The Crisis Before Trump
From the Global Financial Crisis to the 2016 breakthrough, the pandemic and the 2024 return: how disaffection became political coordination — and what the eight functions of CAMS can reveal
The crisis preceded the candidate
Donald Trump did not create the fractures that helped carry him to the presidency. He recognised disaffection already present in American society and converted it into a durable form of political coordination. The 2008 financial crash was a crucial accelerant, but not the sole source. Deindustrialisation, uneven geography of prosperity, mistrust of institutions, immigration anxieties, changing cultural hierarchies and partisan media developed on overlapping timelines.
The Complex Adaptive Model State (CAMS) offers a disciplined way to ask a different question from conventional political commentary: which functions were under strain, which recovered, and what happened to the relationships among them? This is a structural lens, not a claim that an election can be derived from eight numbers.
An electoral victory may be evidence of successful political mobilisation even when it is not evidence of recovered social cohesion.
The long loosening of the American bargain
For decades, the promise of the post-war American settlement was that expanding production and national prestige would be broadly shared. The settlement was never universal: race, region, gender and bargaining power sharply shaped participation. By the 2000s, the distance between finance-led asset appreciation and many working communities had become a central political vulnerability.
CAMS would look for a divergence between Stewards (allocation and control of resources), Hands (labour and material production), Flow (exchange and circulation) and Lore (the shared story of why arrangements are legitimate). The annual CAMNATIONS5 scores span 1995–2025 and let us inspect this hypothesis before the GFC. They are retrospective model scores, not contemporaneous measurements; the separate quarterly series begins in 2015 and must not be spliced onto them as though it had identical calibration.
Figure 1 · The long arc: 1995–2025
Annual CAMNATIONS5 panel. Separate scoring series from the quarterly plots.
Figure 2 · Thirty years of functional divergence
Annual node-value gap. Not a measure of household wealth inequality.
The rescue and its unequal clock
The Great Recession ran from December 2007 to June 2009. US unemployment eventually reached 10%, home prices fell dramatically, and household net worth contracted by trillions. The financial system was stabilised sooner than many households regained secure livelihoods. US payroll employment did not exceed its January 2008 peak until May 2014 [1–3].
The CAMS historical comparison reports a Stewards recovery within two years but a Hands recovery only after six, implying a four-year asymmetry. The 1995–2025 annual test file independently reproduces these quoted scores (but is itself a retrospective scoring dataset, not an independent real-world observation). What is independently documented is the drawn-out labour recovery, even after broad output and financial indicators turned upwards.
Figure 3 · The GFC: financial repair, deferred labour recovery
Annual node values show the Stewards collapse in 2008 and the lagging Hands trough in 2009.
Figure 4 · Unequal restoration after 2008
Selected annual crisis snapshots; horizontal spacing between labelled years is not proportional to elapsed time.
The political consequence cannot be simply read off a balance sheet. A rescue may restore system function while leaving an enduring perception that different groups face different rules. This is the legitimacy debt hypothesis: short-term coordination success may coexist with a later erosion of consent. CAMS should test it, not presume it.
Figure 5 · Lore legitimacy and governing capacity
Annual Lore coherence versus Helm coherence, not composite V. Movement is descriptive, not a causal demonstration.
How institutional disappointment became a political force
Disaffection is not itself a political movement. Millions can distrust institutions yet remain isolated, resigned, politically divided or loyal to existing parties. To become a consequential electoral force, disappointment must be recognised, narrated, connected and organised. This is the missing transition between the GFC charts and Trump's victories. CAMS supplies a way of locating the pressures and testing the proposed mechanisms without assuming that the scores mechanically caused voters to choose a candidate.
Stage one: the grievance reservoir was unevenly distributed
Before 2008, economic changes were already experienced differently across place, class, educational attainment and race. The financial crash added a vivid comparison: institutions capable of rapidly protecting critical financial operations seemed less capable of restoring security to households. CAMS records the 2008 Stewards trough and the later Hands trough, followed by an unequal recovery. That sequence is consistent with a legitimacy debt: restoration of system throughput without an equally persuasive restoration of fairness. It does not establish that every later grievance came from the GFC; cultural, racial, immigration-related and partisan antagonisms have their own histories.
Stage two: Trump supplied an interpretation rather than a new grievance
The innovation of Trump's 2016 campaign was political compression. Trade losses, immigration anxiety, distrust of Washington, anti-establishment anger, cultural displacement and resentment of expert authority could be made legible through a small set of repeated propositions: the country had been betrayed; established leaders served someone else; a forceful outsider could restore control. For some voters the appeal centred on economic insecurity; for others it involved cultural status, immigration or partisan allegiance. Treating supporters as motivated by one grievance would obscure how the coalition worked.
CAMS separates Lore (shared narratives and legitimacy) from Helm (direction and executive coordination). A campaign can gain coherence within its own coalition by offering a compact counter-narrative even while society-wide Lore becomes more contested. Trump's rhetoric was not simply a symptom of weakening legitimacy; it was an intervention into the contest over who speaks for the nation.
Stage three: a movement solved coordination problems
Anger becomes politically useful only when supporters expect other supporters to show up. Repetition creates common knowledge; rallies make numbers visible; partisan outlets reinforce a shared vocabulary; party structures translate attention into nominations, registrations, donations and votes. The Electoral College and geographically concentrated support then translate a coalition into institutional power in ways that are not equivalent to winning a national popular-vote majority. This explains why 2016 is a successful coalition-coordination event without being evidence that Trump's account was shared by most Americans.
Stage four: governing power and social repair can diverge
The 2016 victory transferred control of Helm, but it did not automatically settle disputes over the legitimacy of courts, the media, expertise, immigration or democratic rules. In fact, a movement can maintain mobilisation by continuing to depict itself as besieged even while its leader occupies the presidency. The crucial CAMS distinction is therefore between within-coalition synchronisation and cross-coalition coordination. They need not move together. Polarised camps can each be highly coherent internally while the bridges between them weaken.
Why 2024 matters differently from 2016
The 2024 victory is more than repetition. Trump returned after a full governing term, defeat in 2020, the January 6 attack on the Capitol, legal conflicts, a pandemic and an inflation shock. His coalition drew support from some voter groups beyond its 2016 profile; economic dissatisfaction and immigration remained salient. The relevant explanatory test shifts from How did an outsider first aggregate dissatisfaction? to How did an established political actor preserve and recombine a mobilisation network across defeat, changing conditions and repeated electoral cycles? Trump had become not merely a spokesman for grievances but an enduring organisational focal point. This does not mean that all voters endorsed every statement or action by him.
What would disprove or refine this explanation?
A persuasive test would require more than placing election markers over annual scores. Compare the timing of CAMS changes with independently observed trust, earnings, unemployment, local exposure to industrial change, voter turnout, switching patterns and media networks. Examine counties and demographic groups where material distress was high but Trump support remained low, and places where his support was high despite economic resilience. Test alternative explanations, including partisan sorting, racial and cultural attitudes, campaign organisation, immigration, incumbent dissatisfaction and electoral institutions. CAMS presently represents the national functional context; it does not model individual voting choices or directly measure coalition-level ties.
The decisive distinction: Trump's success may reveal successful political coordination under conditions of contested institutional legitimacy, rather than national recovery from the conditions that created the opening.
Trump turned disaffection into an organising language
In 2016, Trump offered a political grammar that fused complaints about trade, immigration, elites, national decline and exclusion from decision-making. He presented himself as the actor who could name the grievance, identify its supposed authors and promise restoration. His opponents often heard incoherence; his supporters frequently heard recognition.
The dissatisfaction was measurable before voting day. Pew found 75% of Trump’s Republican primary supporters believed life for people like themselves was worse than 50 years earlier; half said they were angry at government. These are perceptions and political identities, not proof that each perceived cause was objectively correct [4].
His victory was institutionally consequential but electorally specific: 304 electoral votes after faithless electors, while Hillary Clinton won more popular votes. Trump became an unusually effective coordinator of a distinct coalition, not a consensual national voice [8].
Figure 6 · Election victories, not a straight-line trend
Official National Archives Electoral College totals; 2016 includes faithless electors.
Figure 7 · The executive and the story
Helm measures coordination authority; Lore measures shared narrative coherence through the model’s composite score.
CAMS names the likely interaction: weakening confidence in Lore makes an alternative narrative attractive; Helm promises decisive command; Shield supplies threat and protection language; Flow and Hands provide the material vocabulary. This is an explanatory interpretation, not a model-estimated causal chain.
The state gains an organiser; the society does not regain consensus
The key distinction is between a leader’s ability to mobilise a constituency and a system’s ability to coordinate across constituencies. Trump’s first presidency did not end the conflict over what counted as national identity, expertise, fair distribution or legitimate authority. Nor were all those conflicts caused by him. In CAMS terms, political Helm can be energised even while Lore remains contested.
Figure 8 · Eight functions, one society
Quarterly CAMS node values, 2015Q1–2026Q3. Vertical guides mark elections and the pandemic.
Figure 9 · Narrative coherence
Lore coherence C, distinct from the Lore composite V score.
A second shock, a faster rebound, and a fragile repair
COVID sharply interrupted labour and circulation. In the CAMS quarterly scores, 2020Q2 records Hands at −1.0, Stewards at 8.0, and a pronounced decline in available implementable energy (capacity minus stress). Emergency aid reached households and businesses as well as markets. Both Hands and Stewards first regained their 2019Q4 node-value baselines by 2021Q2.
Figure 10 · Finance and labour
Stewards and Hands node values. Both crossed their 2019Q4 baselines in 2021Q2, but that crossing did not persist.
Figure 11 · Stress migrates through the system
Model stress scores for Hands, Stewards, Lore and Helm.
Figure 12 · Implementable energy
K − S reveals strain concealed by a positive composite node value.
Figure 13 · Recovery is not permanence
Quarterly deviations from the 2019Q4 level; a return above zero need not endure.
That is not equivalent to permanent repair. The two functions subsequently fell below their previous baselines. Their equal time to first recovery concealed unequal cumulative shortfalls: over 2020Q2–2021Q1, Hands accumulated 23.9 node-point quarters below its 2019Q4 baseline, versus 15.8 for Stewards, about 51% more. These are model-score integrals, not dollars or household welfare estimates.
The return: grievance became a repeatable political capacity
Trump’s 2024 victory, 312 electoral votes to Kamala Harris’s 226, cannot be explained solely as an echo of 2016. Inflation, affordability and migration were salient, and his winning coalition was more racially and ethnically diverse than in his earlier campaigns. Pew found 93% of Trump supporters regarded the economy as very important to their vote and documented important shifts in turnout and coalition composition [6–7, 9].
CAMS asks how such a heterogeneous collection of anxieties becomes a common political force. An effective movement turns separate local disappointments into a shared story, then links that story to organised participation, leadership and institutional strategy. The same process may deepen a cleavage between those who see restoration and those who see threat.
Figure 14 · A functional gap is not a wealth gap
Difference between two CAMS function scores, not a measure of income inequality.
A model warning, not a verdict of national collapse
The strongest new feature in the quarterly panel is broad, synchronised movement around 2025Q1. All eight node values fall from 2024Q4 to 2025Q1; the sum of absolute node changes is 32.6 points, the largest non-pandemic quarter-to-quarter contraction in this file. By 2025Q4 the Stewards–Hands node-value gap reaches 10.4 points, compared with 6.3 in 2021Q2. In 2026Q3 it is 8.4 points.
Figure 15 · Synchronised movement
Total quarter-to-quarter absolute movement across the eight node-value scores; not an independent crisis index.
Figure 16 · The eight-node landscape
Colour shows higher or lower CAMS composite node values by function and quarter.
These scores should not be presented as proof that Americans’ material living conditions collapsed in 2025. Nor is the node difference a wealth-inequality series. The scores may be capturing political strain, social legitimacy and functional coordination; competing external evidence must decide how far that interpretation travels. Gallup in July 2026 measured average confidence in 14 core institutions at just 27%, while noting substantial partisan differences [10].
What CAMS adds — and what it cannot yet claim
CAMS scores eight coordination functions — Helm, Shield, Lore, Stewards, Craft, Hands, Archive and Flow — on coherence, capacity, stress and abstraction. The composite node value is V = C + K + 0.5A − S. The framework also examines how functions couple, but the pairwise coupling values quoted in the original comparison are not present in the quarterly CSVs and are therefore not graphed here.
Three distinctions matter. (1) A quarterly node score is a modelled assessment, not a directly measured economic quantity. (2) Correlation between model strain and political outcomes is not proof of causation. (3) A change in scoring across multiple nodes may reflect genuine system-wide disturbance, a shared source narrative or model/scorer dependence; blind re-scoring and autocorrelation-aware null tests are necessary.
The most valuable next experiment is a blind re-scoring of the now-available 1995–2025 annual series and the overlapping quarterly panel, with explicit cross-series calibration, spanning at least 2005–2026, scoring both crisis eras under identical rules and independently comparing the results to elections, jobs, real income, household balance sheets and institutional trust. Pre-register the claimed lagged relationship between asymmetric recovery and Lore deterioration, and test it against rival explanations.
Sources and limitations
The 16 graphics use four CAMS datasets: two annual CAMNATIONS5 files (1995–2025) and two quarterly files (2015–2026Q3). Annual and quarterly scores are shown separately: differences in their levels must not be silently interpreted as economic changes. The election chart uses certified results. The 2008–2014 section draws directly on the historical annual scores alongside external historical references. Last updated 13 October 2026.
- Federal Reserve History — The Great Recession
- Federal Reserve — 2017 Monetary Policy Report, recovery
- BLS — Payroll employment finally surpasses January 2008 (2014)
- Pew — 2016 disaffection and perceptions of decline
- Pew — A divided and pessimistic electorate (2016)
- Pew — 2024 election issues
- Pew — 2024 voter coalition retrospective
- US National Archives — 2016 Electoral College
- US National Archives — 2024 Electoral College
- Gallup — Confidence in institutions (July 2026)
Publication note: editorial interpretation, not institutional endorsement or a validated causal model.
Data. Every chart on this page is drawn from these files in the repository:
USA_CAMS_m5n_quarterly_2015-2026Q3_block1.csv— quarterly ensemble means, 2015Q1–2026Q3USA_CAMS_m5n_quarterly_2015-2026Q3_block2.csv— quarterly scorer envelopeUSA_1995_2025_CAMNATIONS5_test1.csv— annual CAMNATIONS5 test run, 1995–2025USA_1995_2025_CAMNATIONS5_raw_envelope1.csv— annual raw envelope
The annual and quarterly series are separate scoring runs with different calibration; they are never spliced. Charts re-rendered for this site from the same CSVs as the original standalone edition.