Will Bolivian president Rodrigo Paz Pereira remain in power through the end of 2026?

TL;DR

I put the chance of YES at 84%. Paz has survived the strongest ouster campaign of his presidency, nationwide protest activity has fallen sharply, and the military still backs constitutional continuity (ACLED, Los Tiempos). The main danger is a September–October cascade in which fuel protests and the Cerimedo scandal reunite street movements, Congress, and parts of the security apparatus against him (AP).

Context

Paz remains president as of August 22, 2026 and has 131.3 days left to reach the resolution deadline; he took office on November 8, 2025 (Bolivian Presidency). He is much weaker than a normal nine-month incumbent. His governing bloc is divided, most of his original cabinet has gone, Congress forced out his economy minister, and an influential former adviser is in pretrial detention while prosecutors investigate allegations that could reach the presidential circle (AP, El País).

The immediate situation is still short of a removal crisis. The May–June blockade coalition broke apart, Evo Morales’s platform shifted toward organizing for the 2030 election, and a fuel-tanker strike due to begin August 24 was suspended after negotiations on August 22 (Visión 360, Radio Splendid).

Evidence

The historical prior favors survival over a four-month window, though Bolivia is riskier than most presidential systems. The full sequence of presidential spells since elected civilian rule returned is:

Adding Paz’s first 287 days gives about 43.86 president-years of exposure. Four political interruptions imply a mechanical 131-day exit probability of 3.2%; adding Banzer’s health resignation raises it to 4.0%. Among the 11 previous spells that reached Paz’s current tenure, none had a non-routine exit during the exact next 131 days, although Sánchez de Lozada’s second administration fell 150 days after reaching day 287. These are small samples, so 3%–4% is a floor, not the forecast.

The comparative literature explains why the current risk is higher. A study of 116 Latin American presidents found that street protests and legislative support interact: a congressional shield can neutralize protests, while a popular shield can deter legislators (Pérez-Liñán). Bolivia’s successful interruptions in 2003, 2005 and 2019 followed the convergence of mass mobilization with elite, legislative, police or military abandonment—not economic pain or scandal alone (CRS).

Recent protest data are the strongest survival signal. The full twelve-month history from the weekly updated ACLED country-month dataset, retrieved through August 22, shows:

*August is partial through August 22. These are reported demonstration events, not participant counts.

The May–June movement was dangerous. It spread across all nine departments, demanded Paz’s resignation and at one point involved about 150 roadblocks. Sectoral agreements and public exhaustion reduced the count to 12 by June 20, after which the state of exception and joint military-police operations cleared the remaining roads (ACLED, EFE). Paz therefore has revealed resilience against the exact mechanism most likely to remove him.

The coercive and diplomatic environment also favors continuity. On August 7, armed-forces commander Víctor Hugo Balderrama said the military and police would defend the stability of the democratically elected government (Los Tiempos). A July mission involving the OAS and seven member states likewise backed the elected government and constitutional order (OAS). Public pledges can reverse, but there is no current counterpart to the police mutiny and military demand that preceded Morales’s 2019 resignation.

The negative evidence is still serious. The PDC won 65 of the 166 domestic legislative seats—49 deputies and 16 senators—so it began without a majority and has since fractured (OEP). Congress assembled the votes to censure Economy Minister José Gabriel Espinoza, and Paz dismissed him on August 21 after initially resisting (El País). Unidad ended its alliance with Paz on July 27, although its leader explicitly said it would not conspire against presidential stability (ANF).

The Cerimedo affair is the largest new downside. A judge ordered 180 days of pretrial detention after forensic evidence allegedly linked the adviser to an attempted killing, while his former partner accused him and people close to Paz of corruption involving fuel, gold and insurance. The evidence reported so far supports the criminal case against Cerimedo; it does not yet prove criminal conduct by Paz (El País, AP).

Formal removal remains slow. The Constitution says a president ceases through death, resignation, definitive incapacity, a final criminal conviction or recall; a presidential criminal trial requires a prosecutor-general request and authorization from two-thirds of legislators present (Bolivian Constitution, Articles 169–170 and 184). Recall cannot start until half of Paz’s five-year term has passed, placing it outside 2026 (OEP). The practical path to NO is therefore resignation under pressure, not a routine congressional no-confidence vote.

Economic policy supplies several possible triggers. On August 20, the central bank projected a 3.6% real GDP contraction and 9.22% year-end inflation for calendar 2026 (BCB report coverage). July CPI nevertheless fell 2.79% from June, leaving twelve-month inflation through July at 4.93%, largely because food supply normalized after the blockades (INE, published August 4). The July 29 staff-level agreement for a 36-month, US$1.9 billion IMF facility offers relief, but remained subject to prior actions and IMF Executive Board approval at the cutoff (IMF).

My synthesis raises the historical exit prior to about 16%. Roughly ten percentage points come from a renewed protest-and-elite resignation pathway, three from a scandal-driven collapse without a full national uprising, two from a security rupture or coup, and one from legal, medical and other tails. This leaves a final forecast of 84% YES.

What's non-obvious

The missing ingredient is alignment. Paz already has scandal, recession and a hostile legislature. What he does not currently have is a simultaneous national street campaign and security-force abandonment. Coverage that lists each crisis separately can make removal look imminent, but past Bolivian exits occurred when these pressures synchronized. Current ACLED data and the August 7 military statement show that this synchronization has not happened (ACLED, Los Tiempos).

Vice President Edmand Lara is a threat, but succession also divides Paz’s opponents. Paz’s departure would place Lara in the presidency under Article 169, rather than hand power to Libre, Unidad or Morales’s movement (Bolivian Constitution). Lara said during the May crisis that Paz’s resignation was not the solution, and the latest limited urban online poll gave Lara only 12% approval against 75% disapproval (Unitel, Ipsos CIESMORI coverage). I read this as a brake on a unified elite removal coalition, not as protection Paz can rely on indefinitely.

Uncertainties

• The Cerimedo investigation is only days old. Authenticated financial records or testimony directly implicating Paz or his family would sharply reduce this forecast; the present public record contains allegations, not proof against the president (AP).
• The state of exception runs for 90 days from June 20, concentrating risk around its September expiry and the COB agreement’s similar implementation deadline (UN treaty notification). A new state of exception within a year requires prior legislative authorization, reducing Paz’s unilateral room to respond (Constitutional rule). Internal COB assessments of whether the June agreement has been fulfilled would close much of this gap.
• Public approval is poorly measured. An online four-city survey of 400 respondents found 32% approval and 55% disapproval, while a regional vendor reported 50.5% positive and 44.9% negative days later (Ipsos CIESMORI, CB Global Data coverage). A nationally representative rural-and-urban probability survey would be more useful.
• The military’s public loyalty is confirmed; its private cohesion is not. Evidence of police refusal, command resignations or reluctance to clear renewed blockades would be the strongest signal toward NO. Continued unity through October would move the forecast materially toward YES.