we-solve (beta)

Econ-539 · economics · Economics

Economic origins and consequences of populism

LLM status Open unverified

Statement

Q4 | Political economy | Editorial research problem family

Definitions and assumptions

Fix regions $r$, a horizon $h$, a preregistered definition of a populist party, and vote share $P_{rh}\in[0,1]$. Let the vector $a\in\mathcal A$ describe feasible trade, technology, migration, redistribution and identity-salience interventions. Let $P_{rh}(a)$ and $Y_{rh}(a)$ be potential vote shares and economic outcomes, allowing regional spillovers and strategic party entry. Fix a baseline $a^0$ and comparison $a^1$. For $K$ intervention components and $S\subseteq\{1,\ldots,K\}$, let $a^S$ adopt the comparison component for indices in $S$ and the baseline component otherwise; require these mixed regimes to be feasible.

Formal research question

Identify or bound $v(S)=\mathbb E[P_{rh}(a^S)]$ and the subsequent economic contrast $\mathbb E[Y_{rh}(a^1)-Y_{rh}(a^0)]$. If a component attribution is required, specify the convention, for example the Shapley attribution \[ \phi_k=\sum_{S\subseteq\{1,\ldots,K\}\setminus\{k\}} \frac{|S|!(K-|S|-1)!}{K!}\,[v(S\cup\{k\})-v(S)]. \] Under this definition $\sum_k\phi_k=v(\{1,\ldots,K\})-v(\varnothing)$. Determine which comparisons are identified under the declared model class, assignment mechanism and transport assumptions.

Interpretation and scope

The decomposition is an editorial accounting convention for specified interventions, not a unique fraction of history caused by each mechanism. It accommodates interaction effects by averaging their allocation across orderings. If mixed interventions are infeasible, the Shapley target must be abandoned or replaced. Trade shocks do not manipulate culture independently, and a populist vote need not imply a particular economic policy. Electoral and policy consequences therefore remain distinct estimands.

Source audit and status

Rodrik (2018) is verified, while the atlas link is the 2017 working paper. Colantone--Stanig (2018) is ambiguous between two verified papers; both are supplied. The undated phrase "Autor et al. political China-shock research" is resolved to a selected 2020 article, with its 2016 working-paper antecedent distinguished. These sources support mechanisms and specific evidence, not the editorial attribution formula or a universal causal decomposition.

References

    }},leftmargin=3em]
  1. Dani Rodrik (2018). Populism and the Economics of Globalization. Journal of International Business Policy 1, 12--33.

    Locator and source role: Abstract and globalization mechanisms. The linked NBER Working Paper 23559 is dated 2017, not 2018.

    https://www.hks.harvard.edu/publications/populism-and-economics-globalization-0

  2. Italo Colantone and Piero Stanig (2018). Global Competition and Brexit. American Political Science Review 112(2), 201--218.

    Locator and source role: Abstract and trade-exposure design; one of two plausible works matching the original author-year citation.

    https://www.cambridge.org/core/journals/american-political-science-review/article/abs/global-competition-and-brexit/C843990101DB9232B654E77130F88398

  3. Italo Colantone and Piero Stanig (2018). The Trade Origins of Economic Nationalism: Import Competition and Voting Behavior in Western Europe. American Journal of Political Science 62(4), 936--953.

    Locator and source role: Abstract; second plausible author-year match, retained to expose the original ambiguity.

    https://onlinelibrary.wiley.com/doi/10.1111/ajps.12358

  4. David Autor, David Dorn, Gordon Hanson, and Kaveh Majlesi (2020). Importing Political Polarization? The Electoral Consequences of Rising Trade Exposure. American Economic Review 110(10), 3139--3183.

    Locator and source role: Abstract and local import-exposure identification; resolves the atlas phrase political China-shock research to a selected identifiable paper.

    https://www.aeaweb.org/articles?id=10.1257/aer.20170011

Common conventions: Source mathematical conventions

Unless an entry states otherwise, agent and firm sets are finite. State and action spaces are measurable, strategies and outcome maps are measurable, probability laws are specified, and expectations exist for the targets under discussion. Infinite-horizon discount factors lie in $(0,1)$ and discounted objectives are finite. Norms, tolerances and complexity input sizes must be specified for computational comparisons. Constants or primitives that appear locally are inputs, not empirical facts asserted by this catalogue.

Identification. A statistical model $m\in\mathcal M$ implies an observable law $P_m$ and target $\theta(m)$. At law $P$, its identified set is \[ \Theta(P;\mathcal M)=\{\theta(m):m\in\mathcal M,\ P_m=P\}. \] Point identification holds when this set is a singleton, or equivalently when $P_{m_1}=P_{m_2}$ implies $\theta(m_1)=\theta(m_2)$ for all compatible models. Sharp bounds mean bounds attained, or approached when the boundary is not attained, by compatible models. Writing this set is a definition, not a solution: the substantive task is to establish informative restrictions and derive or estimate the set. An unrestricted-confounding model may give only trivial bounds.

Inference. Identification, estimability and valid uncertainty quantification are separate questions. Fix $\alpha\in(0,1)$. A confidence procedure $C_n$ over a declared law class $\mathcal P$ has asymptotic uniform coverage at level $1-\alpha$ when \[ \liminf_{n\to\infty}\inf_{P\in\mathcal P}\Pr_P\{\theta(P)\in C_n\}\geq1-\alpha. \] For set-identified targets the coverage event must be defined separately, for example coverage of the full identified set. If an entry uses identification-indexed models instead of uniquely indexed laws, the same coverage convention is applied over those models. Pointwise limits do not establish uniform validity, and asymptotic validity does not guarantee finite-sample precision. Sample sizes, dependence structures and weak-identification sequences are part of each application.

Counterfactuals. $Y_i(a)$ is a potential outcome under a specified intervention, including its timing, implementation and versions. It is not $Y_i$ conditional on voluntarily choosing $a$. A full intervention vector permits interference; shorthand scalar treatment notation does not silently impose no interference. Assignment, selection, attrition, anticipation and measurement must be specified. A claim about an instrument's local effect is not automatically a population policy effect.

Economic equilibrium. A counterfactual model includes best responses, constraints, feasibility, market clearing, state transitions and expectations consistent with the stated information. When several equilibria exist, either a declared selector is an input or the target is set-valued. Comparisons across policy regimes must use the stated selection convention. Reduced-form relationships are not presumed invariant after an equilibrium-changing intervention.

Optimization and welfare. Optimization is over the stated feasible set. If attainment is not established, $\sup$ or $\inf$ and $\varepsilon$-optimal choices replace an asserted maximizer or minimizer. For example, with value $V=\sup_{a\in\mathcal A}W(a)<\infty$, an $\varepsilon$-optimal set is $\{a\in\mathcal A:W(a)\geq V-\varepsilon\}$ for $\varepsilon>0$. Benefits, costs, risk and health or environmental outcomes can be aggregated only using declared units and conversion weights. Interpersonal utility weights, the treatment of fiscal transfers and foreign profits, discounting, and the behavioral welfare criterion are explicit inputs. A comparative-static derivative additionally requires existence and appropriate differentiability of the selected counterfactual.

Sources. Local labels [S1], [S2], and so forth restart in every question. Locators identify the relevant abstract, model, section or result at the level actually checked; a bibliographic or abstract-level verification is not represented as inspection of an entire proof. Working papers are distinguished from later publications and changing versions. Source summaries are paraphrased. All URL checks and editorial formulations refer to the audit date stated above.

Statement licence: CC-BY-4.0 · Source file

Contributions 1

Econ-539-A1 · Exploration · LLM-generated · gpt 6 astra ultra ·

GPT 6 Astra Ultra attempt on Econ-539

Sharp attribution bounds and strategic entry in populist voting

Problem ID: D72-535. Status: Unresolved. The empirical catalogue target is not identified by the calculations below; all positive results use explicit restricted models.

Interventions rather than an undifferentiated historical share

The target fixes a definition of a populist party, regions, horizon and feasible intervention components. It asks for vote-share responses, an optional Shapley allocation of the total response, and subsequent economic consequences. Rodrik's discussion of globalization and different forms of political backlash motivates the economic setting [rodrik]; it does not identify the editorial Shapley target. The present attempt gives a sharp identified set for two components and separates party entry from preferences and governing effects.

Take two feasible binary intervention components, such as a specified trade-adjustment policy and a specified redistribution policy. Denote their four expected vote shares by \[ v_{00}=a,\qquad v_{10}=x,\qquad v_{01}=y,\qquad v_{11}=b. \] All lie in $[0,1]$. Each policy is a full regime for the stated regions, including spillovers and an equilibrium-selection rule if necessary. Suppose only the baseline and joint reform have been experimentally assigned, so their means $a,b$ are known at the population level. Sampling uncertainty is set aside initially to isolate the identification question.

The two Shapley attributions are \[ \phi_1=\tfrac12\{x-a+b-y\},\qquad \phi_2=\tfrac12\{y-a+b-x\},\qquad \Delta=b-a. \tag{1} \] Their sum is $\Delta$ because each intermediate mean appears with opposite signs. The formula averages the two possible orders of implementation. It is a convention about these specified interventions, not a claim that history has a unique additive causal partition.

Sharp bounds from endpoint data

Proposition (Joint attribution set): With unrestricted missing cell means in $[0,1]$, the sharp identified set is \[ \mathcal I(a,b)=\left\{(z,\Delta-z): \frac{\Delta-1}{2}\le z\le\frac{\Delta+1}{2}\right\}. \tag{2} \] If component interventions are additionally mean-monotone and $b\ge a$, so $a\le x,y\le b$, it becomes \[ \mathcal I_{\rm mon}(a,b) =\{(z,\Delta-z):0\le z\le\Delta\}. \tag{3} \]
Proof: Equation (1) gives $\phi_1=(\Delta+x-y)/2$. Unrestricted cell means allow every difference $x-y\in[-1,1]$, establishing the coordinate interval in (2). For any such difference choose admissible $x,y$ with that difference; together with the fixed endpoints these define a possible response surface. For example, take a uniform random variable $U$ and potential binary vote indicators ${\mathbf1}\{U\le v_{ts}\}$. Their means are the desired values, and randomizing only endpoints cannot distinguish the missing cells. This constructs every point in the displayed line segment, proving sharpness. Under monotonicity the difference instead ranges over $[-\Delta,\Delta]$, with endpoints attained by $(x,y)=(a,b)$ or $(b,a)$. The same uniform construction also satisfies pointwise component monotonicity because each intermediate threshold lies between $a$ and $b$. ∎

The joint set is a line segment, not a rectangle formed by combining the two coordinate intervals freely. For example, $a=0.2,b=0.6$ gives $\Delta=0.4$. Without monotonicity either component's attribution can range from $-0.3$ to $0.7$, with the other constrained to make the sum $0.4$. With monotonicity the range narrows to $[0,0.4]$, but endpoint data still permit either component to receive the whole effect. A precise total effect need not yield precise attribution.

One might add an interaction bound \[ |b-x-y+a|\le\Gamma. \tag{4} \] This restricts the sum $x+y$ while (1) depends on their difference. It need not resolve attribution. In the numerical example, even $\Gamma=0$ permits $(x,y)=(0.2,0.6)$ and $(0.6,0.2)$, both monotone, yielding attributions $(0,0.4)$ and $(0.4,0)$. Thus exact additivity of this two-component response surface is insufficient when only the joint intervention is observed. Separate component assignments or other justified restrictions are needed.

If either mixed regime is institutionally impossible, assigning a free mean to it is not a valid policy experiment. In that case (1) itself lacks the intended counterfactual meaning. The feasible joint contrast remains well-defined, but one must choose a decomposition on feasible paths or refrain from reporting component shares.

Strategic entry can create a discontinuous vote response

Here is a fully specified illustration of why vote share is not simply an attitude measure. Conditional on an eligible populist party entering, let its support be $s(a)\in[0,1]$. Entry provides gross organizational benefit $B s(a)$ and costs $f$, with $B>0$ and $0<f<B$. Assume a single potential entrant, deterministic tie-breaking in favor of entry, and no support for a classified populist party when it does not enter. Its equilibrium vote share is \[ P(a)=s(a){\mathbf1}\{s(a)\ge f/B\}. \tag{5} \] The entry rule follows directly by comparing net payoff $Bs(a)-f$ with the zero payoff from staying out. Even continuous latent support produces a jump of size $f/B$ in observed populist voting at the entry threshold.

For $f/B=0.3$, an intervention moving latent support from $0.29$ to $0.31$ raises recorded populist vote share from zero to $0.31$. Only two percentage points of this change are a change in conditional support. Alternatively, reducing entry costs from $0.32B$ to $0.28B$ at fixed $s=0.3$ raises the observed share by $0.3$ without any preference change. Both are hypothetical calculations. They show why preregistered party labels, entry and ballot access belong in the outcome model. Relabeling parties after seeing economic shocks would change the target rather than explain it.

Votes, power and economic outcomes remain distinct

An intervention's effect on $P$ does not identify its effect on governing power. For instance, in a deliberately simple majority-election rule $G(a)={\mathbf1}\{P(a)>1/2\}$, a move from $0.2$ to $0.4$ changes votes but not government, whereas a move from $0.49$ to $0.51$ changes both. Coalition systems require a different mapping. A government then chooses policies, and economic outcomes may also respond directly to the original economic intervention.

Two models can agree on all four vote-share laws while assigning different economic outcomes under the new government. One can set $Y(a)=0$ in every regime; another can set $Y(a)=G(a)$, with identical electoral behavior. If only voting is observed, both are observationally compatible but yield different economic contrasts when governing status changes. Even observing economic outcomes at baseline does not identify the missing policy-regime outcome. Randomizing the original intervention would identify its total economic effect under the declared assignment assumptions, yet interpreting that effect as entirely mediated by populist governance requires an additional exclusion restriction.

With observed factorial vote means and simultaneous error bounds $|\widehat v_{ts}-v_{ts}|\le\epsilon$, (1) has error at most $2\epsilon$ per attribution by the triangle inequality. This addresses sampling once the four means are identified; it cannot repair missing mixed regimes. No actual factorial measurements, credible structural restrictions or economic follow-up data are supplied here. Equations (2)--(5) delimit the identified set and demonstrate a supply-of-parties mechanism, leaving the full empirical origins-and-consequences target unresolved.

Completion Estimate

55%

This is a post hoc, heuristic estimate for the full catalogue target. The attempt characterizes sharp two-component Shapley sets from endpoint data and derives a strategic-entry mechanism separating latent support from recorded populist votes. Full origins and consequences still require feasible mixed-policy comparisons, credible regional assignment and spillover models, and identified subsequent governing and economic responses.

References

  • [rodrik] D. Rodrik (2017), Populism and the Economics of Globalization, HKS Faculty Research Working Paper RWP17-026, June. Primary working-paper record. This cites the 2017 working paper rather than conflating it with the 2018 journal version. The sharp sets and entry model are self-contained illustrations, not attributed findings.

Go to contribution →