model 05 — US heterogeneous-agent NK model · us-hank · US · hosted

See who responds to a US shock, not just how much.

Run stylized monetary, fiscal-spending, and productivity shocks through a two-asset heterogeneous-agent New Keynesian model — aggregate impulse responses plus first-order distributional output: MPCs by wealth quartile and the hand-to-mouth share.

with the microsimulation

Two distributional cuts of the same shock: the model's and the microsimulation's.

→Connection. Shock-to-incidence bridge: hank_shock_incidence scales the US microsimulation's employment income with the model's wage and labor impulse responses — the model's MPC-by-wealth story on one side, statutory tax-and-transfer incidence on the other. Experimental illustration. score_reform refuses model="hank": three stylized instruments map to no statute.

→Worked example. One monetary shock, two answers: the HANK block reports consumption responses by wealth position; the microsimulation reports the same earnings change after taxes, SNAP, and the EITC, decile by decile.

→Validation. Replicates Auclert et al. (2021) via the authors' toolkit — gated on targets, market clearing, and shock responses.

→Results. Run hank_shock from any connected client.

01 — what it is

The distributional member for the US.

A validated replication of the two-asset heterogeneous-agent New Keynesian (HANK) model of Auclert, Bardóczy, Rognlie and Straub (Econometrica 2021, "Using the Sequence-Space Jacobian to Solve and Estimate Heterogeneous-Agent Models"), built directly on the authors' sequence-jacobian toolkit at the paper's production grid sizes, not the packaged demo grids.

Where frb-us answers "how much does US output move after a shock", this model answers who moves. Households differ in liquid and illiquid wealth — and so in marginal propensity to consume — and that whole distribution carries through the general-equilibrium response. A VAR-free, sequence-space HANK: no estimated reduced form, no expectations shortcut — first-order impulse responses from the model's own general-equilibrium Jacobian.

Not a forecaster, and no detailed tax-reform scoring: it scores stylized shocks — monetary, fiscal spending, productivity — around a calibrated steady state, with distributional outputs that are first-order approximations from steady-state policies.

02 — what it can answer

Stylized US shocks, with the household distribution attached.

Its job is the transmission of a stylized aggregate shock through a realistic household wealth distribution: a monetary easing raising output, consumption, investment and inflation; a tax-financed spending expansion raising output while the endogenous labor tax leans on consumption; the response concentrating among high-MPC, low-liquid-wealth households — the point of a HANK model, which a representative-agent model cannot say.

It cannot:

Forecast — responses are deviations around a calibrated steady state, not a projection of the US economy.
Score a PolicyEngine reform — score_reform refuses model="hank" outright: three stylized instruments and an endogenous labor tax mean no mapping from statute exists. For statutory reforms scored end to end, use the micro-to-macro workflow; for who wins and loses at household resolution, the microsimulation answers directly.
Go beyond first order — responses are linear in the shock, with no state dependence, and the distributional cuts are first-order approximations from steady-state policies, not full household-level dynamics.

It does connect to the microsimulation in one direction: hank_shock reports the pre-tax real wage and labor impulse responses, and hank_shock_incidence scales the US microsimulation's employment income with them — the model's own MPC-by-wealth story on one side, the statutory tax-and-transfer incidence of the same shock on the other. Two complementary distributional cuts of one experiment, clearly labelled an experimental illustration.