Parallel tracks: multilateral development bank sanctions proceedings and the criminal law of corruption in Latin America

Rhys Davies
Temple Garden Chambers, London
rhysdavies@tgchambers.com

Cristian González Ruiz
International Human Rights Advisors, The Hague
 

The Inter-American Development Bank’s (IDB) sanctions system is invisible to most criminal practitioners in Latin America. It should not be. Over the past decade, it has become a parallel track of corruption enforcement, reaching every IDB-financed contract in the region, debarring firms and individuals across the multilateral development bank system and increasingly feeding material to national prosecutors.

Five practices trigger the system: corrupt, fraudulent, coercive, collusive and obstructive practices, each defined in Section 2.2 of the IDB’s Procedures and mirrored in the IDB’s Invest Integrity Framework. Sanctions range from a letter of reprimand to permanent debarment, with restitution available where appropriate. The limitation period is ten years. Every sanction is published, by name, in the IDB’s public register.

In practice, investigation files, debarment findings and supporting evidence are shared more freely among national prosecutors, multilateral development bank (MDB) integrity offices and procurement agencies than at any point in the region’s history. The IDB Group is one of the most prolific sources of that material. Its sanctions regime has become, in practice, a parallel enforcement system: useful in the regional fight against corruption and yet operating according to standards that no criminal court would accept.

An interesting feature of the system is the fact that debarment does not stop at the IDB. Under the 2010 Agreement for Mutual Enforcement of Debarment Decisions, the World Bank, the African Development Bank (AfDB), the Asian Development Bank (ADB) and the European Bank for Reconstruction and Development (EBRD) automatically enforce any IDB debarment of more than one year. By 2017, over 500 entities had been cross-debarred under the regime and the figure has continued to climb. Unlike a criminal conviction, no court tests the underlying finding before the sanction takes effect. From the day the IDB publishes a debarment, it becomes enforceable across all of the signatory institutions. That speed and reach are what make the system work. It is also what makes it dangerous.

The regional anti-corruption fight needs the IDB system more than most criminal lawyers realise. Domestic prosecutors across the Americas don’t always have the procurement expertise or the cross-border reach to track bribery and collusion in regard to the kind of infrastructure deals the MDBs finance. The Office of Institutional Integrity (OII) has both. Its findings are public, reasoned and enforceable across the MDB system, making them an obvious starting point for any national authority. The Procedures expressly contemplate referrals in which the conduct is also potentially criminal, and the route is well used: the World Bank’s Integrity Vice Presidency, whose model the IDB largely tracks, made 26 such referrals in fiscal year 2025 alone.

The published Sanctions Committee decisions show what this looks like in practice. In decisions SNC63 and SNC64, Ecuadorian firms and their chief officer were debarred for 132 months, with conditional release and automatic cross-debarment, after being found on a preponderance of the evidence to have laundered and repatriated bribes paid to a public official in connection with an IDB-financed project. In decision SNC71, an individual was debarred for knowingly misrepresenting an existing World Bank debarment, his company’s bankruptcy and a contract transfer to win IDB contracts in Haiti’s transport sector. Each decision is a self-contained dossier that any competent district attorney (fiscalía) or Attorney-General’s office (ministério público) could fold into a domestic investigation.

Despite its effectiveness, the system also has serious problems that warrant reflection. The first is how the IDB defines a corrupt practice. In regard to the definition set out in the Procedures, it is the offering, giving, receiving or soliciting, directly or indirectly, of anything of value to improperly influence another party. No public official is required. No de minimis threshold applies. No specific quid pro quo need be shown. ‘Another party’ plainly captures private-to-private bribery, alongside classical public sector corruption. Compare that with what the region’s criminal codes actually criminalise. Colombia’s Criminal Code sets out the principal bribery offences in Articles 405 to 407 (cohecho propio, cohecho impropio and cohecho por dar u ofrecer): each requires a public official, an undue or unjustified utility and conduct tied to that official’s functions. Soborno transnacional, the foreign bribery offence under Article 433, applies only to foreign public officials and only in connection with international commercial transactions. The same picture, with local variation, recurs across the region. A respondent can, therefore, be debarred for a decade for conduct that no Colombian prosecutor would have raised charges against.

The second is what counts as proof. Section 3.3 of the Procedures defines preponderance of the evidence as ‘more likely than not’. This is the civil burden. It sits well below the beyond-a-reasonable-doubt standard required in every Latin American criminal jurisdiction and well below the tighter conviction beyond-any-reasonable-doubt formulations used across the region’s accusatorial codes. The Sanctions Committee is not bound by formal rules of evidence. It has wide discretion over the weight it gives to material. There is no live cross-examination as of right, no judicial review beyond the Committee itself and no confidentiality regime comparable to the protections that are attached to a national criminal investigation.

The consequences, however, are criminal in everything but name. Debarments routinely run for three to 13 years. They are public, listed by name, disclosable in tender documents and replicated across all of the other MDBs through cross-debarment. They generate referrals to national authorities, regulatory disclosures and disqualification under domestic procurement statutes. And they very often precede, rather than follow, a national criminal investigation.

The IDB system exists to protect bank funds and it has done more for the regional anti-corruption fight than most criminal practitioners realise. In practice, it has become a parallel, quasi-criminal enforcement regime for firms and professionals across the region. Mapping the overlap and gaps between the MDB definition of corrupt practice and each jurisdiction’s penal code is doctrinal work that has barely begun, and there is a real opportunity here for the criminal and international Bar. That is not work for compliance lawyers alone.