rof-tables-figures

Review of Finance Tables And Figures

Use this when exhibits are too dense, under-explained, or wasting the 60-page budget. At RoF

exhibits are not free space: the hard 60-page cap explicitly counts appendices, bibliography,

figures, and tables, so every exhibit must earn its place and be self-contained for a

top-three-standard referee.

When to trigger

• The main table has too many columns and buries the headline result.
• Figures lack confidence bands, units, or self-contained notes.
• The manuscript is near or over 60 pages and exhibits are part of why.

Exhibit rules

• Lead with one clean headline exhibit a reader grasps in seconds; keep the main paper focused

on exhibits that carry the finance contribution or address likely top-three-standard

objections.

• Use table notes to document sample, period, variable construction, fixed effects,

clustering, controls, units, and economic magnitude — a referee should read the exhibit

without hunting through the text.

• Prefer figures where a relationship is visual: event-time cumulative returns, sorted

portfolios, factor loadings. Show confidence bands; avoid chartjunk; use vector output

(PDF/EPS).

• Move mechanical robustness grids to the supplementary replication package rather than padding

in-paper exhibits, keeping the main identification logic self-contained.

• For theory papers, use figures to clarify comparative statics or equilibrium regions, not to

decorate algebra; keep proofs compact so they do not eat the page budget.

• Respect the formatting baseline (at least 12-point font, at least 1.5 line spacing) and

re-check that appendices, bibliography, figures, and tables still fit the 60-page cap.

Page-budget worked example

Illustrative. A corporate-finance draft carries 14 tables and 5 figures — roughly 19

exhibit pages, nearly a third of the 60-page ceiling before a word of prose. The

RoF-shaped cut keeps in the body: summary statistics, the headline result, the

identification exhibit (an event-study figure), one mechanism table, and one robustness

table — five to seven exhibits. Everything else moves to the internet appendix behind

one-sentence pointers ("Table IA.4 repeats column 3 under alternative winsorization").

The body lands near eight exhibit pages, leaving room for the argument the referees grade.

Exhibit calibration against the journal (hedged)

• Accepted general-interest finance papers tend toward six to ten body exhibits with a

substantial internet appendix carrying the robustness battery; conventions differ by

subfield, so sample recent RoF issues before fixing your own count.

• Supplementary material is hosted via OUP alongside the article and the replication

package; confirm appendix naming and hosting mechanics against the journal's current

author guidelines.

Referee objections to exhibits and the conforming fix

| Objection | Fix that satisfies an RoF report |

|---|---|

| "Table 5 is uninterpretable" | split into panels; isolate the headline column; add the magnitude sentence to the note ("one s.d. of X moves Y by Z bp") |

| "The figure carries no inference" | add 95% bands and state the clustering level in the figure note |

| "Where is the robustness for claim 3?" | a visible internet-appendix pointer — robustness may leave the body but never the paper trail |

| "European and US samples are not comparable" | harmonization note: source per sample (Datastream vs. CRSP), screens applied, currency conversion, matching filters |

| "Exhibit numbers do not match the text" | regenerate all exhibits from the replication scripts (one numbered script per exhibit) before resubmitting |

Execution bridge (StatsPAI / Stata MCP)

Generate exhibits from the fitted result, not by retyping numbers. Full map:

[execution-with-mcp](../../../shared-resources/empirical-methods/execution-with-mcp.md). Review of Finance is the EFA flagship — corporate-causal chain for corporate papers, factor-zoo haircut for asset pricing.

• Tables: etable (multi-model) or did_summary_to_latex straight from the result_id.
• Figures: plot_from_result / enhanced_event_study_plot / event_study_table —

axis units and the SE/clustering note baked in.

• Every note names the estimator + clustering and states the magnitude in interpretable units.

See a full fitted-result → exhibit chain in the [JF execution walkthrough](../../../Journal-of-Finance-Skills/resources/worked-examples/02-execution-walkthrough.md).

Anti-patterns

• A nine-column kitchen-sink table as the main result.
• Figures with no units, no bands, or notes that require the body text to interpret.
• Letting appendix tables and the bibliography silently push the paper past 60 pages.

Output format

[Exhibit status] keep / compress / move / delete
[Finance role] <main result, identification, mechanism, robustness, or theory>
[Missing note] <sample, variables, controls, SEs, units, or data source>
[Page-budget action] <specific cut or move>