Topping Up and Optimal Redistribution

Mitchell Watt; Zi Yang Kang

arxiv: 2606.28919 · v1 · pith:QBPY37SAnew · submitted 2026-06-27 · 💰 econ.GN · econ.TH· q-fin.EC

Topping Up and Optimal Redistribution

Zi Yang Kang , Mitchell Watt This is my paper

Pith reviewed 2026-06-30 08:27 UTC · model grok-4.3

classification 💰 econ.GN econ.THq-fin.EC

keywords topping upin-kind transfersoptimal redistributionnonlinear pricingscreeningredistributive prioritydemand correlationpublic intervention

0 comments

The pith

Topping up leaves the optimal redistribution mechanism unchanged when demand correlates positively with redistributive priority, but weakens screening and reduces redistribution when the correlation is negative.

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

This paper examines how allowing consumers to supplement in-kind transfers in a private market changes what a social planner can achieve with its own nonlinear price schedule. The central result is that the impact turns on the sign of the correlation between a consumer's redistributive priority and their demand for the good. A reader would care because real redistribution programs often face this topping-up choice, and the finding shows when the choice alters the amount or targeting of transfers. The analysis also maps out how topping up shrinks the cases where any public intervention is worthwhile and changes who receives support inside those cases.

Core claim

When the correlation between redistributive priority and demand is positive, topping up does not affect the optimal mechanism. When the correlation is negative, topping up weakens screening and reduces redistribution. At the extensive margin, topping up reduces the set of environments in which intervention is optimal. At the intensive margin, topping up weakly reduces both the scope of a free public option and the mass of consumers served, and shifts redistribution away from the consumers with the highest redistributive priority. The paper fully characterizes the optimal mechanisms and the changed comparative statics with respect to redistributive priorities.

What carries the argument

The sign of the correlation between redistributive priority and demand, which governs whether topping up preserves or erodes the planner's screening power through its nonlinear price schedule.

If this is right

When priority and demand correlate positively, the planner's optimal nonlinear schedule remains the same whether topping up is allowed or banned.
When priority and demand correlate negatively, allowing topping up strictly lowers total redistribution and shifts it away from highest-priority consumers.
Topping up shrinks the range of parameter values in which any public intervention is better than leaving the market alone.
Inside environments where intervention occurs, topping up reduces the size of any free public option and the number of consumers served by it.

Where Pith is reading between the lines

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

Policy designers facing negative correlation may prefer to restrict topping up to protect targeting of high-priority groups.
The same logic could apply to other public-private competition settings, such as public health insurance alongside private markets.
Empirical work could test whether observed topping-up rules align with the sign of the priority-demand correlation in each market.

Load-bearing premise

Consumers can buy the good in a competitive private market while the planner offers a separate nonlinear price schedule, and redistributive priority is a fixed attribute that can be correlated with demand.

What would settle it

Measure the empirical correlation between demand and redistributive priority for a transferred good, then compare the planner's optimal transfers and who receives them in otherwise identical environments with and without topping up allowed.

Figures

Figures reproduced from arXiv: 2606.28919 by Mitchell Watt, Zi Yang Kang.

**Figure 2.** Figure 2: Structure of the optimal allocation function under positive correlation characterized by Theorem [PITH_FULL_IMAGE:figures/full_fig_p018_2.png] view at source ↗

**Figure 3.** Figure 3: Structure of the optimal allocation function under negative correlation characterized by Theorem [PITH_FULL_IMAGE:figures/full_fig_p025_3.png] view at source ↗

**Figure 4.** Figure 4: Proof sketch of Theorem 1 with topping up. 38 [PITH_FULL_IMAGE:figures/full_fig_p039_4.png] view at source ↗

**Figure 5.** Figure 5: Proof sketch of Theorem 1 without topping up. No topping up. We now consider the case of no topping up. As before, we consider a small subsidy Δ𝜎 > 0 that is targeted at consumers of types 𝜃 ∈ [𝜃,ˆ 𝜃ˆ + Δ𝜃], as shown in [PITH_FULL_IMAGE:figures/full_fig_p042_5.png] view at source ↗

read the original abstract

This paper studies how topping up -- allowing recipients of in-kind transfers to supplement subsidized consumption in a private market -- affects optimal redistribution. Consumers can access a competitive private market, while a social planner offers an alternative nonlinear price schedule. We show that the effect of topping up depends on the correlation between redistributive priority and demand. When the correlation is positive, topping up does not affect the optimal mechanism. When the correlation is negative, topping up weakens screening and reduces redistribution. At the extensive margin, topping up reduces the set of environments in which intervention is optimal. At the intensive margin, topping up weakly reduces both the scope of a free public option and the mass of consumers served, and shifts redistribution away from the consumers with the highest redistributive priority. We characterize the optimal mechanisms and show how topping up changes the comparative statics of optimal redistribution with respect to redistributive priorities.

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.

Desk Editor's Note private letter to a colleague

The paper's main result is that topping up leaves the optimal mechanism unchanged under positive correlation between priority and demand but weakens screening and redistribution under negative correlation.

read the letter

The colleague should know that this paper isolates a clean comparative-static distinction: when redistributive priority and demand correlate positively, allowing topping up in a competitive private market does not change the planner's optimal nonlinear schedule. When the correlation is negative, topping up relaxes some incentive constraints, reduces the power of screening, and lowers the amount of redistribution achieved. It also shows that topping up shrinks the set of environments where any intervention is optimal and, inside those environments, trims the free public option, the mass of consumers reached, and the share going to the highest-priority types.

The work does a solid job of embedding the topping-up option inside a standard mechanism-design model with a separate planner schedule and then deriving how the sign of the correlation flips the outcome. The characterization of the optimal mechanisms in each case is stated directly and the extensive-margin and intensive-margin effects are tracked separately. The modeling choices line up with existing literature on in-kind transfers, so the distinction by correlation sign comes through without obvious internal contradictions.

The soft spots are mostly about verification rather than fatal gaps. The abstract and stress-test note give no derivations, so it is not yet possible to check how many regularity conditions are needed for the positive-correlation invariance result or exactly which constraints bind in the negative case. If the proofs rely on strong assumptions about the joint distribution of types, that would narrow the practical reach. No free parameters or invented entities appear in the summary, which helps.

This is a paper for public-economics and mechanism-design readers who work on welfare-program design and nonlinear pricing. Anyone already using type correlations in redistribution models will find the comparative statics usable. It shows clear engagement with the setup and deserves a serious referee because the result is specific enough to be worth checking the proofs even if revisions follow.

Referee Report

0 major / 1 minor

Summary. The paper studies how allowing topping up of in-kind transfers via a competitive private market affects a social planner's optimal nonlinear pricing mechanism for redistribution. It establishes that the effect hinges on the sign of the correlation between redistributive priority and demand: positive correlation leaves the optimal mechanism unchanged, while negative correlation weakens screening and reduces redistribution. The analysis covers extensive-margin effects (smaller set of environments where intervention is optimal) and intensive-margin effects (reduced scope of free public options, lower mass served, and shifts away from highest-priority consumers), with full characterization of the mechanisms and altered comparative statics.

Significance. If the results hold, the paper delivers a clean, correlation-sign-dependent comparative static that clarifies when private-market topping up is neutral versus distortionary for redistribution. This distinction is policy-relevant for the design of in-kind transfers versus cash or mixed mechanisms and extends standard screening models by incorporating redistributive weights explicitly correlated with demand.

minor comments (1)

The abstract refers to 'characteriz[ing] the optimal mechanisms' but does not indicate whether closed-form solutions, numerical examples, or explicit conditions on the correlation parameter are provided; adding a brief statement on the form of the characterization would improve readability.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive summary, significance assessment, and recommendation of minor revision. No specific major comments were listed in the report.

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The provided abstract and skeptic summary describe a standard theoretical characterization of optimal nonlinear pricing mechanisms under topping-up, with results conditioned on the sign of a correlation parameter between redistributive priority and demand. No equations, fitted parameters, self-citations, or ansatzes are exhibited that would reduce any claimed prediction to an input by construction. The derivation chain is therefore self-contained against external benchmarks and receives the default non-finding.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract supplies no explicit free parameters, axioms, or invented entities; analysis is limited to high-level claims.

pith-pipeline@v0.9.1-grok · 5671 in / 1030 out tokens · 61265 ms · 2026-06-30T08:27:46.639362+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

5 extracted references

[1]

Observe that the social planner’s objective is continuous in(𝑈, 𝜈)

Existence of solution.To see that an optimal solution to each problem exists, we endowI with 36 the metric 𝑑 (𝜈1, 𝜈2) = ∥𝜈1 − 𝜈2∥ 𝐿1 + | 𝜈1(𝜃) − 𝜈2(𝜃)| . Observe that the social planner’s objective is continuous in(𝑈, 𝜈). Without loss of generality, we focus on the set𝐾 ≔ [𝑈LF, 𝜃𝑣 ( 𝐴)] × n ℎ ∈ I : [𝜃, 𝜃] → [ 𝑣 (0), 𝑣 ( 𝐴)] o . By the Helly selection theo...
[2]

Wenowarguethattheoptimalsolutiontoeachproblemisunique when E[𝜔] ≠ 𝛼

Generaluniquenessofsolution. Wenowarguethattheoptimalsolutiontoeachproblemisunique when E[𝜔] ≠ 𝛼. Suppose that (𝑈1, 𝜈1) and (𝑈2, 𝜈2) are distinct optimal solutions with distinct allocation functions 𝜈1 ≠ 𝜈2, and consider(𝑈∗, 𝜈∗) = (𝑈1 + 𝑈2)/2, (𝜈1 + 𝜈2)/2 . Clearly, 𝜈∗ is nondecreasing,and (𝑈∗, 𝜈∗)satisfiestheconstraints(sincetheconstraintsarelinear). How...
[3]

For any stochastic mechanism, let𝜈†(𝜃) be the expected subutility allocated to type𝜃, and let𝑈† be the expected utility of the lowest type

Optimality of deterministic mechanisms.Next, we observe that the restriction to deterministic mechanisms entails no loss of generality. For any stochastic mechanism, let𝜈†(𝜃) be the expected subutility allocated to type𝜃, and let𝑈† be the expected utility of the lowest type. The constraints are preserved by linearity, and the deterministic mechanism(𝑈†, 𝜈...
[4]

Toillustrate,suppose 37 that the (IR) constraint binds for the highest type

Convex program with majorization constraints.Finally, we point out that the problem without toppingupcanbewrittenasaconvexprogramwithmajorizationconstraints. Toillustrate,suppose 37 that the (IR) constraint binds for the highest type. Then the (IR) constraints can be rewritten as ∫ 𝜃 𝜃 𝜈(𝑠) d𝑠 ≤ ∫ 𝜃 𝜃 𝜈LF(𝑠) d𝑠, for all𝜃 ∈ [ 𝜃, 𝜃]. Since 𝜈 and 𝜈LF are non...

2021
[5]

Moreover,for 𝜃 ∈ [ 𝜃 𝐿, 𝜅] and 𝑧 ≥ 𝜃,choose 𝑦 = 𝜃 inthemax-minformula

implies that, for𝜃 ∈ [ 𝜅, 𝜃], 𝐽𝜇∗ | [𝜃 𝐿,𝜃] (𝜃) = inf 𝑧∈[ 𝜃,𝜃] sup 𝑦∈[ 𝜃 𝐿,𝜃] ∫ [𝑦,𝑧] 𝐽𝜇∗ d𝐹 𝐹 (𝑧) − 𝐹 (𝑦) ≥ inf 𝑧∈[ 𝜃,𝜃] sup 𝑦∈[ 𝜅,𝜃 ] ∫ [𝑦,𝑧] 𝐽𝜇∗ d𝐹 𝐹 (𝑧) − 𝐹 (𝑦) = 𝐽𝜇∗ | [𝜅,𝜃] (𝜃) ≥ 𝜃. Moreover,for 𝜃 ∈ [ 𝜃 𝐿, 𝜅] and 𝑧 ≥ 𝜃,choose 𝑦 = 𝜃 inthemax-minformula. If 𝑧 ≤ 𝜅,equation(14) gives ∫ 𝑧 𝜃 𝐽𝜇∗ (𝑠) d𝐹 (𝑠) ≥ 𝜃 [𝐹 (𝑧) − 𝐹 (𝜃)] . If 𝑧 > 𝜅 , equation (14) gi...

1969

[1] [1]

Observe that the social planner’s objective is continuous in(𝑈, 𝜈)

Existence of solution.To see that an optimal solution to each problem exists, we endowI with 36 the metric 𝑑 (𝜈1, 𝜈2) = ∥𝜈1 − 𝜈2∥ 𝐿1 + | 𝜈1(𝜃) − 𝜈2(𝜃)| . Observe that the social planner’s objective is continuous in(𝑈, 𝜈). Without loss of generality, we focus on the set𝐾 ≔ [𝑈LF, 𝜃𝑣 ( 𝐴)] × n ℎ ∈ I : [𝜃, 𝜃] → [ 𝑣 (0), 𝑣 ( 𝐴)] o . By the Helly selection theo...

[2] [2]

Wenowarguethattheoptimalsolutiontoeachproblemisunique when E[𝜔] ≠ 𝛼

Generaluniquenessofsolution. Wenowarguethattheoptimalsolutiontoeachproblemisunique when E[𝜔] ≠ 𝛼. Suppose that (𝑈1, 𝜈1) and (𝑈2, 𝜈2) are distinct optimal solutions with distinct allocation functions 𝜈1 ≠ 𝜈2, and consider(𝑈∗, 𝜈∗) = (𝑈1 + 𝑈2)/2, (𝜈1 + 𝜈2)/2 . Clearly, 𝜈∗ is nondecreasing,and (𝑈∗, 𝜈∗)satisfiestheconstraints(sincetheconstraintsarelinear). How...

[3] [3]

For any stochastic mechanism, let𝜈†(𝜃) be the expected subutility allocated to type𝜃, and let𝑈† be the expected utility of the lowest type

Optimality of deterministic mechanisms.Next, we observe that the restriction to deterministic mechanisms entails no loss of generality. For any stochastic mechanism, let𝜈†(𝜃) be the expected subutility allocated to type𝜃, and let𝑈† be the expected utility of the lowest type. The constraints are preserved by linearity, and the deterministic mechanism(𝑈†, 𝜈...

[4] [4]

Toillustrate,suppose 37 that the (IR) constraint binds for the highest type

Convex program with majorization constraints.Finally, we point out that the problem without toppingupcanbewrittenasaconvexprogramwithmajorizationconstraints. Toillustrate,suppose 37 that the (IR) constraint binds for the highest type. Then the (IR) constraints can be rewritten as ∫ 𝜃 𝜃 𝜈(𝑠) d𝑠 ≤ ∫ 𝜃 𝜃 𝜈LF(𝑠) d𝑠, for all𝜃 ∈ [ 𝜃, 𝜃]. Since 𝜈 and 𝜈LF are non...

2021

[5] [5]

Moreover,for 𝜃 ∈ [ 𝜃 𝐿, 𝜅] and 𝑧 ≥ 𝜃,choose 𝑦 = 𝜃 inthemax-minformula

implies that, for𝜃 ∈ [ 𝜅, 𝜃], 𝐽𝜇∗ | [𝜃 𝐿,𝜃] (𝜃) = inf 𝑧∈[ 𝜃,𝜃] sup 𝑦∈[ 𝜃 𝐿,𝜃] ∫ [𝑦,𝑧] 𝐽𝜇∗ d𝐹 𝐹 (𝑧) − 𝐹 (𝑦) ≥ inf 𝑧∈[ 𝜃,𝜃] sup 𝑦∈[ 𝜅,𝜃 ] ∫ [𝑦,𝑧] 𝐽𝜇∗ d𝐹 𝐹 (𝑧) − 𝐹 (𝑦) = 𝐽𝜇∗ | [𝜅,𝜃] (𝜃) ≥ 𝜃. Moreover,for 𝜃 ∈ [ 𝜃 𝐿, 𝜅] and 𝑧 ≥ 𝜃,choose 𝑦 = 𝜃 inthemax-minformula. If 𝑧 ≤ 𝜅,equation(14) gives ∫ 𝑧 𝜃 𝐽𝜇∗ (𝑠) d𝐹 (𝑠) ≥ 𝜃 [𝐹 (𝑧) − 𝐹 (𝜃)] . If 𝑧 > 𝜅 , equation (14) gi...

1969