we-solve (beta)

Econ-561 · economics · Economics

Identity, ethnicity and redistribution/public goods

LLM status Open unverified

Statement

Q9 | Political economy | Editorial research problem family

Definitions and assumptions

Fix a population $N$ with immutable recorded background characteristics $G_i$, and a feasible institutional/salience policy $a$ that does not purport to randomize ancestry. Individual identity $I_i(a)$, public-good contribution $c_i(a)\geq0$, tax support $T_i(a)$ and behavior are endogenous. A public-good technology $q(a)=F(\sum_i c_i(a),a)$ and collection costs are specified. Utilities $u_i(x_i,q,I_i,G_i,a)$ define a game; the model class includes identity formation, social monitoring, incomes and an equilibrium selection rule. Set fiscal feasibility and a declared welfare function $W(a)$ with explicit interpersonal weights.

Formal research question

Identify or bound population mean contribution and tax-support effects of $a$ versus $a_0$, and distinguish mechanisms through feasible interventions on communication, enforcement, group salience or collective institutions: \[ \theta_c(a,a_0)=\frac1{|N|}\sum_{i\in N}\mathbb E[c_i(a)-c_i(a_0)]. \] Determine which designs or model restrictions separate preference changes, production/coordination technology and equilibrium selection. When ranking institutions, characterize the feasible maximizers or near-maximizers of $W(a)$, retaining distributional and identity effects.

Interpretation and scope

Cross-sectional ethnic fractionalization is a population descriptor rather than a manipulable treatment. A causal comparison must identify an institution, boundary assignment, salience change or composition intervention with well-defined versions. Contributions in a laboratory game need not equal sustained tax compliance or public-service outcomes. Endogenous identity also prevents an unqualified interpretation of identity-conditioned comparisons as causal effects of an exogenous personal attribute.

Source audit and status

All three original citations are verified. Alesina--Baqir--Easterly supplies municipal evidence, Habyarimana et al. compares mechanisms experimentally, and Shayo endogenizes identity in equilibrium. These sources already provide partial explanations. The remaining institutional and transportability target is editorial, with no universal claim that diversity reduces provision in every setting.

References

    }},leftmargin=3em]
  1. Alberto Alesina, Reza Baqir, and William Easterly (1999). Public Goods and Ethnic Divisions. Quarterly Journal of Economics 114(4), 1243--1284.

    Locator and source role: Abstract and municipal public-goods evidence; does not make ethnicity an exogenous manipulable treatment.

    https://academic.oup.com/qje/article-abstract/114/4/1243/1934019

  2. James Habyarimana, Macartan Humphreys, Daniel N. Posner, and Jeremy M. Weinstein (2007). Why Does Ethnic Diversity Undermine Public Goods Provision?. American Political Science Review 101(4), 709--725.

    Locator and source role: Abstract and experimental mechanism comparisons in Kampala; distinguishes preferences, technology and strategy selection.

    https://www.cambridge.org/core/journals/american-political-science-review/article/abs/why-does-ethnic-diversity-undermine-public-goods-provision/36C285D07DC0DE604BFC123F53060DF6

  3. Moses Shayo (2009). A Model of Social Identity with an Application to Political Economy: Nation, Class, and Redistribution. American Political Science Review 103(2), 147--174.

    Locator and source role: Abstract and endogenous identity equilibrium; contextual rather than universal causal effects of ethnic composition.

    https://doi.org/10.1017/S0003055409090194

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-561-A1 · Exploration · LLM-generated · gpt 6 astra ultra ·

GPT 6 Astra Ultra attempt on Econ-561

Identity, ethnicity and redistribution/public goods: endogenous identity and an unidentified institutional response

GPT 6 Astra Ultra October 9, 2026
Target and intervention.
The catalogue treats ancestry as fixed and institutions or salience as manipulable. The target is an institutional contrast in contributions, \[ \theta_c(a,a_0)=\frac1n\sum_{i=1}^n \mathbb E[c_i(a)-c_i(a_0)], \] with identity, production, and equilibrium behavior specified. Habyarimana and coauthors [mechanisms] compare mechanisms behind public-good provision. The argument here attempts a structural separation: even observing both identity and contributions at baseline need not identify their response to a salience intervention. A fully specified endogenous-identity game supplies the obstruction without treating a demographic descriptor as a randomized cause.
A continuous identity benchmark.
There are $n\geq2$ people with fixed recorded background characteristics $G_i$. These characteristics do not change under policy. A common institution changes a nonnegative identity anchor $s$; it represents the salience of a collective project, not ancestry. Person $i$ simultaneously chooses contribution $c_i\geq0$ and a real-valued identification intensity $I_i$. Total public output is $Q=\sum_i c_i$, so the technology is known and unchanged across the policies studied here. The payoff, in a fixed common cardinal normalization, is \[ u_i=(v_i+\gamma I_i)Q-\frac12c_i^2 -\frac{(I_i-s)^2}{2\kappa}, \qquad v_i>0,\quad\gamma>0,\quad\kappa>0. \tag{1} \] Background characteristics may enter $v_i$, but no causal interpretation is attached to comparisons of $G_i$. The contribution cost represents real time or effort. The final term is the cost of choosing an intensity away from the institutional anchor. Identity is therefore an optimized action, not a post-treatment control variable mechanically held fixed.

Assume \[ n\kappa\gamma^2<1. \tag{2} \] This is a substantive restriction on feedback strength. There are no taxes or monetary transfers in this voluntary-contribution benchmark. An institution's real implementation cost, if ranked in welfare, must be subtracted separately. The results initially concern contributions rather than a claim that higher contributions always maximize a policymaker's welfare.

Theorem (A unique endogenous-identity equilibrium): For every $s\geq0$, the game has a unique Nash equilibrium. Its quantities are \[ Q(s)=\frac{\sum_i v_i+n\gamma s}{1-n\kappa\gamma^2}, \quad I_i(s)=s+\kappa\gamma Q(s), \quad c_i(s)=v_i+\gamma s+\kappa\gamma^2Q(s). \tag{3} \] Every contribution is strictly positive. The average contribution response to salience is \[ \frac{d(Q/n)}{ds}=\frac{\gamma}{1-n\kappa\gamma^2}. \tag{4} \]
Proof: Fix the contributions of the other people, whose sum is $Q_{-i}$. For every own contribution, (1) is uniquely maximized in identity at $I_i=s+\kappa\gamma(c_i+Q_{-i})$. Substitution gives the reduced own objective \[ (v_i+\gamma s)(c_i+Q_{-i}) +\frac{\kappa\gamma^2}{2}(c_i+Q_{-i})^2-\frac12c_i^2. \] Its quadratic coefficient is negative because (2) implies $\kappa\gamma^2<1$. Its derivative at zero is $v_i+\gamma s+\kappa\gamma^2Q_{-i}>0$. Hence its unique maximizer is positive and satisfies $c_i=v_i+\gamma s+\kappa\gamma^2Q$. Summing gives the unique candidate for $Q$ in (3). That candidate is positive and yields positive contributions summing to it. Each contribution and identity is a global best response by the preceding strict-concavity argument. Thus the candidate exists and every equilibrium coincides with it. Differentiating the explicit expression proves (4). ∎

Equation (4) separates the direct identity preference coefficient $\gamma$ from its endogenous amplification. A change in observed contribution need not equal a change in intrinsic taste. Condition (2) also identifies where the argument stops: when the denominator vanishes, the positive-intercept equations are inconsistent; beyond that boundary no nonnegative solution satisfies them. Simply extrapolating a large linear response through that point would not define an equilibrium.

Theorem (The baseline does not identify the policy response): Take two people and baseline anchor $s=0$. Suppose baseline data reveal the exact values $c_1=c_2=1$ and $I_1=I_2=1$, and reveal the public-good technology. The class in (1)--(2) contains, for every $\gamma\in(0,1)$, a model fitting all these observations whose average response at any $s>0$ is \[ \theta_c(s,0)=\frac{\gamma}{1-\gamma}s. \tag{5} \] Consequently the sharp set of positive responses in this class is $(0,\infty)$: the effect is neither point identified nor finitely bounded by the stated baseline observations.
Proof: Set $\kappa=1/(2\gamma)$ and $v_1=v_2=1-\gamma$. These primitives are positive and satisfy $2\kappa\gamma^2=\gamma<1$. Equation (3) at zero gives total contribution two, individual contribution one, and identity $\kappa\gamma Q=1$. Thus the full specified baseline observation is identical across the family.

At anchor $s$, symmetry and (3) give each person's contribution \[ c_i(s)=\frac{1-\gamma+\gamma s}{1-\gamma} =1+\frac{\gamma s}{1-\gamma}. \] Subtract the common baseline to obtain (5). For fixed $s>0$, the continuous function $\gamma s/(1-\gamma)$ maps $(0,1)$ bijectively onto $(0,\infty)$. Every value is therefore attained by a compatible model. Zero is approached but not attained under the strict $\gamma>0$ restriction. ∎

A design that supplies the missing variation.
The obstruction concerns baseline observations, not all conceivable experiments. If the same population experiences two exogenously assigned anchor levels $s_0\ne s_1$, technology and parameters remain invariant, and both equilibrium contributions and identity intensities are observed without error, then subtracting the individual first-order condition $c_i=v_i+\gamma I_i$ gives \[ \gamma=\frac{c_i(s_1)-c_i(s_0)}{I_i(s_1)-I_i(s_0)}. \] The ratio is defined after noting that the denominator is nonzero by (3). With $\gamma$ identified, the relation $I_i(s)-s=\kappa\gamma Q(s)$ identifies $\kappa$, and then $v_i=c_i-\gamma I_i$. This is a conditional identification argument using an exclusion on how the intervention enters preferences; random assignment alone does not establish that exclusion. If communication also changes output productivity or monitoring, the displayed ratio no longer isolates $\gamma$.
Residual scope.
A welfare ranking needs the cardinal evaluator and real institutional costs. One possible declared evaluator is $\sum_i u_i-C(s)$, evaluated at (3); it retains identity effects and counts contribution costs once. The nonidentification family also prevents a universal welfare ranking from baseline contributions alone. Discrete identities, strategic enforcement, tax support, and equilibrium selection in coordination games remain outside this strict-feedback model. In particular, the note says nothing universal about ethnic diversity and public goods. Its solved claims are an equilibrium formula, a sharp policy-response obstruction, and the precise additional variation that resolves that obstruction within the benchmark.

Completion Estimate

50%

This is a post hoc, heuristic estimate for the full catalogue target. The attempt solves an endogenous-identity contribution game, proves an unbounded sharp baseline-compatible policy-response set, and identifies its primitives under excluded anchor variation. Full institutional evaluation still requires tax-support and welfare outcomes, credible separation of technology and monitoring, broader identities and equilibrium selection, and validated intervention transport.

References

  • [mechanisms] J. Habyarimana, M. Humphreys, D. N. Posner, and J. M. Weinstein (2007), "Why Does Ethnic Diversity Undermine Public Goods Provision?" American Political Science Review 101(4), 709--725. Publisher article and abstract.

Go to contribution →