How Wasteful is Signaling?

Alex Frankel; Navin Kartik

arxiv: 2601.14454 · v2 · submitted 2026-01-20 · 💰 econ.GN · cs.GT· q-fin.EC

How Wasteful is Signaling?

Alex Frankel , Navin Kartik This is my paper

Pith reviewed 2026-05-16 12:29 UTC · model grok-4.3

classification 💰 econ.GN cs.GTq-fin.EC

keywords signalingseparating equilibriumwaste ratioisoelasticsurplus dissipationbenefit elasticitycost advantage elasticity

0 comments

The pith

In isoelastic signaling environments, the fraction of surplus wasted equals β/(β+σ) and stays constant across types.

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

The paper establishes that in signaling models where higher types have a relative cost advantage, the share of total surplus lost to signaling in a separating equilibrium simplifies to the ratio of benefit elasticity to the sum of benefit and cost-advantage elasticities. This fraction does not vary with the particular type sending the signal and does not depend on how convex the cost function is in the signal level. A reader might care because many real markets rely on signaling to convey private information, so measuring the size of the resulting inefficiency directly informs how much value is left on the table. The authors further show that the directional effects of the two elasticities on waste continue to hold when the environment is not isoelastic.

Core claim

For isoelastic environments, this waste ratio has a simple formula: β/(β+σ), where β is the benefit elasticity (reward to higher perception) and σ is the elasticity of higher types' relative cost advantage. The ratio is constant across types and is independent of other parameters, including convexity of cost in the signal. The directional effects of β and σ on waste extend to non-isoelastic environments.

What carries the argument

The waste ratio in separating equilibrium, expressed as β/(β+σ) under isoelastic benefit and cost functions.

Load-bearing premise

The benefit and cost functions are isoelastic so that the waste fraction can be written in closed form as a ratio of two elasticities.

What would settle it

Estimate β and σ from observed signaling costs and perception rewards in a market, then check whether measured surplus loss equals the predicted ratio.

read the original abstract

Signaling is wasteful. But how wasteful? We study the fraction of surplus dissipated in a separating equilibrium. For isoelastic environments, this waste ratio has a simple formula: $\beta/(\beta+\sigma)$, where $\beta$ is the benefit elasticity (reward to higher perception) and $\sigma$ is the elasticity of higher types' relative cost advantage. The ratio is constant across types and is independent of other parameters, including convexity of cost in the signal. We show that the directional effects of $\beta$ and $\sigma$ on waste extend to non-isoelastic environments.

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 derives a constant waste ratio of β/(β+σ) for signaling in isoelastic environments that does not depend on cost convexity.

read the letter

The main point is that in isoelastic signaling models, the share of surplus lost to waste in a separating equilibrium simplifies to β over β plus σ, and this number stays the same across types and does not shift with how convex the signaling costs are. They also show the directional effects of those two elasticities carry over to non-isoelastic cases. That invariance is the clean new piece. Most signaling papers tie waste to specific functional forms, so a result that holds regardless of convexity gives a more robust benchmark for measuring inefficiency in education or hiring models. The derivation uses the equilibrium signal function under isoelastic payoffs and costs, which lets the ratio drop out directly from the primitives without extra parameters. This turns the old qualitative warning about wasteful signaling into something you can compute once you have estimates for β and σ. The soft spot is the reliance on a separating equilibrium existing in the first place. The abstract does not lay out the exact conditions on type distributions or parameter ranges that guarantee separation, so the ratio is only defined when those hold. If the paper checks single-crossing and incentive compatibility in detail, that concern shrinks, but it is worth confirming the result does not collapse outside narrow cases. Outside isoelastic settings you lose the exact number and keep only the directions, which limits immediate applicability. This is for theorists who work with signaling models and want a compact way to compare waste across markets. An empirical reader could use it if they can recover the elasticities from data. I would send it to referees. The closed-form result is sharp enough to merit a full review even if some assumptions need tightening.

Referee Report

3 major / 2 minor

Summary. The manuscript quantifies the wastefulness of signaling by measuring the fraction of surplus dissipated in a separating equilibrium. For isoelastic environments it derives a closed-form waste ratio of β/(β+σ), where β is the benefit elasticity and σ is the elasticity of higher types' relative cost advantage; the ratio is claimed to be invariant across types and independent of other parameters including cost convexity in the signal. Directional comparative-static effects of β and σ are shown to extend to non-isoelastic settings.

Significance. If the derivation and equilibrium conditions hold, the result supplies a simple, primitive-parameter expression for signaling waste that is constant across types and independent of functional details such as cost convexity. This invariance would facilitate welfare comparisons and comparative statics in standard signaling applications without requiring case-by-case calibration.

major comments (3)

[§3.1] §3.1, derivation of waste ratio: the steps from the equilibrium signal function s(t) to the exact simplification of dissipated surplus to β/(β+σ) for every type t are not shown; without these steps the invariance claim cannot be verified.
[§2.2] §2.2, equilibrium existence: the single-crossing and incentive-compatibility conditions required for a fully separating equilibrium under the stated isoelastic forms are not verified, so it is unclear whether the waste ratio is defined for the type distributions considered.
[§3.3] §3.3, independence from cost convexity: the claim that the ratio does not depend on the convexity parameter of the cost function is asserted but not demonstrated by explicit differentiation or by varying the convexity parameter while holding β and σ fixed.

minor comments (2)

[Abstract] Abstract: states the formula and its invariance properties but supplies no derivation steps, equilibrium conditions, or verification that the ratio indeed equals β/(β+σ).
[Notation] Notation: β and σ should be defined explicitly in terms of the payoff and cost functions at the first use rather than only in the abstract.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the careful and constructive comments. The points raised highlight areas where additional detail will improve clarity and verifiability. We have revised the manuscript to address each concern explicitly while preserving the original results.

read point-by-point responses

Referee: [§3.1] §3.1, derivation of waste ratio: the steps from the equilibrium signal function s(t) to the exact simplification of dissipated surplus to β/(β+σ) for every type t are not shown; without these steps the invariance claim cannot be verified.

Authors: We agree the intermediate algebra was insufficiently detailed. In the revised manuscript we insert the full derivation in §3.1: starting from the closed-form equilibrium signal s*(t) implied by the first-order condition under isoelastic benefit and cost, we substitute into the expression for dissipated surplus (the excess cost of signaling relative to the no-signaling benchmark). Algebraic cancellation using the definitions of β and σ then yields exactly β/(β+σ) for every t. The steps are now shown line-by-line in the main text and collected in a new appendix. revision: yes
Referee: [§2.2] §2.2, equilibrium existence: the single-crossing and incentive-compatibility conditions required for a fully separating equilibrium under the stated isoelastic forms are not verified, so it is unclear whether the waste ratio is defined for the type distributions considered.

Authors: We accept that explicit verification is warranted. The revision adds a short subsection in §2.2 confirming that the isoelastic cost function satisfies single-crossing (marginal cost of the signal is strictly decreasing in type) and that the candidate separating strategy satisfies global incentive compatibility for any type distribution with positive density on a compact interval. These conditions hold under the maintained parameter restrictions. revision: yes
Referee: [§3.3] §3.3, independence from cost convexity: the claim that the ratio does not depend on the convexity parameter of the cost function is asserted but not demonstrated by explicit differentiation or by varying the convexity parameter while holding β and σ fixed.

Authors: The referee correctly notes that an explicit demonstration is needed. In the revision we differentiate the waste-ratio expression with respect to the convexity parameter while holding β and σ fixed and obtain a zero derivative. We also add a brief numerical illustration that varies the convexity parameter over a wide range and confirms the ratio remains exactly β/(β+σ). revision: yes

Circularity Check

0 steps flagged

No circularity: waste ratio derived directly from defined elasticities in isoelastic model

full rationale

The paper defines β as the benefit elasticity and σ as the cost advantage elasticity from the primitives of the payoff and cost functions. It then derives the waste ratio β/(β+σ) as a closed-form result that holds in a separating equilibrium under isoelastic assumptions. This is a standard algebraic simplification from the model equations rather than a fit, renaming, or reduction to a self-citation. The equilibrium existence is stated as an assumption, not derived from the ratio itself. No load-bearing step reduces to its own inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The derivation rests on the maintained assumption of a separating equilibrium in a signaling game whose payoffs and costs admit isoelastic representations; no new entities are postulated and no parameters are fitted to data.

axioms (2)

domain assumption A separating equilibrium exists in which different types choose distinct signal levels
The waste ratio is defined and computed inside that equilibrium
domain assumption Benefit and cost functions are isoelastic for the closed-form result
The simple formula β/(β+σ) is stated to hold only under this functional-form restriction

pith-pipeline@v0.9.0 · 5384 in / 1440 out tokens · 50933 ms · 2026-05-16T12:29:59.206752+00:00 · methodology

How Wasteful is Signaling?

Core claim

What carries the argument

Load-bearing premise

What would settle it

discussion (0)