1{"text": "{\"genre\":\"empirical\",\"subfield\":\"corporate_finance\",\"subfield_secondary\":\"financial_intermediation_banking\",\"primary_method\":\"reduced_form_regression\",\"identification\":{\"estimand\":\"Association between a model's share of captive (manufacturer) financing and its resale price depreciation, LTVs and production quantities; also the effect of a rise in captive share after an acquisition\",\"strategy\":\"Cross-sectional regressions of model-level depreciation on captive share. Depreciation is decomposed into new-price versus productivity-driven components to separate the commitment and information-asymmetry hypotheses. A natural experiment uses Volvo's 2007 acquisition of an Ingersoll Rand unit. Further tests use within-manufacturer regressions of quantities and time-series variation in loan-officer collateral tightening.\",\"key_assumptions\":[\"Information asymmetry about quality would show up in productivity-driven (cross-sectional age) depreciation rather than in new-price depreciation\",\"The acquisition shifted captive share for reasons unrelated to the machines' ex-ante quality\",\"Used machine prices measured at auction reflect market resale values\"],\"threats_addressed\":[\"ex-ante machine quality / signaling alternative\",\"market power confound\",\"bank credit pull-back as an alternative explanation for machine choice\"],\"causal_language_strength\":3,\"uses_causal_claim\":true},\"abstract_moves\":{\"question\":{\"present\":true,\"sentence_index\":0},\"gap\":{\"present\":false,\"sentence_index\":-1},\"method\":{\"present\":true,\"sentence_index\":1},\"results\":{\"present\":true,\"sentence_index\":1},\"magnitude\":{\"present\":false,\"sentence_index\":-1},\"contribution\":{\"present\":true,\"sentence_index\":4}},\"introduction\":{\"moves\":[{\"paragraph_index\":0,\"move\":\"motivation\"},{\"paragraph_index\":0,\"move\":\"tension_or_puzzle\"},{\"paragraph_index\":1,\"move\":\"mechanism_or_theory_sketch\"},{\"paragraph_index\":2,\"move\":\"mechanism_or_theory_sketch\"},{\"paragraph_index\":3,\"move\":\"preview_of_approach\"},{\"paragraph_index\":3,\"move\":\"preview_of_results\"},{\"paragraph_index\":4,\"move\":\"identification_defense\"},{\"paragraph_index\":4,\"move\":\"preview_of_results\"},{\"paragraph_index\":5,\"move\":\"preview_of_results\"},{\"paragraph_index\":6,\"move\":\"preview_of_results\"},{\"paragraph_index\":6,\"move\":\"policy_or_broader_implication\"},{\"paragraph_index\":7,\"move\":\"related_literature_positioning\"},{\"paragraph_index\":7,\"move\":\"contribution_statement\"},{\"paragraph_index\":7,\"move\":\"roadmap\"}],\"contribution_paragraph_index\":7,\"contribution_types\":[\"new_evidence\",\"new_theory_or_model\"],\"roadmap_present\":true,\"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\":\"conclusion\"}],\"results_presentation\":{\"economic_significance_discussed\":true,\"statistical_significance_language\":\"unclear\",\"magnitude_statements_in_intro\":false},\"robustness\":{\"location\":\"none_seen\",\"types\":[]},\"tables\":{\"convention_notes\":\"Table notes describe the sample, the unit of analysis and how depreciation rates are estimated (model-level log price on log age). Variable definitions such as the down payment ratio are given in the notes. Excerpts are truncated, so the standard-error treatment is not visible.\",\"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\":[\"'consistent with' framing\",\"'strongly suggestive evidence'\",\"'move closer to causal inference'\",\"'may be reasonable to think'\",\"explicit statement that the paper does not rule out other motives\"]},\"claim_evidence\":[{\"claim\":\"Manufacturer-financed shares of new durable investment are large across equipment industries\",\"evidence_type\":\"figure\",\"evidence_ref\":\"Figure 1\"},{\"claim\":\"Models with more captive financing depreciate less in resale markets\",\"evidence_type\":\"table\",\"evidence_ref\":\"Table I; Figure 2\"},{\"claim\":\"Captive finance is unrelated to productivity-driven depreciation, undermining the quality-signaling view\",\"evidence_type\":\"table\",\"evidence_ref\":\"Table II\"},{\"claim\":\"Resale performance of Ingersoll Rand machines improved after the Volvo acquisition raised captive share\",\"evidence_type\":\"table\",\"evidence_ref\":\"Table IV; Figure 4\"},{\"claim\":\"Equipment types with a history of captive backing show lower future production within a manufacturer\",\"evidence_type\":\"table\",\"evidence_ref\":\"Table V\"},{\"claim\":\"Captive-backed models carry higher LTVs, even when financed by banks\",\"evidence_type\":\"table\",\"evidence_ref\":\"Table VI\"},{\"claim\":\"Bank-financed buyers shift toward captive-backed models when collateral requirements tighten\",\"evidence_type\":\"table\",\"evidence_ref\":\"Table VII; Figure 5\"}],\"structural_summary\":\"The paper opens with stylized facts on the scale of manufacturer lending and asks why manufacturers finance their own sales. It proposes a Coase-commitment explanation and contrasts it with an information-asymmetry alternative, then separates the two by decomposing depreciation. A background and hypothesis section precedes a brief data section. A long results section moves from correlations to the alternative-hypothesis test, market power, an acquisition-based natural experiment, quantity effects, and pledgeability spillovers. A short discussion restates the findings and frames captive finance as a costly hedge.\",\"writing_profile_keywords\":[\"captive finance\",\"Coase conjecture\",\"durable goods\",\"hypothesis discrimination\",\"natural experiment\",\"pledgeability\",\"heavy equipment auction data\",\"competing hypotheses\"],\"confidence\":0.8,\"notes\":\"No explicit robustness section appears in the excerpts. An internet appendix is mentioned but its contents are not visible. Statistical significance conventions and standard-error treatment are not visible.\"}", "json": {"genre": "empirical", "subfield": "corporate_finance", "subfield_secondary": "financial_intermediation_banking", "primary_method": "reduced_form_regression", "identification": {"estimand": "Association between a model's share of captive (manufacturer) financing and its resale price depreciation, LTVs and production quantities; also the effect of a rise in captive share after an acquisition", "strategy": "Cross-sectional regressions of model-level depreciation on captive share. Depreciation is decomposed into new-price versus productivity-driven components to separate the commitment and information-asymmetry hypotheses. A natural experiment uses Volvo's 2007 acquisition of an Ingersoll Rand unit. Further tests use within-manufacturer regressions of quantities and time-series variation in loan-officer collateral tightening.", "key_assumptions": ["Information asymmetry about quality would show up in productivity-driven (cross-sectional age) depreciation rather than in new-price depreciation", "The acquisition shifted captive share for reasons unrelated to the machines' ex-ante quality", "Used machine prices measured at auction reflect market resale values"], "threats_addressed": ["ex-ante machine quality / signaling alternative", "market power confound", "bank credit pull-back as an alternative explanation for machine choice"], "causal_language_strength": 3, "uses_causal_claim": true}, "abstract_moves": {"question": {"present": true, "sentence_index": 0}, "gap": {"present": false, "sentence_index": -1}, "method": {"present": true, "sentence_index": 1}, "results": {"present": true, "sentence_index": 1}, "magnitude": {"present": false, "sentence_index": -1}, "contribution": {"present": true, "sentence_index": 4}}, "introduction": {"moves": [{"paragraph_index": 0, "move": "motivation"}, {"paragraph_index": 0, "move": "tension_or_puzzle"}, {"paragraph_index": 1, "move": "mechanism_or_theory_sketch"}, {"paragraph_index": 2, "move": "mechanism_or_theory_sketch"}, {"paragraph_index": 3, "move": "preview_of_approach"}, {"paragraph_index": 3, "move": "preview_of_results"}, {"paragraph_index": 4, "move": "identification_defense"}, {"paragraph_index": 4, "move": "preview_of_results"}, {"paragraph_index": 5, "move": "preview_of_results"}, {"paragraph_index": 6, "move": "preview_of_results"}, {"paragraph_index": 6, "move": "policy_or_broader_implication"}, {"paragraph_index": 7, "move": "related_literature_positioning"}, {"paragraph_index": 7, "move": "contribution_statement"}, {"paragraph_index": 7, "move": "roadmap"}], "contribution_paragraph_index": 7, "contribution_types": ["new_evidence", "new_theory_or_model"], "roadmap_present": true, "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": "conclusion"}], "results_presentation": {"economic_significance_discussed": true, "statistical_significance_language": "unclear", "magnitude_statements_in_intro": false}, "robustness": {"location": "none_seen", "types": []}, "tables": {"convention_notes": "Table notes describe the sample, the unit of analysis and how depreciation rates are estimated (model-level log price on log age). Variable definitions such as the down payment ratio are given in the notes. Excerpts are truncated, so the standard-error treatment is not visible.", "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": ["'consistent with' framing", "'strongly suggestive evidence'", "'move closer to causal inference'", "'may be reasonable to think'", "explicit statement that the paper does not rule out other motives"]}, "claim_evidence": [{"claim": "Manufacturer-financed shares of new durable investment are large across equipment industries", "evidence_type": "figure", "evidence_ref": "Figure 1"}, {"claim": "Models with more captive financing depreciate less in resale markets", "evidence_type": "table", "evidence_ref": "Table I; Figure 2"}, {"claim": "Captive finance is unrelated to productivity-driven depreciation, undermining the quality-signaling view", "evidence_type": "table", "evidence_ref": "Table II"}, {"claim": "Resale performance of Ingersoll Rand machines improved after the Volvo acquisition raised captive share", "evidence_type": "table", "evidence_ref": "Table IV; Figure 4"}, {"claim": "Equipment types with a history of captive backing show lower future production within a manufacturer", "evidence_type": "table", "evidence_ref": "Table V"}, {"claim": "Captive-backed models carry higher LTVs, even when financed by banks", "evidence_type": "table", "evidence_ref": "Table VI"}, {"claim": "Bank-financed buyers shift toward captive-backed models when collateral requirements tighten", "evidence_type": "table", "evidence_ref": "Table VII; Figure 5"}], "structural_summary": "The paper opens with stylized facts on the scale of manufacturer lending and asks why manufacturers finance their own sales. It proposes a Coase-commitment explanation and contrasts it with an information-asymmetry alternative, then separates the two by decomposing depreciation. A background and hypothesis section precedes a brief data section. A long results section moves from correlations to the alternative-hypothesis test, market power, an acquisition-based natural experiment, quantity effects, and pledgeability spillovers. A short discussion restates the findings and frames captive finance as a costly hedge.", "writing_profile_keywords": ["captive finance", "Coase conjecture", "durable goods", "hypothesis discrimination", "natural experiment", "pledgeability", "heavy equipment auction data", "competing hypotheses"], "confidence": 0.8, "notes": "No explicit robustness section appears in the excerpts. An internet appendix is mentioned but its contents are not visible. Statistical significance conventions and standard-error treatment are not visible."}, "model": "claude-opus-5-5", "stop_reason": "end_turn", "latency_s": 18.1, "usage": {"input": 6612, "output": 2195, "cache_read": 3968, "cache_write": 0}, "cost_usd": 0.071142, "request_id": "req_011CggqHH9dcjTCJ3Jp2ymKg", "cached": false, "ts": "2026-11-03T23:06:29+00:00", "effort": "medium"}