---
name: jfi-contribution-framing
description: Use when framing the contribution of a Journal of Financial Intermediation (JFI) paper around what it teaches about intermediation — the institution, the friction, the mechanism, and the economic consequence — so it clears the desk screen and reads as banking, not generic finance. It frames; it does not write the paper.
---

# Contribution Framing (jfi-contribution-framing)

## When to trigger

- Drafting the abstract and the contribution paragraph of the introduction
- A reader cannot tell, in one sentence, why intermediaries are central to your result

## The JFI framing bar

Because JFI runs an **active desk-rejection** screen, the contribution must be legible in the abstract and
first pages. Frame it as a statement about **intermediation**, not finance in general. A strong frame names
four things:

1. **Institution / intermediary** — banks, lenders, dealers, insurers, the specific actor.
2. **Friction** — the economic problem the intermediary faces or resolves (information asymmetry,
   monitoring, liquidity transformation, capital constraints, agency).
3. **Mechanism** — how that friction produces the result; the causal or theoretical channel.
4. **Consequence** — why it matters for credit, stability, welfare, or policy.

"Banks do X" is weak; "Because of friction F, intermediary type I responds via mechanism M, with
consequence C for credit/stability" is a JFI contribution.

## Empirical vs. theory framing

- **Empirical:** lead with the identified fact and the mechanism it reveals, then the broader lesson for
  intermediation — not a coefficient in search of a story.
- **Theory:** lead with the friction and the new prediction the model generates; state the testable
  implication so the contribution is not purely formal.

## Worked vignette: framing a capital-shock transmission paper

A hypothetical submission (all numbers illustrative): using a supervisory credit register, the authors
find that banks hit by a 1-percentage-point capital shortfall after a stress-test redesign cut credit to
the **same firm** by 3.4% more than unaffected banks (firm×time fixed effects), with the cut twice as
large for relationship borrowers lacking alternative lenders.

- **Weak frame:** "We study the effect of stress tests on bank lending." No friction named; could run in
  any banking outlet.
- **Better, still short:** "Capital regulation reduces credit supply." Mechanism missing — why capital,
  and why these borrowers?
- **JFI-ready:** "Because raising equity is costly (friction), constrained banks deleverage where their
  information monopoly is strongest (mechanism — relationship borrowers cannot switch), so capital
  regulation taxes precisely the borrowers intermediation theory says banks exist to serve (consequence)."
- The last frame clears the screen because it connects an identified estimate to a Bhattacharya–Thakor-style
  account of what intermediaries do — that dialogue with intermediation theory is the JFI bar.

## Frame-strength ladder for the desk screen

| Rung | The abstract reads as | Likely JFI desk outcome |
|---|---|---|
| 1 | A correlation about banks | High desk-reject risk |
| 2 | An identified effect, mechanism unnamed | Vulnerable: "fine design — what is the intermediation lesson?" |
| 3 | Effect + named friction + channel | Survives triage; referees then test the channel |
| 4 | Rung 3 plus theory dialogue (which intermediation model is disciplined or rejected) | Strongest JFI frame |

Aim for rung 3 at minimum; reach rung 4 whenever the literature offers competing intermediation models
your estimate can separate.

## Referee pushback on the frame, and the JFI fix

- "This could be a demand-side story" → re-anchor the claim to the supply channel your within-firm design
  isolates, and say so in the abstract, not only in Section 5.
- "Banks are incidental here" → make the intermediary's balance sheet or information role load-bearing,
  or concede the paper belongs at a general-finance outlet.
- "The welfare claim outruns the design" → downgrade the consequence from "welfare" to the credit,
  stability, or real-outcome margin the design actually measures.

As a calibration (reading-based, not a rule): accepted JFI introductions usually state the friction and
mechanism within the first two paragraphs and the headline magnitude by paragraph three; the contribution
paragraph names what the result teaches intermediation theory, not just what the regression found.

## Anti-patterns

- A contribution that would fit any finance journal (no intermediary mechanism)
- Burying the mechanism past page 3, where the desk screen never reaches it
- Over-claiming policy or welfare beyond what the design or model supports
- Listing results without saying what they teach about intermediation

## Output format

```
【Institution】<the intermediary>
【Friction → Mechanism】<problem → channel>
【Consequence】<why it matters>
【One-sentence contribution】<the abstract-ready claim>
【Next skill】jfi-tables-figures
```