludes, was proportional due to ‘the seriousness of [the applicant’s] offending, his conduct, the risk he was found to pose to the public and the resulting imposition of a harsh sentence of indeterminate detention’ (para 133).
Hora’s application required the Court to address the challenging situation in which the CM had approved administrative measures as very minimalistic execution measures. As noted, these measures arguably did not fully comply with the previous judgments. Given the heated domestic debate that took place after the adoption of the previous judgments, not siding with the CM was probably not an option if the Court was to avoid jeopardising its relationship with the CM and the UK. The Court gives two reason for accepting the administrative measures in Hora.
The first reason is that its Article 46 indication in Greens and M.T. ‘did not stipulate that legislative amendment was required’ (para 121). However, this explanation is not in line with the text of the indication to ‘bring forward […] legislative proposals’ (para 6(a) of the operative provisions). Arguably, this reason was unnecessary because the other reason is more convincing, meaning that the Court did need to deny the text of the indication.
The second reason is that the CM must retain the flexibility to take account of new developments during the execution phase, including when the Court gave an Article 46 indication. As the Court previously explained, the CM needed this flexibility when supervising the execution of Al-Saadoon and Mufdhi v. the UK, for example. In that case, the Court asked the UK, under Article 46, to take ‘all possible steps to obtain an assurance from the Iraqi authorities that [the applicants] will not be subjected to the death penalty’ (para 171). However, the CM did not require the UK to take these steps once it became clear that the applicants had been acquitted. Yet, ignoring the Court’s indication in the pilot judgment Greens and M.T. is of a different nature compared to ignoring the indication in Al-Saadoon and Mufdhi. While the indication was included in the operative part of the pilot judgment, this was not the case in the other judgment. Furthermore, changed factual circumstances meant that it was no longer necessary to comply with the individual measure in Al-Saadoon and Mufdhi; however, this did not apply to Greens and M.T., which, moreover, included general measures. In view of these differences, the CM has opened the door somewhat wider for states wishing to argue that they do not have to implement an Article 46 indication.
If the CM had referred the question of interpretation of the operative provision in Greens and M.T. to the Court, it would have given the Court a bit more control over this challenging situation. Article 46(3) of the Convention empowers the CM to refer a question of interpretation with a two-thirds majority if that question hinders the execution of a judgment. While it would still have been difficult for the Court not to side with the CM, at least it would been part of a dialogue and consulted prior to the final resolution being presented to it as a fait accompli.
In cases concerning the disenfranchisement of prisoners, the Court essentially has to decide on two issues: whether to review the impugned legislation in the abstract or as it was applied to the applicant, and whether the disenfranchisement was proportionate. Whether an interference took place is usually not in dispute, and the Court has accepted that the disenfranchisement of prisoners pursues different legitimate aims, including preventing crime and ‘enhancing civic responsibility and respect for the rule of law’ (para 76).
The Court’s approach to the first issue is not entirely consistent (as Gerards has observed more broadly). In Hirst, Greens and M.T. and Kulinski and Sabev v. Bulgaria, for example, the Court conducted in abstracto review. However, in Kalda v. Estonia (No 2), the Court engaged in concrete review, which revealed that, despite the existence of a blanket ban, the applicant’s right to vote had not been violated, because national judges had reviewed the proportionality of applying the ban to the applicant. In Scoppola v. Italy (No 3) and Myslihaka and Others v. Albania, the Court conducted both abstract and concrete review: although it concluded that there had been no violation ‘in the circumstances of the present case’, it also reviewed the applicable legal framework.
In Hora, when explaining its decision to conduct concrete review, the Court referred to the ‘process of dialogue’ regarding the applicable legal framework that took place in the UK. This process ‘plays a crucial role in securing the practical and effective protection of Convention rights in accordance with the principle of subsidiarity’, according to the Court. The Court conducted a concrete review because ‘that dialogue has reached its conclusion’ (para 129). Additionally, the Court referred to the ‘wide margin of appreciation applicable in this area’ to justify concrete review (para 130). In other cases, the quality of the parliamentary process or dialogue has influenced the calibration of the margin of appreciation, rather than the choice between concrete or abstract review. Since it cannot be assumed generally that in concreto review makes a finding of a violation less likely than in abstracto review, applying a wide margin of appreciation in the way the Court did in Hora should not be repeated because it does not necessarily mean more leeway for the domestic authorities.
Despite its reference to the subsidiarity principle (para 129), the Court must review the proportionality of the interference with Hora’s rights itself and cannot rely on proportionality review carried out by the domestic authorities. As the Court explained in Hirst, the legislature did not carry out such a review when adopting the 1983 Act. In Hora’s case, no domestic judge had assessed the proportionality of applying the law to the applicant. This demonstrates that, if domestic authorities do not fulfil their primary responsibility (for example, to assess the proportionality of an interference), the Court must conduct a review that arguably should be avoided in order to function as a true subsidiarity institution.
It is striking that, while the margin of appreciation played an important role in the abstractreview of the interference’s proportionality in Hirst, the Court does not mention the doctrine at all when assessing the proportionality of Hora’s inability to vote. As noted above, the margin of appreciation was wide, but not all embracing, in Hirst due to the lack of parliamentary debate when the 1983 Act was adopted. Other factors that the Court mentioned prior to determining the extent of the margin of appreciation in Hirst are the lack of proportionality review by domestic judges and the lack of European consensus about the disenfranchisement of convicted prisoners.
The Court could have structured its arguments in Hora more along the lines of those in Hirst. It could have engaged in abstract review and relied on the same two factors to calibrate the margin of appreciation, while still concluding that no violation had taken place. While in Hirst the Court wrote that ‘there is no evidence that Parliament has ever sought’ to assess the proportionality of the blanket ban (para 79), in Hora it could have referred to the (parliamentary) dialogue about the 1983 Act that took place in response to the previous judgments. Furthermore, whereas in Hirst the Court noted that the national courts had not engaged in proportionality review, it could have explained in Hora that it now considers this to be less important. More specifically, in Scoppola (No 3), the Court expressly departed from its finding in Frodl v. Austria that ‘the decision on disenfranchisement should be taken by a judge’ (para 34). Finally, the Court could have argued that the system has become less disproportionate due to the administrative amendments. However, from a rights perspective and based on the assumption that the Court was not prepared to overrule the CM’s approval of the administrative execution measures, the individual assessment is to be welcomed, given that the Court can still conclude that the application of the ban is disproportionate in a case with different facts from Hora. When leaving that assumption aside, and departing from the same perspective and that of legal certainty, it would clearly have been preferable for the Court to assess the execution measures in the abstract and to find a violation in the case of Hora on the same grounds as in Hirst.
Hirst was sentenced to discretionary life imprisonment for serious crimes, while Hora received an indeterminate prison sentence for comparable offences. Both were disenfranchised under the 1983 Act. In Hirst, the Court found a violation of Article 3 P1 in 2005. In Hora, it found no such violation approximately twenty years later, despite the fact that the 1983 Act had not been amended. Although administrative measures had been implemented as a result of a domestic dialogue about disenfranchisement, they were irrelevant to either applicant. This difference in outcome stems only from Court’s distinct approaches in the two cases – abstract review in Hirst and concrete review in Hora – and is challenging to justify in terms of legal certainty.
The Court notes that the process of dialogue about the 1983 Act has ‘reached its conclusion’ in the UK (para 129). I doubt that that the same applies to the dialogue between the Court, the CM and the UK regarding disenfranchisement of convicted prisoners. This doubt stems not only from the concrete review in Hora, but also from the clear-cut facts of the case in terms of proportionality. Given the gravity of Hora’s offences (rape and sexual assault) and the resulting indeterminate prison sentence, it was predictable, based on previous case law, that the Court would deem the interference proportionate. However, it is unclear what conclusion the Court will reach in the case of a convicted prisoner serving a ten-month sentence, for example. The UK parliament’s Joint Committee had proposed enfranchising persons sentenced to a term of twelve months or less. Furthermore, this proposal has been implemented in Scotland for local elections and elections to the Scottish Parliament. These developments demonstrate that a case involving the aforementioned facts is far from clear-cut and that the Court might find the interference with Article 3 P1 disproportionate in such a case.
A longer version of this blog will be published as a case note in Dutch on EHRC Updates.
Strasbourg Observers is an academic blog that discusses recent developments at the European Court of Human Rights. The editorial team of Strasbourg Observers is based at the Human Rights Centre of Ghent University and the Centre for Government and Law of Hasselt University.
Strasbourg Observers would not be possible without the valued work of countless guest bloggers. Visit the “Contact” page on our website to find out how to become one!
