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

TL;DR

I put YES at 87%. Paz has recent legislative support and active police-military compliance, making survival through year-end the clear base case. The main NO path is a diesel-price protest wave that becomes a national resignation movement, produces deaths, and triggers legislative or security-force defections.

Context

At the cutoff, Bolivia's official presidency page still listed Rodrigo Paz Pereira as president and recorded that he took office on November 8, 2025. The immediate trigger is the government's decision to raise diesel from Bs 9.80 to Bs 17.95 per liter and link it to international prices, while offering cash payments and preferential credit to offset part of the impact (Ministry of Hydrocarbons and Energy). Interdepartmental transport leaders announced a 24-hour stoppage for September 21 and threatened longer action after talks failed (Red Uno).

Bolivia is not in routine unrest. Paz already survived a 53-day campaign that demanded his resignation, generated more than 100 roadblocks and was later reported to have caused 22 deaths (El País). But the new crisis starts with a weaker government: by August, only six of the original 15 ministers remained, and Vice President Edmand Lara's relationship with Paz had become openly hostile (Associated Press).

Evidence

The best historical backbone is a short-horizon conversion of presidential-exit research, not the raw full-term failure rate. Among 44 elected South American presidential terms from 1978 through 2005, 17 presidents were challenged and 11 left office prematurely (Hochstetler and Edwards). A broader study estimated a 4.23% annual interruption incidence in Latin America and found that protests explicitly aimed at the president were associated with roughly nine to ten times higher odds of interruption (Kim and Bahry). Converted mechanically to the remaining horizon, the unconditional risk is about 1%, while the protest-adjusted benchmark is about 11%. That is an anchor, not a direct model of Bolivia in 2026.

Bolivia's closest analogues show why security forces and elite behavior matter more than protest size alone. Gonzalo Sánchez de Lozada's 2003 fall followed lethal repression, Vice President Carlos Mesa's break and wider institutional abandonment (Journal of Latin American Studies). Evo Morales's 2019 resignation followed mass protests and loss of control over the security forces (International Journal of Constitutional Law). Paz has the protest and vice-presidential risk factors, but the evidence at the cutoff showed operational compliance: about 500 police and military personnel were deployed to keep the Cochabamba–Oruro corridor open after the diesel decision (La Razón).

Paz lacks a personal legislative majority. His PDC won 49 of 130 lower-house seats and 16 of 36 Senate seats (IPU election record). Yet revealed legislative behavior is stronger than the seat count suggests. The Assembly supported a 90-day extension of emergency rule with 100 votes, versus 58 opposing or abstaining (El País). The IMF financing law then passed by more than two-thirds in both the Chamber of Deputies and the Senate. These are issue-specific coalitions, not personal loyalty. They still show that legislators had not converged on replacing him days before the cutoff.

Formal removal is a weak near-term route. The Constitution lists death, resignation, definitive incapacity, final criminal conviction and recall as grounds for presidential cessation. Recall cannot begin until half the term has elapsed, while a presidential criminal trial requires a prosecutor-general request, authorization by two-thirds of legislators present and a final conviction (Bolivian Constitution, Articles 169–171, 184 and 240). No advanced proceeding satisfying those stages was visible at the cutoff. A politically forced resignation is far more plausible than completed legal removal.

The downside is concentrated in the new diesel shock. At least five organized sectors—including the COB, drivers, heavy transport, fuel-tanker operators and the Six Federations of the Cochabamba Tropics—rejected the measure or prepared mobilization (Brújula Digital). Paz also has little political cushion. An online Ipsos CIESMORI survey of 400 internet-connected adults in four major urban areas, fielded August 19–31, found 21% approval and 65% disapproval, with a referential margin of error of 4.9 percentage points (Brújula Digital). The survey is not nationally representative, but it shows severe urban erosion.

There are economic buffers, though their benefits arrive later than the costs. The government announced Bs 874 million for roughly 2.9 million cash-transfer recipients and up to Bs 800 million in preferential credit (Ministry of Hydrocarbons and Energy). Official August CPI was 1.10% month-on-month, 3.01% for January–August and 5.02% over twelve months, but that release predates the final diesel increase (Bolivian National Statistics Institute). The IMF agreement covers about $1.9 billion over 36 months but still required Executive Board approval and prior actions at the cutoff (IMF).

My central event tree is:

This tree produces about 14% NO. I combine it with an 11% historical-model estimate and a 17% institutional stress case, weighted 60%, 25% and 15% respectively. The combined result is about 13% NO and 87% YES.

What's non-obvious

The approval rating is less informative than the legislature's last two votes. Presidents can survive deep unpopularity when lawmakers and coercive institutions still prefer continuity. The constitutional successor is also a poor focal point: the same urban online poll put Lara at 10% approval and 75% disapproval (Ipsos CIESMORI reporting). Legislators may prefer a weakened Paz to empowering Lara.

The transport movement is not one organization acting under one command. Some departmental, intermunicipal and interprovincial representatives reported an agreement and withdrew their stoppage threat (Bolivia TV), while interdepartmental leaders separately left talks and confirmed a 24-hour strike (Red Uno). That fragmentation matters. It gives the government room to settle one sector at a time, as it did during the earlier crisis.

Uncertainties

The largest gap is that the cutoff came before the first full day of the announced September 21 stoppage. The forecast therefore rests on mobilization plans, not observed nationwide paralysis.

• Actual roadblock counts, geographic spread and duration would sharply improve the estimate.
• The key political question is whether transporters, the COB, miners, farmers, teachers and the September 26 coca-grower meeting form one command and renew an explicit resignation demand (El País).
• Any deaths caused by enforcement, refusal of orders by police or military commanders, or a legislative demand for immediate succession would move the forecast down fast.
• The emergency extension expires around December 20, leaving a short year-end gap under ordinary rules (Red Uno). IMF Board approval and actual disbursement also remained unresolved at the cutoff (IMF).

Some crisis figures are vintage-sensitive. Reporting on June 20 cited 14 deaths, while a September retrospective cited 22 (June account; September account). I treat the later figure as a cumulative total, not a clean measure of deaths directly caused by security forces. My reasonable range for YES is 78%–93%, with most downside concentrated in the next several weeks.