Background: What the PITF Actually Is
In 1994, at the direction of Vice President Al Gore, the CIA commissioned a group of academic researchers to build a predictive model: given data on a country today, could the model forecast political breakdown or stability two years out? The project became the Political Instability Task Force (PITF), housed at George Mason University's Center for Global Policy and funded primarily by the CIA's Directorate of Intelligence.[1]
Researchers gathered data on every major episode of political violence (revolutionary wars, ethnic conflicts, adverse regime changes, and mass atrocities) in countries with populations above 500,000 from 1955 onward. Jack Goldstone of George Mason University and colleagues published the model's definitive methodology in 2010 in the American Journal of Political Science, testing more than 100 potential predictors and isolating four that reliably forecast instability: regime type, infant mortality rates, state-sponsored discrimination against defined groups, and conflict in neighboring states.[2]
The PITF was never a CIA "watch list" for adversaries. It was an academic research program producing open-access datasets and papers, funded by the CIA as a policy-relevant research investment. Crucially, it was designed to assess foreign countries, not the United States. Federal law prohibits the CIA from conducting domestic intelligence.[1]
PITF's funding of the Polity data series (the democracy-scoring project it relied upon) ended in early 2020, and the Polity project was discontinued as a living dataset at that point.[5]
The Model's Highest-Risk Category
The most consequential finding in the Goldstone et al. 2010 paper was a five-category regime typology. Contrary to the common assumption that democracies are stable and autocracies are fragile, the model showed that full democracies and full autocracies are both relatively stable. The danger zone is the middle: anocracies, or partial democracies. Among anocracies, the most volatile are those with what Polity codes as factional political competition.
The PITF Five Regime Types — Ranked by Instability Risk
| Regime Type | Polity Range | Relative Risk |
|---|---|---|
| Full democracy | +6 to +10 | Lowest |
| Full autocracy | −10 to −6 | Low |
| Partial autocracy | −5 to −1 | Moderate |
| Partial democracy (no factionalism) | +1 to +5 | Elevated |
| Partial democracy with factionalism | +1 to +5, PARCOMP=3 | Highest — ~30× autocracy baseline |
The PARCOMP indicator (Competitiveness of Participation) in the Polity dataset classifies political competition as competitive (score 5), transitional (4), factional (3), suppressed (2), or repressed (1). A factional score means political competition is organized around parochial or identity-based groupings rather than broad ideological or policy coalitions.[4]
Goldstone et al. found that a country fitting this description (a partial democracy with factional competition) carried roughly 30 times the odds of political instability onset compared to a full autocracy baseline. That is the highest-risk category in the model.[2]
The model was accurate more than 80 percent of the time two years in advance across dozens of countries from 1955 to 2003, though predictive power for magnitude and specific outcome type was lower.[1]
How the US Was Scored — and Who Did It
The task force never scored the United States. It legally could not: the CIA cannot conduct domestic political analysis, and the task force existed specifically to assess foreign countries at risk of collapse or civil conflict.
What actually happened involves three separate moments:
2016–2020: The Polity Recodings
The Center for Systemic Peace, which maintains the Polity dataset, made two significant changes to the US score between 2016 and 2020. First, in 2016, it recoded US political competition from PARCOMP=5 (competitive) to PARCOMP=3 (factional), reflecting partisan polarization around the Trump campaign and election. That change cut the US Polity score from +10 to +8.[3]
Second, in 2020, after recording executive actions including the systematic exclusion of political opponents from administration roles, pressure on election officials, and public undermining of trust in electoral institutions, the Center for Systemic Peace dropped the US score further, to +5. That score placed the US in the anocracy range (−5 to +5), below the democracy threshold of +6, for the first time since 1800.[3]
With a Polity score of +5 (partial democracy) and a PARCOMP of 3 (factional), the US in 2020 met the regime-type criteria the PITF model identifies as highest-risk. The Polity dataset was not updated after 2020. Informal analyses of the Biden-era period placed the US score at approximately +8, back above the democracy threshold, though this was never formally published in Polity format.
2022: Barbara Walter's Independent Application
Barbara Walter, a political scientist at UC San Diego, served on the Political Instability Task Force from 2017 to 2021. In her 2022 book How Civil Wars Start: And How to Stop Them, she applied its methodology to the United States on her own authority — which is what the task force itself could never do.[6]
Walter found the US met both conditions the model identifies as most dangerous: anocracy (the Polity +5 score) and ethnic factionalism (the Republican Party's shift toward organizing around white identity rather than cross-demographic policy coalitions). She estimated this combination produced roughly a 4 percent annual probability of civil conflict onset. Per Walter's own calculation, that figure compounds to approximately 40 percent cumulative risk over 10 years of unchanged conditions.[6]
Walter was explicit that this was her independent analysis, not a CIA or task force finding. The headline claim collapses this distinction.
2026: The Armitage Article
The immediate source of the Reddit post circulating this claim is a September 13, 2026 Substack article by Christopher Armitage, "Scored on the CIA's Instability Model, the United States Is in the Highest-Risk Category for Civil War, Coup, or Collapse."[7] Armitage is an independent writer, not affiliated with the task force. His article updates Walter's analysis with more recent democracy indicators and argues the US position has worsened since 2020. The article itself acknowledges the CIA restriction. The Reddit headline, derived from the article's title, dropped this caveat entirely.
The Right Question: Attribution vs. Risk
The headline frames this as a question about risk level. That question has a largely supported answer for 2016–2020: yes, per Polity data mechanically applied to the PITF regime typology, the US met the highest-risk criteria. But there is a more important question embedded here: who made this assessment, and on what authority?
The framing "CIA's instability model" implies the CIA evaluated the United States and produced a threat assessment. That did not happen. The CIA funded the development of a research model applied to foreign countries; academics then applied the same methodology to the US on their own. A "CIA threat assessment" carries institutional authority and legal sanction; an academic's independent application of a CIA-funded model carries academic authority only.
A second embedded question (the one the editor's revision addresses) is whether the broader pattern of US democratic decline, viewed comparatively, supports the elevated risk claim regardless of attribution. That question has a clearer answer, covered in the sections below.
The Case Against: Why Critics Push Back
Strongest Case That the Risk Is Overstated
The PITF model was built on data from countries substantially different from the United States: post-colonial states, developing economies with weak institutions, countries undergoing first-wave democratization. Critics argue the model's calibration may not translate to a country with centuries of constitutional continuity, an independent federal judiciary, a professional military with strong norms against political involvement, and deep civil society infrastructure.
The January 6, 2021 Capitol attack, the most severe challenge to the transfer of power in modern American history, produced no military coup attempt, no contested assumption of office, and ultimately a lawful succession. The 2024 election was contested but ended in a peaceful transfer. These events suggest American institutions absorbed severe stress without collapse, which is not the pattern in countries the PITF model was built to predict.
The Polity +5 score is also a snapshot, not a verdict. Some researchers argue the 2020 downgrade overstated democratic erosion relative to the actual legal and institutional record. The Biden-era informal reassessment put the US at approximately +8, back in formal democracy territory. The Polity dataset was discontinued before capturing either Trump's return or any potential institutional recovery.
Finally, the 30× odds ratio is a global average across regime types in the full dataset, not a US-specific calibration. A model trained on Somalia, Ivory Coast, and Pakistan in the 1960s–1990s may produce risk estimates that are structurally miscalibrated for a country with the US's institutional depth.
These are legitimate objections. The PITF model's accuracy rate (80 percent across historical cases) still implies a 20 percent error rate, and the US sits in a structural and institutional context the model was not trained to assess. The counter-arguments score as genuine methodological caveats (score 3 on direct refutation), not as evidence that risk is absent. They explain why the model's output should be treated as an upper-bound estimate, not a certainty.
Corroborating Evidence From Other Indices
Where multiple independent democracy-measurement systems agree, convergence strengthens any single finding. In 2025–2026, several major indices registered their most significant US downgrades on record:
- V-Dem (2026): The US was downgraded from liberal democracy to electoral democracy. The US Liberal Democracy Index score fell 24 percent in a single year, from 0.75 to 0.57, dropping the country's global ranking from 20th to 51st out of 179 nations. V-Dem described this as "the sharpest one-year fall in our record reaching back more than two centuries" among established democracies. The US score of 0.57 matches the level last seen in 1965, the year the Voting Rights Act was signed.[8]
- Freedom House (2026): The US scored 81 out of 100, its lowest score since the organization began its 100-point system. The US has lost 12 points over 20 years of measurement, the steepest long-term decline of any country Freedom House still rates as "Free." In 2025 alone, the US dropped 3 points, a single-year drop matched only by Bulgaria.[11]
- The Economist Intelligence Unit (2025): Rated the US a "flawed democracy" with a score of 7.65 out of 10 — its lowest since the index launched in 2006. The US fell from 28th to 34th globally, where it has remained a "flawed democracy" since 2016.[9]
- Century Foundation Democracy Meter (2025): A new tool assessing 23 sub-indicators rated US democracy at 57 out of 100 in 2025, down from 79 in 2024 — a 28 percent single-year drop. The State Institutions sub-category (checks and balances constraining executive power) scored 10 out of 30 in 2025, down from 22 out of 30 in 2024, losing more than half its prior score in a single year.[10]
- Pew Research Center (2026): Multiple survey-based and institutional indicators showed measurable decline in executive constraint, judicial independence, and press freedom metrics during 2025. Among G7 nations using V-Dem scores, the US at 0.57 fell below Canada (0.74), France (0.80), Germany (0.78), Italy (0.64), Japan (0.73), and the UK (0.69).[9]
These indices measure democratic quality rather than civil war probability. Democratic erosion and civil conflict risk are correlated but not equivalent; many countries with declining democracy scores do not experience civil war or coup. The convergence of indicators does support the claim's underlying premise that US democratic health has deteriorated across multiple independent measurement frameworks.
What Is Driving the Decline
The indices above measure outcomes; the specific factors they flag explain why the US scores are falling. Across Freedom House, V-Dem, and the Century Foundation assessments, five categories of institutional erosion appear consistently.
1. Collapse of Executive Constraints
V-Dem's measures of legislative and judicial constraints on executive power fell in 2025 to their lowest levels in over 100 years. The Century Foundation's State Institutions category, which specifically measures checks and balances limiting presidential authority, fell from 22 to 10 out of 30 points (73 percent to 33 percent of ideal capacity) in a single year. This is the factor with the highest weight in the PITF model and the one that most directly connects the democracy indices to the instability framework.[8][10]
2. Press and Academic Freedom
V-Dem found that freedom of expression in the US "is now at its lowest level since the end of World War II." The 2026 report cited "suppression and intimidation of media and dissenting voices" as a primary driver of the US downgrade. Freedom House docked points specifically for "a multiyear rise in threats and reprisals for political expression, and the new administration's efforts to punish nonviolent speech by noncitizens." The US Press Freedom Index separately placed the US 64th globally in 2026, behind Botswana — down 7 places from the prior year.[8][11]
3. Anticorruption Safeguards
Freedom House reduced its corruption-safeguards sub-score specifically citing "weakening of anticorruption safeguards and enforcement efforts," including presidential family business ventures during office and clemency decisions benefiting political allies. This category had been a relative strength of US democratic governance and is among the factors distinguishing full democracies from anocracies in the PITF framework.[11]
4. Congressional Dysfunction
Freedom House noted "escalation of a multiyear trend of partisan gridlock and dysfunction in Congress, culminating in a funding crisis and the longest government shutdown in US history" as a driver of its 2026 downgrade. Congressional capacity to check executive action — one of the PITF model's central regime-type variables — is measured partly through effective legislative oversight, which multiple indicators show has weakened.[11]
5. Immigration Enforcement and Due Process
Freedom House reduced the US's equal-treatment score citing immigration enforcement actions characterized by "disregard for or avoidance of due process protections." This factor intersects with the PITF model's government-discrimination-against-defined-groups variable — one of the four predictors Goldstone et al. identified as independently forecasting instability onset.[11]
Taken together, four of the PITF model's four independent predictors of instability (regime type, state discrimination, executive constraints, and by extension factionalism) are now moving in the adverse direction, per multiple sources. The model's regime-type predictor, its strongest single variable, requires formal Polity coding to apply mechanically, and that data is unavailable after 2020. But the underlying dynamics the Polity score was designed to capture are visible across every active tracking system.
Nations That Rank Near the United States
Prior to 2017, V-Dem rated the US as more democratic than most G7 peers. By 2026, the US scored below all of them. More striking is the company the US now keeps at rank 51 in the overall Liberal Democracy Index: countries in that neighborhood include Eastern European states that have been categorized as troubled democracies for decades.
G7 Comparison
G7 Nations — V-Dem Liberal Democracy Index Score (2025 data)
| Country | LDI Score (0–1) | V-Dem Classification | Trend |
|---|---|---|---|
| France | 0.80 | Liberal Democracy | ↓ Slight decline |
| Germany | 0.78 | Liberal Democracy | → Stable |
| Canada | 0.74 | Liberal Democracy | → Stable |
| Japan | 0.73 | Liberal Democracy | → Stable |
| United Kingdom | 0.69 | Liberal Democracy | ↓ Declining |
| Italy | 0.64 | Liberal Democracy | ↓ Declining |
| United States | 0.57 | Electoral Democracy | ↓↓ Unprecedented drop |
Source: Pew Research Center, April 2026, citing V-Dem Institute 2026 data.[9]
Countries at a Similar Classification Level
The V-Dem categories (liberal democracy, electoral democracy, electoral autocracy, closed autocracy) are based on a combination of scores, not a single threshold. The US's shift from "liberal democracy" to "electoral democracy" places it in the same tier as countries including Romania and Bulgaria — former Warsaw Pact states with histories of post-Soviet institutional fragility. That classification gap matters: "liberal democracy" describes a system where civil liberties and rule of law constrain power even between elections; "electoral democracy" describes a system with competitive elections but insufficient protection of those civil liberties.[8]
Long-Run Trajectories and Comparable Cases
Long-Term Freedom House Declines: 20-Year Comparison
| Country | FH Score Change (20 yr) | Current FH Score | Current Status | Notable Factor |
|---|---|---|---|---|
| Hungary | −28 pts | 65/100 | Partly Free | Systematic media and judiciary capture under Orbán since 2010 |
| India | −14 pts | ~67/100 | Partly Free | Press freedom restrictions, minority discrimination, executive centralization |
| United States | −12 pts | 81/100 | Free | Executive dominance, congressional dysfunction, press freedom erosion |
| Brazil | −4 pts | ~80/100 | Free | Declined during Bolsonaro (2019–2022); partially recovering since |
| Italy | ±0 | ~90/100 | Free | V-Dem flagged for recent backsliding; FH score relatively stable |
Sources: Freedom House 2026; figures for India and Brazil are approximate based on long-term trend data reported in Freedom House's 20-year comparison.[11]
Speed of Decline: The Autocrat Comparison
Perhaps the most arresting comparative finding comes from V-Dem's speed analysis. The 2026 report found that the pace of US democratic erosion in 2025 "outpaces not only Trump 1.0 but also the most prominent autocrats of the last 25 years," specifically naming Russia under Putin, Turkey under Erdoğan, and Hungary under Orbán as reference points. Those three cases are the benchmark comparators in academic autocratization research because they represent the fastest documented democratic breakdowns in countries with functional pre-existing institutions. On raw speed of decline, the US exceeded all three in 2025.[12]
For context: the Armitage Substack article also benchmarks the Republican Party's V-Dem anti-pluralism score at 0.43 (2018 coding), compared to Hungary's Fidesz at 0.88 and Turkey's AKP at 1.0 on the same 0–1 scale. That score measures how strongly a governing party rejects the legitimacy of political opponents. At 0.43, the Republican Party scored higher than any governing party in Western Europe but lower than the parties that produced Hungary's and Turkey's full autocratization.[7]
What Peer Cases Suggest About Trajectory
The Hungarian and Turkish cases are the most instructive peer references because both started in the "partial democracy with factionalism" zone before full autocratization. Hungary's Freedom House score fell 28 points over 20 years to reach its current "Partly Free" rating of 65. Turkey lost its democratic classification entirely between 2013 and 2017. Both trajectories took years, and both featured similar institutional-erosion sequences to what the US indices are currently recording: media capture, judiciary pressure, executive constraint collapse, then formal reclassification.
Brazil offers a counter-case: it entered similar autocratization dynamics under Bolsonaro (2019–2022), retained its "Free" classification, held free elections in 2022, and has since partially recovered on some indicators. Brazil's resilience is cited by critics of the instability-model application as evidence that institutional stress does not inevitably lead to collapse.
Verdict: Why This Rates Mixed
The factual core of the claim is substantially supported for the 2020 period. The PITF model identifies partial democracy with factionalism as its highest-risk regime type. The US met that description per Polity data for 2020–2021. Barbara Walter, with direct task-force experience, applied the methodology to the US and found an elevated annual conflict probability. Multiple democracy indices corroborate continued and deepening institutional stress since then.
The comparative analysis in this revision strengthens the underlying concern: the US has now fallen below all G7 peers, dropped from 20th to 51st globally, and is declining faster than the historical reference cases used to define autocratization. These are not predictions of civil war; they are measures of the institutional conditions the PITF model treats as precursors.
The claim fails on three specific points. First, the central attribution: the CIA never scored the United States and is prohibited by law from doing so. The assessment is academic, not governmental. Second, the Polity data underlying the formal "highest-risk" designation is from 2020 and was not updated. The project was discontinued and we cannot formally apply the PITF regime-type classification to 2025–2026 conditions. Third, the probabilistic-to-categorical collapse: the model produces odds ratios, not a binary verdict. A 4 percent annual conflict probability is elevated; it is not a prediction of imminent civil war.
The Mixed rating stands. The underlying risk concern is well-supported across independent measurement systems. The specific framing ("CIA scored the US") is not.
What to Watch
Forward-Look Signals — What Would Settle or Shift the Picture
- November 2026 midterm elections (8 weeks away): Whether results are accepted and transitions proceed without legal challenge is the most direct near-term test. V-Dem's designation of the midterms as a "critical test" for US democratic classification reflects this: disputed results leading to institutional conflict would support the risk assessment; a clean process with broad acceptance would be evidence against the factionalism coding. V-Dem has indicated it will update the US classification following the midterms.
- Polity project revival: Scholars have proposed continuing the Polity dataset under different sponsorship. A new Polity score covering 2021–2026 would provide the first model-native update to the "highest-risk" finding, settling whether the Biden-era partial recovery was real and whether Trump 2.0 conditions re-trigger the anocracy and factionalism codings.
- V-Dem's 2027 annual report: V-Dem has now downgraded the US twice in succession. Whether the 2025 decline continues, stabilizes, or reverses will indicate whether the institutional erosion is a temporary shock or a sustained structural shift. A second consecutive sharp decline would formally move the US from the bottom of the "electoral democracy" band toward "electoral autocracy" territory.
- Court rulings on executive power: Several cases concerning presidential immunity, agency independence, and judicial enforcement are working through federal courts. Outcomes will shape whether the institutional constraints Polity measures as "executive limitations" hold or erode further. The Century Foundation's State Institutions score — which fell from 22 to 10 out of 30 in a single year — will track this most directly.
- Hungary post-2026 election: Hungary's parliamentary election in April 2026 produced an opposition victory after 12 years of Orbán's rule, offering a potential case study in democratic recovery. Whether Hungarian institutions rebuild is a live test of whether autocratization is reversible once index scores reach the levels the US is approaching — and how long recovery takes.
What Is Being Checked?
A post shared widely online claims: "Scored on the CIA's instability model, the United States falls into the highest-risk category for civil war, coup, or collapse." The title comes from a September 2026 Substack article by Christopher Armitage, which drew on political scientist Barbara Walter's 2022 book How Civil Wars Start.
Verdict: Mixed. Part of the claim is supported. The US did score in the highest-risk group when a CIA-funded research tool is applied to its recent political data. But the framing is misleading. The CIA never scored the US, and it is against the law for it to do so. Academics applied the model on their own.
What Is the Political Instability Task Force?
In 1994, the CIA paid a group of academic researchers to build a tool for predicting political collapse. The researchers looked at countries around the world and tried to find patterns that showed up before civil wars, coups, or government failures. This program was called the Political Instability Task Force, or PITF for short.
The PITF was not a CIA spy program. It was academic research, published in open journals. The key paper came out in 2010, written by a team led by Jack Goldstone of George Mason University. The CIA cannot legally analyze political conditions inside the United States. That is a federal restriction. The task force only studied other countries.
What Did the Model Find?
The model tested more than 100 possible warning signs. Only four turned out to be strong predictors of political collapse: regime type, infant mortality rates, government discrimination against specific groups, and conflicts in neighboring countries. Regime type was the most important factor.
The single most dangerous category was a partial democracy where political parties compete along identity lines rather than policy lines. The model called this "factional" competition. Countries in this category were about 30 times more likely to experience political breakdown than pure dictatorships. That is the highest-risk category in the model.
How Was the US Scored?
The US was scored, but not by the CIA. Two separate researchers applied the model to the US.
First, the data itself changed. A long-running research project called the Polity dataset tracks how democratic each country is on a scale from -10 (fully autocratic) to +10 (fully democratic). In 2016, researchers changed the US code for political competition from "competitive" to "factional," cutting the US score from +10 to +8. Then in 2020, they dropped the score again to +5. A score of +5 puts the US in the anocracy range — a mixed democracy-autocracy. The US had not been there since 1800.
Barbara Walter, who had served on the task force, then applied the model to the US in her 2022 book. She estimated about a 4 percent annual chance of civil conflict. After 10 years without improvement, cumulative risk reaches about 40 percent by her calculation.
What Is Driving the Decline?
Several big forces pushed the US democratic scores lower in 2025 and 2026, according to multiple research groups:
- Checks and balances collapsed. One group found that the ability of Congress and courts to limit presidential power fell to the lowest level in over 100 years in 2025. Another group scored this category at just 10 out of 30 points, down from 22 the year before.
- Press freedom declined. One research group found that freedom of expression in the US is now at its lowest since World War II. The US fell to 64th in the world on press freedom in 2026.
- Anti-corruption protections weakened. Freedom House reduced the US score because of actions it described as undermining safeguards against corruption in government.
- Congress became less effective. The US experienced its longest government shutdown in history. Congress's ability to check the executive branch was reduced, according to multiple groups.
- Due process concerns. Freedom House docked points for immigration enforcement actions that did not follow normal legal protections.
Which Countries Are Near the US Now?
Before 2017, the US ranked among the most democratic countries in the world — above most of its closest allies. By 2026, that had changed. The US is now below all six other G7 nations on the main international democracy score:
- France: 0.80 out of 1.0
- Germany: 0.78 out of 1.0
- Canada: 0.74 out of 1.0
- Japan: 0.73 out of 1.0
- United Kingdom: 0.69 out of 1.0
- Italy: 0.64 out of 1.0
- United States: 0.57 out of 1.0
The US score of 0.57 puts it in the same category — "electoral democracy" — as countries like Romania and Bulgaria, which have been categorized as troubled democracies for decades. It is no longer classified as a "liberal democracy."
One research group found that the speed of US democratic decline in 2025 was faster than the decline seen in Russia under Putin, Turkey under Erdogan, or Hungary under Orbán — three of the fastest-moving cases of democratic breakdown in recent history. Hungary's Freedom House score has fallen 28 points over 20 years. The US has fallen 12 points in the same period, with most of that recent drop happening in 2025.
The Case Against: Why Critics Disagree
The model was built using data from developing countries and newly independent states, not from long-established democracies like the US. The US has more than 200 years of continuous constitutional government. It has an independent federal judiciary, a professional military with strong traditions of neutrality, and deep civic institutions.
When the Capitol was attacked on January 6, 2021, no military coup happened. The law prevailed. The 2024 election ended in a peaceful transfer of power. Brazil went through a similar period of democratic stress under Bolsonaro (2019–2022) and largely recovered. These cases suggest institutional stress does not automatically lead to collapse.
Verdict Summary
The core risk finding is supported. The US met the PITF model's highest-risk criteria per 2016–2020 data. Multiple other indexes confirm continued democratic decline since then — faster than comparable historical cases. The comparative picture is alarming: the US now ranks 51st globally, below all G7 peers, in the same classification as Romania and Bulgaria.
The framing is misleading. Saying the CIA "scored" the US implies an official intelligence assessment that never happened and could not legally happen. The scoring was done by an academic applying the model on her own authority.
The risk is real and documented. The attribution is wrong. Both things are true at the same time, which is why the overall rating is Mixed.
What to Watch
- November 2026 midterms (8 weeks away): Whether elections proceed without serious legal challenges will be the most direct near-term test. V-Dem has called them a "critical test" for the US democratic classification.
- Polity project revival: If researchers restart the Polity dataset, the updated US score will show whether the 2020 highest-risk classification still holds.
- V-Dem 2027 report: A third consecutive downgrade would signal the trend is continuing. A reversal would be evidence the risk is stabilizing.
- Court rulings on executive power: Cases testing presidential authority will shape the key Polity-coded variable of executive constraints.
- Hungary's recovery: Hungary's opposition won parliamentary elections in April 2026 after 12 years of authoritarian rule. Whether its democratic institutions recover is a live test of whether democratic backsliding is reversible — and how long it takes.
Sources
- Political Instability Task Force
- A Global Model for Forecasting Political Instability (open-access PDF)
- USA Downgraded from Democracy to Anocracy
- PITF Democracy Indicator (PARCOMP values)
- Polity Project — Center for Systemic Peace
- Q&A with Barbara Walter, author of How Civil Wars Start
- Scored on the CIA's Instability Model, the United States Is in the Highest-Risk Category for Civil War, Coup, or Collapse
- Democratic Backsliding Reaches Western Democracies, with US Decline "Unprecedented"
- Multiple Indicators Show a Decline in the Health of America's Democracy in 2025
- New Democracy Index Finds American Democracy 'Collapsed' as Trump Government 'Slid Into Authoritarianism' in 2025
- United States: Freedom in the World 2026 Country Report
- Democracy Under Trump Is Eroding Faster Than It Did in Russia Under Putin
