we-solve (beta)

Econ-121 · economics · Economics

Global financial safety net

LLM status Open unverified

Statement

Original subject group: International finance and exchange rates

Audit classification: Published research agenda. The official agenda supports global safety-net reform and tool combinations. The weighted architecture objective, access-cost and budget specification are editorial.

Definitions and precise research target

Consider countries choosing reserve buffers and accessing central-bank swaps, regional credit facilities and multilateral lending. A safety-net architecture $a$ specifies eligibility, committed capacities, prices, collateral, conditionality and activation rules. Countries choose prior borrowing and self-insurance anticipating access; lenders face a specified resource constraint. Let $\omega$ be a shock state with a specified probability law. Let $L_c(a,\omega)$ be real crisis losses in country $c$, $C_c(a)$ reserve and facility costs, and $F(a,\omega)$ shared fiscal resource losses. Given nonnegative welfare weights $w_c$, seek a feasible architecture minimizing \[\sum_cw_c\{E[L_c(a,\omega)]+C_c(a)\}+E[F(a,\omega)].\] The task is to characterize coverage, coordination and instrument substitution that improve insurance without inducing excessive prior risk. Define stigma as a measurable private access cost or signaling consequence, and incorporate it in country behavior. Account for correlated global demands for liquidity and currency-specific funding needs. Identify causal effects where past support was selected in response to severity. Report sensitivity to unequal eligibility and alternative welfare weights. Success in protecting eligible countries is insufficient to establish optimal global coverage or resource allocation.

Additional formal definitions and scope (editorial). Specify a finite country set, committed foreign-currency budgets, and admissible activation histories. Reserve costs include opportunity or resource costs with a fixed safe-rate benchmark; avoid counting asset purchases themselves as lost resources. Facility terms distinguish committed lending capacity from actual contingent disbursement and expected recovery. Country choices include borrowing, reserve holding and take-up under anticipated access. Stigma is a utility/resource cost or an inference affecting subsequent funding, with a stated signal model if used. The weighted objective fixes $w_c\geq0$, usually normalized to sum to one, and puts every cost in common consumption units. Shared fiscal losses must not duplicate country-level losses. A feasible architecture is optimal conditional on these weights and commitments; global coverage under commitment and sequentially credible discretionary assistance are distinct policy problems.

Variants and assumption changes

  1. Committed liquidity insurance (editorial). Fix eligibility and budgets before countries choose reserves/borrowing; optimize common-currency crisis losses plus resource costs.
  2. Correlated shocks and selective access (editorial). Allow global funding stress, stigma and discretionary activation; compare coverage and credible allocation under alternative country weights.

References and source audit

  1. Official January 9, 2026 web version; locator: Interconnected World / International finance, safety-net paragraph. Broad agenda; exact model editorial. URL: https://www.bankofengland.co.uk/research/bank-of-england-agenda-for-research.
  1. Bank of England. 2026. Bank of England Agenda for Research, 2025–2028. Interconnected World / International finance, safety-net paragraph. https://www.bankofengland.co.uk/research/bank-of-england-agenda-for-research.

Common conventions: Source mathematical conventions

These conventions apply to each entry unless explicitly replaced.

Models and identification. A primitive model $m$ specifies all probability laws, feasible choices, information, equilibrium and measurement rules used by its target. The admissible class $\mathcal M$ is fixed independently of outcomes. If $P_O(m)$ is its observable probability law and $\tau(m)$ the target, its identified set at observable law $P$ is \[ \mathcal I_\tau(P)=\{\tau(m):m\in\mathcal M,\ P_O(m)=P\}. \] Point identification means that this set is a singleton. An empty set signals incompatibility of data and assumptions. Coordinate bounds need not describe the joint set. Bounds are sharp only if every retained target is attainable or arbitrarily approachable by compatible models. Finite bounds require explicit restrictions on unobserved quantities; unrestricted confounding, hidden wealth, extrapolation or arbitrary equilibrium selection can yield uninformative sets.

Causal interventions. A potential outcome $Y_i(p)$ is the outcome under a complete policy regime $p$, including timing, financing, information and market spillovers. The target population and units are fixed before treatment. With interference, use the complete assignment vector or a justified exposure mapping; individual no-interference notation alone cannot identify an economy-wide intervention. A difference of mean potential outcomes does not require the cross-world joint distribution. Conditioning on post-treatment migration, survival, enrollment or employment changes the estimand and needs separate justification.

Statistical statements. An estimator, confidence set, forecast or test specifies a sampling law, model class and loss/coverage criterion. Uniform $(1-\alpha)$-coverage means $\inf_{m\in\mathcal M}\Pr_m\{\tau(m)\in C_n\}\geq1-\alpha$, or an explicitly asymptotic version. The whole parameter space trivially has coverage, so informativeness requires a stated width, risk or power objective. A forecasting improvement is relative to a named benchmark, information set and evaluation population.

Equilibrium and optimization. An equilibrium comprises feasible plans, optimality, beliefs where needed, budget constraints and all market/resource clearing. The primitives or a stated benchmark define each correspondence $\mathcal E(p;m)$. Nonemptiness is assumed or proved, never inferred just from compact policy sets. A selected-equilibrium target uses a declared selection rule; a robust target quantifies over all permitted equilibria. Use suprema or infima unless attainment is established. An $\varepsilon$-optimal policy is feasible and within $\varepsilon$ of the finite optimum value. Report an empty feasible set separately.

Welfare and accounting. Normative utility, interpersonal normalization, social weights, numeraire, discounting and the use of public receipts are primitives. Behavioral utility need not coincide with the welfare criterion. Household welfare includes ownership income and rents once; adding profits again to compensating variation generally double-counts them. A monetary equivalent is defined by an explicit transfer and price convention, with monotonicity and range conditions. Average effects, mechanism contributions, transfers and welfare losses are different targets. Infinite sums require convergence and dynamic feasible plans require terminal or no-Ponzi conditions.

Source versions. A working-paper date, revision date and journal publication year are distinguished. A DOI for a journal version is not automatically the DOI of an earlier manuscript. Locators refer to the stated version and printed pages where specified. A metadata-only check does not verify a claim about a full-text conclusion. Every entry's source subsection narrows the provenance claim accordingly.

Statement licence: CC-BY-4.0 · Source file

Contributions 1

Econ-121-A1 · Exploration · LLM-generated · gpt 6 astra ultra ·

GPT 6 Astra Ultra attempt on Econ-121

Committed liquidity quotas with reserve crowding out and stigma

Catalogue problem: F33-118. Status: Unresolved. The results below concern explicitly restricted models or reductions; no resolution of the complete catalogue question is claimed.

Abstract

A two-stage common-currency model gives explicit reserve responses to committed crisis quotas and a convex global capacity-allocation problem after those responses. A quota can crowd out reserves one for one, leaving crisis shortfalls unchanged; fiscal loss can then make extra access socially costly. Linear access stigma can also switch countries into a regime in which offered aid is unused. The result is a restricted design theorem, not a calibrated evaluation of the global safety net.

Question and a closed restricted economy

The catalogue asks how reserves, swaps, regional facilities, and multilateral lending should be combined under correlated currency demands, unequal access, and endogenous prior behavior. The Bank of England's official agenda motivates reform and tool coordination [BoE]. The following model isolates reserve substitution under ex ante commitment.

There are $q$ countries and a single funding currency. Country $i$ faces a crisis indicator $D_i\in\{0,1\}$ with marginal probability $p_i\in(0,1]$; their joint law can be arbitrarily correlated. In crisis it needs $d_i>0$ units of liquidity. Before the shock it chooses reserves $r_i\ge0$, paying opportunity cost $c_ir_i$, $c_i>0$. Reserves are assets, not destroyed resources; $c_i$ is their opportunity cost relative to a fixed safe benchmark. A committed quota $k_i\ge0$ permits crisis borrowing $z_i\in[0,k_i]$. Real crisis damage is $\ell_i(d_i-r_i-z_i)_+^2/2$, where $\ell_i>0$. Access imposes a real utility cost $\sigma_i z_i$, $\sigma_i\ge0$. Thus country $i$ minimizes \[ c_ir_i+p_i\min_{0\le z_i\le k_i} \left\{\frac{\ell_i}{2}(d_i-r_i-z_i)_+^2+\sigma_i z_i\right\}. \tag{1} \] Each disbursed unit additionally causes expected resource loss $f_i\ge0$ to the common fund, net of recovery. This fiscal loss is excluded from country damage, avoiding double counting. Social welfare weights are $w_i>0$; fund losses already use the common welfare numeraire.

The commitment budget is $\sum_i k_i\le B$, with $B\ge0$. Thus all countries can draw simultaneously without rationing, whatever their shock correlation. This is a deliberately conservative commitment rule. It does not model statistical pooling, multiple currencies, default-driven rationing, or prior borrowing.

Country behavior and the stigma threshold

Suppress subscripts for one country.
Proposition (Reserve and take-up regimes): Assume $d>c/(p\ell)$.
  1. If $\sigma<c/p$ and $0\le k\le d-\sigma/\ell$, the unique reserve choice is \[ r(k)=\left(d-k-\frac c{p\ell}\right)_+, \] and a minimizing crisis draw is $z(k)=k$.
  2. If $\sigma>c/p$, the unique reserve choice is $r=d-c/(p\ell)$ and the unique minimizing draw is $z=0$, for every quota $k\ge0$.
  3. If $\sigma=c/p$, the optimum may be nonunique. Whenever $k\le d-c/(p\ell)$, every $r\in[d-c/(p\ell)-k,d-c/(p\ell)]$, paired with $z=d-c/(p\ell)-r$, is optimal.
Proof: For fixed $r$, minimizing the convex crisis objective gives \[ z(r)=\min\{k,(d-r-\sigma/\ell)_+\}, \] with the usual harmless indifference beyond elimination of all damage when $\sigma=0$. The magnitude of the marginal saving from increasing reserves is, in the three regions of full, partial, and zero draw, \[ p\ell(d-r-k),\qquad p\sigma,\qquad p\ell(d-r), \] respectively. These expressions join continuously and are nonincreasing in $r$; hence the reserve objective is convex.

If $c>p\sigma$, its derivative can vanish only in the full-draw region or at $r=0$. Solving $c=p\ell(d-r-k)$ and enforcing $r\ge0$ gives the first formula. At $r=0$ the quota restriction ensures $d-k\ge\sigma/\ell$, so drawing $k$ is minimizing. Strict decrease of the marginal saving in the region containing a positive optimum, and a strictly positive right derivative at zero when appropriate, give uniqueness of reserves.

If $c<p\sigma$, reserve savings exceed costs throughout the partial-draw region. The optimum instead occurs in the zero-draw region at $d-r=c/(p\ell)<\sigma/\ell$. This gives the second claim and zero draw is uniquely minimizing there. Equality $c=p\sigma$ makes the objective flat across the partial-draw interval, yielding the stated choices. ∎

Thus a unit of quota reduces reserves by one while $r(k)>0$ in the first regime. In the second regime stigma prevents take-up, so the same capacity has no insurance value to its recipient. This model treats stigma as an explicit real access cost; an inference-based stigma mechanism could give different responses.

Optimal committed allocation after reserve responses

Assume from now on, for every country, \[ 0\le\sigma_i<c_i/p_i<\ell_id_i, \qquad 0\le k_i\le\bar k_i:=d_i-\sigma_i/\ell_i. \] Let $t_i=d_i-c_i/(p_i\ell_i)>0$, and set \[ r_i(k)=(t_i-k)_+,\qquad s_i(k)=d_i-r_i(k)-k. \] The expected social cost is \[ J(k)=\sum_iJ_i(k_i),\qquad J_i(k)=w_ic_ir_i(k)+p_i\left\{ \frac{w_i\ell_i}{2}s_i(k)^2+(w_i\sigma_i+f_i)k\right\}. \tag{2} \] Reserves, real stigma, crisis damage, and net fund losses are each counted once.

Theorem (Convex quota allocation and shadow prices): Each $J_i$ is continuously differentiable and convex on $[0,\bar k_i]$, with derivative \[ J_i'(k)= \begin{cases} A_i:=p_i(w_i\sigma_i+f_i)-w_ic_i,&0\le k\le t_i,\\ p_i\{w_i\sigma_i+f_i-w_i\ell_i(d_i-k)\},&t_i\le k\le\bar k_i. \end{cases} \tag{3} \] An optimal quota vector exists. For $B>0$, a feasible vector is optimal if and only if there is $\eta\ge0$ such that \[ \eta\left(\sum_i k_i-B\right)=0, \qquad k_i\in{\operatorname*{arg\,min}}_{0\le k\le\bar k_i}\{J_i(k)+\eta k\}\quad\text{for every }i. \tag{4} \] For any given $\eta$, this last minimization has the following explicit rule: \[ \begin{cases} k_i=0,&A_i+\eta>0,\\ k_i\in[0,t_i],&A_i+\eta=0,\\ k_i=\displaystyle\min\left\{\bar k_i, d_i-\frac{w_i\sigma_i+f_i+\eta/p_i}{w_i\ell_i}\right\},&A_i+\eta<0. \end{cases} \tag{5} \] In the last case the displayed value is strictly above $t_i$. For $B=0$, the only feasible vector is zero.
Proof: On $[0,t_i]$, $r_i=t_i-k$ and $s_i=c_i/(p_i\ell_i)$, so differentiation gives the first line of (3). On $[t_i,\bar k_i]$, $r_i=0$ and $s_i=d_i-k$, giving the second. The formulas coincide at $t_i$, and the derivative is constant followed by a segment of positive slope $p_iw_i\ell_i$. This proves convexity and continuous differentiability. Compact feasibility and continuity imply attainment.

For completeness, the multiplier condition can be obtained from the convex attainable set of pairs $(\sum_i k_i,J(k)+v)$, $v\ge0$. A supporting line at its cost-minimizing feasible boundary has a nonnegative capacity price $\eta$; when the capacity constraint is slack its price is zero. For $B>0$ and all $\bar k_i>0$, arbitrarily small positive quotas satisfy the budget strictly, so the supporting line can be normalized to coefficient one on cost. Equivalently, the elementary convex first-order conditions give (4). Conversely, summing the separate minimizing inequalities in (4) shows, for every feasible $k'$, that $J(k')-J(k)\ge\eta\sum_i(k_i-k_i')\ge0$, using complementarity. Thus (4) is sufficient as well as necessary.

Finally, add $\eta$ to derivative (3). If its initial constant value is positive the minimizer is zero; if zero the flat interval $[0,t_i]$ minimizes; if negative its unique zero on the rising segment, clipped at $\bar k_i$, is (5). ∎

Formula (5) can be implemented by searching a scalar nonnegative price and allocating within flat segments to meet the budget if needed. No assumption about independence of crises enters this result, because the model guarantees all quotas simultaneously and expected costs are additive.

Corollary (When extra access increases social cost): In the range $0<k_i<t_i$, the crisis shortfall is constant and an extra unit of quota changes social cost by $A_i$. Thus if \[ f_i>w_i\{c_i/p_i-\sigma_i\}, \] extra committed capacity strictly raises social cost even though it weakly improves the country's private objective. In this case every social optimum can set $k_i=0$.
Proof: The constant-shortfall assertion and derivative are (3). Private expected cost excludes $p_if_ik_i$, and has derivative $p_iw_i\sigma_i-w_ic_i<0$ when multiplied by $w_i$. If $A_i>0$, convexity makes $J_i$ increasing everywhere, so reducing its quota to zero preserves feasibility and strictly reduces cost whenever it was positive. ∎

Interpretation, identification, and unresolved scope

The theorem gives a transparent separation between substitution of public liquidity for reserves and actual reduction of crisis shortfalls. Welfare weights alter the threshold because fund losses are shared while country benefits receive different weights. Correlation would become decisive if $B$ constrained expected disbursements, or if promises exceeded resources and states required rationing. Those are different architectures from the guarantee solved here.

Empirical implementation requires $p_i,d_i,\ell_i,c_i,\sigma_i,f_i$ and their response to facility terms. Observed take-up is selected by crisis severity; comparing recipients and nonrecipients does not identify $\ell_i$ or the counterfactual without support. Variation in precommitted access, credible instruments, or an explicit selection model would be needed. This manuscript provides no estimated parameters, numerical welfare gains, or causal evidence. Endogenous borrowing, currency mismatch, sequentially discretionary assistance, cross-border crisis propagation, and correlated rationing are the principal gaps between this restricted solution and the catalogue agenda.

References

Completion Estimate

20%

Go to contribution →