{"id":"d61ad97e-f50e-4f22-8377-bfb0ca64dea1","arxiv_id":"2606.09223","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":5.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"Develops sharp partial identification bounds and semiparametrically efficient inference for complier treatment effects in nonparametric sample selection models with endogenous treatment.","lead":"This paper develops sharp bounds for treatment effects in sample selection models with endogenous treatment and weak monotonicity in selection. It also introduces a debiased machine learning procedure for root-n inference under high-dimensional covariates.","discovery_kind":"new_method","skeptic_critique":{"model":"grok-4.3","headline":"Sharpness claim for complier intensive-margin bounds rests entirely on weak sample selection monotonicity","rationale":"The reader's weakest_assumption correctly isolates the single modeling choice that delivers both the sharpness result and the improvement over the literature. No other internal inconsistency is visible from the abstract-level description, and the inference procedure (orthogonal moments + debiased ML) is standard once the identification region is taken as given. Because the full text was not supplied in the query, a deeper technical audit of the derivation is impossible; the monotonicity step remains the only load-bearing point that can be identified without the manuscript.","tokens_in":1637,"tokens_out":356,"duration_ms":15232,"concrete_test":"Re-derive the upper and lower bounds on the complier intensive-margin effect while dropping the weak monotonicity restriction entirely (i.e., allow arbitrary dependence between treatment and selection for the never-takers); if the resulting interval coincides with or exceeds the Chen-Flores interval, the claimed sharpness gain is an artifact of the maintained assumption.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The headline result states that the proposed bounds on intensive-margin effects for compliers are sharp and strictly tighter than Chen-Flores (2015). This tightening is obtained only after imposing the weak sample selection monotonicity assumption (never-selected units are unaffected by treatment assignment). The abstract and the reader's note both flag this assumption as the key restriction. If the assumption fails even mildly (e.g., treatment can affect selection probability for the never-takers in either direction), the identification region reverts to the wider Chen-Flores set; the paper supplies no sensitivity analysis or set of primitive conditions under which the monotonicity is implied by the economic model.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"This paper develops partial identification and inference methods for treatment effects in nonparametric sample selection models with endogenous treatment (noncompliance) and weak sample selection monotonicity. The central contribution is a set of sharp bounds on intensive-margin treatment effects among compliers that are strictly tighter than the Chen-Flores (2015) bounds; these are paired with semiparametrically efficient orthogonal moments and a debiased machine-learning estimator that delivers root-n inference under high-dimensional covariates or flexible functional forms. Simulations and two empirical applications (Job Corps, Oregon Health Insurance Experiment) are provided.","tokens_in":1734,"tokens_out":518,"duration_ms":18883,"significance":"If the sharpness result and the efficiency of the orthogonal moments hold, the paper supplies a practically useful tightening of the identified set for a policy-relevant parameter (complier intensive-margin effects) together with an implementable inference procedure that accommodates modern high-dimensional settings. The explicit use of orthogonal moments and DML is a methodological strength that facilitates credible empirical work.","major_comments":[{"comment":"The claim that the bounds are sharp and strictly tighter than Chen and Flores (2015) rests entirely on the weak sample selection monotonicity assumption (never-selected units unaffected by treatment). The manuscript provides no sensitivity analysis showing how the identified set expands when this assumption is relaxed even mildly, nor primitive conditions under which the assumption is implied by an economic model. This is load-bearing for the headline result on tighter bounds.","section":"Abstract; §3 (Identification)"},{"comment":"§4.2, the construction of the orthogonal moments: it is stated that the moments are semiparametrically efficient, but the efficiency bound is not explicitly derived or referenced under the joint maintained assumptions (including monotonicity and the selection model). Without this step it is difficult to verify that the proposed moments attain the bound.","section":"§4.2"}],"minor_comments":[{"comment":"Table 1 (simulation design) reports coverage probabilities but does not include the width of the confidence intervals; adding this column would make the finite-sample comparison with existing methods more informative.","section":"Table 1"},{"comment":"The notation for the selection indicator and the observed outcome is introduced without an explicit comparison table to the Chen-Flores setup; a short notational concordance would improve readability.","section":"§2"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the constructive and detailed comments. We address each major comment below.","responses":[{"response":"The sharpness and tightening relative to Chen and Flores (2015) are established under the maintained weak sample selection monotonicity assumption, which is stated explicitly in the paper. We agree that robustness checks would strengthen the presentation. In the revision we will add a short subsection to §3 that (i) reports how the identified set expands under mild relaxations of the assumption and (ii) supplies primitive economic conditions (e.g., treatment does not affect selection for units that would never be selected under either treatment status) under which the assumption is implied by standard selection models. This addition directly addresses the load-bearing nature of the assumption while preserving the headline result under the stated conditions.","revision_made":"partial","referee_comment":"[Abstract; §3 (Identification)] The claim that the bounds are sharp and strictly tighter than Chen and Flores (2015) rests entirely on the weak sample selection monotonicity assumption (never-selected units unaffected by treatment). The manuscript provides no sensitivity analysis showing how the identified set expands when this assumption is relaxed even mildly, nor primitive conditions under which the assumption is implied by an economic model. This is load-bearing for the headline result on tighter bounds."},{"response":"We thank the referee for this observation. The orthogonal moments were derived to attain the semiparametric efficiency bound under the joint maintained assumptions, but an explicit derivation of that bound was omitted from the current draft. In the revision we will add to §4.2 (i) the derivation of the efficiency bound under the nonparametric selection model together with the weak monotonicity restriction and (ii) a verification that the proposed moments achieve this bound. This will make the efficiency claim fully verifiable.","revision_made":"yes","referee_comment":"[§4.2] §4.2, the construction of the orthogonal moments: it is stated that the moments are semiparametrically efficient, but the efficiency bound is not explicitly derived or referenced under the joint maintained assumptions (including monotonicity and the selection model). Without this step it is difficult to verify that the proposed moments attain the bound."}],"tokens_in":1296,"tokens_out":478,"duration_ms":22593,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The headline result is that the bounds on intensive-margin effects for compliers become sharp and strictly narrower than Chen-Flores once you add the weak monotonicity restriction that treatment does not affect selection for never-takers. That is the only source of the improvement. The paper also supplies semiparametrically efficient orthogonal moments plus a debiased machine-learning estimator that delivers root-n inference under flexible or high-dimensional nuisance functions.\n\nThe inference part is the stronger contribution. The orthogonal-moment construction looks standard but correctly adapted to the selection-plus-endogeneity setting, and the simulations show reasonable coverage. The two applications (Job Corps and Oregon Health Insurance Experiment) demonstrate that the intervals can be noticeably shorter than the earlier bounds.\n\nThe identification claim rests entirely on the monotonicity assumption. If that assumption is even mildly violated, the identified set reverts to the wider Chen-Flores region, and the paper provides no sensitivity analysis or primitive conditions under which the assumption is plausible in the labor or health settings it studies. Without seeing the full proof, it is also impossible to confirm that the bounds are indeed sharp rather than just valid.\n\nThis is useful reading for applied researchers who already work with partial identification in selection models and want a ready-to-use inference tool. A serious referee should see it because the inference method has standalone value even if the monotonicity step is treated as an optional extra assumption.","headline":"The paper tightens complier intensive-margin bounds via weak sample selection monotonicity and adds a debiased ML inference procedure that handles high-dimensional covariates.","tokens_in":2184,"tokens_out":352,"would_cite":false,"duration_ms":12242,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"The paper derives sharp bounds on intensive margin treatment effects among compliers in nonparametric sample selection models with endogenous treatment under weak monotonicity and supplies debiased machine learning inference that achieves r","keywords":["sample selection","treatment endogeneity","partial identification","sharp bounds","debiased machine learning","complier effects","intensive margin","nonparametric models"],"falsifier":"Empirical evidence that the selection probability falls with treatment assignment in a manner that violates monotonicity, or data in which the proposed bounds are crossed while point estimates from alternative identification strategies lie outside them.","tokens_in":2523,"feed_emoji":"","tokens_out":685,"duration_ms":18483,"temperature":0.7,"pith_summary":"This paper seeks to bound and infer treatment effects when outcomes are observed only for a selected subsample and treatment assignment suffers from noncompliance. It focuses on effects for compliers on the intensive margin and shows that the new bounds are sharp and strictly tighter than earlier proposals. A sympathetic reader would care because sample selection and endogeneity arise in many program evaluations, and tighter bounds with valid inference under flexible specifications can produce more decisive conclusions about policy impacts. The work combines partial identification arguments with semiparametrically efficient orthogonal moments and debiased machine learning to maintain root-n consistency even when covariates are high-dimensional or functional forms are left nonparametric.","feed_headline":"Sharp bounds tighten complier treatment effects under selection","feed_subtitle":"New method produces tighter intervals than earlier work and supports root-n inference with high-dimensional covariates in program evaluation","key_machinery":"Sharp bounds on intensive margin complier treatment effects derived from weak sample selection monotonicity, implemented through semiparametrically efficient orthogonal moments and debiased machine learning for inference.","core_discovery":"Under the weak sample selection monotonicity assumption, the proposed bounds for intensive margin treatment effects among compliers are sharp and tighter than those of Chen and Flores (2015). Semiparametrically efficient orthogonal moments and a debiased machine learning procedure permit valid root-n inference under high-dimensional covariates and flexible functional forms. Simulation results indicate good finite sample performance. Applications to Job Corps and the Oregon Health Insurance Experiment show that the method can deliver substantially tighter effect bounds and confidence intervals than existing alternatives.","pith_inferences":["The same monotonicity-based bounding logic could be applied to labor-market or education settings that feature both noncompliance and selective observation.","Orthogonal-moment constructions of this type may extend to other partial-identification problems that mix endogeneity with selection.","Testing the monotonicity assumption itself with the same data would be a natural next diagnostic step left open by the identification result."],"forward_implications":["The bounds yield substantially tighter intervals than prior methods when applied to job training and health insurance experiments.","Root-n inference remains valid even after using machine learning to estimate high-dimensional nuisance functions.","Simulations confirm reliable coverage and length properties for the resulting confidence intervals in moderate samples.","The partial identification strategy applies directly to other nonparametric models that combine endogenous treatment with nonrandom sample selection."],"fun_headline_variants":["Sharper bounds on complier effects under sample selection","Tighter bounds for treatment effects with endogenous selection","Efficient moments enable root-n inference in selection models","Sharp partial identification in models with treatment endogeneity"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"The weak sample selection monotonicity assumption that the probability of selection into the observed sample is nondecreasing in treatment assignment.","fun_headline_variants_meta":{"raw":{"variants":["Sharper bounds on complier effects under sample selection","Tighter bounds for treatment effects with endogenous selection","Efficient moments enable root-n inference in selection models","Sharp partial identification in models with treatment endogeneity"]},"model":"grok-4.3","cost_usd":0.007044,"raw_usage":{"total_tokens":3218,"prompt_tokens":586,"num_sources_used":0,"completion_tokens":57,"cost_in_usd_ticks":70437000,"prompt_tokens_details":{"text_tokens":586,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":2575,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":586,"tokens_out":57,"duration_ms":21441,"temperature":1.0,"reasoning_tokens":2575,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-06-27T14:25:24.721456+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"Empirical evidence that the selection probability falls with treatment assignment in a manner that violates monotonicity, or data in which the proposed bounds are crossed while point estimates from alternative identification strategies lie outside them.","supporting_citations":[],"review_version":1}