CoChat

Modeling Social Pressures Toward Political Instability

Open access
Turchin, Peter · 2013 · Cliodynamics: The Journal of Quantitative History and Cultural Evolution

The DOI matches its Crossref record; 30 of the 36 references it cites were found in a registry or an index; 3 of the 32 sources its claims cite were checked for support.

Abstract

Structural-demographic theory models long-term social pressures that lead to revolutions, civil wars and outbreaks of instability.

Overview

CoChat Summary

TL;DR

Peter Turchin develops a structural-demographic model that quantifies political instability as the product of pressures from the mass population, elites, and the state, then applies it to the antebellum United States and contemporary America. In both case studies, the model is presented as a leading indicator of rising instability: antebellum PSI rises after about 1840 and “literally exploded” during the 1850s, while the contemporary PSI begins increasing after 1980 and grows rapidly after 2000.

Key findings

  • The paper defines the Political Stress Indicator as the product of three components: Mass Mobilization Potential (MMP), Elite Mobilization Potential (EMP), and State Fiscal Distress (SFD), with MMP based on relative wages, urbanization, and youth bulges; EMP based on relative elite income and elite numbers; and SFD based on debt/GDP and distrust in government.
  • In the antebellum model, MMP is “essentially flat until 1820” and EMP declines until the 1830s, then rises sharply in the 1840s and especially 1850s; the combined PSI stays low until about 1840, then increases and “literally exploded during the 1850s.”
  • The wage regression results improve as more structural factors are added: GDP per capita alone yields \(R^2 = 0.73\), GDP plus labor demand/supply yields \(R^2 = 0.93\), and adding the minimum-wage proxy yields \(R^2 = 0.98\). The best ARIMA model includes GDP per capita, labor demand/supply, and real minimum wage, with estimates 0.60, 1.65, and 0.45, respectively.
  • For the contemporary US, the best model is said to fit the post-1970 stagnation regime, and the estimated PSI begins increasing after 1980 and has grown very rapidly after 2000; the paper also reports that union coverage fell from over 25% between 1945 and 1960 to 12% currently, and private-sector union coverage fell from 35% in the 1950s to 7.6% in 2008.

Methods

  • The paper is a theoretical and empirical modeling study using structural-demographic theory; it builds a set of differential-equation-based submodels for population, wages, elite dynamics, and state fiscal distress, then combines them into PSI.
  • For the antebellum case, it models the four Northeastern states (Massachusetts, Connecticut, New York, and New Jersey), using historical data on wages, GDP, population, urbanization, and elite-related quantities; it compares predicted trajectories with observed instability indicators and civil-war-related violence.
  • For contemporary America, it uses data from sources including MeasuringWorth, the Bureau of Labor Statistics, the US Census Bureau, the Historical Statistics of the United States, Pew Research Center, and the Treasury, and estimates the wage model by regression with autocorrelated errors (ARIMA), using exponential smoothing of predictor variables with bandwidth \(h = 4\) years.

Limitations

  • The paper explicitly notes that the antebellum model simplifies reality by ignoring immigration from Europe and emigration to the West, and by assuming constant per-capita population growth; it says these simplifications “largely cancel each other out” before 1880.
  • It also states that data for distrust in state institutions are unavailable before 1958, so the contemporary PSI can only be estimated from 1958 onward.
  • Evident from the text: several formulas are not decoded in the provided full text, so the exact mathematical specification of some model components cannot be fully reconstructed from this excerpt alone.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.21237/c7clio4221333

Cites

30 of 30 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 The main logical components of the structural-demographic theory.
Figure 1 The main logical components of the structural-demographic theory.From the PDF
Figure 2 Population dynamics (a) generated by the model and (b) observed in the four Northeastern states (MA, CT, NY, and NJ). Data sources: calculations by the author and the HSUS (Carter et al. 2004). Model parameters assumed in calculations: r rur = 0.025 y -1 , r urb = 0.015 y -1 , K = 3.5 million, θ = 5, β = 0.5, a = 1, ρ = 0.03 y -1 , γ = 0.01 y -1 .
Figure 2 Population dynamics (a) generated by the model and (b) observed in the four Northeastern states (MA, CT, NY, and NJ). Data sources: calculations by the author and the HSUS (Carter et al. 2004). Model parameters assumed in calculations: r rur = 0.025 y -1 , r urb = 0.015 y -1 , K = 3.5 million, θ = 5, β = 0.5, a = 1, ρ = 0.03 y -1 , γ = 0.01 y -1 .From the PDF
Figure 3 Elite dynamics in the Antebellum model in relation to relative wages ( w ): elite numbers relative to the urban population ( e ) and average elite incomes ( ε ). Parameter values: same as given in the caption of Figure 2, plus μ 0 = 0.002 y -1 , w 0 = 1.
Figure 3 Elite dynamics in the Antebellum model in relation to relative wages ( w ): elite numbers relative to the urban population ( e ) and average elite incomes ( ε ). Parameter values: same as given in the caption of Figure 2, plus μ 0 = 0.002 y -1 , w 0 = 1.From the PDF
Figure 4 Proportion of the cohort aged 20-29 years among the total population of American white males. Data Source: Calculations by the author based on data in (Carter et al. 2004: Table Aa287-364).
Figure 4 Proportion of the cohort aged 20-29 years among the total population of American white males. Data Source: Calculations by the author based on data in (Carter et al. 2004: Table Aa287-364).From the PDF
Figure 5 Dynamics of the two components of Political Stress Indicator, Mass Mobilization Potential (MMP) and Elite Mobilization Potential (EMP), predicted by the Antebellum model. Both indicators were scaled to minimum = 1.
Figure 5 Dynamics of the two components of Political Stress Indicator, Mass Mobilization Potential (MMP) and Elite Mobilization Potential (EMP), predicted by the Antebellum model. Both indicators were scaled to minimum = 1.From the PDF
Figure 6 Predicted Political Stress Indicator compared with the observed dynamics of political instability: number of people killed in political instability events per million people and a qualitative index of sectional conflict, which includes Bleeding Kansas and John Brown's Harpers Ferry raid (1855-1860) and the American Civil War (18611865). Note that the two empirical indicators of instability are not shown to scale (the scale of casualties resulting from the Civil War exceeded that of the largest riot by two orders of magnitude).
Figure 6 Predicted Political Stress Indicator compared with the observed dynamics of political instability: number of people killed in political instability events per million people and a qualitative index of sectional conflict, which includes Bleeding Kansas and John Brown's Harpers Ferry raid (1855-1860) and the American Civil War (18611865). Note that the two empirical indicators of instability are not shown to scale (the scale of casualties resulting from the Civil War exceeded that of the largest riot by two orders of magnitude).From the PDF
Figure 7 Trends in labor demand and supply (the United States, 1927-2012). Both time series have been scaled so that 1927 = 100. Source: calculations by the author based on data from MeasuringWorth, HSUS, and BLS.
Figure 7 Trends in labor demand and supply (the United States, 1927-2012). Both time series have been scaled so that 1927 = 100. Source: calculations by the author based on data from MeasuringWorth, HSUS, and BLS.From the PDF
Figure 8 Labor union dynamics. The curve indicates the proportion of workers covered by unions, 1930-2010. The bar chart traces the proportion of union election campaigns in which pro-union workers were illegally fired, 1951-2007. Data sources: Union coverage (Mayer 2004), supplemented by BLS data; illegal firing (Schmitt and Zipperer 2009).
Figure 8 Labor union dynamics. The curve indicates the proportion of workers covered by unions, 1930-2010. The bar chart traces the proportion of union election campaigns in which pro-union workers were illegally fired, 1951-2007. Data sources: Union coverage (Mayer 2004), supplemented by BLS data; illegal firing (Schmitt and Zipperer 2009).From the PDF
Figure 9 Dynamics of the real minimum wage, 1938-2012. Data source: US Department of Labor.
Figure 9 Dynamics of the real minimum wage, 1938-2012. Data source: US Department of Labor.From the PDF
Figure 10 Results of fitting various forms of the regression model (2) to the data: (a) including only the effect of GDP per capita (no time lag), (b) including both GDP and labor demand/supply ratio (no time lag), (c) including all three factors (no time lag), and (d) the full model (time lag = 5 years).
Figure 10 Results of fitting various forms of the regression model (2) to the data: (a) including only the effect of GDP per capita (no time lag), (b) including both GDP and labor demand/supply ratio (no time lag), (c) including all three factors (no time lag), and (d) the full model (time lag = 5 years).From the PDF
Figure 11 Dynamics of the elite submodel of the general structural-demographic model, 1927-2012: relative wage, relative elite numbers, and relative elite income. All relative variables were scaled to mean = 1 for the Pre-War period (1927-40).
Figure 11 Dynamics of the elite submodel of the general structural-demographic model, 1927-2012: relative wage, relative elite numbers, and relative elite income. All relative variables were scaled to mean = 1 for the Pre-War period (1927-40).From the PDF
Figure 12 The cost of winning an election to the House, 1986-2012 (in thousands of inflation-adjusted 2012 dollars) and the total amount (in millions of 2012 dollars) spent by major party candidates. Data source: The Campaign Finance Institute.
Figure 12 The cost of winning an election to the House, 1986-2012 (in thousands of inflation-adjusted 2012 dollars) and the total amount (in millions of 2012 dollars) spent by major party candidates. Data source: The Campaign Finance Institute.From the PDF
Figure 13 National debt scaled by GDP and proportion of people who distrust government institutions. Debt/GDP data are taken from the US Department of the Treasury. Distrust in government data are taken from the Pew Research Center (proportion responding 'some of the time' or 'never' to the question, 'How much of the time do you trust the government in Washington?').
Figure 13 National debt scaled by GDP and proportion of people who distrust government institutions. Debt/GDP data are taken from the US Department of the Treasury. Distrust in government data are taken from the Pew Research Center (proportion responding 'some of the time' or 'never' to the question, 'How much of the time do you trust the government in Washington?').From the PDF
Figure 14 Estimated Political Stress Indicator, 1958-2012.
Figure 14 Estimated Political Stress Indicator, 1958-2012.From the PDF

Verification

Claims the paper makes

3 Valid · 19 Not valid · 30 Not checked · 121 Own results and interpretations

65 more claims

Create a free account to see all 173 claims and to verify the 1 reference still open.

See all claims — free account
  • Probably not a substantial claim. 58
  • It may state two separate claims as one. 60
  • The quote only partly states this claim. 1

Create a free account to see these claims.

Citations

Cites

Research works printed but not linked: 2

  1. Blanchard, O. 1997. Macroeconomics. Prentice Hall, Upper Saddle River, NJ.
  2. Samuelson, P. A. and W. D. Nordhaus. 1998. Economics. 16th edition. McGraw-Hill, New York.

Cited by

See all 42

Metadata

Authors
Turchin, P.
Affiliations
University of Connecticut (US)
Corresponding author
Turchin, Peter
Venue
Cliodynamics: The Journal of Quantitative History and Cultural Evolution
Published
2013-12-31
Volume
4
Issue
2
ISSN
2373-7530 (ISSN-L)
OpenAlex
W276290970
Cited by
42
References
36
Citations by year
2026: 2 · 2025: 2 · 2024: 5 · 2023: 5 · 2022: 4 · 2021: 4 · 2020: 5 · 2019: 4 · 2018: 4 · 2017: 4 · 2016: 2 · 2014: 1
FWCI
3.10
Type
journal article
Language
en
Open access
Licence
cc-by
Locations
  • Cliodynamics The Journal of Quantitative History and Cultural Evolution (publishedVersion, cc-by)
  • eScholarship (California Digital Library) (submittedVersion, cc-by)
Funders
  • National Science Foundation (REF RES-060-25-0085)
  • Economic and Social Research Council (RES‐060‐25‐0085, REFRES-060-25-0085)
Topics
  • Social Sciences › Sociology and Political Science › Political Conflict and Governance
  • Physical Sciences › Physics and Astronomy › Statistical and Nonlinear Physics › Opinion Dynamics and Social Influence
  • Social Sciences › Political Science and International Relations › Political Theory and Democracy
Keywords
social pressure, political instability, political violence, civil war

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (qwen3.8-omni-flash).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.