vii. Enhanced recording/data collection 106. A record of the terms of the negotiation leading to a trial waiver (where such negotiations occur), or the details of the resulting benefit conferred on the defendant, is not always produced. In Hong Kong, for example, a record of the terms of the negotiation is only produced when the defendant is unrepresented. Similarly, as earlier noted in relation to Kenya, guidelines suggest that trial waiver negotiations involving defendants without a lawyer should be electronically recorded. In systems featuring only sentence incentives with statutory control over benefits conferred, the production of a record may not be as vital as in jurisdictions featuring charge or fact incentives, or in cooperation agreements, where the benefits offered and concessions made by the parties are more variable. Without such a record, appeals become practically difficult, as does tracking the effect of prosecutorial guidelines and other instruments. 107. Furthermore, opacity of negotiations (where they take place as part of the trial waiver system) contributes to public distrust in trial waivers. There is particular concern about the lack of openness of terms of agreements when they involve the payment of financial penalties as a condition of sentence reduction (as is common in Georgia and Russia, for example, and has also been raised as a feature of the trial waiver system (delação premiada) used in Brazil). Where it is perceived that highly resourced defendants are able to evade justice through the payment of fines, the purpose of using trial waivers to tackle public corruption is seriously undermined. 108. Where trial waiver systems are newly adopted or have recently undergone reforms, authorities are in the unique position of being able to track more easily the effects of the change in procedure and to adjust policy accordingly. Bosnia and Herzegovina and Georgia are both examples of countries where the implementation of the trial waiver system was heavily monitored, and reforms were made in response to findings. However, much of this monitoring was conducted by civil society organisations, rather than government actors, potentially diminishing the scope and sustainability of the data collection.107 109. There are a few examples of countries that employ enhanced recording to ensure the integrity of trial waiver systems. In Finland, for example, trial waivers take place in open court, and the subsequent judgment will state what the sentence would have been if the conviction had occurred following a full trial rather than as a result of a trial waiver. In England and Wales, the sentence discount granted following a trial waiver (known as a “guilty plea”) in line with the sentencing guidelines is calculated and applied separately, after adjustments have been made to reflect other mitigating factors, and so should be clear on the record. In New Zealand, negotiations between the defence and prosecution are memorialised in a case management memorandum. viii. Limitations on benefits 110. It may seem counter-intuitive to frame limitations on benefits provided in exchange for trial waivers as a safeguard for defendants, since such limitations restrict what they could gain from a trial waiver. However, conversations with US experts consistently highlighted the coercive effect of the large sentencing differential in the US between the potential sentences faced after conviction at trial as opposed to those available through plea deals. The coercive effect of this trial penalty has been well-documented.108 111.  The wide discretion US prosecutors have, through plea deals, to offer sentences drastically shorter than those possible after conviction at trial relies on several features that are distinct to US law and practice, principally the unusually long sentences and heavy use of mandatory minimums, coupled with a criminal legal culture characterised by powerful and largely unaccountable prosecutors. Many other jurisdictions, even those with substantial procedural similarities to the US criminal justice system, control the extent of sentencing discounts either by statute or in practice. The Federal Constitutional Court of Germany has recognised that a significant sentence differential between trial waiver and trial can act as an “illegal influence” on the defendant’s free will.109 107. See, e.g., “Plea Bargaining in Georgia: Negotiated Justice”, Transparency International: Georgia (Dec. 15, 2010). Available at: http://www.transparency.ge/en/ node/1061. See also “Plea Agreements in Bosnia and Herzegovina: Practices before the Courts and their Compliance with International Human Rights Standards”, Organization for Security and Co-operation in Europe (OSCE): Mission to Bosnia and Herzegovina (2006). 108. See, e.g., Fellner, “An Offer You Can’t Refuse”, supra at n. 34. 109. ‘Plea Bargaining and its History’, Albert Alschuler, Columbia Law Review, Vol 79, No. 1, January 1979. See also Altenhain, K., “Absprachen in German Criminal Trials”, from World Plea Bargaining: Consensual Procedures and the Avoidance of the Full Criminal Trial, Thaman, S. C., ed., Carolina Academic Press, Durham, NC. (2010), pp. 169–170. 56.