Wissenschaftliche Quellen
Townsend (1979) -- Optimal Contracts and Competitive Markets with Costly State Verification, the micro foundation for the CSV debt contract.
Journal of Economic Theory · 1979
Gale and Hellwig (1985) -- Incentive-Compatible Debt Contracts: The One-Period Problem, which shows standard debt is the optimal contract under costly state verification.
Review of Economic Studies · 1985
Bernanke and Gertler (1989) -- Agency Costs, Net Worth, and Business Fluctuations, the original net-worth mechanism in an overlapping-generations model.
Carlstrom and Fuerst (1997) -- Agency Costs, Net Worth, and Business Fluctuations: A Computable General Equilibrium Analysis, the CSV contract in a real business cycle model.
Kiyotaki and Moore (1997) -- Credit Cycles, the parallel collateral-constraint tradition with land as the durable asset.
Journal of Political Economy · 1997
Bernanke, Gertler, and Gilchrist (1999) -- The Financial Accelerator in a Quantitative Business Cycle Framework, the canonical DSGE implementation.
Handbook of Macroeconomics · 1999
Christiano, Motto, and Rostagno (2014) -- Risk Shocks, which makes the dispersion of idiosyncratic returns a shock and estimates it to be a leading driver of US business cycles.
American Economic Review · 2014
Iacoviello (2005) -- House Prices, Borrowing Constraints, and Monetary Policy in the Business Cycle, housing-collateral variant.
American Economic Review · 2005
Gertler and Kiyotaki (2010) -- Financial Intermediation and Credit Policy in Business Cycle Analysis, moved the friction to the banking sector.
Handbook of Monetary Economics · 2010
Del Negro, Giannoni, and Schorfheide (2015) -- Inflation in the Great Recession and New Keynesian Models, an estimated Smets-Wouters model with a BGG block, the basis of the New York Fed's DSGE model.
American Economic Journal: Macroeconomics · 2015
Jermann and Quadrini (2012) -- Macroeconomic Effects of Financial Shocks, identified financial shocks as a major business cycle driver.
American Economic Review · 2012
Forschungsgrundlage
Evidenz und Daten
Track credit spreads, net worth, collateral values, investment, and defaults as the amplification channel.
Kalibrierung oder Messung
External finance premium, net-worth sensitivity, and shock persistence drive the strength of amplification.
Grenzen
- Bank runs and liquidity spirals need separate mechanics.
- Distribution within firms is compressed.
- Identification requires credit-market evidence.
Hinweise zur Verwendung
Wann das Modell ausreicht
Credit-driven amplification of real shocks when the external finance premium moves inversely with borrower net worth and collateral values. The Bernanke-Gertler-Gilchrist 1999 (Handbook of Macroeconomics) mechanism is the right tool when declining asset prices tighten credit, raise the cost of external funds, and depress investment beyond what the underlying productivity shock alone would produce. The Kiyotaki-Moore 1997 (JPE) collateral channel adds the endogenous asset-price feedback loop.
Wann es nur eine Skizze ist
Do not use for bank-run dynamics, fire-sale spirals in shadow banking, or the full pathology of a systemic crisis. The model amplifies real shocks through borrower balance sheets; it does not model bank capital impairment, interbank-market freezes, or the wholesale-funding runs that characterized 2007 to 2009. Those mechanics require explicit bank-liability modeling.
Wann ein Modellwechsel sinnvoll ist
Switch to a model with explicit bank capital constraints (Gertler-Kiyotaki 2010 Handbook; Gertler-Karadi 2011 JME) when the binding friction is on the bank's own balance sheet rather than the borrower's. Switch to a sudden-stop model (Mendoza 2010 AER) when emerging-market capital-flow reversals and sovereign risk define the crisis.
Widerlegungssignal
A deep credit contraction that produces only a mild output decline would falsify the accelerator mechanism, since the model's defining property is that credit tightening amplifies the real shock beyond the direct effect. A chart of credit spreads rising sharply and persistently without a corresponding widening of the output decline relative to trend would indicate the amplification channel is not operating as specified.