The New IS-LM Model: Language, Logic, and Limits

This article gives a simple, self-contained exposition of what King calls the “New IS-LM model” – a small, three-equation macroeconomic system, built from optimizing microfoundations and analyzed under rational expectations, consisting of a forward-looking IS equation (current output depends on expected future output and the real interest rate), a Fisher equation (the nominal rate equals the real rate plus expected inflation), and an expectational (“New Keynesian”) Phillips curve (current inflation depends on expected future inflation and the current output gap). King situates the model historically as the outgrowth of a “New Neoclassical Synthesis” that answers the rational-expectations-era critique of the original Hicksian IS-LM framework and its 1970s descendants, while explicitly noting the model is not itself derived from first principles in this article but is instead a distillation used to communicate results from more fully articulated, microfounded models. Working through the model’s implications, King shows that a “neutral” monetary policy – one that always keeps output at its capacity level – implies a specific, and in general history-dependent, inflation target: inflation should be exactly zero if there are no exogenous “inflation shocks,” and otherwise the target should absorb the persistence properties of those shocks while never responding to shocks to aggregate demand, capacity growth, or money demand. He derives the forward-looking New Keynesian Phillips curve explicitly from Calvo-style staggered, forward-looking price-setting by monopolistically competitive firms, under an admittedly “heroic” assumption linking real marginal cost to the output gap, and shows the resulting curve implies essentially no long-run trade-off between inflation and output, even though nominal disturbances can still generate output effects that persist for many periods – resolving an apparent tension between the model’s long-run classical neutrality and the empirically persistent business cycles that motivated earlier “old Keynesian” IS-LM analysis. Turning to policy-rule design, King shows that interest rate rules of the Taylor type must satisfy restrictive parameter conditions – broadly, an “aggressive” response of more than one-for-one to inflation – to guarantee a unique, stable rational-expectations equilibrium rather than a continuum of self-fulfilling “sunspot” equilibria, and demonstrates the specific and somewhat counterintuitive result that this zone of admissible, determinacy-preserving rules is actually smaller (the zone of indeterminacy larger) once prices are sticky than in the flexible-price benchmark, with the exact boundary conditions differing sharply depending on whether the rule responds to current or expected future inflation. He closes by explicitly flagging the model’s own limits: it cannot, from within itself, justify why output stabilization at capacity is welfare-improving, define what an “inflation shock” actually is at a structural level, or evaluate the consequences of omitting investment and capital altogether – questions that, King stresses, can only be answered by stepping outside the New IS-LM model into the fully articulated, microfounded models it is meant to summarize.

Robert G. King | Macro Paper Warehouse

The New IS-LM Model: Language, Logic, and Limits