1{"text": "{\"genre\":\"mixed_theory_empirical\",\"subfield\":\"asset_pricing\",\"subfield_secondary\":\"behavioral_finance\",\"primary_method\":\"reduced_form_regression\",\"identification\":{\"estimand\":\"Degree of expectation stickiness (lambda) in analyst forecasts, and how strongly profitability-anomaly returns vary with stickiness and with profit persistence\",\"strategy\":\"Model-derived tests. Forecast errors are regressed on forecast revisions (Coibion-Gorodnichenko style) to estimate stickiness. Forecast errors are regressed on profitability signals. Portfolios are double-sorted on firm-level stickiness or persistence and on the anomaly signal.\",\"key_assumptions\":[\"Analyst forecasts represent market participants' beliefs\",\"Profits are predictable from persistent public signals\",\"Sticky-expectation updating nests rational expectations\"],\"threats_addressed\":[\"look-ahead bias in firm-level stickiness estimation\",\"mechanical correlation under rational-expectations null (simulations)\",\"past trend in profits as alternative explanation\",\"risk adjustment via factor models\"],\"causal_language_strength\":2,\"uses_causal_claim\":false},\"abstract_moves\":{\"question\":{\"present\":true,\"sentence_index\":0},\"gap\":{\"present\":false,\"sentence_index\":-1},\"method\":{\"present\":true,\"sentence_index\":3},\"results\":{\"present\":true,\"sentence_index\":3},\"magnitude\":{\"present\":false,\"sentence_index\":-1},\"contribution\":{\"present\":true,\"sentence_index\":0}},\"introduction\":{\"moves\":[{\"paragraph_index\":0,\"move\":\"motivation\"},{\"paragraph_index\":0,\"move\":\"related_literature_positioning\"},{\"paragraph_index\":1,\"move\":\"motivation\"},{\"paragraph_index\":1,\"move\":\"preview_of_approach\"},{\"paragraph_index\":1,\"move\":\"mechanism_or_theory_sketch\"},{\"paragraph_index\":2,\"move\":\"related_literature_positioning\"},{\"paragraph_index\":2,\"move\":\"contribution_statement\"}],\"contribution_paragraph_index\":2,\"contribution_types\":[\"new_theory_or_model\",\"new_evidence\",\"new_data_or_measure\"],\"roadmap_present\":false,\"literature_positioning_location\":\"inside_introduction\"},\"section_roles\":[{\"idx\":0,\"role\":\"introduction\"},{\"idx\":1,\"role\":\"theory\"},{\"idx\":2,\"role\":\"data\"},{\"idx\":3,\"role\":\"results\"},{\"idx\":4,\"role\":\"robustness\"},{\"idx\":5,\"role\":\"conclusion\"},{\"idx\":6,\"role\":\"appendix\"}],\"results_presentation\":{\"economic_significance_discussed\":true,\"statistical_significance_language\":\"unclear\",\"magnitude_statements_in_intro\":false},\"robustness\":{\"location\":\"main_text_section\",\"types\":[\"placebo\",\"alternative_specification\",\"other\"]},\"tables\":{\"convention_notes\":\"Table notes define variables using model notation and describe the regression or sort procedure. Portfolio tables report excess returns and CAPM, three-factor and four-factor alphas, plus double-sorts. Units are partially stated in the notes (returns/alphas are given in some unit, but the excerpt is truncated).\",\"units_stated_in_notes\":true,\"standard_errors_description\":\"unclear\"},\"figures\":{\"uncertainty_shown\":\"unclear\"},\"conclusion\":{\"bounded_claims\":true,\"policy_implications\":false,\"open_questions\":false,\"restates_contribution\":true},\"hedging\":{\"level\":\"moderate\",\"typical_devices\":[\"conditional assumptions (assuming analyst forecasts are representative)\",\"'in line with' interpretations\",\"'suggests'\"]},\"claim_evidence\":[{\"claim\":\"The profitability anomaly carries large economic significance: high Sharpe ratio, no crash risk, high capacity\",\"evidence_type\":\"prior_literature\",\"evidence_ref\":\"Novy-Marx (2013, 2015); Lemperiere et al. (2017); Landier, Simon, Thesmar (2015)\"},{\"claim\":\"Analyst expectations are sticky: forecast errors are positively related to past forecast revisions\",\"evidence_type\":\"table\",\"evidence_ref\":\"Table III, Figure 1\"},{\"claim\":\"Analysts are too pessimistic about future profits of high-profit firms\",\"evidence_type\":\"table\",\"evidence_ref\":\"Table VI, Figure 2\"},{\"claim\":\"The profitability anomaly is stronger for stocks followed by stickier analysts\",\"evidence_type\":\"table\",\"evidence_ref\":\"Table VII\"},{\"claim\":\"The profitability anomaly is stronger for firms with more persistent profits\",\"evidence_type\":\"table\",\"evidence_ref\":\"Table VIII\"},{\"claim\":\"Double-sort results are not mechanical under the rational-expectations null\",\"evidence_type\":\"figure\",\"evidence_ref\":\"Figure 3\"},{\"claim\":\"Less experienced and busier analysts have stickier beliefs, consistent with limited attention\",\"evidence_type\":\"table\",\"evidence_ref\":\"Table V\"},{\"claim\":\"In the sticky-belief model, past profits forecast future returns\",\"evidence_type\":\"equation_or_proof\",\"evidence_ref\":\"Section I; Appendix proofs\"}],\"structural_summary\":\"The paper opens with the risk-versus-behavioral debate over return anomalies and narrows to the profitability anomaly. It builds a simple sticky-expectations pricing model borrowed from the macro information-rigidity literature, which yields testable propositions. The empirical part then follows the propositions in order. First, it measures stickiness from I/B/E/S analyst forecasts in pooled, analyst-level and firm-level estimates. Second, it shows that past profits predict forecast errors. Third, it double-sorts anomaly returns on stickiness and on profit persistence. A separate robustness section addresses look-ahead bias and uses simulations under the rational-expectations null. A brief conclusion restates the model, its predictions, and the limited-attention determinants of stickiness.\",\"writing_profile_keywords\":[\"proposition-driven testing\",\"behavioral asset pricing\",\"analyst forecasts\",\"portfolio sorts\",\"model-to-data mapping\",\"sticky expectations\"],\"confidence\":0.8,\"notes\":\"Section III is labeled 'theory' by the role rule but contains the empirical tests, so it is annotated as results. Significance-reporting conventions are not visible in the excerpts. The introduction excerpt is truncated and no roadmap is visible.\"}", "json": {"genre": "mixed_theory_empirical", "subfield": "asset_pricing", "subfield_secondary": "behavioral_finance", "primary_method": "reduced_form_regression", "identification": {"estimand": "Degree of expectation stickiness (lambda) in analyst forecasts, and how strongly profitability-anomaly returns vary with stickiness and with profit persistence", "strategy": "Model-derived tests. Forecast errors are regressed on forecast revisions (Coibion-Gorodnichenko style) to estimate stickiness. Forecast errors are regressed on profitability signals. Portfolios are double-sorted on firm-level stickiness or persistence and on the anomaly signal.", "key_assumptions": ["Analyst forecasts represent market participants' beliefs", "Profits are predictable from persistent public signals", "Sticky-expectation updating nests rational expectations"], "threats_addressed": ["look-ahead bias in firm-level stickiness estimation", "mechanical correlation under rational-expectations null (simulations)", "past trend in profits as alternative explanation", "risk adjustment via factor models"], "causal_language_strength": 2, "uses_causal_claim": false}, "abstract_moves": {"question": {"present": true, "sentence_index": 0}, "gap": {"present": false, "sentence_index": -1}, "method": {"present": true, "sentence_index": 3}, "results": {"present": true, "sentence_index": 3}, "magnitude": {"present": false, "sentence_index": -1}, "contribution": {"present": true, "sentence_index": 0}}, "introduction": {"moves": [{"paragraph_index": 0, "move": "motivation"}, {"paragraph_index": 0, "move": "related_literature_positioning"}, {"paragraph_index": 1, "move": "motivation"}, {"paragraph_index": 1, "move": "preview_of_approach"}, {"paragraph_index": 1, "move": "mechanism_or_theory_sketch"}, {"paragraph_index": 2, "move": "related_literature_positioning"}, {"paragraph_index": 2, "move": "contribution_statement"}], "contribution_paragraph_index": 2, "contribution_types": ["new_theory_or_model", "new_evidence", "new_data_or_measure"], "roadmap_present": false, "literature_positioning_location": "inside_introduction"}, "section_roles": [{"idx": 0, "role": "introduction"}, {"idx": 1, "role": "theory"}, {"idx": 2, "role": "data"}, {"idx": 3, "role": "results"}, {"idx": 4, "role": "robustness"}, {"idx": 5, "role": "conclusion"}, {"idx": 6, "role": "appendix"}], "results_presentation": {"economic_significance_discussed": true, "statistical_significance_language": "unclear", "magnitude_statements_in_intro": false}, "robustness": {"location": "main_text_section", "types": ["placebo", "alternative_specification", "other"]}, "tables": {"convention_notes": "Table notes define variables using model notation and describe the regression or sort procedure. Portfolio tables report excess returns and CAPM, three-factor and four-factor alphas, plus double-sorts. Units are partially stated in the notes (returns/alphas are given in some unit, but the excerpt is truncated).", "units_stated_in_notes": true, "standard_errors_description": "unclear"}, "figures": {"uncertainty_shown": "unclear"}, "conclusion": {"bounded_claims": true, "policy_implications": false, "open_questions": false, "restates_contribution": true}, "hedging": {"level": "moderate", "typical_devices": ["conditional assumptions (assuming analyst forecasts are representative)", "'in line with' interpretations", "'suggests'"]}, "claim_evidence": [{"claim": "The profitability anomaly carries large economic significance: high Sharpe ratio, no crash risk, high capacity", "evidence_type": "prior_literature", "evidence_ref": "Novy-Marx (2013, 2015); Lemperiere et al. (2017); Landier, Simon, Thesmar (2015)"}, {"claim": "Analyst expectations are sticky: forecast errors are positively related to past forecast revisions", "evidence_type": "table", "evidence_ref": "Table III, Figure 1"}, {"claim": "Analysts are too pessimistic about future profits of high-profit firms", "evidence_type": "table", "evidence_ref": "Table VI, Figure 2"}, {"claim": "The profitability anomaly is stronger for stocks followed by stickier analysts", "evidence_type": "table", "evidence_ref": "Table VII"}, {"claim": "The profitability anomaly is stronger for firms with more persistent profits", "evidence_type": "table", "evidence_ref": "Table VIII"}, {"claim": "Double-sort results are not mechanical under the rational-expectations null", "evidence_type": "figure", "evidence_ref": "Figure 3"}, {"claim": "Less experienced and busier analysts have stickier beliefs, consistent with limited attention", "evidence_type": "table", "evidence_ref": "Table V"}, {"claim": "In the sticky-belief model, past profits forecast future returns", "evidence_type": "equation_or_proof", "evidence_ref": "Section I; Appendix proofs"}], "structural_summary": "The paper opens with the risk-versus-behavioral debate over return anomalies and narrows to the profitability anomaly. It builds a simple sticky-expectations pricing model borrowed from the macro information-rigidity literature, which yields testable propositions. The empirical part then follows the propositions in order. First, it measures stickiness from I/B/E/S analyst forecasts in pooled, analyst-level and firm-level estimates. Second, it shows that past profits predict forecast errors. Third, it double-sorts anomaly returns on stickiness and on profit persistence. A separate robustness section addresses look-ahead bias and uses simulations under the rational-expectations null. A brief conclusion restates the model, its predictions, and the limited-attention determinants of stickiness.", "writing_profile_keywords": ["proposition-driven testing", "behavioral asset pricing", "analyst forecasts", "portfolio sorts", "model-to-data mapping", "sticky expectations"], "confidence": 0.8, "notes": "Section III is labeled 'theory' by the role rule but contains the empirical tests, so it is annotated as results. Significance-reporting conventions are not visible in the excerpts. The introduction excerpt is truncated and no roadmap is visible."}, "model": "claude-opus-5-5", "stop_reason": "end_turn", "latency_s": 17.07, "usage": {"input": 5901, "output": 2185, "cache_read": 3968, "cache_write": 0}, "cost_usd": 0.068098, "request_id": "req_011CggqGABeYE9wTYTzhNLup", "cached": false, "ts": "2026-11-03T23:06:13+00:00", "effort": "medium"}