Criminal negotiation in Brazil is no longer an exception. Since Law 9.099/1995 opened room for civil composition, penal transaction, and suspended prosecution, the topic has matured and gained new scale with the Non-Prosecution Agreement (ANPP), introduced nationwide by the CNMP (Resolução 181/2017) and later codified in practice. Unlike the small-claims arrangements, ANPP requires a formal, detailed confession; collaboration agreements are complex procedural deals whose benefits depend on results. In both, the premise is an admission of responsibility — which is why technical calculation is mandatory before sitting at the table.
A solid starting point comes from Getting to Yes, by Roger Fisher and William Ury, a classic associated with the Program on Negotiation at Harvard. The book popularized two simple, powerful yardsticks for evaluating proposals: BATNA and WATNA. BATNA is the best alternative to a negotiated agreement; WATNA is the worst alternative to a negotiated agreement. With them, you compare the offer with what would likely happen without a deal.
Criminal negotiation: when does it make sense?
The practical question is straightforward: does it make sense to negotiate now? The criterion is to compare the offer with your no-deal scenario. If the package beats your BATNA and reduces your WATNA in measurable terms, negotiating tends to be rational. If it does not, the urge to “settle quickly” usually proves costly.
How to compute the best and worst no-deal alternatives (defense and prosecution)
For the defense, the best alternative without a deal may be a likely acquittal, a charge downgrade, a more lenient regime, sentence substitution, limitation period, a meaningful procedural nullity, or even a conviction with manageable collateral effects; the worst alternative is a harsher conviction (penalty, regime, loss of office), with spillovers on compliance, reputation, and business. For the prosecution, the best alternative is a robust case with a high likelihood of conviction; the worst is weak evidence, risk of acquittal, and nullities.
Bargaining power flows from this symmetry: the stronger the prosecutor’s evidence, the more concessions the defense must present; the more vulnerable the proof, the greater the room to reduce demands or simply decline a deal.
ANPP and collaboration agreements: differences that matter at the table
In ANPP, the starting point is the formal and detailed confession of the facts. That choice has legal and practical effects beyond the criminal sphere, because there is no general rule shielding the confessed content from other jurisdictions (administrative, civil, regulatory). Hence, the scope, form, and custody of the confession belong in the deal’s architecture.
In collaboration agreements, the deal does not operate by itself: benefits are applied at sentencing (or related decisions) if the information proves useful and corroborated, with clear limits on the judge’s role during negotiations. It is a guilty-defendant agreement with performance requirements — potentially generous, but not a safe-conduct.
Criminal negotiation: practical steps to decide
1. Map evidence and risks. What is already in the file? What can emerge? Which nullities are plausible?
2. Quantify the no-deal scenario. List your BATNA and WATNA — including financial and regulatory effects (compliance, contracts, market).
3. Infer the prosecution’s scenario. What are the prosecutor’s BATNA and WATNA? Where is the evidence strong? Where is it vulnerable?
4. Design verifiable terms. In ANPP: proportional conditions, timeline, form/scope of the confession, access and custody of sensitive documents. In collaboration: subject-matter, consideration, usefulness metrics, verification and review mechanisms.
5. Compare coldly. Sign only if the package beats your BATNA and reduces your WATNA in measurable terms.
6. Keep coherence across fronts. Strategy must align with crisis management, communications, and evidentiary defense — a topic we cover in “Summons to testify: what to do and how to defend yourself”.
A good deal is a compared deal
Investigation pressure invites shortcuts. But a good deal is a compared deal — grounded in evidence, risks, and real alternatives. When the proposal improves your no-deal scenario and reduces concrete damage, negotiating makes sense. When it doesn’t, the best strategic decision may be to litigate.
