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

TL;DR

I assign an 86% chance that Rodrigo Paz Pereira remains continuously President through December 31, 2026. YES is the base case because only 101 days remain, he survived a larger 53-day blockade crisis, and Congress and the security chain still support the government’s core survival tools. The main NO path is a diesel-triggered national blockade that causes deaths and then legislative or security-force defections; the Lara succession dispute is a smaller but real legal tail.

Context

At the September 22, 2026, 03:47 UTC cutoff—23:47 on September 21 in Bolivia—approximately 101 days remained. Bolivia’s official presidency page still listed Paz as president, while an official September 21 release described him conducting presidential business in the United States.

Two fresh shocks dominate the forecast. Paz raised diesel from Bs9.80 to Bs17.95 per liter, an 83% increase, immediately after Congress approved a US$1.9 billion IMF program that still required IMF Executive Board approval before disbursement. Vice President Edmand Lara then invoked Article 169 and declared himself president in exercise during Paz’s trip, while the executive said Paz continued governing remotely under Supreme Decree 5515 and had not transferred authority. (AP; La Razón; official decree).

Evidence

The historical backbone starts with a low short-horizon base rate. Academic research identifies four Bolivian presidencies shortened by protest since democratic restoration—1985, 2003, 2005, and 2019—while Hugo Banzer’s 2001 health resignation gives a broader fifth event. Using four to five exits over roughly 44 president-years produces a random 101-day exit probability of about 2.5%–3.1%: 1 - exp[-(4 to 5)/44 × 101/365.24]. Paz’s present exposure is much worse than a random quarter, but this baseline prevents treating turmoil as equivalent to imminent removal. (Cambridge study; Bolivian historical overview).

The closest policy analogues favor survival, often through retreat. Bolivia’s 2010 gasolinazo involved a one-off fuel increase of roughly 60%–80%, produced mass protests, and was reversed without removing Evo Morales. Ecuador’s 2019 subsidy removal paralyzed the country for 11 days, but Lenín Moreno rescinded the measure and completed his presidency. These cases show that a fuel revolt can defeat a policy without defeating the president. (IMF working paper; Ecuador account). (imf.org)

Paz’s best evidence is his survival of the May–June crisis. That campaign lasted 53 days, exceeded 100 roadblocks, caused severe shortages and at least 22 reported deaths under the broadest later count, and included explicit demands for his resignation. Paz negotiated with the COB, declared emergency rule, used joint police-military deployments, and remained in office after all national routes were reported open on June 23. (El País; Bolivian Public Works Ministry).

The institutional shield is stronger than Paz’s party label implies. On September 17, the Assembly approved a 90-day extension of emergency rule with 100 votes among 158 legislators present. On September 17–18, both chambers approved the IMF financing bill by more than two-thirds, despite Paz lacking a stable party majority. These are issue-specific coalitions rather than personal loyalty, but they show that most legislators still prefer an operating Paz government to uncontrolled succession or renewed paralysis. (Interior Ministry; Economy Ministry—Deputies; Economy Ministry—Senate).

The security evidence also favors survival. Police and military personnel were deployed on strategic routes after the diesel announcement, and Bolivia’s fundamental road network had no socially motivated blockades in the early September 21 official transit update. The government reached technical-talk agreements with parts of the transport sector, while military commanders had recently defended the institutional chain of command against Lara’s attempt to direct an Army officer. This does not prove future loyalty, but it is affirmative evidence of current obedience rather than mere silence about defection. (El Deber; Public Works Ministry; ERBOL). (oopp.gob.bo)

The downside remains large. Diesel is central to buses, freight, mining, farming, and food distribution. The national drivers’ organization, COB, teachers, campesino groups, and Morales-aligned coca growers were preparing meetings or demonstrations, with the drivers’ decision due September 22 and a major cocalero meeting planned for September 26. The first response was fragmented, but those groups have already demonstrated that they can combine into a national blockade coalition. (Visión 360; La Razón).

Paz is also politically weak. An Ipsos CIESMORI online poll of 400 internet-connected adults in La Paz, El Alto, Cochabamba, and Santa Cruz, fielded August 19–31, reported 21% approval and 65% disapproval; Lara was worse, at 10% approval and 75% disapproval. The survey is urban and online rather than nationally representative, but the deterioration is real. Only six of Paz’s original 15 ministers reportedly remained by August, and an INESAD estimate published September 14 put first-half real GDP contraction at 3.34%, while official August CPI was 1.10% month over month and 5.02% year over year. (Ipsos CIESMORI; AP; INESAD; INE).

The Lara dispute adds less risk than the headlines suggest, but more than zero. Article 169 says the vice president assumes the presidency during a temporary absence, while DS 5515 says Paz can exercise authority electronically and requires an express written transfer when remote exercise is impossible. The decree remains in force, its suspension was denied, and Bolivia’s Constitutional Court has only four of nine magistrates and had issued no merits rulings on constitutional challenges for months because five votes were normally required. Senate President Diego Ávila—the person Lara proposed as the next temporary successor—was a senior aide and close ally of Paz in Tarija, making his participation in a rival transfer unlikely. (Constitution; decree status; court quorum; Ávila profile).

My central event tree assigns a 40% chance of another sustained national crisis before year-end and a 28% chance of permanent exit conditional on that crisis. I assign a 2% political-exit risk if no such crisis develops, a separate 1.2% risk from an officially recognized temporary succession or adverse legal decision, and 0.6% for death, incapacity, a final criminal judgment, or another idiosyncratic route. The political branch is 0.40 × 0.28 + 0.60 × 0.02 = 12.4%; combining the other branches gives P(NO) = 1 - (1 - 0.124)(1 - 0.012)(1 - 0.006) = 14.0%, leaving 86% YES.

What's non-obvious

Paz’s 21% approval rating overstates the public opposition to the specific reform now driving the crisis. The same Ipsos CIESMORI research program reported in August that 63% of respondents in Bolivia’s urban axis still supported eliminating fuel subsidies, although only 50% accepted higher prices as the cost of ending shortages. That makes a broad anti-Paz uprising less automatic than the approval number suggests: organized transport groups can paralyze roads, but they may not command a national majority for removing the president. (Ipsos CIESMORI).

The other overlooked point is that Lara has symbolic access to succession but little executable power. No cabinet minister, police commander, military commander, major business organization, or cross-party legislative majority had recognized him as president by the cutoff. His announced plan to travel to Paraguay and hand temporary authority onward to Ávila looks like constitutional theater and rival-brand building, not an effort to seize and retain the executive apparatus. The OAS’s September report also found broad support for respecting elected mandates and constitutional mechanisms, raising the cost of irregular removal. (OAS; La Razón).

Uncertainties

The highest-value missing information is what the drivers, COB, and cocalero organizations decide between September 22 and September 26. A coordinated indefinite blockade would move the forecast down quickly; separate negotiations, marches, and short strikes would move it above 90%.

The polling evidence is thin. The main approval and subsidy results come from a monthly online panel of 400 connected urban adults, not a national rural-and-urban probability sample. Rural campesino and cocalero attitudes matter more for road-blockade capacity than their share of the national population.

The legal status of temporary succession remains unresolved. There was no formal transfer, institutional recognition of Lara, or final Constitutional Court ruling by the cutoff, but Article 169’s text is direct enough that an unexpected court action or formally accepted transfer could trigger the question’s strict continuity rule. September protest-event data also lag the diesel decision, so existing event databases cannot yet show whether a second national wave has begun.