we-solve (beta)

Econ-202 · economics · Economics

Simple lending contracts preserving group-finance benefits

LLM status Open unverified

Statement

ECON-086 | R | Source-backed agenda; exact formalization is editorial

Definitions and formal target

A lending group is a network $G$ with borrower types $\theta_i$, repayment actions $r_i$, and peer information or support. Contract $a$ determines liability, meeting requirements, and repayment rules. Complexity cost is $K(a)$; borrower welfare and lender return are $U(a,G)$ and $\Pi(a,G)$.

Characterize the frontier \[ \max_{a:\,K(a)\leq\overline K}\mathbb E[U(a,G)] \quad\text{s.t.}\quad\mathbb E[\Pi(a,G)]\geq0, \] including endogenous group formation and the value of social support. Determine which simplifications preserve valuable features of group finance rather than removing them inadvertently.

Clarifications and required assumptions

Predeclare a complexity metric $K$, such as the number of state-contingent rules, measured comprehension burden or implementation cost; these are not interchangeable. Define each borrower's utility, information, outside option and repayment behavior, liability and lender costs. Group formation, peer actions and repayment require a stated equilibrium concept and selection, especially with strategic interactions. Include borrower participation and lender viability; positive average welfare does not guarantee voluntary membership. A fixed network $G$ differs from a network formed under the contract.

Variants and changes of scope

  • Joint liability, meetings, information-sharing and mutual insurance are separate group-finance components.
  • Fixed groups versus endogenous formation change selection and incentives.
  • Simplicity can be a constraint or valued outcome, but its measure must be independently defined.

Audit conclusion and source relationship

[S1] and [S2] support simple contracts preserving valuable group/social-capital features. This frontier is editorial and overlaps mechanism design and game theory.

References and exact source roles

[S1] Jing Cai, Muhammad Meki, Simon Quinn, Erica Field, Cynthia Kinnan, Jonathan Morduch, Jonathan de Quidt, and Farah Said (2025). Microfinance. VoxDevLit 3(3), January 2025.

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/microfinance

[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: Microfinance. 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-202-A1 · Exploration · LLM-generated · gpt 6 astra ultra ·

GPT 6 Astra Ultra attempt on Econ-202

Simple lending contracts preserving group-finance benefits A separation result for liability and mutual insurance

Problem ID: G21-198. Status: Unresolved; restricted results and explicit gaps.

Scope and a predeclared complexity metric

The target asks which lending simplifications preserve social-support and group-formation benefits subject to borrower participation and lender viability. The VoxDev research-gap compilation [gaps], under microfinance, explicitly raises flexible contracts that retain simplicity, low costs and social capital; the accompanying review [micro] supplies the microfinance context. Our model isolates liability from mutual insurance. It proves a welfare comparison and an endogenous formation result for homogeneous borrowers, not a universal ranking of microfinance products.

There are finitely many ex ante identical borrowers, each with a project requiring external funding $I>0$. A project pays $H>0$ with probability $q\in(0,1)$ and zero otherwise; outcomes are independent across borrowers. Borrowers have a protected consumption endowment $b>0$ and a common increasing concave utility $u$ on $[b,b+H]$. There is no hidden effort, project selection or strategic default: project cash flows are verifiable, and stipulated payments can be legally enforced up to project output. The lender is risk neutral with zero safe rate and no outside subsidy.

Groups have two members. A separate, pre-existing mutual-insurance technology splits their residual project output equally after lender payments. It is an enforceable reciprocal agreement entered before risks are realized; the protected endowments are not pooled. It requires group membership but imposes no cross-liability to the lender. This strong commitment assumption is the precise social-support benefit retained in our simplification. It is not a claim that selfish borrowers voluntarily make all such payments after observing outcomes.

Define complexity $K$ as the number of distinct positive repayment rules for a successful borrower in the lender's contract, distinguished by whether they condition on peer outcomes. The zero repayment on own failure is a common rule and is not counted. Thus individual liability has $K=1$; the joint-liability contract below has $K=2$. The mutual-insurance rule is held fixed and excluded from this lender-contract metric. A metric counting every peer obligation would be a different problem.

Two break-even contracts with identical support

Initially take zero meeting and administration costs. Under individual liability the borrower pays $R_I=I/q$ on own success and zero on failure. Under joint liability the borrower pays $R_J$ if both succeed, $2R_J$ if she alone succeeds, and zero if she fails, where \[ R_J=\frac{I}{q(2-q)},\qquad R_I=(2-q)R_J. \tag{1} \] Assume $H\geq2R_J$, so every prescribed payment is feasible and $R_I<H$. Both contracts are followed by the same residual-output insurance. The resulting consumption of each group member is \[ \begin{array}{c|ccc} &\text{both succeed}&\text{one succeeds}&\text{neither succeeds}\\ \hline I&b+H-R_I&b+(H-R_I)/2&b\\ J&b+H-R_J&b+(H-2R_J)/2&b. \end{array} \tag{2} \] The state probabilities are $q^2$, $2q(1-q)$ and $(1-q)^2$.
Theorem (A simpler liability contract preserves and improves insurance): Both contracts have zero expected lender profit. Under the assumptions above, each borrower's expected utility under individual liability with mutual insurance is at least as large as under joint liability with the same insurance. The inequality is strict if $u$ is strictly concave. Individual liability with insurance also weakly improves utility relative to the same individual-liability loan held alone, with strict improvement under strict concavity. All these borrowers voluntarily accept the loan relative to not investing.
Proof: Individual expected repayment is $qR_I=I$. Under joint liability a borrower pays $R_J$ in the both-success state and $2R_J$ when she alone succeeds, so expected repayment is $q^2R_J+2q(1-q)R_J=q(2-q)R_J=I$. Limited liability follows from $H\geq2R_J>R_I$.

From (2), individual liability decreases each borrower's consumption in the high state by $\Delta=R_I-R_J=(1-q)R_J>0$ and increases consumption in the middle state by $\delta=R_J-R_I/2=qR_J/2>0$. The expected changes balance because $q^2\Delta=2q(1-q)\delta$. The entire middle-state increase interval lies below the entire high-state decrease interval: its upper endpoint is $b+(H-R_I)/2$, whereas the latter's lower endpoint is $b+H-R_I$. For a concave function the average slope on a lower interval is at least that on a higher interval. This follows directly from the defining three-point secant-slope inequality for concavity, applying it to the ordered endpoints; shared endpoints follow by continuity. Multiplying the two average slopes by the equal expected changes proves the expected utility comparison. Under strict concavity and disjoint nondegenerate intervals the secant inequality is strict; here $H-R_I>0$ and both changes are positive, so strictness follows.

Let $h=H-R_I>0$. A solo individual-liability borrower has expected utility $S=q\,u(b+h)+(1-q)u(b)$. In a pair with insurance, it is $G=q^2u(b+h)+2q(1-q)u(b+h/2)+(1-q)^2u(b)$. Subtracting gives \[ G-S=q(1-q)[2u(b+h/2)-u(b+h)-u(b)]\geq0, \tag{3} \] by concavity, strictly under strict concavity. Since every consumption in (2) is at least $b$, expected utility is at least $u(b)$ under either contract, and individual liability gives a positive gain for an increasing strict utility and $h>0$. Thus voluntary acceptance is satisfied. ∎

The comparison does not remove the support agreement. Joint liability collects more in the low-output mixed state and less in the both-success state; individual liability moves consumption toward the former while holding expected financing costs constant. Once mutual insurance is available separately, that extra lender contingency is unfavorable to concave borrower welfare in this restricted environment.

Meeting costs, participation, and endogenous formation

Now a group incurs a real meeting/administration cost $m\geq0$ per borrower, financed as part of the loan. Thus replace $I$ by $I+m$ for group contracts, but a solo borrower still finances only $I$. Assume $H\geq2(I+m)/[q(2-q)]$. Define \begin{align*} h_m&=H-(I+m)/q,\\ G_m&=q^2u(b+h_m)+2q(1-q)u(b+h_m/2)+(1-q)^2u(b),\\ S_0&=q\,u(b+H-I/q)+(1-q)u(b). \end{align*} Group lender receipts cover project and meeting resources exactly. The previous theorem, with $I+m$ in place of $I$, still compares the two group liability contracts, but joining now requires $G_m\geq S_0$.
Proposition (An exact and a sufficient participation test): The individual-liability group is voluntarily preferred to solo borrowing if and only if $G_m\geq S_0$. If additionally $u$ is differentiable with $0\leq u'\leq L$ on $[b,b+H]$, a sufficient condition is \[ q(1-q)[2u(b+h_m/2)-u(b+h_m)-u(b)]\geq Lm. \tag{4} \]
Proof: The exact criterion is the comparison of the two specified available payoffs. Let $S_m=q\,u(b+h_m)+(1-q)u(b)$ be solo-style utility at the group financing charge. Formula (3) gives $G_m-S_m$ as the left side of (4). Moreover \[ 0\leq S_0-S_m =q[u(b+H-I/q)-u(b+h_m)]\leq qL(m/q)=Lm, \] by integration of the derivative bound. Subtracting proves sufficiency. ∎

To define endogenous formation, let each borrower choose membership in at most one pair. Anyone unpaired obtains $S_0$ by solo borrowing; each member of a pair using the simple contract obtains $G_m$. Membership is voluntary before outcomes and the mutual-insurance agreement is binding afterward. A matching is stable when no current member strictly gains by leaving alone, and no two borrowers strictly gain by leaving their current arrangements and pairing with each other. Identities carry no additional cost or benefit in this homogeneous model.

Theorem (Formation and the restricted complexity frontier): If $G_m>S_0$, the stable matchings are exactly those leaving at most one borrower unpaired. Such matchings exist for every finite population, and all have identical aggregate expected utility. If $G_m<S_0$, the unique stable allocation has everyone borrowing alone. If $G_m=S_0$, every matching is stable under the stated strict-blocking convention. Among the menu consisting of solo borrowing, the simple insured group, and the joint-liability insured group, any complexity budget allowing the simple group cannot improve its attainable borrower welfare by also allowing joint liability.
Proof: When $G_m>S_0$, no matched member wants to leave, while any two unmatched borrowers would both strictly gain by pairing. Hence stability requires at most one unmatched borrower. Conversely, in a matching with at most one unmatched borrower, any prospective new pair includes at least one already matched borrower, whose payoff would remain $G_m$ and thus would not strictly improve. There is no blocking pair or unilateral exit. Pairing consecutively in any ordering constructs such a matching; the number of pairs is always $\lfloor n/2\rfloor$, so aggregate utility is identical. If $G_m<S_0$, every matched member strictly prefers exit, whereas no unmatched pair wishes to form. At equality neither exit nor pairing strictly improves either payoff, proving the last stability case.

For the menu comparison, Theorem 1 applies to the two group contracts at the same cost $I+m$ and proves weak dominance of the simple group for each member, with equal lender viability. If joint-liability membership beats solo borrowing, simple membership does also. If it does not, it cannot improve the menu's attainable utility. Thus adding the more complex contract does not improve the frontier, including endogenous membership. ∎

The theorem concerns strict pairwise blocking, homogeneous borrowers and an anonymous support technology. With heterogeneous partner preferences, matching existence and welfare are separate problems. In particular it would be incorrect to maximize welfare on a fixed network and treat that network as automatically formed under the contract.

Why preserving the support technology is substantive

Proposition (No unilateral sharing with selfish terminal preferences): Suppose the mixed-outcome insurance transfer is not enforceable, there is no future interaction or social sanction, and the successful borrower maximizes only her own strictly increasing terminal utility. Under either loan contract, if her residual output is positive, her unique voluntary transfer to the failed peer is zero. The equal-sharing allocation is then not an equilibrium of that transfer stage.
Proof: If residual output is $h>0$, giving $t\in[0,h]$ leaves consumption $b+h-t$. Strictly increasing utility makes every $t>0$ strictly worse than zero. The equal-sharing transfer is $h/2>0$, so it fails individual optimality. The conclusion uses no claim about earlier membership choices. ∎
This obstruction identifies the main remaining mechanism-design problem: a simple lender rule can preserve insurance only when an independent support arrangement is credible. Social sanctions, altruism, repeated interaction or a formal reciprocal agreement require their own incentives and resource costs. Hidden effort could also make joint liability valuable through monitoring, which this model excludes. Estimating those channels, allowing heterogeneous endogenous groups and validating a comprehension metric remain unresolved. No experimental effect or numerical simulation is claimed.

References

Completion Estimate

30%

Go to contribution →