Crises and Contestations of the Liberal Script

Macartan Humphreys

1 Overview

Next stages of SCRIPTs research may contain a theme group on the connection between crises\(^*\) and liberal orders (leads: Heike Klüver, Alexander Libman).

I am interested especially in the impact of crises on liberalism.

  • Rich topical topic
    • Multiple crises, simultaneous and extended: migration, corona, climate, conflicts
    • Theoretically complex

\(^*\) Crisis defined, following Brinks and Ibert (2020), defined by uncertainty, urgency and threat

1.1 Today: Groundclearing

Here some thoughts and some evidence:

  1. Some conceptual reflections on crisis and contestations

  2. Some data points from:

    • Corona studies with Heike Klüver, Johannes Gieseke, Felix Hartman, Ferdinand Geissler
    • SCRIPTs survey analysis with Jonah Foong, Lukas Antoine, Heiko Giebler, Ramus Ollroge

  3. Thoughts looking forward

    • Rich agenda
    • Theoretical work to do as very likely heterogeneous processes and contradictory logics
    • Will require conceptual disaggregation

1.2 Crisis (\(X\))

1.3 Crisis (\(Y\))

1.4 Crises

Crises: Multiple, simultaneous.

Themselves contested, often along lines along which liberalism is also contested:

  • Denial of climate change
  • Corona skeptics
  • Rejection that migration influx is a crisis
  • (There was even recently a rejection that there is a humanitarian crisis in Gaza)

In part, perhaps, because all elements of the definition: uncertainty, urgency and threat depend on perceptions

1.5 Is there a crisis of liberalism itself?

Possibly exaggerated, and possibly poorly posed (some cold water on this proposition later)

1.6 Challenge

Challenge: Can we get a definition less prone to contestation?

2 Theoretical thoughts

2.1 Types of responses we’re seeing

Diverse effects:

Crisis Government Society
Migration UK limitations on labor movement, undermining of asylum norms (Fear of) Importation of non liberal norms (“dissolving force”—Borrell) / Responses: Remigration master plans
Corona Obvious restrictions on movements and trade Rejections of government legitimacy, scientific authority
Climate Increased regulation Last generation interruptions of business-as-usual
Conflict Current and recent curtailment of speech in Germany, Israel Rising intolerance, antisemitism, islamophobia

2.2 Types of responses

  • These span many dimensions of liberalism:

    • Markets
    • Tolerance
    • Speech
    • Science
    • …

Challenge: Can there be a theory for them all?

  • It’s a big ask.

2.3 What logics connect crises to contestations?

Two transitory logics for observed accompanying contestations:

  • Exceptionalism: Desperate times call for desperate measures: no change in values, only an exceptional change in strategies
  • Unmasking: Crises expose otherwise invisible sources of control, otherwise invisible sources of disaffection

2.4 What logics connect crises \(\rightarrow\) contestations

Two permanent logics:

  • Carryover: Exceptions become normalized or have long lasting effect: delegitimizing of international institutions or scientific authority
  • Re-evaluation, revolution:
    • Crises create moments of weakness and open up opportunities to effect change
    • They create moment to re-evaluate structures and values; (a) hitherto underappreciated weaknesses exposed or (b) crises themselves be driven by failings of liberal ideas

2.5 What logics connect crises to contestations?

Not bad… But:

  • These seem like general logics connecting crisis to regime change

  • Is there anything particular about the vulnerabilities of liberalism.

    • Particular vulnerabilities? particular robustness? Trust dependence? Open fora? Trust reservoirs?

  • If not, then, via Przeworski logic, crises could as easily provide opportunities for liberalism (ambiguity in effects)

Challenge: Does the effect of crises on liberalism have to be studied in comparison with effects of crises on other systems?

3 Some data

3.1 Corona Surveys

A series of studies supported by SCRIPTS with Heike Klüver, Johannes Gieseke, Felix Hartman, Ferdinand Geissler

Headlines:

  1. People very willing to trade off rights for welfare, without implying an abandoning (supports transient logic)
  2. Cosmopolitan ideas quite strong, trust in science quite strong (supports ambiguity in effects)

3.2 Corona Surveys

Moment to display illiberal tendencies (except among AfD supporters!); but we see strategies changing, not values.

Attitudes to restrictions (Germany), Hartmann et al (2023)

3.3 Corona Surveys

But also an occasion to display liberal values (Humanitarian concern / cosmopolitanism, Trade linkages, Cooperation with states)

Broad (non-strategic) support for vaccine sharing (Germany), Geissler et al. (2022)

3.4 Corona Surveys

Broadly these patterns support resilience and unmasking logics, though in both cases sharply heterogeneous patterns

4 Some data II

PALs survey analysis (with Jonah Foong, Lukas Antoine, Heiko Giebler, Ramus Ollroge)

  • We study the level and correlates of contestations cross nationally.
  • Core strategy uses two conjoint experiment items capturing valuation and interpretation of different aspects of liberalisms

Headlines:

  • Puts in question general contestation of liberalism
  • Puts in question utility of overarching concept

5 Design

Key features of strategy:

  1. Very tangible, contextualized, notion of support: whether institutions that reflect principles render a society more deireable

  2. Disaggregation allows exploration of rival logics for rival components

  3. Disaggregation protects us from definitional vulnerability:

  4. Scope to use subjective definition rather than stipulated definition

5.1 Conjoint table

Attributes Illiberal treatment Liberal treatment
Constrained Government The government is free to make decisions that it thinks are good for society as a whole even if these go against the rights of minority groups. The government is not free to make decisions that it thinks are good for society as a whole if these go against the rights of minority groups.
Democratic Government Most major policy decisions are controlled by government experts and not by elected representatives. Most major policy decisions are controlled by democratically elected representatives not by government experts.
Free Markets The government tries to ensure that the economy is strong by actively controlling major industries. The government tries to ensure that the economy is strong by putting few controls on major industries.
Low Taxes Taxes are relatively high so that the government can ensure greater equality in society. Taxes are kept low so that individuals, and not the government, get to decide how best to use their money.
Tolerance Homosexual relationships are penalized. Homosexual couples have the same rights as heterosexual couples.
Openness The government makes sure that immigration is kept to a minimum to protect the nation's culture. The government encourages talented foreigners to come to work as this enriches the nation's culture.
Income The income per capita is around (3,500 / 23,000 / 43,000 / 63,000) USD. For comparison: in (COUNTRY), the income per capita is (NATIONAL GDP PER CAPITA IN USD) USD per year.

5.2 Results I

  • Very broad global support for liberalism

  • However

    • support varies a lot by subcomponent
    • support varies a lot by region
    • willingness to trade-off liberalism and welfarism (Sen)

5.3 Index

5.4 Subcomponents

5.5 Tradeoffs

Again, willingness to trade off liberal values against welfare

Figure 1: ?(caption)

5.6 Heterogeneity 1

Nature of contestation varies by dimension:

  • Broad support for: Democratic government, Low Taxes
  • Ambivalence for: Constrained government (minority rights), free markets
  • Sharp heterogeneity for: Tolerance

5.7 Correlates of contestations I

Figure 2: Effects of conjoint features | individual and national level moderators.

5.8 Correlates of contestations I

Liberal ideas…

  • Most stongly supported: in wealthier countries

  • Contested especially for: politically excluded in liberal societies

5.9 Return to Crises

Does exposure to crises—climate, corona, migration—“explain” ambivalence to liberalism?

Headline:

  • yes, a bit: but the direction!
  • and it depends on dimensions

5.10 Migration

  • Bivariate relationship does not support the claim migration \(\rightarrow\) contestation
  • Obvious selection concern here: need for intertemporal data and identification strategy

5.11 Climate

  • No clear bivariate relationship; and negative with core controls
  • Should we expect it given global nature of crisis?

5.12 Covid

  • Bivariate relationship does not support the claim: migration associated with more support (also with controls)
  • Obvious selection concern here and need for intertemporal data and identification strategy

5.13 Full analysis: patterns relatively strong

5.14 Full analysis: patterns relatively strong

  • But, again variation across dimensions

  • e.g. Bigger Covid shock is associated with less support for free markets (more support for intervention) but more openness

5.15 Implications

  • Broadly: little evidence for a global crisis of liberalism

  • Empirical grounds for skepticism regarding utility of overarching concept

  • Triple heterogeneity:

    • Heterogeneity across crisis
    • Heterogeneity across measures
    • Heterogeneity in the direction of correlations with crises

5.16 Conclusions

This will be a rich and exciting agenda but a lot more needed to answer the basic question: “do exogeneous crises induce a crisis of liberalism?”

Think progress in this area will require:

  • deepening of theoretical logics
  • disaggregating liberalism and focussing on more cleanly specified concepts
  • repeated measurement (!)

Almost certainly: strong qualitative orientation of research, likely in early stages

5.17 References

5.18 What logics connect crises \(\rightarrow\) contestations?

Two transitory logics for observed accompanying contestations:

5.19 Heterogeneity 1

  • which dimensions are threatened varies across regions
Geissler, Ferdinand, Felix Hartmann, Macartan Humphreys, Heike Klüver, and Johannes Giesecke. 2022. PLOS ONE 17 (12): e0278337.