© 2026 Cambridge Governance Labs. All rights reserved.
Working Draft — Not for Distribution
All data and analysis from the Political Topology Project.
Dataset: 91 countries, 225 years (1800–2025), 1,656 observations.
This work draws on data from Freedom House (Freedom in the World), the Varieties of Democracy Institute (V-Dem), the Fund for Peace (Fragile States Index), the World Bank (World Development Indicators and Worldwide Governance Indicators), the United Nations Development Programme (Human Development Index), and the International Monetary Fund (World Economic Outlook). The authors gratefully acknowledge these organisations for making their data publicly available. Any errors of interpretation are our own.
No part of this working draft may be reproduced, distributed, or transmitted in any form, or by any means without the prior written permission of the publisher, except for brief quotations in critical reviews, and certain other noncommercial uses permitted by copyright law.
This book represents working research. All findings should be treated as preliminary. The models, probabilities, and projections contained herein are offered as tools for structured thinking, not as predictions of specific outcomes.
First working edition: February 2026
Typeset in IBM Plex Serif and IBM Plex Sans.
Designed and composed by Cambridge Governance Labs, Cambridge, United Kingdom.
For the institution-builders—past, present, and yet to come—
who understood that freedom is not a natural resting state
but a structure that must be constructed, maintained,
and, when it fractures, rebuilt.
And for those who defend democratic institutions
in the hours when defence is most costly
and least rewarded.
Something is happening to political freedom that we do not yet fully understand. Over the past two decades, democratic institutions in every region of the world have weakened, hollowed out, or collapsed entirely. This much is widely reported. What has not been widely understood is that the decline may be far harder to reverse than the optimistic post-Cold War consensus assumed. This book presents evidence that political freedom behaves less like a pendulum—swinging back after each retreat—and more like a ball on a landscape of hills and valleys, where rolling past certain thresholds makes return not merely difficult but statistically improbable.
Political Topology is the product of a two-year research programme at Cambridge Governance Labs. We assembled a dataset covering 91 countries across 225 years, from 1800 to 2025, yielding 1,656 country-period observations. Each observation was scored on three dimensions—Liberty (L), Tyranny (T), and Chaos (C)—constrained to sum to 100. This deceptively simple framework, which we call the LTC model, allows us to map every country at every point in time onto a single triangular coordinate space, a political topology in the literal sense: a landscape with peaks, valleys, basins of attraction, and thresholds from which recovery becomes vanishingly rare.
The book opens by establishing the theoretical architecture. Chapter 1 introduces the tristable basin model: the observation, grounded in data, that political systems tend to cluster around three attractors rather than distributing smoothly along a single democracy-autocracy continuum. Liberty, Tyranny, and Chaos are not merely labels; they are gravitational wells in the political landscape. Countries near the centre of each basin tend to stay there. Countries between basins are unstable, and the direction of their drift is more predictable than conventional political science typically acknowledges. Chapter 2 identifies what we call the event horizon—a Liberty threshold, located empirically at approximately L=52, below which the probability of a country recovering to full democratic governance in any subsequent decade drops to 3.0 percent. This is not a theoretical conjecture. It is a statistical regularity observed across two centuries of data. Chapters 3 and 4 detail the eight institutional mechanisms that drive the descent from Liberty towards Tyranny, and the methodological foundations of the analysis respectively.
Part II presents the empirical core. Chapter 5 surveys 225 years of political topology, showing how the global distribution of countries across the LTC space has shifted over time—the great expansion of Liberty after 1945, the second wave of democratisation after 1989, and the steady contraction since approximately 2006 that has now erased nearly two decades of gains. Chapter 6 documents what we regard as the most strategically consequential finding in the dataset: the great decoupling of state capability from political freedom. As of 2025, 39 countries qualify as capable autocracies—states with high institutional effectiveness, advanced infrastructure, and growing economies, but with Liberty scores below 35. This number now equals the count of states that are both free and capable. For the first time in the modern era, autocracy has achieved parity with democracy as a governance model for delivering material prosperity. The implications for the ideological competition between open and closed societies are profound. Chapter 7 maps the atlas of political freedom across regions, and Chapter 8 provides deep dives into the trajectories of specific regions, and country clusters.
Part III turns to the intersection of governance and sovereign credit. Chapter 9 asks why credit markets have been so poor at pricing the risk of democratic erosion, finding that bond spreads lag measurable governance decline by an average of 4.7 years. This delay is not a minor inefficiency; it represents a systematic blind spot in the way sovereign risk is assessed. Chapter 10 models the strategic interaction amongst three players—governments, creditors, and citizens—showing how each can be captured or sidelined during institutional decay. Chapter 11 identifies four distinct pathways from democratic decline to sovereign default, and Chapter 12 presents a prototype sovereign credit model that incorporates governance trajectory alongside conventional fiscal, and macroeconomic indicators.
Part IV applies the full analytical framework to the United States. Chapter 13 examines the "American exception"—the long-standing assumption, embedded in political science, and market pricing alike, that the United States is structurally immune to the patterns of democratic erosion observed elsewhere. We test this assumption against the data and find it wanting. The institutional safeguards that have historically distinguished the American system—an independent judiciary, a free press, federalism, a professional civil service—are eroding at rates consistent with early-stage democratic backsliding in other consolidated democracies. Chapter 14 measures the velocity of this decline and finds that the United States is moving away from its historical Liberty baseline faster than any other consolidated democracy in the dataset. Chapter 15 constructs probability cones for the American trajectory over the next decade under various policy scenarios, and Chapter 16 stress-tests these findings against the strongest available counter-arguments.
Part V subjects our own work to the scrutiny we apply to others. Chapter 17 identifies the findings that survived rigorous stress-testing: the tristable basin structure, the event horizon threshold, the great decoupling, and the credit market lag. Chapter 18, with equal candour, details what was refuted, or weakened upon closer examination—including certain overly precise probability estimates and some claims about the irreversibility of decline that the data ultimately could not support at high confidence. Chapter 19 presents the recalibrated framework: the model as it stands after subjecting every claim to adversarial review. Chapter 20 looks forward, identifying the research agenda that follows from these findings, and the policy interventions that the evidence suggests are most likely to matter.
This is a stress test, not a prediction. Our purpose is to map the terrain, identify the thresholds, and measure the distances—so that those who wish to defend democratic institutions can do so with clear eyes and better tools.
We wish to be explicit about what this book does and does not claim. It does not predict that any particular country, including the United States, will collapse into authoritarianism. Probabilistic models describe likelihoods, not certainties. A 3 percent recovery rate means that 3 countries in 100 do recover; it does not mean recovery is impossible for any given state. What the data does show, with considerable robustness, is that the terrain of political freedom is not flat. There are hills and valleys. There are thresholds beyond which the physics of recovery change dramatically. Ignoring this topography—assuming that democratic norms will reassert themselves automatically, that institutions will self-correct, that the arc of history bends reliably towards justice—is not optimism. It is a navigational error, made by travellers who have not consulted the map.
This book is that map. It is imperfect, as all maps are. Some of its contours will be revised as new data arrives and as the research community engages with its methods. But we believe it is substantially more accurate than the flat-earth model of politics that has governed most democratic complacency for the past three decades. We offer it in the spirit of informed vigilance—not as counsel of despair, but as equipment for the work of institutional defence.
This is a long book. It spans mathematics, political history, financial markets, and case-study analysis across five continents and more than two centuries. Not every reader will need every chapter, and not every chapter demands the same kind of attention. We have designed the book to support multiple reading paths, depending on your time, your interests, and the questions you bring to the material. Below are four suggested routes through the text, each self-contained, each offering a different depth of engagement with the research.
Before choosing a path, a word about the book's structure. The twenty chapters are organised into five parts, each of which builds on the preceding one but can also be read semi-independently. Part I establishes the theoretical framework—the tristable basin model, the event horizon concept, and the methodology. Part II presents the global evidence. Part III explores the financial implications. Part IV applies everything to the American case. Part V audits the entire project, separating what held up from what did not. The front matter you are now reading and the back matter that closes the volume provide orientation and reference.
Each chapter opens with a brief summary box that states, in three to four sentences, what the chapter argues, and what evidence it marshals. If you are reading selectively, these boxes will help you decide whether a given chapter merits a full reading, or a skim. Cross-references throughout the text point you forward and backward to related arguments; you need not follow them linearly.
Route: Executive Summary → Chapter 1 → Chapter 6 → Chapter 13 → Chapter 17
This path is for the reader who needs the core argument and its most important evidence in a single sitting. The Executive Summary provides the full architecture in compressed form. Chapter 1 introduces the tristable basin model—the book's central theoretical contribution. Chapter 6 presents the great decoupling, our most strategically significant empirical finding. Chapter 13 applies the framework to the United States, the case that will matter most to the majority of readers. Chapter 17 tells you what survived our own stress tests. At the end of this path, you will understand the book's central claims, its strongest evidence, and its honest assessment of its own limitations. You will not have the full evidentiary base or the financial analysis, but you will have enough to engage meaningfully with the argument, and to decide whether a deeper reading is warranted.
Route: Executive Summary → Part I (Chapters 1–4) → Part IV (Chapters 13–16) → Chapter 20
This path is designed for policymakers, civil servants, journalists, and anyone whose primary concern is the health of democratic institutions. Part I gives you the complete theoretical framework, including the methodology, so you can evaluate the claims on their merits. Part IV applies the framework to the United States in granular detail, covering the velocity of decline, scenario projections, and the counter-arguments that challenge our conclusions. Chapter 20 closes with the forward-looking agenda: what interventions the evidence suggests might matter, and what institutional defences have historically proven effective. This path skips the global evidence chapters and the financial analysis; if time permits, adding Chapter 6 (the great decoupling) and Chapter 8 (regional deep dives) will substantially enrich the reading.
Route: Executive Summary → Chapter 6 → Part III (Chapters 9–12) → Chapter 16
This path is for sovereign credit analysts, portfolio managers, and anyone who prices country risk for a living. Chapter 6 establishes the great decoupling—the divergence of state capability from political freedom—which has direct implications for how governance should be weighted in credit models. Part III is the financial heart of the book: the 4.7-year credit market lag, the three-player game between governments, creditors, and citizens, the four roads to default, and the prototype sovereign credit model that incorporates governance trajectory. Chapter 16 provides the stress tests, including the arguments against our framework that an investor would want to evaluate before adjusting any model. This path does not require the theoretical chapters in Part I, though readers who find the credit model compelling may wish to return to Chapters 1, and 2 for the underlying mechanics.
Route: Cover to cover
The book was written to be read in sequence, and the full reading experience is, we believe, the most rewarding. Each part builds on the preceding one. The theoretical framework in Part I gains force from the evidence in Part II. The financial analysis in Part III is more compelling once the global patterns have been established. The American case study in Part IV carries greater weight when read against the cross-national data. And the audit in Part V—where we turn our methods on ourselves—is most meaningful for the reader who has followed the full argument and can judge for themselves what should have survived and what should not have. This is also the reading that will leave you best equipped to engage with, critique, or extend the research. We encourage scholars, graduate students, and dedicated generalists to take this path. The chapters are designed to be read in one or two sittings per part, with natural resting points at the part boundaries.
The book uses mathematics, but it is not a mathematics book. We have kept formal notation to a minimum in the main text, relying instead on visual representations—scatter plots, trajectory maps, probability cones—to convey the quantitative findings. Readers comfortable with statistics will find that the Technical Appendix provides the full formal treatment, including regression specifications, robustness checks, and code references. Readers without a quantitative background should find the main text fully accessible; the core arguments can be followed without engaging the equations. Where a formula appears in the main text, it is because the elegance of the mathematics illuminates the argument in a way that words alone cannot. The constraint L + T + C = 100 is the most important equation in the book, and it requires no mathematics beyond arithmetic to understand.
A few terms recur throughout the book and are worth defining at the outset. Liberty (L) refers to the composite score measuring political rights, civil liberties, institutional constraints on executive power, and the rule of law. Tyranny (T) measures the degree of centralised authoritarian control, repression, and institutional capture. Chaos (C) captures state fragility, civil conflict, institutional collapse, and the absence of effective governance. These three dimensions are constrained to sum to 100 for each country at each observation point, creating a zero-sum compositional framework. A country that gains Liberty must lose either Tyranny or Chaos (or some of both); a country that falls into Chaos does so at the expense of whatever order—free or authoritarian—previously prevailed. The event horizon refers to the Liberty threshold (approximately L=52) below which recovery rates drop dramatically. The great decoupling denotes the post-2000 divergence between state capability and political freedom. These terms are defined more precisely in Chapters 1 and 4.
We have aimed throughout for clarity over jargon. Where we have borrowed terminology from physics, topology, or dynamical systems, we have done so because the analogies are precise, not merely decorative. The political landscape genuinely behaves like a dynamical system with multiple attractors. To describe it otherwise would be to sacrifice accuracy for false modesty.