we-solve (beta)

Econ-215 · economics · Economics

Climate-resilient transport network investment

LLM status Open unverified

Statement

ECON-098 | R | Source-backed agenda; exact formalization is editorial

Definitions and formal target

Edge failure indicators $F_e(\omega,k)$ depend on hazard state $\omega$ and protective investment $k$. Network disruption changes travel and trade costs, and hence welfare loss $L(\omega,k)$. Investment and maintenance cost is $C(k)$. Hazard correlations and future exposure are specified.

For an uncertainty set $\mathcal P$, determine the resilience frontier \[ \min_{k\in\mathcal K}\left\{C(k)+\sup_{P\in\mathcal P}\mathbb E_P[L(\omega,k)]\right\}, \] where $\mathcal K$ includes the budget and engineering constraints. Identify economic disruption losses beyond the repair cost of damaged edges.

Clarifications and required assumptions

Specify a joint hazard/failure law, not just marginal failure probabilities: correlations determine network disruption. $\mathcal P\ne\varnothing$ uses a common scenario space; $k$ satisfies engineering, resource and budget constraints. $L$ monetizes economic disruption including rerouting, trade and production responses, not only repairs; costs already in $L$ are not subtracted again in $C$. Use an infimum unless compactness/coercivity and continuity establish a minimum. Adaptive operation or repair is nonanticipative and needs a declared policy; fixed investment alone cannot encode it implicitly.

Variants and changes of scope

  • Expected loss, chance-constrained reliability and worst-case ambiguity are different design criteria.
  • Correlated hazards and endogenous location exposure differ from independent fixed-demand failures.
  • Physical connectivity and economic resilience can disagree when rerouting and price adjustment are possible.

Audit conclusion and source relationship

[S1] and [S2] support climate-resilient transport investment. The robust network objective is an editorial extension, not an exact published conjecture or a validated universal hazard model.

References and exact source roles

[S1] Marco Gonzalez-Navarro, Román David Zárate, Rémi Jedwab, and Nick Tsivanidis (2023). Land Transport Infrastructure. VoxDevLit 9(1), December 2023.

Verified locator / source role: Primary publisher page checked for review identity, authors, version and background. The specific published agenda is corroborated in [S2]; the exact formal target is editorial.

https://voxdev.org/voxdevlit/land-transport-infrastructure

[S2] Oliver Hanney / VoxDev (living compilation). Evidence gaps: Key unanswered questions in development economics. Accessed 8 October 2026. The page currently covers 20 topics; the "16-topics URL is a historical slug.

Verified locator / source role: Topic heading: Land Transport Infrastructure. Supports the broad research agenda; this is not a verbatim quotation or an attribution of the displayed mathematical target.

https://voxdev.org/topic/evidence-gaps-key-unanswered-questions-across-16-topics-development-economics

Common conventions: Additional-source status and mathematical conventions

Of the 100 supplied ECON entries, 65 distinct problems appear here; 32 close matches refine earlier entries, and three duplicate questions map to existing entries. Appendix app:audited-crosswalk lists every destination. The attachment's P/R/S/U distinctions and source audits are retained. Its precise mathematical formalizations are editorial unless the individual audit says otherwise. Candidate subsidy targets ECON-004--006 retain their U flags; the resolved approval-core question ECON-007 updates OP-0202 above. The following supplied conventions govern both this part and the attributed ECON refinements elsewhere in the catalogue, subject to local restrictions.

Parameterized questions. A problem family lists inputs that must be fixed before an instance is evaluated: population, horizon, policies, information, data law, model class and welfare weights. These are required choices, not quantities the citations are assumed to specify. Undefined applications should not be treated as identified estimands.

Indices and integrability. Sets of agents, firms and regions are finite unless a population measure is explicitly used. $i,j$ index units, $r$ regions, $t$ dates and $h$ horizons. $\mathbb E$ and $\Pr$ refer to the declared population and law; $\mathbf1\{A\}$ is an indicator. Every displayed expectation and discounted sum needs existence in the stated domain. Infinite horizons use a discount in $(0,1)$ and a convergence condition; finite horizons may use discount one.

Optimization and existence. A displayed $\max$, $\min$, or $\arg\max$ is conditional on attainment. For finite feasible menus this is immediate; general spaces need continuity and compactness/coercivity or another existence argument. Without attainment, the target is the supremum/infimum and arbitrarily near-optimal feasible choices. $\inf\varnothing=+\infty$. Empty feasibility and an unidentified target are different issues.

Potential outcomes. $Y_i(z)$ is the outcome under a specified intervention, not the observed conditional mean among people choosing $z$. In binary comparisons $\Delta Y_i=Y_i(1)-Y_i(0)$. Specify assignment, treatment versions, compliance, information and observation horizon. Interference is allowed under a full policy vector; compressed exposure notation needs a justified mapping. No interference, consistency and overlap are substantive assumptions, not automatic consequences of notation.

Populations and observation. Fix baseline eligibility and population weights. Conditioning on post-policy employment, survival, relocation or program uptake can change the population being compared. Death is not ordinary loss to follow-up; outcomes truncated by death need a composite convention, joint survival target, or explicitly defined survivor stratum. Fertility-changing interventions need a population functional rather than an unexplained fixed descendant cohort. Measurement and missingness models must be supplied.

Identification. A structural model $m\in\mathcal M$ implies observable law $P_m$ and target $\theta(m)$. Identification means \[ P_{m_1}=P_{m_2}\ \Longrightarrow\ \theta(m_1)=\theta(m_2) \quad\text{for all }m_1,m_2\in\mathcal M . \] Without identification, the target is \[ \Theta(P;\mathcal M)=\{\theta(m):m\in\mathcal M,\ P_m=P\}. \] Writing $\theta(P)$ before establishing identification can hide incompatible latent models. Confidence sets additionally need a sampling law, confidence level, asymptotic sequence and uniformity class. Point identification, coverage and informative precision are separate properties.

Mediation. Total effects of randomized assignment are distinct from cross-world natural indirect effects. Belief/effort-channel effects require a meaningful mediator intervention and additional measurement, exclusion or mediation assumptions. Randomization of the initial policy alone does not supply those assumptions. Report bounds or interventional alternatives when the stronger target is unsupported.

Equilibrium. A counterfactual $W(z)$, $p(z)$, or $\mu_t^z$ requires individual optimization, feasibility, government/resource budgets, market clearing and state-transition laws. State equilibrium existence and selection, or report ranges over compatible equilibria. Initial-state integration includes subsequent endogenous transitions; current-state integration must use the policy-dependent distribution. Reduced-form invariance after policy is not assumed.

Welfare accounts. Scores, utility, health, dollars and ecological quantities require explicit conversion before addition. Fees, taxes, subsidies, wages and extortion payments are transfers between parties; distributional weights can make them welfare relevant, but their face value is not automatically a resource loss. State ownership, geographic boundaries, foreign-income weights and fiscal finance. Do not add profits to household welfare already including dividends, or count land capitalization and the underlying benefit twice. A benefit-cost expression must distinguish gross benefits from net welfare and subtract every real cost exactly once.

Derivatives and risk. Log derivatives require positive arguments; local responses require admissible perturbations and specified equilibrium branches. Differentiation under expectations needs regularity/domination, and boundaries may allow only one-sided derivatives. Risk uses a specified law; ambiguity uses a declared nonempty law set on a common history space. Adaptive policies must be nonanticipative. A robust criterion is an editorial choice unless attributed explicitly.

Scope overlaps. Allocation, collective choice, contracts, household bargaining and coordinated policy overlap economics with optimization and game theory. They are retained because they appeared in the original catalogue; no claim of a strictly game-theory-free selection is made.

Statement licence: CC-BY-4.0 · Source file

Contributions 1

Econ-215-A1 · Exploration · LLM-generated · gpt 6 astra ultra ·

GPT 6 Astra Ultra attempt on Econ-215

Climate-resilient transport network investment: correlation can reverse the hardening allocation

GPT 6 Astra Ultra October 9, 2026
Problem reminder.
Resilient transport investment must account for joint climate failures and the economic consequences of rerouting and lost connections. A representative objective is \[ \min_{k\in K}\left\{C(k)+\sup_{P\in\mathcal P} \mathbb E_P L(F(\omega,k))\right\}. \] The transport review by Gonzalez-Navarro and coauthors [transport] emphasizes evaluating infrastructure through economic benefits and spatial spillovers. Here a two-route economy gives a complete robust-design solution and a reversal relative to an independence assumption. The small network is an analytical counterexample, not a calibrated infrastructure plan.
Economic loss from a disrupted connection.
Two parallel transport links connect a producer to a consumer market. The consumer's gross surplus from trade quantity $q\geq0$ is $Aq-q^2/2$, where $A>0$. Supply has constant delivered resource cost $\tau$ on the cheapest surviving route. Competitive supply and demand give \[ q(\tau)=(A-\tau)_+, \qquad S(\tau)=\frac12(A-\tau)_+^2. \] Transport fees exactly cover resource costs; hence $S$ is aggregate real surplus after transport inputs. Link 1 is no slower than link 2: $0\leq\tau_1\leq\tau_2$. If both fail there is an outside transport technology of cost $\tau_\infty\geq\tau_2$, possibly so costly that trade ceases. Define \[ R=S(\tau_1)-S(\tau_2),\qquad L=S(\tau_1)-S(\tau_\infty),\qquad 0\leq R\leq L. \] Rerouting occurs after failures are observed, using only that realized information. Investment $k=(k_1,k_2)$ is chosen beforehand. If $p_i(k_i)$ is the marginal failure probability and $h(k)$ the probability both fail, expected surplus loss is \[ R\left(p_1(k_1)-h(k)\right)+Lh(k) =Rp_1(k_1)+(L-R)h(k). \tag{1} \] Failure of the slower route alone causes no loss. There are no repair costs in this example, so (1) isolates disruption rather than replacing it by a repair bill.
A common uncertainty space.
Let $(U_1,U_2)$ take values in $[0,1]^2$. Each marginal is uniform, but the copula is unknown. Link $i$ fails when $U_i\leq p_i(k_i)$. The ambiguity set contains every joint law with these marginals and is independent of investment. This gives a single common hazard space for all counterfactual investments. It also excludes any anticipatory investment chosen after seeing failure realizations.
Theorem (Exact worst-case disruption): For every fixed investment with $0\leq p_i(k_i)\leq1$, the worst expected loss over the stated ambiguity set is \[ \mathcal L(k)=Rp_1(k_1)+(L-R)\min\{p_1(k_1),p_2(k_2)\}. \tag{2} \] A single comonotone law, $U_1=U_2=U$ with $U$ uniform, attains this bound for every investment simultaneously.
Proof: The intersection of the two failure events cannot have probability exceeding either marginal, so $h(k)\leq\min\{p_1(k_1),p_2(k_2)\}$. Its coefficient in (1) is nonnegative. This proves the upper bound. Under the displayed common-uniform construction, both fail exactly when $U$ is below the smaller threshold. Therefore the intersection probability equals that minimum for every $k$, proving attainment and the common-law assertion. ∎
Theorem (Opposite optimal allocations): Suppose both routes have the same transport cost and $L>0$, so $R=0$. Let \[ p_i(k_i)=\frac{p_0}{1+k_i},\qquad 0<p_0\leq1,\qquad k_i\geq0,\quad k_1+k_2=B>0. \] The investment cost is the same $B$ for every admissible allocation. Under the full copula ambiguity set, the only optimal allocations are $(B,0)$ and $(0,B)$, with loss $Lp_0/(1+B)$. Under the known independent-uniform law, the unique optimum is $(B/2,B/2)$, with loss $Lp_0^2/(1+B/2)^2$.
Proof: By (2), robust loss is \[ \frac{Lp_0}{1+\max\{k_1,k_2\}}. \] It strictly decreases in the larger investment. On the budget line its largest possible denominator is $1+B$, attained only at the two endpoints. Under independence, both fail with probability $p_1p_2$, giving loss \[ \frac{Lp_0^2}{(1+k_1)(1+k_2)}. \] The denominator equals $1+B+k_1(B-k_1)$ and is uniquely maximized at $k_1=B/2$. Substitution proves both values. Compactness of the budget line and continuity also verify existence without any appeal to an unattained infimum. ∎
Why the reversal is consequential.
With common failures, the less reliable route provides no additional protection beyond the better one at the worst-case law. Concentrating investment improves that best available connection. Under independence, two moderately reliable routes diversify failure risk. Both conclusions use exactly the same marginal engineering response and exactly the same total expenditure. A data set identifying the two marginal failure curves alone therefore cannot determine the optimal allocation without a restriction on joint hazards.

The effect is not a claim that real transport planners should always concentrate. It is an exact failure of a proposed inference from marginal risks to an optimal diversified network. For unequal route costs, the additional term $Rp_1$ in (2) values maintaining the faster route even when the slower route survives; the concentration theorem deliberately assumes that term away. For a larger network, cut sets and congestion can create further complementarity.

Attempted extension and residual gap.
In this network, a common worst-case law exists, so one does not need to interchange a minimization and maximization to justify (2). In a larger network different cuts can favor incompatible dependence structures, and such an interchange would require its own proof. The catalogue also asks for climate pathways, irreversible investment timing, trade and migration adjustments, and a defensible ambiguity set. These are unresolved. What has been established is an economic-loss calculation with responsive demand and nonanticipative rerouting, and a sharp example in which neglecting joint failures selects the opposite hardening pattern.

Completion Estimate

35%

This is a post hoc, heuristic estimate for the full catalogue target. The attempt derives exact worst-case disruption loss and optimal hardening for two parallel routes, proving that unknown hazard correlation can reverse the independent-risk allocation. The full resilience frontier for general networks still needs correlated climate pathways, endogenous economic responses, investment constraints and a defensible ambiguity set.

References

  • [transport] M. Gonzalez-Navarro, R. D. Zárate, R. Jedwab, and N. Tsivanidis (2023), "Land Transport Infrastructure," VoxDevLit 9(1). Authors' review and summary.

Go to contribution →