Submission Preflight (jfi-submission)

When to trigger

• "Submitting tomorrow" — last check before pressing submit in Editorial Manager
• Confirming what Editorial Manager expects at initial submission
• Checking the fee, abstract/keyword metadata, data links/statements, and required declarations

Process facts (verified 2026-06-22; re-confirm on the official Guide for Authors)

• JFI is published by Elsevier (ISSN 1042-9573); submission is via Editorial Manager, which converts uploaded files into a single PDF for peer review. Editable files (Word/LaTeX) are needed only for typesetting after acceptance.
• A US$500 submission fee applies to all new submissions, paid in Editorial Manager before the paper is considered; desk-rejected papers are not refunded. Transferred manuscripts do not pay another submission fee.
• Requires a concise factual abstract not exceeding 250 words and 1-7 English keywords.
• "Your-paper-your-way": no strict reference style at submission as long as it is internally consistent; the author–date Elsevier style is applied at proof.
• A Declaration of Generative AI and AI-assisted technologies in the writing process must appear before the References if such tools were used; AI cannot be an author.
• Optional but encouraged: Highlights (3–5 bullets, max 85 characters each) and free SSRN preprint posting during submission (not prior publication).

Preflight checklist

• Manuscript merges cleanly to one review PDF; numbered sections (1, 1.1, 1.1.1)
• References in any internally consistent style with clean author–year fields
• Concise, stand-alone abstract, <=250 words, no avoidable references
• US$500 fee ready in Editorial Manager (non-refundable on desk-reject)
• 1-7 English keywords chosen for indexing
• Generative-AI declaration placed before References (or omitted if not used)
• Decided on optional SSRN preprint and optional Highlights
• Option C data plan ready: deposit/cite/link research data, or explain why data cannot be shared
• Data Statement prepared; datasets linked in EM where applicable; [dataset] tags in references
• Conflict-of-interest / funding statements ready; not under review elsewhere

Fee-gated go/no-go before pressing submit

Because the fee is non-refundable on desk-reject, treat submission as a gated decision, not a formality:

• Gate 1 — fit: the jfi-topic-selection verdict is "strong," not "borderline"; a borderline intermediation fit means you are likely paying for a desk rejection.
• Gate 2 — frame: the abstract names the intermediary, friction, mechanism, and consequence (jfi-contribution-framing) — the desk screen reads no further.
• Gate 3 — design: the first objection a banking referee will raise (usually a demand-side confound) is already pre-empted in the introduction, not saved for Section 6.

Cover-letter calibration for this desk screen

Three short paragraphs: what the paper shows about intermediation (one identified or proved claim, with the headline magnitude or proposition); why this journal (name the sub-literature — relationship lending, capital regulation, deposit competition — not "fits the aims and scope"); and any editor-expertise match if you request a Managing Editor. Do not paste the abstract.

Last-mile glitches that cost real money here

• The EM-built review PDF scrambles landscape tables — proof the system's PDF, not your local compile.
• Keywords are generic ("banking," "finance") rather than naming the intermediation mechanism and data.
• Highlights written as full sentences exceeding the 85-character limit, truncated badly by the system.
• Fee paid but Data Statement or declarations missing, triggering an administrative return that delays triage. Confirm the current declaration list against the journal's live author guidelines before final filing.

Anti-patterns

• Submitting without paying the fee, or expecting a refund after desk-reject
• Submitting an abstract over 250 words or fewer than 1 / more than 7 English keywords
• Omitting the generative-AI declaration when such tools were used
• Anonymizing the manuscript as if it were double-blind — JFI is single-blind

Output format

【Review PDF】one clean file? [Y/N]
【Fee】US$500 ready in Editorial Manager? [Y/N]
【Abstract / Keywords】<=250 words and 1-7 English keywords? [Y/N]
【AI declaration】present-if-used, before References? [Y/N]
【Option C data plan / Data Statement / [dataset] tags】[Y/N]
【Next step】single-blind review → jfi-rebuttal on R&R