Query

What experiences of using amnesties for corruption exist? What have been the outcomes and conditions of success of such initiatives?

Potential risks and benefits of using amnesties for corruption

Amnesties refer to general absolutions of certain categories of crime granted to groups of people or individual offenders or to a decision whereby the state chooses not to prosecute crimes related to certain periods or groups of persons (Roth-Arrazia 2014). As such, amnesties are a legal mechanism adopted by the legislature or the government that exempt certain categories of offences and/or offenders from prosecution and provide some form of immunity (UNODC 2004). However, contrary to immunities, which can also protect a person from criminal or civil liability for crimes that may not have occurred yet, amnesties exclusively relate to past offences (Katz 2017).

The literature primarily discusses the use of amnesties within the framework of transitional justice in connection with human right abuses at the sensitive time of political transition or regime change.

While there are few studies looking at the use of amnesties for economic crimes, some lessons can be drawn from the experience of using them on issues related to human rights abuses. Some authors argue that transitional justice should handle economic rights in the same way it approaches human rights as corruption and human rights violations are mutually reinforcing forms of abuse (Carranza 2008). Focusing exclusively on human right violations and not violations of economic and social rights such as corruption would result in leaving accountability for economic rights to ineffective domestic institutions or nascent anti-corruption mechanisms (Carranza 2008).

The use of amnesty – whether in the case of economic crimes or human right abuses – is a politically controversial measure that challenges the principles of criminal responsibility and accountability. By cancelling the effect of a judicial decision and its consequences and/or stopping judicial investigations, amnesties may also undermine the deterrent effect of sanctioning and weaken the rule of law (Ardigo 2017). In its article 30, the United Nations Convention against Corruption calls states parties to impose sanctions that “take into account the gravity of the offence”, including taking all necessary measures to ensure law enforcement and deterrence in relation to corruption offences and making provision for the removal of corrupt officials from positions or offices where such offences are likely to be repeated. In recent years, however, there has been a growing interest in granting amnesties under certain conditions, although rarely for economic crimes. A recent study of amnesties adopted since 1990 indicates that over 75% of them are related to conflict and that 49% of the peace agreements adopted in the same period provided for some form of amnesty (Mallinder 2018).

As a highly sensitive political decision, amnesties are usually granted on a case by case basis (UNODC 2004). Granting them in cases of corruption, however, has been exceptional and also mainly envisaged for contexts of political transitions when the new governments are willing to have a fresh start and make a clean break from the past (UNODC 2004). Such an approach entails a series of risks and challenges that should be taken into account when assessing the relevance and potential of using amnesties for corruption offences. This answer elaborates on a previous U4 Helpdesk answer on the Use of Amnesties in Tackling Corruption that was published in January 2006 (U4 2006).

The rationale for using amnesties for corruption offences

Dealing with past offences should help achieve accountability for past crimes and prevent future new crimes. In the context of anti-corruption, leniency measures such as amnesties for corrupt practices may be considered – and are often perceived by the public – as measures that are not consistent with the goal of deterrence, accountability and criminal responsibility (UNODC 2004). There is also the fear that they may foster a culture of impunity. Some scholars argue, however, that amnesties can play an important anti-corruption role and contribute to the transformation of the political culture of a particular society (David 2010). In certain cases, the recourse to amnesty laws for economic crimes may also be justified, either for pragmatic reasons or by the specific political circumstances of a given country.

Promoting future anti-corruption reforms and overcoming political resistance to change

The first rationale is to resort to amnesties during times of political transition. In such a critical period, new governments may also primarily need to target resources to implementing new anti-corruption reforms to prevent future acts of corruption rather than on enforcing past standards and regulations. In addition, endemic corruption combined with high public expectations placed on the new government may result in the costly process of prosecuting an overwhelming number of cases, paralysing overstretched judicial systems. This can also potentially lead to the removal of large numbers of civil servants from the public sector and undermine the provision of public services.

Amnesties can then be considered a realistic form of transitional justice that prevents overburdening the court system and allows limited resources to be focused on the most serious violations (Tahir Institute of Middle East Policy 2018). In such contexts, there is also some recognition that civil servants may have been forced into corruption by low wages and economic hardships. When large numbers of low-level officials are involved in corruption, a general amnesty followed by re-training may therefore be preferable to the prosecution costs and need to replace a large number of public officials (UNODC 2004).

One of the most compelling arguments for granting amnesties is the need to achieve peace, stability, reconciliation and democratic governance in countries emerging from conflicts (Han no date). Most studies of amnesties find that they have a positive effect on peace when used as part of negotiated peace processes (Mallinder 2018).

In times of political transition, the need for influential officials to overcome the resistance to change may be directly affected by anti-corruption reforms and can also justify the use of amnesties to gain support for reform. In such cases, amnesty may be granted to senior officials as part of the negotiation process associated with ensuring a smooth and non-violent transfer of power and the return of stolen assets (UNODC 2004).

Encouraging cooperation with law enforcement authorities

By nature, corruption occurs behind closed doors, and successful detection, investigation and prosecution of corrupt offences often depend on insiders or involved individuals reporting corrupt deals and cooperating with the law enforcement authorities. As a result, some forms of amnesty, immunity or mitigation measures may be granted in certain situations for economic crimes, where offenders cooperate with the prosecuting authorities. Article 37 of the UNCAC specifically allows state parties to consider mitigating punishment or granting immunity from prosecution to individuals who “provide substantial cooperation in the investigation or prosecution of an offence”. Granting some form of immunity or leniency provides incentives for persons who participate in crime to cooperate in the investigation/prosecution process and may lead to more information on the forms, nature and extent of corrupt practices and identify weaknesses in current laws and procedures. Immunity provisions are usually left to the discretion of the prosecutors and provide a certain degree of flexibility.

Promoting accountability through the voluntary disclosure of offence

Immunity from prosecution may be granted in exchange for self-reporting or confessing the offence. Such measures, similar to “effective regret”, which applies when offenders report the crime shortly after its commission, seek to promote accountability through truth telling. This approach has been used in several countries, mainly but not exclusively for human right abuses, usually in the form of a public process where perpetrators are required to make full disclosure of their criminal actions. The well-known South African Truth and Reconciliation5c9634126164 Commission is an illustration of this strategy (Kushleyko 2015).

Voluntary disclosure provides the perpetrators with an opportunity to clear the past at a relatively low cost while revealing useful information on corrupt deals and practices. This can help uncover corruption cases that may never have come to light otherwise. Accountability is ensured through the process of publicly admitting guilt and facing public humiliation and stigmatisation. Immunity seekers are encouraged to reveal all information on corrupt practices and identify all other persons involved in the offence.

Recovering the proceeds of corruption

In addition to truth telling, the UNDOC anti-corruption toolkit recommends the redistribution of the corruption proceeds in exchange for immunity from prosecution (UNODC 2004). This approach provides an opportunity to recover – at least partially – ill-gotten gains. In some cases, where the offender cannot fully restore the stolen funds, a tax could be levied on illicit wealth, as has been used in some cases of tax amnesties (World Bank 2001).

While “amnesty-for-truth” is not per se an asset recovery mechanism, it can contribute to establishing a perpetrator’s complicity in large-scale corruption schemes and to tracking hidden assets. There have been some examples of a judicious use of “conditional amnesties” contributing to asset recovery processes. In the Philippines, for example, immunity offered by an inquiry agency led to the disclosure of hidden assets by Ferdinand Marco and supported asset recovery litigation in Switzerland (Carranza 2008).

Potential risks and challenges

The strategy of creating a fresh start by granting amnesty for past corruption offences has some moral, practical and political implications. Prosecution has deterrence, retribution and incapacitation functions – i.e. removing dangerous criminal from society – which may be undermined by the use of amnesties. Amnesties that are too broad create a precedent and are likely to undermine deterrence and rule of law, fostering a culture of impunity where potential offenders simply assume that malpractice will eventually be ignored or amnestied.

Rule of law and victims’ access to redress

It is clear that granting amnesties may undermine the rule of law by providing protection to offenders while depriving victims of access to justice (Han No date). Amnesties also deny victims the right to redress, and such approaches may erode the trust of the public in the government commitment to fight corruption and render justice, ultimately undermining the legitimacy of the regime and the rule of law.

Legitimacy of the regime

Within the framework of “transitional justice” following violent conflicts, amnesties, reconciliation or similar attempts to deal with the past have only been implemented in a few exceptional cases for economic crimes as most serious economic crimes can ultimately affect the legitimacy of the regime (Mallinder 2008). In fact, many “new” governments excluded economic crimes such as embezzlement, extortion and bribery from their amnesty laws, probably willing to distance themselves from the corrosive effect of corrupt practices on regime legitimacy and image (Mallinder 2008). Several amnesty laws covered a range of crimes, including murder and physical injury but excluded fraudulent crimes committed by public officials. For example, extortion practices were not included in the 1987 Salvadorian amnesty law (Mallinder 2008).

Misuse of amnesties by the political elite

The rationale and political readiness for using amnesties in countries where the whole political class is, to a greater or lesser extent, widely involved in corruption, could also be questionable. In such contexts, there could be a general interest among politicians across the board to promote (and misuse) the use of blanket amnesties to cover past or present acts of corruption and preserve a certain form of status quo that benefits the elite.

Intersection between economic and human rights abuses

Both economic and human rights forms of abuse are connected and mutually reinforcing. For example, providing amnesties for corruption abuses leave corrupt dictators access to their ill-gotten assets that they can use to finance destabilisation and intimidation tactics to obstruct justice, delay trials, fight extradition and ultimately undermine efforts to hold them accountable for human rights abuses or corruption. As such, impunity for economic crimes may fuel impunity for human rights abuses, and some authors recommend handling economic crimes as human rights violations (Carranza 2008).

Impunity for international crimes

A duty to prosecute international crimes exists in international law under a number of international treaties and conventions (Han No date). Since granting amnesties implies that there will be no criminal trials in the domestic sphere, this may prevent alleged offenders from being prosecuted in international courts, which can be problematic when dealing with crimes against humanity. Sierra Leone and Uganda provide examples of how national blanket amnesties can conflict with international actions. In 2004, the Special Court for Sierra Leone rendered the first decision of an international criminal tribunal unequivocally stating that amnesties do not bar the prosecution of international crimes before international or foreign courts (Meisenberg 2004). In 2005, the International Criminal Court (ICC) challenged the blanket amnesty that had been granted to the Lord’s Resistance Army (LRA) and issued warrant arrests against five members of the LRA, arguing that “domestic amnesties are strictly a matter for national authorities and do not bar an investigation by the ICC” (The New Humanitarian 2005).

A recent study of the Amnesties, Conflict and Peace Agreements (ACPA) dataset – a dataset of amnesties in all regions of the world from 1990 to 2016 – indicates that 22% of amnesties grant immunities for international crimes such as genocide, crimes against humanity and serious human rights violations, while 23% exclude them. Less than half of these amnesties relate to international crime (Mallinder 2018).

The use of amnesties for corruption offences may also challenge the recovery of assets from other countries as such procedures are generally linked to criminal prosecution and a final court decision. Furthermore, the granting of amnesties should be consistent with international commitments regarding extradition and prosecution.

Forms of amnesties

The diversity of amnesty design

There is considerable diversity in the design of amnesties. They can, for example, be either broad or limited in scope, conditional or unconditional, generous or punitive in their legal effects. There is a growing trend to move away from unconditional blanket amnesties to “smarter” conditional amnesties (Mallinder 2018). In fact, only 37% of the amnesties in the above-mentioned ACPA dataset were unconditional (Mallinder 2018).

Amnesty design can vary greatly in terms of the enactment process, the categories of beneficiaries, the nature of the crimes covered, the conditions attached to the amnesty, and the amnesty’s legal effect and implementation process. The analysis of the ACPA datasets reveals considerable disparity in State practice relating to amnesties, with some aiming to provide victims with a remedy, and others seeking to create complete impunity for perpetrators. To date, few legal trends relating to amnesty laws are emerging, although it appears that amnesties offering unconditional “blanket amnesties” that apply across the board without requiring any initial inquiry into the facts or application on the part of the beneficiary have declined (Mallinder 2018).

Conditions that allow for some form of accountability or strive to address the risks associated with such measures are increasingly attached to amnesties along with the introduction of complementary programmes to repair the harm and prevent a repetition of the crimes (Mallinder 2008 and 2018). Such amnesties are typically attached to a series of conditions such as admitting guilt, providing information or cooperating with law enforcement authorities, paying reparations, a commitment to renounce violence and non-recidivism, participating in community-based justice mechanisms and/or being removed from office. (Mallinder 2014).

In terms of categories of crimes, political offences such as treason, sedition, espionage, rebellion, human rights violations etc. are by far the most amnestied type of offences while only between 2% and 7% of amnesties adopted between 1990 and 2016 cover economic crimes (Mallinder 2018).

Different categories of actors can benefit from amnesties, including State agents, evaders and deserters, nationals outside borders, foreigners, participants in coup d’état, non-state armed groups, etc. It is interesting to note that State actors are more likely to benefit from unconditional forms of amnesties than other actors (Mallinder 2018).

Amnesty design can also vary greatly in terms of its legal effects, ranging from barring new criminal investigations, stopping ongoing trials and investigations, immunity from administrative sanctions, removing administrative penalties as well as any negative consequences linked to the conviction to imposing alternative sanctions and barring civil liability (Mallinder 2014).

Truth and reconciliation commissions

Amnesties granted through a transparent process involving all parties and guaranteeing some form of accountability are more likely to gain public support and achieve the intended purposes, as shown by the experience of truth and reconciliation commissions.

As of 2008, more than 30 countries had established truth and reconciliation commissions (International Center of Transitional Justice 2008). For example, in addition to the famous South African Truth and Reconciliation Commission, truth and reconciliation commissions have been established in countries as diverse as Ghana (2002), Sierra Leone (2005), Morocco (2004), Timor-Leste (2002), Liberia (2006) and Rwanda (1999). Such mechanisms have mainly been established for serious human rights violations in the context of transitional justice for post-conflict countries, but some of the lessons learnt in the process could be applied to economic crimes. The potential of such an approach to corruption-related crimes was discussed extensively in the above-mentioned answer on the use of amnesties to tackle corruption (U4 2006).

In spite of this potential, most truth commissions have ignored corruption and economic crimes (Carranza 2008). Some authors strongly recommend that transitional justice mechanisms also engage with economic crimes as such an approach would help to reveal the full scale of the damages caused by perpetrators beyond violence (Carranza 2008). In practice, there have been very few examples where such commissions have reported on corruption, such as the Chad commission, which had the mandate to identify the financial operations and bank accounts as well as other assets of former President Hissène Habré or the Truth, Justice and Reconciliation Commission established in Kenya in 2008 (Carranza 2008).

As part of the lessons learnt from these examples, the International Centre of Transitional Justice (2008) emphasises the need to create such commissions only through broad national consultation and to give them a clear and appropriate mandate. Civil society should be involved in the establishment process to ensure legitimacy and diversity of the membership. The success of such approaches also depends on their credibility, transparency and independence from undue governmental or societal pressure. A truth commission must therefore have full autonomy to control its resources, conduct its investigation, build alliances and propose policy changes. The presence of a genuine political will to allow an independent and robust inquiry is an important prerequisite for the success of this process.

Leniency and immunity programmes as an alternative approach

Leniency programmes that mitigate sanctions rather than grant full amnesty for economic crimes have been considered as an alternative approach likely to have a deterrent effect on corrupt deals (Kukutschka and Chêne 2017). Such programmes offer immunity or reduced sentences to offenders if they blow the whistle, self-report criminal activities or cooperate with law enforcement. Research into this area is still at an early stage, and there is little empirical evidence to demonstrate the impact of such approaches on the prevention of corrupt deals (Kukutschka and Chêne 2017). However, such voluntary disclosure programmes were introduced in the US in the context of anti-cartel law enforcement and extended to other forms of collusion in view of their effect on the number of successful prosecutions (Nell 2008).

Such leniency programmes have been extended to the fight against corruption, and countries such as Brazil and Mexico use such approaches to encourage whistleblowing and cooperation with law enforcement in corruption cases (Luz and Spagnolo 2016). In the United States, a pilot leniency programme has been implemented since 2016, granting credits in matters related to the enforcement of the Foreign Corrupt Practices Act (FCPA). Companies and individuals that self-disclose wrong-doing, fully cooperate with law enforcement and implement remediation, including improving the company’s compliance programme, receive credits that can grant a reduction in the applicable sentencing (McFadden et al. 2016).

A few conditions must be met for leniency provisions to act as a deterrent to corrupt deals. For example, mitigating sanctions in exchange for valuable information or self-reporting should not be left to the discretion of the prosecutors but be universally and automatically applied in codified situations to establish a sense of certainty and provide stronger incentives to report (Nell 2008).

Experience of using amnesties for corruption

The use of amnesties for economic crimes has been the exception rather than the rule and is often accompanied by massive public controversy and resistance. However, there have been a number of examples of individuals who have been absolved for corruption (Ardigo 2017). These examples tend to illustrate that granting amnesties or some form of impunity are highly sensitive decisions, especially when they are granted unilaterally by the government for political bargaining.

Romania

In 2017, a decree granted a general amnesty for people sentenced for corruption crimes under a value of €34,000. The leader of the ruling party, the Social Democrats party, who was involved in a corruption case himself that bars him from becoming prime minister, is a fervent supporter of this law (Matomoros 2019). A series of mass demonstrations – the largest demonstrations in the country since 1989 – resulted in the withdrawal of this decree (Ardigo 2017). The debate has been rekindled since the end of 2018 amid the ruling coalition’s plan to draft and pass an emergency ordinance on amnesty and pardoning, which would grant amnesty to individuals with a sentence shorter than five years for certain crimes. Although some politicians justify this measure as a way to relieve overcrowded prisons, this measure is largely perceived as a way for politicians to avoid prison for corruption (BBC 2019). A study revealed that most Romanians (91% of respondents) oppose the pardoning of criminals convicted for corruption, as well as the amnesty for acts of corruption (Marica 2018). In early 2019, the European Commission also warned Romania against passing such a decree (Gotev 2019).

Moldova

In July 2018, a law of “capital amnesty” was adopted in Moldova, allowing the legalisation of money with fraudulent provenance, in spite of a 2017 legal analysis by the National Anticorruption Centre (Ursu 2018). The law allows individuals and businesses to legalise their currently hidden assets – even if the assets were acquired from illicit wealth – by paying a 3% fee from the declared value of the asset. This declaration of assets will come with a guarantee that they will not be prosecuted (Lupusor 2018). The president justified this measure as “economically, socially and politically reasoned” to bolster business and diminish the informal sector and increase tax revenues (Necsutu 2018; Expert Grup 2018). However, the short duration of the capital amnesty - only 2-3 months compared to similar programmes that usually run for at least one year appears suspicious. The concern is that the purpose of this measure is more to legalise illegally obtained and owned assets by a narrow group of people than extend the tax base(Expert Grup 2018). This controversial law triggered protest from civil society and is largely seen in the international community as a law fuelling impunity for the corrupt (Transparency International 2018).

Tunisia

In 2017, a law granting amnesty to Tunisian officials accused of corruption under former President Zine al-Abidine Ben Ali was passed by parliament, triggering angry protests from the opposition and activists who had been resisting the legislation since a version was presented in 2015 (The Guardian 2017). The law gives amnesty as long as civil servants did not personally benefit from embezzling public funds. While promoted as a necessary step to restore trust, the amnesty law is widely perceived as a means for President Essebi’s political party to reward the business leaders that supported its campaign in 2014, undermining the credibility of the anti-corruption campaign launched by the government in August 2016 (Washington Post 2017).

Mongolia

A 2015 law terminated 45 out of the 55 cases that the Independent Agency against Corruption in Mongolia was investigating and granted amnesty to the accused. The alleged crimes involved more than 32 billion Mongolian Togrog (US$16.2 million). Observers also saw this law as an attempt to undermine the Independent Agency against Corruption. The law also clearing criminal records was promoted by the politicians who were under investigation by the Independent Agency against Corruption in order to continue their political careers (Transparency International 2015). This decision was met with controversy in the international NGO community who saw this decision as a “blatant attempt by politicians to grant themselves impunity” (Transparency International 2015). Civil society organisations called the President of Mongolia to veto the law and he finally issued a partial veto specifying that amnesty would not apply to those accused of corruption, abuse of power, illegal enrichment, embezzlement of budget funds, among others (OECD 2015).

Nigeria

In 2013, President Goodluck Jonathan granted a pardon to ex-Bayelsa State governor and former ally, Diepreye Alamieyeseigha who was convicted of corruption during his time in office because he had been “remorseful”. Alamieyeseigha was also declared free to run for elections again (Agbiboa 2013). This decision was perceived by activists as amajor blow to efforts to curb corruption in Nigeria (BBC 2013).

Philippines

In September 2007, ex Philippines leader Estrada was convicted of plunder and theft of public funds and ordered to return US$16.7 million of ill-gotten assets. He first challenged the court’s decision but finally dropped the appeal of his conviction for large-scale corruption at the end of Octber 2007, seeking an unconditional pardon from President Gloria Macapagal Arroyo instead. His pardon was granted in October 2007 and met with controversy, with some suspecting that Arroyo was trying to curry favour with the opposition.

Conditions of success for granting amnesties

To be “smart”, conditional amnesties need to satisfy key accountability requirements while facilitating a peaceful transition and reconciliation process (Kushleyko 2015). A few principles emerge from the literature and experiences of granting amnesties for economic and other crimes (Kushleyko 2015; Mallinder 2008, 2014 and 2018; World Bank 2001).

Last resort strategy

The option of using amnesties should be taken as a last resort after other less extreme alternatives have been considered. The decision to proceed should be definitive and mechanisms to determine exceptions should be trusted. It is also worth keeping in mind that overly broad or repeated amnesties may be counterproductive, erode the goal of deterrence as they reinforce a signal of impunity. Amnesties should be carefully circumscribed by law, in accordance with the intended policy, and only be used once (UNODC 2004).

Transparency and participation

“Transparency in the clemency process can prevent arbitrariness, discrimination, and political favouritism by allowing added public scrutiny” (Nowak 2016). The amnesty granting process should be transparent, open to scrutiny and fully trusted. Decision makers should enjoy the confidence of the public, comply with the highest standards of integrity and be perceived as representative and unbiased. The rationale for using amnesties should be made public for citizens to fully understand the need for and benefits of such a reconciliation mechanism. This could be achieved through strategic communication and awareness raising activities accompanying the granting of full or conditional amnesty.

Scope of amnesties

The scope of the amnesty law should agree with international laws and commitments, and be restricted with regards to eligible persons, crimes or acts. Those having committed especially serious or prejudicial offences (e.g. war crimes, crimes against humanity, etc) should be excluded from the amnesty process. Amnesties should also be conditional on good behaviour, granted only for specific offences, limited in time and not applicable to future acts of malfeasance. They should also be reversible if the beneficiary repeats the type of offences covered by the amnesty.

Victim empowerment and redress

Amnesties granted in exchange for truth telling should involve victims, offenders and communities in the decision and be publicised. Public hearings can be held in accessible public halls and publicised on TV. Victims should be given a chance to challenge an individual claim to amnesty and be provided with some form of reparation. When individuals fail to comply with the requirements of the truth commission, prosecution should be pursued. The granting of amnesties should be accompanied by a reparation programme that takes into account the needs of the victims and society. Ideally, amnesty for economic crimes should be granted in exchange for the redistribution of the proceeds.

Balanced use of transitional justice mechanisms

The need to balance accountability with stability and peaceful political transitions support a balanced approach to transitional justice, with a combination of trials, amnesties and truth commissions. Trials, amnesties or truth and reconciliation commissions cannot alone and by themselves provide a successful pathway to peace, democracy and human rights. Instead of promoting a single mechanism, holistic approaches should combine a balanced use of trials and amnesties, with or without truth commissions (Olsen, Payne and Reiter 2010).