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

TL;DR

Rodrigo Paz Pereira is more likely than not to remain president through 31 December 2026; my estimate is 80% YES. The June crisis has started to move in his favor because the COB signed a truce, the Túpac Katari and Bartolina Sisa organizations paused La Paz mobilizations, the Assembly ratified the emergency decree, and the road-blockade count fell to 17 active points on 20 June (AP, 21 Jun 2026, Correo del Sur, 20 Jun 2026). The main NO path is a lethal confrontation in Chapare or a renewed economic blockade wave that causes the military, Vice President Edmand Lara, or Congress to abandon him.

Context

Paz is the incumbent president: Bolivia’s official presidency page says he assumed office on 8 November 2025, after the TSE proclaimed the Paz–Edmand Lara ticket winner of the 2025 runoff with 3,519,534 votes, or 54.96%, against Jorge Quiroga’s 45.04% (Presidency of Bolivia, EFE, 27 Oct 2025). The forecast window now has a little over six months left, and the question resolves NO if he loses the presidency even temporarily before 23:59 Bolivia time on 31 December 2026.

The immediate crisis is severe. Protests and blockades that began in early May demanded his resignation, isolated La Paz and El Alto, produced food, fuel, medicine, and oxygen shortages, and led Paz to declare a nationwide 90-day state of exception on 20 June 2026 (AP, 20 Jun 2026, El País, 20 Jun 2026). But the latest facts at forecast time are not a straight line toward collapse: AP reported on 21 June that roads were reopening, one major rural union had ordered a pause, the Assembly had ratified the decree by majority, and only the Evo Morales-aligned cocalero sector remained in full protest posture in Chapare (AP, 21 Jun 2026).

Evidence

The historical base rate is dangerous but not decisive. Since democratic restoration in October 1982, Bolivia has had several early or abnormal exits, but also long stretches where presidents survived serious unrest. Counting hard early exits by death/illness, resignation under protest pressure, or security-force/constitutional rupture gives four cases before Paz: Hugo Banzer in 2001, Gonzalo Sánchez de Lozada in 2003, Carlos Mesa in 2005, and Evo Morales in 2019. Counting Hernán Siles Zuazo’s shortened 1985 handover as a softer interruption gives five cases over roughly 43.7 country-years, an unconditional hazard around 11% per year and about 6% for the remaining half-year. Paz is not in an unconditional half-year, so this is a floor, not the forecast.

The broader South American reference class says the risk jumps once a president faces a serious removal challenge. Hochstetler and Edwards found that 17 of 44 elected South American presidents in their sample faced serious challenges, and 11 of those 17 failed to finish their terms; they also found that successful removals were almost always accompanied by large mobilizations against the president (Hochstetler and Edwards, 2009). Paz is clearly in the challenged class. The adjustment back down is that the data cover full presidential terms, while this question has only about six months left, and the latest coalition signals are less favorable to removal than they were in early June.

The strongest current evidence for a YES is coalition fragmentation. On 19–20 June, the COB signed a truce with the government and explicitly split from campesino and cocalero groups still demanding resignation (AP, 20 Jun 2026). On 20 June, the La Paz Túpac Katari and Bartolina Sisa organizations declared a cuarto intermedio in their departmental mobilizations after the COB lifted pressure measures and after the state of exception was declared (Oxígeno, 20 Jun 2026). By 20 June at 10:15, the road agency map cited by Correo del Sur showed 17 active blockade points in three departments, down from 42 the previous day and far below early-June reports near 90 or more (Correo del Sur, 20 Jun 2026). AP’s 21 June update said a return to normality had begun slowly and that only the Morales-aligned cocalero union in Chapare remained firmly in protest (AP, 21 Jun 2026).

The strongest current evidence for a NO is the damage already done and the risk of a bloody emergency operation. AP reported at least 365 arrests, 37 injuries, and at least 17 deaths, mostly tied to lack of medical care caused by transport disruptions; El País cited 14 deaths from the Defensoría del Pueblo on 20 June, so the exact count is noisy but the humanitarian harm is real (AP, 20 Jun 2026, El País, 20 Jun 2026). The decree orders the armed forces to support police in reopening roads and protecting supply routes, which is a survival tool if it works and a removal trigger if it produces massacre-scale casualties (AP, 20 Jun 2026). This is why I do not move the YES probability into the high 80s.

The legal path to removal is narrow in 2026. Article 170 of the Bolivian Constitution ends a presidential mandate through death, resignation to the Plurinational Legislative Assembly, definitive absence or impediment, final criminal sentence, or recall; Article 169 puts the vice president first in the succession line (Constitute Project, Bolivia 2009 Constitution). Recall is not a practical path this year because the electoral law allows a recall initiative only after at least half the elected term has passed and not in the final year; Paz’s five-year term began on 8 November 2025, so the ordinary recall window starts around May 2028 (Infoleyes, Electoral Law Article 27). A final criminal conviction before 31 December 2026 also looks implausible absent an already mature case.

Institutional and security-force signals still favor survival. The PDC is not a majority party, but IPU’s election data show it became the largest bloc in the 130-member Chamber of Deputies with 49 seats, and the legislature has spent June enabling emergency powers rather than organizing a removal process (IPU Parline, 2025 Deputies election, Swissinfo/EFE, 7 Jun 2026). Vice President Lara is a real tail risk because he publicly opposed the state of exception, but he also said he could not vote and treated approval as the Assembly’s decision rather than presenting himself as a replacement (Oxígeno, 20 Jun 2026). I found no public evidence of police or military defection; the public evidence points to compliance, with police and soldiers clearing roads under the emergency decree (AP, 21 Jun 2026).

The economy keeps the back half of 2026 risky. INE reported May 2026 CPI inflation of 2.13% month-on-month, 2.62% year-to-date, and 12.51% over 12 months, with the La Paz conurbation up 5.41% in May alone (INE Bolivia, 3 Jun 2026). IMF DataMapper’s April 2026 WEO vintage projects Bolivia’s 2026 real GDP growth at -3.3%, average consumer-price inflation at 20.7%, and end-period inflation at 26.1% (IMF DataMapper, April 2026 WEO). The central bank’s Q1 2026 reserve report showed international reserves of $3.6125 billion on 31 March 2026, with only $144.4 million in SDRs and most reserves in gold, which helps explain why fuel, dollar, and import stress remain politically explosive (Banco Central de Bolivia, Q1 2026 reserve report).

My scenario model gives a 20% NO probability. I assign 55% to normalization of the current crisis with roads mostly reopened and the COB/Túpac split holding, with 8% loss risk inside that branch; 27% to recurrent blockades and grinding economic conflict without elite rupture, with 25% loss risk; 13% to a repression or Chapare confrontation that causes mass-casualty escalation or security/elite fracture, with 65% loss risk; and 5% to other shocks such as death, incapacity, coup, or a legal surprise, with 20% loss risk. The calculation is:

$$P(\mathrm{NO}) = 0.55(0.08) + 0.27(0.25) + 0.13(0.65) + 0.05(0.20) \approx 0.206$$

I pull the NO probability slightly below that raw 20.6% because the freshest 20–21 June evidence is better for Paz than the weekly average: the Assembly ratified the decree, blockade counts fell sharply, and one of the major rural fronts paused. That yields a final YES probability of 80%.

What's non-obvious

The obvious story is that Bolivia has a history of presidents falling to blockades and Paz faced 50 days of blockades. The less obvious update is that blockades are also costly to the protesters once La Paz, El Alto, hospitals, transport workers, and small businesses become victims of the tactic; AP reported supply shortages, stranded trucks, and deaths linked to blocked medical access, and the latest reports show groups peeling away from the maximalist resignation line (AP, 20 Jun 2026, AP, 21 Jun 2026). Paz is not safe because the economy is bad, but the current movement no longer looks like a unified national uprising.

The second missed point is that the clean constitutional exit ramps are mostly closed. Recall rhetoric has political force, but the half-term rule makes a 2026 recall unavailable, and the legislature’s behavior so far is to legalize and ratify emergency tools rather than remove him (Infoleyes, Electoral Law Article 27, AP, 21 Jun 2026). The live NO path is therefore not a routine vote; it is a sudden rupture after casualties, security-force refusal, or a VP/congressional break.

Limitations

The situation is moving faster than normal political data can capture. I could verify a 20 June road-agency count of 17 blockade points and a 21 June AP report of partial normalization, but I could not verify a single official live national blockade dashboard at the exact forecast timestamp (Correo del Sur, 20 Jun 2026, AP, 21 Jun 2026). The casualty data also vary by cutoff and definition, with AP at at least 17 deaths and El País/Defensoría at 14 on 20 June (AP, 20 Jun 2026, El País, 20 Jun 2026).

The biggest hidden variable is security-force sentiment. Public reporting shows compliance, but Bolivian history shows that police and military judgments can shift privately before they become public (Library of Congress, 13 Dec 2019). The second gap is public opinion: I found no high-quality, fresh national poll that cleanly separates support for resignation, support for order, support for the emergency, and anger over prices. The third gap is the economy: the INE inflation data and IMF projections are enough to show stress, but they do not say whether the next wave of anger will be directed more at Paz, at blockaders, or at the previous MAS order (INE Bolivia, 3 Jun 2026, IMF DataMapper, April 2026 WEO).