Rebuttal of "On Nonparametric Identification of Treatment Effects in Duration Models"

Gerard J. van den Berg; Jaap H. Abbring

arxiv: 1907.09886 · v1 · pith:BQL2FGYDnew · submitted 2019-07-20 · 💰 econ.EM

Rebuttal of "On Nonparametric Identification of Treatment Effects in Duration Models"

Jaap H. Abbring , Gerard J. van den Berg This is my paper

Pith reviewed 2026-05-24 18:49 UTC · model grok-4.3

classification 💰 econ.EM

keywords rebuttalduration modelstreatment effectsnonparametric identificationeconometricsrandom variable transformation

0 comments

The pith

Johansson and Lee's claim that Proposition 3 does not hold is incorrect because they made a basic error transforming one random variable into another.

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

The paper shows that a published critique of the main result in Abbring and Van den Berg (2003b) rests on a mistake. Johansson and Lee argue that nonparametric identification of treatment effects in duration models fails, yet this conclusion traces directly to their incorrect handling of a random-variable transformation. Correcting the step restores the original proposition. Readers working with duration data and treatment effects would care because the rebuttal confirms that the identification result remains intact.

Core claim

Johansson and Lee claim that the main result (Proposition 3) in Abbring and Van den Berg (2003b) does not hold. This claim is incorrect because at a certain point within their line of reasoning, they make a rather basic error while transforming one random variable into another random variable, and this leads them to draw incorrect conclusions. As a result, their paper can be discarded.

What carries the argument

The basic error in transforming one random variable into another that invalidates the entire critique of Proposition 3.

If this is right

Proposition 3 in Abbring and Van den Berg (2003b) continues to hold.
Nonparametric identification of treatment effects in duration models is valid under the conditions originally stated.
Any conclusions drawn from Johansson and Lee's argument against the identification result are unsupported.

Where Pith is reading between the lines

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

Econometric papers that rely on similar variable transformations should verify each algebraic step independently.
The rebuttal leaves open whether other, unrelated critiques of duration-model identification might succeed.
Empirical applications that used the original Proposition 3 retain their theoretical foundation.

Load-bearing premise

The identified transformation error is the decisive flaw that invalidates Johansson and Lee's entire line of reasoning against Proposition 3.

What would settle it

A direct check confirming that the random-variable transformation step in Johansson and Lee's paper is in fact correct would support their claim that Proposition 3 fails.

read the original abstract

In their IZA Discussion Paper 10247, Johansson and Lee claim that the main result (Proposition 3) in Abbring and Van den Berg (2003b) does not hold. We show that their claim is incorrect. At a certain point within their line of reasoning, they make a rather basic error while transforming one random variable into another random variable, and this leads them to draw incorrect conclusions. As a result, their paper can be discarded.

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

This rebuttal identifies a specific algebraic slip in Johansson and Lee's transformation of a random variable that they say invalidates their attack on the 2003 Proposition 3.

read the letter

The core point is straightforward: Abbring and van den Berg locate one step in Johansson and Lee's reasoning where a random variable is mishandled during a transformation, and they argue this single error collapses the rest of the critique. That is the entire contribution. No new identification result is offered, and the note does not revisit the original duration model setup beyond restating that the 2003 claim survives once the error is fixed. The paper does what it sets out to do cleanly by naming the exact point of failure rather than offering a broad defense. Readers who have already worked through both the 2003 paper and the IZA discussion paper will be able to check the step in a few minutes. The main limitation is length and presentation. The abstract states the conclusion without showing the transformation itself, so the note functions more as a pointer than a self-contained correction. Anyone not already familiar with the two earlier papers will still need to pull the original sources to confirm the mistake is real and decisive. There is no circularity or hidden assumption in the rebuttal's own logic; it rests only on the claim that the cited transformation is wrong. This is the sort of short correction that matters inside a narrow literature on nonparametric duration models. People who teach or apply the 2003 result, or who are deciding whether to cite Johansson and Lee, will want to know about it. It is not important enough to reorganize the field, but it is precise enough that a journal should consider it as a note rather than desk-rejecting it outright. I would send it to referees who know the original identification argument.

Referee Report

1 major / 0 minor

Summary. The manuscript rebuts Johansson and Lee (IZA DP 10247), who claim that Proposition 3 of Abbring and Van den Berg (2003b) fails. It asserts that Johansson and Lee reach an incorrect conclusion solely because of a basic algebraic error when transforming one random variable into another; once corrected, their line of reasoning against the proposition collapses and the original result stands.

Significance. If the asserted transformation error is correctly identified and shown to be decisive, the note would restore the validity of the nonparametric identification result for treatment effects in duration models. The manuscript supplies no equations, no explicit mapping of the claimed mistake to a specific step in Johansson and Lee, and no verification that the error propagates to their main claim, so the significance cannot yet be assessed.

major comments (1)

Abstract (and entire manuscript): the central claim is that a 'rather basic error' occurs 'while transforming one random variable into another random variable' and that this single mistake invalidates Johansson and Lee's entire argument against Proposition 3. No equation, no random-variable transformation, and no citation to the precise location of the alleged mistake in the target paper are supplied. Without this derivation the assertion remains unsupported and the rebuttal cannot be evaluated.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the detailed report. The referee correctly identifies that the current short note does not supply the explicit equations or the precise mapping of the transformation error to Johansson and Lee. We will revise the manuscript to include this derivation.

read point-by-point responses

Referee: Abstract (and entire manuscript): the central claim is that a 'rather basic error' occurs 'while transforming one random variable into another random variable' and that this single mistake invalidates Johansson and Lee's entire argument against Proposition 3. No equation, no random-variable transformation, and no citation to the precise location of the alleged mistake in the target paper are supplied. Without this derivation the assertion remains unsupported and the rebuttal cannot be evaluated.

Authors: We agree that the manuscript as currently written leaves the location and nature of the algebraic error implicit. In the revision we will add a dedicated paragraph (or short section) that (i) reproduces the relevant transformation step from Johansson and Lee, (ii) shows the incorrect algebraic manipulation, and (iii) demonstrates how the corrected transformation restores the validity of Proposition 3. This will make the argument self-contained and directly address the referee's concern. revision: yes

Circularity Check

0 steps flagged

No circularity; rebuttal identifies external algebraic error

full rationale

The paper's argument chain consists solely of locating and exhibiting a specific transformation error in Johansson and Lee's random-variable manipulation, which directly invalidates their critique of Proposition 3. This is an external algebraic verification against the opposing paper's equations rather than any derivation that reduces to the rebuttal's own inputs, fitted parameters, self-citations, or ansatzes. No load-bearing step invokes self-definition, uniqueness theorems from the authors' prior work, or renaming of known results. The abstract and described structure are self-contained against the external benchmark of the critiqued manuscript.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No free parameters, axioms, or invented entities are introduced; the paper is a short theoretical rebuttal based on the abstract.

pith-pipeline@v0.9.0 · 5600 in / 909 out tokens · 18055 ms · 2026-05-24T18:49:08.645888+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

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

IndisputableMonolith/Foundation/AbsoluteFloorClosure.lean reality_from_one_distinction unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

the subdensities of Y on {Y < W} and W on {W < Y} … equal … (3) and (4)

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.