{
 "validation_errors": [],
 "doc_id": "jof_12728",
 "ontology_version": "1.0",
 "prompt_version": "1.0",
 "model": "claude-opus-5-5",
 "effort": "medium",
 "annotated_at": "2026-11-03T23:05:53+00:00",
 "source_sha256": "bc468fa86ab06455edd10dffa8cd03438e62be388a374345ae554584b1ffaee8",
 "parser": "pymupdf-jof-heuristics-1.3",
 "payload_chars": 27554,
 "usage": {
  "input": 9337,
  "output": 2419,
  "cache_read": 3968,
  "cache_write": 0
 },
 "cost_usd": 0.086522,
 "request_id": "req_011CggqEUyCaGmKuyx6hjE3n",
 "annotation": {
  "genre": "empirical",
  "subfield": "market_microstructure",
  "subfield_secondary": "OTC markets / fixed income (municipal bonds), network analysis",
  "primary_method": "reduced_form_regression",
  "identification": {
   "estimand": "Association between a dealer's network centrality and investor execution quality: round-trip markups, intermediation chain length, prearranged versus principal trading, and inventory duration.",
   "strategy": "Panel regressions of markups and trade outcomes on a composite centrality index (first principal component of network measures), with state and month fixed effects and rich dealer/trade/bond controls. Within-dealer variation comes from dealer fixed effects and state-specific centrality based on tax-segmented state markets. Additional tests address alternative channels: adverse selection, bond riskiness and inventory.",
   "key_assumptions": [
    "Conditional on controls, centrality is not a proxy for dealer size or unobserved dealer quality.",
    "Central dealers do not systematically intermediate a priori riskier or more illiquid bonds.",
    "State tax segmentation creates local submarkets in which state-specific centrality varies independently of aggregate dealer attributes."
   ],
   "threats_addressed": [
    "dealer size / scale effects",
    "bond riskiness and illiquidity selection",
    "inventory risk channel",
    "asymmetric information / adverse selection",
    "time-invariant dealer heterogeneity (dealer fixed effects)"
   ],
   "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": 1
   },
   "results": {
    "present": true,
    "sentence_index": 2
   },
   "magnitude": {
    "present": true,
    "sentence_index": 2
   },
   "contribution": {
    "present": true,
    "sentence_index": 4
   }
  },
  "introduction": {
   "moves": [
    {
     "paragraph_index": 0,
     "move": "motivation"
    },
    {
     "paragraph_index": 0,
     "move": "literature_gap"
    },
    {
     "paragraph_index": 0,
     "move": "preview_of_approach"
    },
    {
     "paragraph_index": 1,
     "move": "preview_of_results"
    },
    {
     "paragraph_index": 1,
     "move": "mechanism_or_theory_sketch"
    },
    {
     "paragraph_index": 2,
     "move": "other"
    },
    {
     "paragraph_index": 3,
     "move": "identification_defense"
    },
    {
     "paragraph_index": 3,
     "move": "preview_of_results"
    },
    {
     "paragraph_index": 4,
     "move": "robustness_mention"
    },
    {
     "paragraph_index": 4,
     "move": "identification_defense"
    },
    {
     "paragraph_index": 5,
     "move": "preview_of_results"
    },
    {
     "paragraph_index": 5,
     "move": "related_literature_positioning"
    },
    {
     "paragraph_index": 5,
     "move": "contribution_statement"
    }
   ],
   "contribution_paragraph_index": 5,
   "contribution_types": [
    "new_evidence",
    "new_data_or_measure"
   ],
   "roadmap_present": false,
   "literature_positioning_location": "both"
  },
  "section_roles": [
   {
    "idx": 0,
    "role": "introduction"
   },
   {
    "idx": 1,
    "role": "literature"
   },
   {
    "idx": 2,
    "role": "data"
   },
   {
    "idx": 3,
    "role": "results"
   },
   {
    "idx": 4,
    "role": "results"
   },
   {
    "idx": 5,
    "role": "results"
   },
   {
    "idx": 6,
    "role": "results"
   },
   {
    "idx": 7,
    "role": "robustness"
   },
   {
    "idx": 8,
    "role": "conclusion"
   },
   {
    "idx": 9,
    "role": "appendix"
   },
   {
    "idx": 10,
    "role": "appendix"
   },
   {
    "idx": 11,
    "role": "appendix"
   },
   {
    "idx": 12,
    "role": "appendix"
   }
  ],
  "results_presentation": {
   "economic_significance_discussed": true,
   "statistical_significance_language": "unclear",
   "magnitude_statements_in_intro": true
  },
  "robustness": {
   "location": "main_text_section",
   "types": [
    "alternative_specification",
    "subsample",
    "alternative_measure",
    "falsification",
    "other"
   ]
  },
  "tables": {
   "convention_notes": "Long descriptive table notes state the estimator (panel OLS with state and month fixed effects, Probit, Tobit), point to Appendix D for control definitions, define the centrality index as a first principal component, and report markups in percent. Tables progress from univariate tier sorts to multivariate regressions and then subsample tests.",
   "units_stated_in_notes": true,
   "standard_errors_description": "unclear"
  },
  "figures": {
   "uncertainty_shown": "no"
  },
  "conclusion": {
   "bounded_claims": false,
   "policy_implications": true,
   "open_questions": false,
   "restates_contribution": true
  },
  "hedging": {
   "level": "moderate",
   "typical_devices": [
    "'consistent with' theory predictions",
    "'suggests that'",
    "framing of results as tests of hypotheses derived from models"
   ]
  },
  "claim_evidence": [
   {
    "claim": "The interdealer network has a stable core-periphery structure with a small set of hubs and thousands of peripheral dealers.",
    "evidence_type": "figure",
    "evidence_ref": "Figure 1, Figure 2, Table II"
   },
   {
    "claim": "Central dealers charge investors higher round-trip markups than peripheral dealers.",
    "evidence_type": "table",
    "evidence_ref": "Table IV, Table V"
   },
   {
    "claim": "Central dealers sell more directly to end customers and are associated with shorter intermediation chains.",
    "evidence_type": "table",
    "evidence_ref": "Table VI, Table VII"
   },
   {
    "claim": "Central dealers prearrange fewer trades and unwind inventory faster, implying more immediacy.",
    "evidence_type": "table",
    "evidence_ref": "Table VIII, Table XIII"
   },
   {
    "claim": "Investors turn more to central dealers in stress times.",
    "evidence_type": "table",
    "evidence_ref": "Table IX"
   },
   {
    "claim": "The adverse-selection component of markups is lower for central dealers.",
    "evidence_type": "table",
    "evidence_ref": "Table X"
   },
   {
    "claim": "State-specific centrality affects markups more than aggregate centrality and adds explanatory power beyond dealer fixed effects.",
    "evidence_type": "table",
    "evidence_ref": "Table XIV"
   }
  ],
  "structural_summary": "The introduction motivates the role of dealers in OTC markets and notes how little empirical work exists for lack of data. It previews the regulatory data, the main results, a state-segmentation identification refinement, tests ruling out alternative explanations, and the contribution relative to the literature. A separate section then derives hypotheses from network models. The paper next describes the data and the round-trip chain construction. Successive results sections cover network structure, the centrality premium in markups, search efficiency, and execution speed and stress-time trading. A section on alternative theories (adverse selection, riskiness, inventory, local information, subsamples) follows. The conclusion restates the cost–speed trade-off and the systemic role of central dealers.",
  "writing_profile_keywords": [
   "hypothesis-driven empirics",
   "theory-motivated tests",
   "network centrality",
   "regulatory audit trail data",
   "alternative explanations section",
   "OTC market microstructure",
   "descriptive-then-multivariate tables"
  ],
  "confidence": 0.8,
  "notes": "Section VII is labelled 'theory' by the role rule but functions as robustness/alternative explanations, so it was assigned robustness. Standard error treatment and significance notation are not visible in the excerpts. No roadmap is visible in the excerpted introduction paragraphs."
 },
 "deterministic_checks": {
  "roadmap_regex": false,
  "contribution_regex_in_intro": true,
  "magnitude_in_abstract": false,
  "abstract_n_sentences": 5,
  "intro_n_paragraphs": 6,
  "intro_n_words": 1281,
  "causal_words_in_abstract": 0,
  "hedge_words_in_abstract": 1,
  "stars_mentioned": false,
  "clustered_se_mentioned": false,
  "internet_appendix_mentioned": false,
  "section_roles_rule": {
   "0": "introduction",
   "1": "literature",
   "2": "data",
   "3": "other",
   "4": "other",
   "5": "other",
   "6": "other",
   "7": "theory",
   "8": "conclusion",
   "9": "appendix",
   "10": "appendix",
   "11": "appendix",
   "12": "appendix"
  }
 },
 "schema_valid": true,
 "code_version": "feb6c9e-dirty"
}