“Corruption-proofing” is a method for assessing whether a draft law or regulation poses a risk of corruption. A independent expert analyzes whether the way a proposed legal norm is drafted or to be implemented is likely to pose a risk of corruption and if so, how it can be amended to minimize or eliminate the risk. First used in the early 2000s by EU Eastern Partnership Countries, it has since spread to other states in Eastern Europe and Central Asia and to South Korea.  

The technique’s immediate value is that it gives lawmakers a chance to revise their drafts to address the corruption risks they might create. Of even greater import, when citizens or civil society have an opportunity to weigh in on corruption risks, it opens the door to public discussion and participation in what is government’s most critical task: the making of legal norms binding on all.

My reading of the experience with corruption-proofing suggests others would benefit from adopting a similar procedure. What I learned about that experience is summarized below. Comments welcome and information on other studies or countries where it has been tried most welcome.

Two studies have evaluated the experience with corruption-proofing, Tilman Hoppe’s 2014 report for the Regional Anticorruption Initiative for Southeastern Europe and Quentin Reed’s 2017 assessment for the Council of Europe and European Union’s Partnership for Good Governance Project. Both reviewed the issues different countries examined, the agency responsible, and the treatment of the assessment; both also offered observations on what has been effective and suggested good practices. Cristina Cojocaru published a method for corruption-proofing Albanian laws in 2010, and several reports on Moldova’s experience with corruption-proofing have appeared (here, here and here).

While the techniques vary from country to country, all include a linguistic analysis of the norm’s text. Are there ambiguities that make it open to different interpretations and so to the risk bribery will determine the interpretation chosen? Besides a textual analysis, in some states the drafting process is assayed to determine whether it was open and transparent. In others corruption-proofing extends to a substantive analysis of the norm, to whether it favors a special interest at the expense of the public interest. Some states require that regional and local governments’ draft norms be corruption-proofed, and in some current laws and regulations are also assessed.

Norms drafted by an executive branch agency are in most cases corruption-proofed by the anticorruption agency or ministry of justice. Parliaments sometimes have their own rules for corruption-proofing draft legislation.

The corruption-proofers’ recommendations are most often advisory. Several states do require the executive agency responsible for a draft norm to respond to a corruption-proofing analysis, explaining and in some cases justifying a decision to reject a recommendation. Where parliament declines to accept a recommendation, some countries require it to explain why as well.

The 2014 Hoppe Report

Hoppe identified 13 countries where corruption-proofing of some or all legal norms is required by law, and a handful more where either legislative drafting guides recommend it or where civil society groups do it on an ad hoc basis. The 13 with a corruption-proofing law are listed in table 1. The countries range from Latvia, Lithuania, and South Korea — nations with a strong commitment to fighting corruption and high scores on cross-national ratings of corruption — to Azerbaijan, Tajikistan, and Uzbekistan, countries where the fight against corruption is not a priority, which score poorly on cross-national measures of corruption, and where leaders reportedly have little if any interest in containing corruption, or at least in corruption that inures to their benefit. 

Table 1. Countries with Corruption Proofing Law
Albania	Kyrgyzstan	Moldova	Tajikistan
Armenia	Latvia	Russia	Ukraine
Azerbaijan	Lithuania	South Korea	Uzbekistan
Kazakhstan

Corruption-proofing laws in the former category are tightly drawn and widely publicized. Agencies must respond to recommendations within a set time, in Latvia three months. In Lithuania the assessment is done by the anticorruption agency and must be posted both on the agency and the parliament’s websites; on the parliament’s site there must be a link to the draft bill.

In the countries where combatting corruption is not a priority, the corruption-proofing statutes contain provisions that reduce their effectiveness. In Kazakhstan, draft laws and decrees authored by the president are exempted from the corruption-proofing requirement. In Armenia, the agency that authored the draft is not required to respond to the assessment. Russia allows certified “corruption experts” to analyze a draft law for corruption risks, but how one becomes a certified expert remains murky.

Hoppe’s report discussed parliament’s role in corruption-proofing in two countries: Kyrgyzstan and Moldova. In Kyrgyzstan all bills the executive submits to parliament must include an assessment of their impact on corruption. Each executive drafted bill is referred first to the Expert Department of the Parliament’s Secretariat which conducts its own assessment. It is then sent to a committee which evaluates it. If the committee amends the bill, the Expert Department conducts a new assessment. Parliamentary rules provide that independent experts and civil society organizations may submit corruption-proofing analysis of the draft too. The responsible committee must hold a hearing on the corruption assessments and explain why it accepted or rejected any of the recommendations in the assessments.

Moldovan law requires its anticorruption agency to corruption-proof all draft legislation. Legislators initially resisted submitting bills they had drafted, arguing that, as a separate branch of government, bills originating in parliament were not subject to the agency’s jurisdiction. In response, parliament amended its rules to require all bills be corruption-proofed by the anticorruption agency no matter where they originated. If an emergency requires immediate action on a bill, the anticorruption agency must provide an assessment within three days.

Hoppe offers his own proposal for corruption-proofing a legal norm in an annex. He divides the analysis into four areas:

1. Ambiguity. Hoppe explains that a text can be ambiguous due to language: the choice of words, the construction of sentences, or because it is not legally coherent. By the latter he means conflicts among provisions within the draft or conflicts between the draft and existing norms. Ambiguity can also arise if the references to other texts are unclear, if terms are used in different ways within the text, and if the draft is not uniformly structured. It is also caused by what he calls “regulatory gaps,” the failure to specify how conduct governed by the draft will be regulated.
2. Prevention gaps (public laws). For norms enforced by a public authority, he recommends a seven step analysis which he divides into several subparts. “Competencies,” the power to decide, is the first and calls for examining who enforces the norm and how, whether the enforcement power is divided among different entities or there are overlaps among them, and whether there is a conflict of interest. The latter appears to refer to situations where either an official responsible for enforcing the law would have a personal conflict or where there is an organizational conflict — as when an association of estate agents is assigned responsibility for enforcing antimoney laundering laws against members.

Whether the enforcement authority’s discretion in taking decisions is “excessive” is a second area for review. A third is an evaluation of the enforcement procedures. What are the steps required to comply with the law? How many? Are there fees involved? Repetitive inspections? The fourth is oversight: judicial review, transparency, civil society monitoring, and sanctions for violations. Finally, corruption-proofers should consider whether the norm affects a sector where specific safeguards are already in force.

• Prevention gaps (private law). When the norm is one governing relations between individuals, Hoppe recommends analysts assess the strength of rules for accounting and documenting transactions, those on exchanges or interests in property, and judicial procedures that can be invoked to resolve disputes.
•  Corrupted legislation. Hoppe notes throughout his report that if the process of drafting legal norms is corrupted, the norm itself will almost surely further corrupt ends. Corruption-proofers should thus include an examination of the drafting process itself. Were rules on the lobbying of drafters and the making political contributions to elected officials involved in drafting observed? Was the drafting process transparent? Free from ethical violations? Included here is an examination of the draft’s substance. Does it favor certain interests or disadvantage others?

    Hoppe identified several good practices. The key ones:

• Ensure broad participation in the exercise by civil society and others with an interest or expertise,
• Give the corruption-proofing body authority to review any law and regulation, whether in draft or already enacted, but give it discretion to select those to examine,
• Require legislature to consider corruption-proofer’s recommendations and explain why it if declines to follow them,

At several points he emphasizes that corruption-proofing will be successful only if the rules surrounding the drafting of legal norms are sound. They must be drafted transparently and with public participation, and provisions on lobbying, political finance, and ethics are necessary to ensure the drafting process itself is not corrupted.

The 2017 Reed Assessment

Reid draws on several sources for his analysis: Hoppe’s paper; a questionnaire on corruption-proofing completed by Armenia, Azerbaijan, Belarus, Georgia, Moldova, and Ukraine; regional workshops with these countries and Lithuania, the 2010 Albanian study, and an in-depth analysis of Moldova’s experience. Like Hoppe, he stresses that corruption-proofing should be part of a larger effort to ensure legal norms are thoughtfully and carefully drafted. Reid provides three examples where corruption-proofing identified corruption risks in proposed laws –

Ukraine: A draft law on ports created an agency to regulate seaports and the rules for its operation. It included a provision granting an entity founded by a private firm the exclusive right to develop and implement the agency’s software and fix subscriber payments for its use. The opportunity to corruptly abuse this power was highlighted in the corruption-proofing analysis.

Moldova: A corruption-proofing report on a draft law on capital liberalization and fiscal stimulus found it would i) allow the registration of property and other assets in the name of another person and ii) cancel fines due for unpaid taxes, social, and health insurance under very generous conditions.

Lithuania: Analysis of a draft law authorizing a public-private partnership to build the Vilnius metro found that it permitted certain companies to partner to avoid competition for the work. Parliament adopted the legislation, but the President vetoed it.

Reid offers his own corruption-proofing methodology. It proceeds in two steps, the first assessing “risk factors” and the second the “institutional setting.” Like Hoppe’s it includes a textual analysis, a review of the openness and ethics of the drafting process, and an assessment of whether special interests are favored.   

Risk factors. The first and most important risk factor Reid identifies is “excessive discretion.” This review covers the same issues as Hoppe includes under “ambiguity”: instances where the text is poorly worded, contains conflicting legal provisions, unclear references, vague or undefined criteria for enforcing the norm, or where the authorities are given “unjustifiably broad criteria for enforcing the norm.  He emphasizes that “excessive discretion” more than any other factor is responsible for corruption risks. At the same time, Reid acknowledges “some discretion is necessary in almost all decision-making processes.” He offers no clear test for when discretion is excessive, however, saying it needs to be judged on a case-by-case basis.

The other risk factors he cites are —

Poor accountability	Insufficient transparency
Providing unjustifiable benefits to a person or group	Violations of procedural rules governing approval of the norm
	Others

“Poor accountability” means those with the power to enforce the norm need not justify or explain their actions. Corruption-proofers should determine if enforcement agents are either responsible to a more senior official for their decisions, the norm provides an appeal and redress mechanism or complaints procedure, or the agents can be sanctioned for failing to follow established rules.

Reid notes that identifying who stands to benefit “unjustifiably” from a norm may require conjecture and even speculation and thus can be politically risky.  Hence, he cautions that it should be done only where “the institution conducting proofing is very well-established and sufficiently respected not to be vulnerable to accusations of bias.” Examples of when a norm “unjustifiably” benefits a particular interest include:

“market/pricing rules that benefit certain interests with no justification, criteria to qualify for financial benefits (e.g. tenders, subsidies) that are irrelevant to the matter and/or distorted to benefit particular interests without justification, unjustified exemptions from obligations, legalization of a dominant or monopoly position without justification.”

He notes that a useful test is whether a convincing explanation for why the benefit is being provided is contained in an accompanying report. Does the report provide a credible, realistic discussion of the reasons why the benefit is justified?

On process, whereas Hoppe says corruption-proofers should examine how lobbying and campaign finance laws might have affected the norm’s provisions, Reid recommends corruption-proofers examine whether procedures required by law or parliamentary rules for approving the norm were followed. Were consultation provisions ignored? Did the appropriate executive-branch agency or agencies review and okay the draft?

Institutional setting. Reid here discusses organizational and process issues, most often offering guidance on good practice rather than hard-and-fast recommendations. Good practice is determined by a country’s laws and institutional capacity. They will dictate how different nations should address questions such as who conducts a corruption-proofing analysis, when it should be done, and how it should be coordinated with other processes. The discussion is drawn from his analysis of the procedures Armenia, Azerbaijan, Belarus, Moldova, and Ukraine follow. For each, he summarized the following criteria:

The law or decree requiring corruption-proofing	What norms are examined
The agency or agencies conducting corruption proofing	What response if any is required to an assessment
Whether corruption-proofing is a stand-alone procedure or part of a more comprehensive analysis or procedure	Whether the assessment is made public

Corruption-proofing is one part of assessing the effects of a proposed norm and might, as in Armenia, feed into a broader assessment of its impact. Depending on the capacity and scope of authority of the entity responsible for corruption-proofing, it might also constitute part of a broader assessment of corruption risks in a sector. Laws and regulations in force might also be corruption-proofed, but this depends upon a nation’s capacity.

Reid recommends those responsible for turning a policy proposal into a draft law or regulations be cognizant of corruption risks when preparing the draft. He cites the example of Albania where the official law drafting manual lists risk factors to avoid and stresses drafting principles that reduce or eliminate risks.

Among the nations studied, Reid found some centralized corruption-proofing responsibility and others diffused it across different agencies. In Moldova and Belarus, it is centralized respectively in the national anticorruption agency and a unit within the prosecutor’s office. In Azerbaijan whatever ministry drafting the bill is responsible; in Ukraine a variety of agencies share responsibility.  While every drafting body should follow corruption-proofing guidelines, he concludes that experience shows that a separate, independent body with sufficient expertise and resources should ultimately be responsible for corruption-proofing.

Like Hoppe, Reid recommends that the corruption-proofing process be transparent and that drafters and legislators be required to respond to the analysis and justify any decision to ignore its recommendations.

Finally, Reid urges that corruption-proofing analyses use a standard template that group risks into pre-determined categories. He cites Moldova where this allows for reports showing those areas where initial drafts create the most risks. For example, one report showed “excessive, improper duties or duties contrary to status of private person/entity” was found in 22% of the norms assessed, conflicting provisions in 15%, and ambiguous wording in 14%.

Moldova

Moldova began corruption-proofing norms and proposed norms in 2006, part of assistance provided by the Joint Project of the Council of Europe and the European Commission against Corruption, Money Laundering and Terrorist Financing in the Republic of Moldova (MOLICO).  The 2006 law called for the identification of “corruptibility factors”, that is, whether provisions of a law or regulation proposed by an executive branch agency would facilitate or further act of corruption, and where the analysis showed one or more might, recommend ways to eliminate them or at least reduce their impact. The analysis is done on all draft legislative and regulatory acts by a staff of 12 within the anticorruption agency.

The Centre for the Analysis and Prevention of Corruption (CAPS) and other NGOs corruption-proofs draft bills once they are submitted to parliament. These reviews are usually done prior to the draft’s second reading in Parliament. This arrangement became possible after the Moldavian Parliament yielded to mounting pressure from various civil society organizations demanding placement of all draft laws on the web site.

Moldova’s anticorruption agency reports regularly on its corruption-proofing activity, its 2014-2015 annual report stating it had assessed 1,329 draft norms during the period. While the number of norms examined is one measure of effectiveness, a recent report by Moldova’s Independent Anticorruption Advisory Committee suggests the agency has overlooked a second, the degree to which the corruption-proofing analysis has prevented enactment of flawed legislation.  It suggests that while the exercise has helped catch bills that might have furthered small-time or petty corruption, provisions in bills that fostered large scale corruption by powerful interests slipped through the process unscathed. One example it cited was a case where the head of the anticorruption agency and the chair of a parliamentary committee had facilitated passage of a law allowing for the suspension of FIU decisions.

The Independent Committee concluded that the agency invests “too much effort and resources” in corruption-proofing and that its resources would be better devoted to investigating allegations of corruption. It recommended corruption-proofing be lodged in the Ministry of Justice. As one of the report’s authors explained after the report was published, while corruption-proofing enthusiasts may have overstated its impact, it has nonetheless played an important role in curbing corruption in Moldova and should be continued.

Moldova has distilled its corruption-proofing methodology into the 37 questions listed below.

Moldova corruption proofing risks factors
Language
1.	Introduction of new terms that do not have a definition in the legislation or in the draft act
2.	Uneven use of terms
3.	Ambiguous wording that allows abusive interpretations
Legislative Coherence
4.	Defective reference rules
5.	Defective blanket rules
6.	Rules of law competition
7.	Gap in the law
Transparency/RTI
8.	Insufficient access to information about the act subordinate to the law
9.	Lack/insufficient transparency of public entities
10.	Lack/insufficient access to public interest information
Citizen rights/obligations
11.	Excessive costs in relation to public benefit
12.	Promoting interests contrary to the public interest
13.	Harm to the interests contrary to the public interest
14.	Excessive requirements for the exercise of excessive rights/obligations
15.	Unfounded derogations from the exercise of rights/ obligations
16.	Unjustified limitation of human rights
17.	Discriminatory provisions
18.	Excessive, improper duties or contrary to the status of the private entity/person
19.	Stimulating unfair competition
20.	Unrealizable norms

Public entities’ duties
21.	Extensive regulatory duties
22.	Excessive, improper duties or contrary to the status of private entity
23.	Parallel duties
24.	Non-determination of the responsible public entity/ subject to which the provision relates
25.	Duties allowing derogations and abusive interpretations
26.	Establishing a public entity’s right instead of an obligation
27.	Defective cumulation of competences to be exercised separately
28.	Non-exhaustive/ambiguous/subjective grounds for a public entity’s refusal or inaction
29.	Lack/ambiguity of administrative procedures
30.	Lack of concrete deadlines/unjustified deadlines/ unjustified extension of deadlines
Control mechanisms
31.	Lack/insufficiency of control and surveillance mechanisms (hierarchical, internal, public)
32.	Lack of/insufficient challenge mechanisms
Liability and sanctions
33.	Confusion/duplication of types of legal liability for the same infringement
34.	Non-exhaustive grounds for liability
35.	Lack of clear responsibility for violations
36.	Lack of clear sanctions
37.	Imbalance between violation and sanction