pith. sign in

arxiv: 2511.07431 · v2 · submitted 2025-10-30 · 💱 q-fin.RM · math.OC· math.PR

Optimal Cash Transfers and Microinsurance to Reduce Social Protection Costs

Pablo Azcue , Corina Constantinescu , Jos\'e Miguel Flores-Contr\'o , Nora Muler This is my paper

Pith reviewed 2026-05-18 03:49 UTC · model grok-4.3

classification 💱 q-fin.RM math.OCmath.PR
keywords cash transferspoverty reductionsocial protectionoptimal controlmicroinsurancedynamic programmingcapital injectioncost minimization
0
0 comments X

The pith

Cash transfers minimize total costs when provided before households reach the poverty line rather than after.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper sets up an optimization problem for the cheapest way to keep a household's capital above the poverty line by choosing when and how much to inject. It models capital changes as a stochastic process with sudden drops and uses dynamic programming to derive the best policy. The resulting equation has a solution that can be computed numerically or exactly in simple cases. Examples show the lowest expected cost occurs by acting while capital remains above the line. This matters in low-resource settings because it points to preventive aid as a way to stretch limited budgets farther than emergency responses alone.

Core claim

Solving the optimal control problem for capital injections yields a value function that satisfies the associated Hamilton-Jacobi-Bellman equation in the viscosity sense, and numerical results indicate that the cost-minimizing policy injects capital at a level strictly above the poverty threshold.

What carries the argument

The Hamilton-Jacobi-Bellman equation obtained from the dynamic programming principle applied to the problem of choosing injection amounts to minimize expected discounted costs of staying above the poverty line.

If this is right

  • Social protection budgets in low-income countries can be used more efficiently by shifting toward preventive capital injections.
  • Closed-form solutions for the optimal injection amounts exist in certain simplified versions of the capital process.
  • Microinsurance products can serve as a complement that reduces the frequency or size of needed capital transfers.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same optimization approach could be adapted to design timing rules for other aid types such as food or health support.
  • Collecting local data on how capital actually fluctuates would allow testing whether the modeled loss process matches observed poverty entries.
  • Combining the injection policy with insurance uptake incentives might produce a hybrid program with even lower overall costs.

Load-bearing premise

Household capital changes over time according to a specific stochastic process that includes sudden proportional losses capable of driving it below the poverty line.

What would settle it

If direct cost calculations or real household data show that injecting capital exactly at or below the poverty line produces lower total expected spending than the policy of acting above it, the optimality claim would not hold.

Figures

Figures reproduced from arXiv: 2511.07431 by Corina Constantinescu, Jos\'e Miguel Flores-Contr\'o, Nora Muler, Pablo Azcue.

Figure 1
Figure 1. Figure 1: (a) Upper boundary of the region defined by the constraint [PITH_FULL_IMAGE:figures/full_fig_p009_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The first row (a)–(c) shows the cost of social protection and the value function of a [PITH_FULL_IMAGE:figures/full_fig_p021_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The optimal threshold y ⋆ when Zi ∼ Beta(α, 1), a = 0.10, b = 3, c = 0.40, λ = 1, x ∗ = 20 for different values of the discount rate δ [PITH_FULL_IMAGE:figures/full_fig_p022_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: The first row (a)–(c) shows the cost of social protection and the value function of a [PITH_FULL_IMAGE:figures/full_fig_p024_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: The optimal threshold y ⋆ when Zi ∼ Kumaraswamy(p, 4), a = 0.10, b = 3, c = 0.40, λ = 1, x ∗ = 20 for different values of the discount rate δ [PITH_FULL_IMAGE:figures/full_fig_p025_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: The value function of a threshold strategy with the optimal threshold, [PITH_FULL_IMAGE:figures/full_fig_p029_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: The value function of a threshold strategy with the optimal threshold, [PITH_FULL_IMAGE:figures/full_fig_p031_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: The value function of a threshold strategy with the optimal threshold, [PITH_FULL_IMAGE:figures/full_fig_p033_8.png] view at source ↗
read the original abstract

Design and implementation of appropriate social protection strategies is one of the main targets of the United Nation's Sustainable Development Goal (SDG) 1: No Poverty. Cash transfer (CT) programmes are considered one of the main social protection strategies and an instrument for achieving SDG 1. Targeting consists of establishing eligibility criteria for beneficiaries of CT programmes. In low-income countries, where resources are limited, proper targeting of CTs is essential for an efficient use of resources. Given the growing importance of microinsurance as a complementary tool to social protection strategies, this study examines its role as a supplement to CT programmes. In this article, we adopt the piecewise-deterministic Markov process introduced in Kovacevic and Pflug (2011) to model the capital of a household, which when exposed to proportional capital losses (in contrast to the classical Cram\'er-Lundberg model) can push them into the poverty area. Striving for cost-effective CT programmes, we optimise the expected discounted cost of keeping the household's capital above the poverty line by means of injection of capital (as a direct capital transfer). Using dynamic programming techniques, we derive the Hamilton-Jacobi-Bellman (HJB) equation associated with the optimal control problem of determining the amount of capital to inject over time. We show that this equation admits a viscosity solution that can be approximated numerically. Moreover, in certain special cases, we obtain closed-form expressions for the solution. Numerical examples show that there is an optimal level of injection above the poverty threshold, suggesting that efficient use of resources is achieved when CTs are preventive rather than reactive, since injecting capital into households when their capital levels are above the poverty line is less costly than to do so only when it falls below the threshold.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. The paper models household capital via the piecewise-deterministic Markov process with proportional losses from Kovacevic and Pflug (2011). It formulates a stochastic control problem to minimize the expected discounted cost of capital injections (cash transfers) that keep capital above a poverty threshold, derives the associated HJB equation via dynamic programming, establishes that the equation admits a viscosity solution approximable numerically, obtains closed-form solutions in special cases, and presents numerical examples indicating that the optimal policy injects capital above the threshold (preventive rather than reactive CTs). Microinsurance is examined as a complementary instrument.

Significance. If the numerical findings are reliable, the work supplies a rigorous justification for preferring preventive cash transfers over purely reactive ones in resource-constrained settings, with potential implications for cost-effective implementation of SDG 1. The derivation of the HJB equation from first principles and the existence result for the viscosity solution constitute a solid technical contribution; the closed-form expressions in special cases are a clear analytical strength.

major comments (1)
  1. [Numerical examples] Numerical examples section: the approximation of the viscosity solution to the HJB equation is described only as 'can be approximated numerically,' with no specification of the discretization scheme (finite differences, policy iteration, etc.), spatial or temporal grid resolution, treatment of the proportional-loss jumps, or any convergence verification, a posteriori error bounds, or sensitivity checks. Because the central claim—that an optimal injection level exists above the poverty threshold and that preventive CTs are less costly than reactive ones—rests directly on these examples, the absence of these details prevents independent confirmation that the reported policy is a property of the true value function rather than a numerical artifact.
minor comments (2)
  1. [Abstract] The abstract states that closed forms exist 'in certain special cases' but does not indicate which parameter regimes or sections contain these expressions.
  2. [Model] A short self-contained recap of the key parameters of the Kovacevic–Pflug PDMP (intensity of proportional losses, discount rate) in the model section would improve readability.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the positive assessment of our technical contributions and the potential policy implications. We address the single major comment below and will revise the manuscript accordingly.

read point-by-point responses

  1. Referee: Numerical examples section: the approximation of the viscosity solution to the HJB equation is described only as 'can be approximated numerically,' with no specification of the discretization scheme (finite differences, policy iteration, etc.), spatial or temporal grid resolution, treatment of the proportional-loss jumps, or any convergence verification, a posteriori error bounds, or sensitivity checks. Because the central claim—that an optimal injection level exists above the poverty threshold and that preventive CTs are less costly than reactive ones—rests directly on these examples, the absence of these details prevents independent confirmation that the reported policy is a property of the true value function rather than a numerical artifact.

    Authors: We agree that additional details are required for reproducibility and to confirm that the reported optimal policy is not a numerical artifact. In the revised manuscript we will expand the numerical examples section to describe the discretization scheme (a finite-difference approximation of the HJB equation solved via policy iteration), the spatial grid (uniform partition of the capital domain with explicit step size and number of points), the handling of proportional-loss jumps (by direct evaluation of the jump integral on the grid), convergence verification (results under successive grid refinements), and sensitivity checks with respect to the discount rate and loss intensity. These additions will directly address the referee’s concern and strengthen the numerical support for the advantage of preventive injections. revision: yes

Circularity Check

0 steps flagged

No circularity; derivation uses external model and standard dynamic programming

full rationale

The paper adopts the piecewise-deterministic Markov process directly from the external reference Kovacevic and Pflug (2011) without modification or self-referential fitting. It then applies standard dynamic programming to formulate the optimal control problem of minimizing expected discounted injection costs, derives the associated HJB equation, proves existence of a viscosity solution, obtains closed-form expressions in special cases, and performs numerical approximation in others. The reported finding of an optimal injection level above the poverty threshold is a computed outcome of this optimization rather than a quantity defined by or equivalent to the inputs by construction. No self-citations appear as load-bearing premises, no parameters are fitted to subsets of data and then relabeled as predictions, and no ansatz or uniqueness result is smuggled in via the authors' prior work. The derivation chain is therefore self-contained against the external benchmark model.

Axiom & Free-Parameter Ledger

2 free parameters · 2 axioms · 0 invented entities

The central claim rests on the imported capital process, standard stochastic-control assumptions, and numerical approximation whose accuracy is not quantified in the abstract.

free parameters (2)
  • discount rate
    Standard in expected discounted cost problems; value not stated in abstract but required for the HJB equation.
  • proportional loss intensity
    Parameter of the piecewise-deterministic Markov process taken from the 2011 reference; controls shock size.
axioms (2)
  • domain assumption Household capital evolves as a piecewise-deterministic Markov process with proportional losses
    Invoked in the second paragraph of the abstract as the modeling choice adopted from Kovacevic and Pflug (2011).
  • standard math The associated HJB equation admits a viscosity solution that can be approximated numerically
    Stated as a result obtained via dynamic programming techniques.

pith-pipeline@v0.9.0 · 5866 in / 1526 out tokens · 44861 ms · 2026-05-18T03:49:02.493623+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

33 extracted references · 33 canonical work pages

  1. [1]

    Albrecher, H., H. U. Gerber, and E. S. W. Shiu (2011). The Optimal Dividend Barrier in the Gamma–Omega Model.European Actuarial Journal 1(1), 43–55

    work page 2011

  2. [2]

    Asmussen, S. and H. Albrecher (2010).Ruin Probabilities. Singapore: World Scientific

    work page 2010

  3. [3]

    Azaïs, R. and A. Genadot (2015). Semi-Parametric Inference for the Absorption Features of a Growth-Fragmentation Model.TEST 24(2), 341–360. 34

    work page 2015

  4. [4]

    Azcue, P. and N. Muler (2014).Stochastic Optimization in Insurance: A Dynamic Programming Approach. New York Heidelberg Dordrecht London: Springer

    work page 2014

  5. [5]

    Systems & Control: Foundations & Applications

    Bardi, M.andI.Capuzzo-Dolcetta(1997).Optimal Control and Viscosity Solutions of Hamilton- Jacobi-Bellman Equations. Systems & Control: Foundations & Applications. New York: Springer Science+Business Media

    work page 1997

  6. [6]

    Bellman, R. (1954). Dynamic Programming and a New Formalism in the Calculus of Variations. Proceedings of the National Academy of Sciences 40(4), 231–235

    work page 1954

  7. [7]

    (2006).Protecting the Poor: A Microinsurance Compendium — Volume I

    Churchill, C. (2006).Protecting the Poor: A Microinsurance Compendium — Volume I. Geneva, Switzerland: International Labour Organization (ILO)

    work page 2006

  8. [8]

    Churchill, C. and M. Matul (2012).Protecting the Poor: A Microinsurance Compendium — Volume II. Geneva, Switzerland: International Labour Organization (ILO)

    work page 2012

  9. [9]

    Crandall, M. G. and P.-L. Lions (1983). Viscosity Solutions of Hamilton-Jacobi Equations. Transactions of the American Mathematical Society 277(1), 1–42

    work page 1983

  10. [10]

    Social Protection in Poor Coun- tries.Social Protection Briefing Note Series 1

    Department for International Development (DFID) (2006). Social Protection in Poor Coun- tries.Social Protection Briefing Note Series 1

    work page 2006

  11. [11]

    Farrington, J. and R. Slater (2006). Introduction: Cash Transfers: Panacea for Poverty Re- duction or Money Down the Drain?Development Policy Review 24(5), 499–511

    work page 2006

  12. [12]

    Pieri, E

    Firpo, S., R. Pieri, E. Pedroso, and A. P. Souza (2014). Evidence of Eligibility Manipulation for Conditional Cash Transfer Programs.EconomiA 15(3), 243–260

    work page 2014

  13. [13]

    Fleming, W. H. and H. M. Soner (2006).Controlled Markov Processes and Viscosity Solutions. United States of America: Springer Science+Business Media, Inc

    work page 2006

  14. [14]

    Flores-Contró, J. M. (2025). The Gerber-Shiu Expected Discounted Penalty Function: An Application to Poverty Trapping.Scandinavian Actuarial Journal, 1–22

    work page 2025

  15. [15]

    Flores-Contró, J. M. and S. Arnold (2024). The Role of Direct Capital Cash Transfers Towards Poverty and Extreme Poverty Alleviation - An Omega Risk Process.Scandinavian Actuarial Journal 2024(8), 781–812

    work page 2024

  16. [16]

    Flores-Contró, J. M., K. Henshaw, S.-H. Loke, S. Arnold, and C. Constantinescu (2025). Subsidizing Inclusive Insurance to Reduce Poverty.North American Actuarial Journal 29(1), 44–73

    work page 2025

  17. [17]

    Garcia, M. and C. M. T. Moore (2012).The Cash Dividend: The Rise of Cash Transfer Programs in Sub-Saharan Africa. Directions in Development - Human Development. Washington, D.C.: The World Bank

    work page 2012

  18. [18]

    Haughton, J

    Habimana, D., J. Haughton, J. Nkurunziza, and D. M.-A. Haughton (2021). Measuring the Impact of Unconditional Cash Transfers on Consumption and Poverty in Rwanda.World Devel- opment Perspectives 23, 100341

    work page 2021

  19. [19]

    Handa, S. and B. Davis (2006). The Experience of Conditional Cash Transfers in Latin America and the Caribbean.Development Policy Review 24(5), 513–536. 35

    work page 2006

  20. [20]

    (2005).Cash and Vouchers in Emergencies

    Harvey, P. (2005).Cash and Vouchers in Emergencies. London, United Kingdom: Overseas Development Institute

    work page 2005

  21. [21]

    Henshaw, K., J. M. Ramirez, J. M. Flores-Contró, E. A. Thomann, S. H. Loke, and C. D. Constantinescu (2023). On the Impact of Insurance on Households Susceptible to Random Pro- portional Losses: An Analysis of Poverty Trapping.Working Paper

    work page 2023

  22. [22]

    Issues Paper on Conduct of Business in Inclusive Insurance

    International Association of Insurance Supervisors (IAIS) (2015). Issues Paper on Conduct of Business in Inclusive Insurance

    work page 2015

  23. [23]

    Keen, M. (1992). Needs and Targeting.The Economic Journal 102(410), 67–79

    work page 1992

  24. [24]

    Kovacevic, R. M. and G. C. Pflug (2011). Does Insurance Help to Escape the Poverty Trap? — A Ruin Theoretic Approach.Journal of Risk and Insurance 78(4), 1003–1027

    work page 2011

  25. [25]

    Owusu-Addo, E., A. M. N. Renzaho, P. Sarfo-Mensah, Y. A. Sarpong, W. Niyuni, and B. J. Smith (2023). Sustainability of Cash Transfer Programs: A Realist Case Study.Poverty & Public Policy 15(2), 173–198

    work page 2023

  26. [26]

    (1992).Stochastic Integration and Differential Equations: A New Approach

    Protter, P. (1992).Stochastic Integration and Differential Equations: A New Approach. New York: Springer-Verlag Berlin Heidelberg

    work page 1992

  27. [27]

    (2007).Stochastic Control in Insurance

    Schmidli, H. (2007).Stochastic Control in Insurance. London: Springer

    work page 2007

  28. [28]

    Slater, R. (2011). Cash Transfers, Social Protection and Poverty Reduction.International Journal of Social Welfare 20(3), 250–259

    work page 2011

  29. [29]

    Slater, R. and J. Farrington (2010).Appropriate, Achievable and Acceptable: A Practical Tool for Good Targeting. London: ODI

    work page 2010

  30. [30]

    Soner, H. M. (1988). Optimal Control of Jump-Markov Processes and Viscosity Solutions. InStochastic Differential Systems, Stochastic Control Theory and Applications, pp. 501–511. Springer-Verlag

    work page 1988

  31. [31]

    Tabor, S. R. (2002). Assisting the Poor with Cash: Design and Implementation of Social Transfer Programs.World Bank Social Protection Discussion Paper 223, 79–97

    work page 2002

  32. [32]

    Number A/RES/70/1

    United Nations (2015).Transforming Our World: The 2030 Agenda for Sustainable Develop- ment. Number A/RES/70/1. New York: United Nations. Appendices A Mathematical Proofs A.1 Proof of Lemma 3.1 Proof.BydefinitionC(x)isnon-increasingandnon-negative. Letusshownowthatlim x→∞ C(x) =

    work page 2015

  33. [33]

    nX k=1 C(X τk)e −δτk 1 τ=τ k # ≤C(0)·E

    We have the following, E e−δτn 1 {τ≥τ n} ≤E e−δτn =E " e −δ nP i=1 (τi−τi−1) # = E e−δτ1 n = δ δ+λ n , 36 which yields:lim n→∞ E e−δτn 1 {τ≥τ n} = 0. SinceC(x)is non-increasing, we have, C(x)≤ nX k=1 C(X τk)e −δτk 1 τ=τ k +C(0)·E e−δτn 1 τ >τn , which leads to the following inequality, 0≤C(x)≤lim n→∞ E " nX k=1 C(X τk)e −δτk 1 {τ=τ k} # . Now, if τ≥τ 1, X...

    work page 1992