Imfer Theory Model

Theory and Model Craft (imfer-theory-model)

When to trigger

• A referee asks "what open-economy mechanism rationalizes this cross-country fact?"
• A reduced-form spillover or capital-flow estimate is credible but its welfare meaning is unclear
• You need a counterfactual a regression cannot deliver (an alternative exchange-rate regime, a CFM rule, a debt-restructuring path)
• A structural open-economy / DSGE model leads the paper and you must show it is disciplined by data, not assumption
• You are tempted to bolt a generic NK-SOE model onto an empirical paper and need to right-size it
• A welfare or optimal-policy number is the headline and you must defend its calibration and invariance

Why theory placement matters at IMFER

IMFER sits between the model-heavy international-finance tradition and the empirical-policy tradition, and the editor will route a paper to referees from whichever side it leads with. That makes the placement decision consequential: a strong reduced-form paper that opens with two pages of model derivation invites a theory referee who will judge the model on its own terms; a model-led paper with a thin empirical section invites an empirical referee who will find the discipline lacking. Decide early which tradition the paper belongs to, signal it in the structure, and make the other component genuinely earn its space rather than apologize for it.

The IMFER theory dial

IMFER publishes both model-led international-finance papers and empirical papers with a supporting framework. The dial is set by what the policy question needs: theory earns its place when it names a transmission channel, maps a reduced-form coefficient to a structural object, or delivers a counterfactual / welfare number a policymaker can use. Because IMFER counterfactuals are read as policy advice, the model's credibility and its policy-invariance matter as much as its elegance. Generic notation that re-labels the empirical result, with no testable prediction or magnitude, is the classic mis-fit.

When the model leads

A model-first IMFER paper must still earn policy relevance: tie each parameter to a data moment or external estimate (an impulse response, a comovement, an external-finance premium), validate against an untargeted moment, and report the policy counterfactual with its uncertainty. State the assumptions under which the counterfactual is policy-invariant — a referee who suspects the Lucas critique bites will not trust the welfare number.

When the model supports

For an empirical-first paper, pick the lightest tool: a sufficient-statistic expression in estimable elasticities often delivers the welfare object while keeping the credibility in the design. State what the sufficient statistic omits.

The open-economy building blocks referees expect

When a model is the lead, IMFER referees look for the canonical international-finance ingredients to be present and motivated, not decorative: a UIP/risk-premium block if exchange rates matter; a collateral or borrowing constraint if you study sudden stops or financial crises; a traded/non-traded split for real-exchange-rate dynamics; a foreign demand / terms-of-trade channel for spillovers; and incomplete markets if the welfare result hinges on risk-sharing failures. Each should map to a moment you target and a counterfactual the policy question needs — a model that includes a block doing no work invites the "decorative" critique.

Calibration vs. estimation — be explicit

IMFER referees distinguish sharply between calibrated and estimated parameters, and want each labeled. State which parameters are set from external sources (and cite them), which are estimated internally (and from what moment), and which are free. A model that quietly calibrates the parameter doing the welfare work — then reports a precise optimal-policy number — invites the charge that the result is assumed. Where the headline magnitude is sensitive to a calibrated value, show the welfare object across a defensible range of that value rather than a single point.