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

TL;DR

I put the chance of YES at 85%. Paz is weak and the September 18 diesel shock can restart nationwide blockades, but the short horizon, fresh congressional backing, and an intact security chain make continuity the base case (AP; Bolivian Ministry of Government; Los Tiempos). The decisive NO path is a transport-led protest wave that unites labor, campesino and Evo-aligned groups, causes casualties, and then pulls Congress or the armed forces away from Paz.

Context

At the September 19 cutoff, Bolivia’s official presidency still identified Rodrigo Paz Pereira as president, with no credible evidence of an earlier interruption. There were 103.97 days left before the resolution deadline. Paz had just announced the immediate removal of the remaining diesel subsidy after Congress approved a three-year, US$1.9 billion IMF financing agreement; IMF Executive Board approval was still required before disbursement (AP).

The government is brittle. Paz’s approval has collapsed, his vice president is openly hostile, and most of the original cabinet has changed. Yet Paz has recently won large congressional votes on emergency powers and the IMF program, while police and military leaders have not broken with him (Ipsos CIESMORI; AP).

Evidence

The historical backbone favors survival. An academic study identifies four Bolivian presidencies shortened by protest after democratic restoration: Hernán Siles Zuazo in 1985, Gonzalo Sánchez de Lozada in 2003, Carlos Mesa in 2005, and Evo Morales in 2019 (Journal of Latin American Studies). Hugo Banzer’s 2001 health-related resignation adds a separate non-political case (BTI). Over the 43.94 years from October 1982 to the cutoff, a constant-hazard calculation gives:

$$p=1-\exp\left[-\frac{k}{T}\times\frac{h}{365.2425}\right]$$

Here, $k$ is the number of interruptions, $T$ is 43.94 president-years, and $h$ is the remaining 103.97 days. Four political interruptions imply a 2.56% baseline risk; including Banzer raises the crude all-cause baseline to 3.18%. Paz’s present risk is several times higher because he faces a live economic shock and organized opponents, but the short horizon remains a strong anchor.

Paz has also passed a severe revealed-behavior test. The May–June confrontation lasted 53 days, involved more than 100 roadblocks, explicit resignation demands and a later reported death toll of 22. The government split the protest coalition, negotiated with part of the labor movement, declared emergency rule and reopened the main roads without a police or military defection (El País).

The full available 2026 monthly ACLED series through August supports the view that the first national wave receded. The table uses ACLED’s monthly public categories; “demonstrations” includes protests and most riots (HDX-HAPI methodology; ACLED Bolivia).

Demonstrations fell sharply after May. That is real evidence against a continuously escalating national campaign. It is not proof of safety: July’s 14 political-violence fatalities show that a lower event count can coexist with dangerous coercion, and the series predates the latest diesel decision.

The institutional picture is stronger than the approval numbers suggest. On September 17, the Assembly extended the state of exception for 90 days by 100 votes to 58 among 158 legislators present, covering most of the remaining forecast window (Bolivian Ministry of Government). Both chambers then approved the IMF agreement by more than two-thirds (La Razón). These votes do not show personal loyalty, but they show that a broad legislative majority still prefers stabilization under the current executive to immediate succession.

Ordinary constitutional removal is hard within 104 days. A recall cannot begin until half of the five-year mandate has elapsed (Bolivian Electoral Authority). A criminal trial requires authorization from at least two-thirds of legislators present, followed by proceedings before the full Supreme Tribunal, while the presidency ends only after a final conviction (Law 044 text; Constitution).

The security forces remain the main stabilizer. On August 7, armed-forces commander Víctor Hugo Balderrama said the military and police would defend the stability of the democratically elected government and that the electorate’s decision was not negotiable (Los Tiempos). The June road-clearance operation also showed operational obedience. There is no verified mutiny, refusal of orders or senior commander calling for succession at the cutoff. Public declarations are not guarantees, but Bolivia’s 2003 and 2019 endgames became decisive only when mass pressure combined with elite or coercive defection.

The fresh downside is large. The diesel decision directly hits trucks, buses and agriculture, and it came only hours before the cutoff. Opposition to the IMF program had already spread among unions, civic groups and some business leaders, while drivers, teachers, campesino organizations and Evo-aligned coca growers had discussed renewed mobilization (Reuters; El País). The September 26 coca-growers’ meeting is the nearest clear escalation test.

Paz’s political position is poor but does not yet point to a replacement coalition. An August four-city survey put his approval at 21% and disapproval at 65%; Vice President Edmand Lara scored worse, at 10% approval and 75% disapproval (Ipsos CIESMORI). AP reported in August that only six of Paz’s original 15 ministers remained (AP). Lara has motive to weaken Paz, but his own ratings and lack of a visible military-congressional coalition make him a weak consensus successor.

My scenario model is:

The weighted NO probability is 15.26%, leaving 84.74% YES. The model gives heavy weight to renewed unrest but requires more than unrest for resolution: resignation, legal succession or a coercive-institution break.

What's non-obvious

The visible indicators are worse than the decisive ones. Approval is awful, cabinet turnover is high, and protest threats are real. But Bolivian presidents usually fall when street pressure joins a congressional or security-force rupture. The May–June campaign failed to produce that rupture, and September’s two major legislative votes moved toward Paz rather than away from him.

Fuel reform is also less uniformly rejected than the headlines imply. In an August online survey of 400 adults in La Paz, El Alto, Cochabamba and Santa Cruz, 63% supported ending fuel subsidies, although support fell to 50% when the question stressed higher living costs; 46% then opposed it (Brújula Digital). Routine foreign travel is not a meaningful NO channel under the question’s rules, and current regulations allow the president to exercise authority remotely while abroad unless that becomes impossible (Bolivian Official Gazette).

Uncertainties

• The diesel decision was only hours old. No reliable post-announcement response from the COB or the national transport federations was available by the cutoff. This is the largest gap because their first coordinated decision could move the forecast by 10 percentage points or more (AP).
• The ACLED series ends in August and cannot capture the new shock. Event counts also measure reported incidents, not crowd size, organization or control of strategic roads (ACLED methodology).
• The main polls are online surveys in four large cities, not national urban-rural samples. They show direction and intensity, but not a precise nationwide level of support (Ipsos CIESMORI).
• Public military support can reverse quickly after a high-casualty confrontation. The missing information is private command cohesion under a second prolonged enforcement campaign, not another ceremonial statement.