Two-asset HANK

US HANK

The model blocks, the steady state and the sequence-space solve.

Steady state, Jacobian, then instant IRFs

From the household structure to the first-order distributional block, built on the paper authors' sequence-jacobian toolkit.

1

Heterogeneous agents, two assets

The model is 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").

Each household holds two assets liquid wealth varies by household illiquid wealth varies by household consumption decision MPC varies with liquid wealth the full distribution enters the general-equilibrium response
click to expand
Households differProduction grids, not demo grids

Household structure

Assets
liquid + illiquid, per household
Grids
50 liquid × 70 illiquid points, 3 income states (plus a 50-point auxiliary solver grid)
Variants
two_asset (default) · one_asset (no capital)

Households differ

Households differ in liquid and illiquid wealth, so their marginal propensities to consume differ, and the model carries that whole distribution through the general-equilibrium response — the thing a representative-agent model cannot say.

Production grids, not demo grids

The model runs at the paper's production grid sizes (above), not the packaged demo grids.

  • It is a VAR-free, sequence-space HANK model throughout, with no estimated reduced form and no expectations shortcut.
  • A fast one-asset textbook variant is solved on the toolkit notebook’s grid: 7 income states × 500 asset points. Before us-hank-model #13 it used the 2 × 10 demo grid.
  • The Use it tab says what it drops.
2

Calibrate and solve the steady state

The first stage solves the general-equilibrium steady state at the production grids.

Calibrate to the paper's targets, solve once, cache calibrate discounting + portfolio-adjustment costs match targets the paper's asset-market targets, parameters checked steady state a few seconds, once, cached at the paper's production grids — no data vintage to refresh
click to expand
Solved parameters checked against the paperSteady-state solve

Steady-state solve

Calibrated
discounting + portfolio-adjustment costs
Targets
the paper's asset-market targets
Cost
a few seconds, computed once, cached

Solved parameters checked against the paper

The seven parameters the model solves for are compared with the paper's Table B.III on the validation page; the calibration targets themselves are inputs. The parameterisation is the published paper's throughout — there is no data vintage to refresh.

3

Differentiate: the sequence-space Jacobian

The second stage builds the general-equilibrium sequence-space Jacobian: the T×T linear map from each exogenous path to each outcome path, with T=300 quarters.

One T×T matrix per shock–outcome pair (T = 300 quarters) outcome path T × 1 = GE Jacobian T × T · a few seconds, cached × shock path T × 1 warm IRF matrix–vector product, effectively instant
click to expand
Instant IRFsJacobian

Jacobian

Object
T×T map: exogenous path → outcome path
Horizon
T=300 quarters
Cost
a few seconds, cached alongside the steady state

Instant IRFs

With both caches warm, an impulse response is a matrix–vector product, effectively instant. The first call per variant pays the steady-state and Jacobian solves; every later call reuses them.

4

Shock: three instruments, endogenous labour tax

A shock is an AR(1) path, εt=size×ρt, on one of three instruments:

An AR(1) shock decays each quarter: size × persistenceᵗ size quarters t → three instruments monetary · fiscal_spending · productivity endogenous labour tax balances the budget each period
click to expand
The design point that mattersThe labour tax is endogenous

Shocks

Path
εt=size×ρt (AR(1))
Instruments
monetary · fiscal_spending · productivity
Fiscal closure
labour tax balances the budget each period

The design point that matters

The labour tax is endogenous. The fiscal block balances the government budget each period.

  • So a spending shock is implicitly tax-financed, and there is no exogenous transfer or tax-rate instrument to shock.
  • Requests for one are refused with that explanation, not answered with an invented mapping.
  • The score_reform tool refuses any model="hank" request outright.
More detail3 points
  • monetary: the Taylor-rule intercept. The policy rate is an outcome, intercept + 1.5 × inflation, so its response depends on the inflation the shock causes and on persistence (measured).
  • fiscal_spending: real government spending, financed by the labour tax.
  • productivity: total factor productivity (TFP).
5

Distribute — within first-order limits

From the steady-state joint distribution and consumption policy, the model reports:

MPCs by liquid-wealth quartile (schematic, not data) Q1 Q2 Q3 Q4 liquid-wealth quartile → hand-to-mouth share from the steady-state distribution impact consumption by total-wealth quartile, allocated by steady-state MPCs first-order: linear in the shock, deviations — not forecasts
click to expand
What "first-order" means hereLinear in the shock, no state dependence

What "first-order" means here

Responses are linear in the shock, with no state dependence: the distributional cuts capture the MPC-heterogeneity channel, not full household-level dynamics along the transition. Every result is a deviation around the calibrated steady state. The replication gate and the full limits are on Validation.

More detail3 points
  • MPCs by liquid-wealth quartile
  • the hand-to-mouth share, counted by liquid wealth
  • the impact consumption response by total-wealth quartile, allocated in proportion to steady-state MPCs
Notation 2 symbols
SymbolMeaningStep
Thorizon of the sequence-space Jacobian, in quarters3
εt, size, ρAR(1) shock path, its impact size and persistence4