Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Original file line number Diff line number Diff line change
@@ -0,0 +1,21 @@
{
"cells": [
{
"cell_type": "markdown",
"id": "7067c25d",
"metadata": {},
"source": [
"Dupor, Bill and Karabarbounis, Marios and Kudlyak, Marianna and Saif Mehkari, M, Regional Consumption Responses and the Aggregate Fiscal Multiplier\n",
"\n",
"Original ballpark: Nathan Robino — February 2026"
]
}
],
"metadata": {
"language_info": {
"name": "python"
}
},
"nbformat": 4,
"nbformat_minor": 5
}
Original file line number Diff line number Diff line change
@@ -0,0 +1,54 @@
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Prior literature summary\n",
"\n",
"{cite:t}`Dupor2023-bn` builds directly on two empirical traditions in fiscal macroeconomics. First, a large ARRA-based regional literature developed causal estimates of local spending effects using plausibly exogenous allocation rules and IV strategies (for example, {cite:t}`Wilson2012-bc`; {cite:t}`Chodorow-Reich2012-bo`; {cite:t}`Conley2013-ca`; {cite:t}`Leduc2017-yn`; {cite:t}`Dupor2018-zh`; {cite:t}`Auerbach2019-pi`; and {cite:t}`Chodorow-Reich2019-zl`). These papers established that cross-region fiscal shocks can raise local employment, wages, and activity, while also emphasizing spillovers and aggregation issues. In parallel, identification work ({cite:t}`Ramey2009-jx,Ramey2016-tn,Ramey2019-go`; {cite:t}`Nakamura2011-dw`; see also {cite:t}`Nakamura2018-wz`) clarified how to interpret local multipliers versus national multipliers and why cross-sectional estimates are informative but not automatically sufficient for aggregate policy conclusions.\n",
"\n",
"The paper also relies on a major theoretical shift from representative-agent models toward heterogeneous-agent, incomplete-markets New Keynesian frameworks. Classic incomplete-markets foundations ({cite:t}`Huggett1993-ft,Aiyagari1994-qp`) and later HANK advances ({cite:t}`Kaplan2014-ee`; {cite:t}`Kaplan2018-sr`; {cite:t}`Auclert2018-ii`; {cite:t}`Auclert2019-dr`; {cite:t}`Hagedorn2019-op`) showed that incomplete risk sharing, liquidity constraints, and MPC heterogeneity are central for fiscal transmission, especially near the zero lower bound (also in {cite:t}`Christiano2011-ro`; {cite:t}`Woodford2011-jp`; {cite:t}`Eggertsson2009-yu`; {cite:t}`Farhi2012-oe`). Empirical household-spending evidence from rebates and balance-sheet shocks ({cite:t}`Sahm2010-sz,Sahm2011-ta`; {cite:t}`Parker2013-py`; {cite:t}`Mian2013-wr`; {cite:t}`Jappelli2014-og`) made these mechanisms quantitatively credible. Together, this literature made Dupor et al.'s strategy possible: estimate local consumption responses in rich micro data, then map them to aggregate multipliers in a multi-region HANK model with trade linkages and incomplete risk sharing.\n",
""
],
"id": "eadff64f"
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 5 foundational papers \n",
"\n",
"1. **Aiyagari (1994), _Uninsured Idiosyncratic Risk and Aggregate Saving_** *(Incomplete-markets foundation)* \n",
" Canonical heterogeneous-agent benchmark showing how uninsurable risk and borrowing constraints shape aggregate outcomes; foundational for later HANK fiscal analysis.\n",
"\n",
"2. **Christiano, Eichenbaum, and Rebelo (2011), _When Is the Government Spending Multiplier Large?_** *(New Keynesian fiscal theory / ZLB)* \n",
" Landmark theoretical result that multipliers can be substantially larger near the zero lower bound, central to post-crisis fiscal-policy modeling.\n",
"\n",
"3. **Nakamura and Steinsson (Fiscal Stimulus in a Monetary Union, NBER/AER line of work)** *(Regional empirical identification)* \n",
" Pioneering use of cross-region variation in a monetary union to identify local spending multipliers; a core template for ARRA-based empirical designs.\n",
"\n",
"4. **Kaplan, Moll, and Violante (2018), _Monetary Policy According to HANK_** *(HANK transmission and redistribution channels)* \n",
" High-impact HANK framework showing that heterogeneity, liquidity, and general-equilibrium income effects dominate representative-agent transmission logic.\n",
"\n",
"5. **Parker, Souleles, Johnson, and McClelland (2013), _Consumer Spending and the Economic Stimulus Payments of 2008_** *(Household spending evidence / MPCs)* \n",
" Influential micro evidence of sizable spending responses to transfers, validating high-MPC mechanisms that heterogeneous-agent fiscal models rely on.\n",
"\n",
"Together, these five papers span the key pillars Dupor et al. (2023) builds on: identification of local fiscal effects, heterogeneous-agent theory, and micro evidence on consumption responses."
],
"id": "c52e9d42"
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"name": "python"
}
},
"nbformat": 4,
"nbformat_minor": 5
}
Original file line number Diff line number Diff line change
@@ -0,0 +1,57 @@
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Subsequent literature (2019+)\n",
"\n",
"Since 2019, the literature most related to {cite:t}`Dupor2023-bn` has moved from asking whether fiscal multipliers are positive to asking which multiplier we mean, in which environment, and through which distributional channels it operates. A first direction is sharper theory of heterogeneous-agent transmission under realistic policy regimes. {cite:t}`fiscal_hagedorn_2019` formalizes a HANK environment with incomplete markets, nominal rigidities, and endogenous labor-market responses, showing that financing and monetary accommodation are first-order determinants of multiplier size. This work helped normalize the idea that a single \"fiscal multiplier\" is not a structural constant: it varies with deficits versus contemporaneous taxation, the policy rule followed by the central bank, and the distribution of marginal propensities to consume across households. That line is extended and systematized by {cite:t}`fiscal_auclert_2024`, which provides a unified modern framework for fiscal and monetary interactions in heterogeneous-agent models. The cutting-edge shift here is methodological as much as substantive: policy analysis is increasingly done in models disciplined by distributional moments and realistic household balance sheets rather than representative-agent shortcuts.\n",
"\n",
"A second direction is stronger emphasis on state dependence and nonlinearities. While post-crisis work had already suggested larger multipliers near the effective lower bound, newer papers investigate richer nonlinear mechanisms and sign asymmetries. {cite:t}`nonlinear_brinca_2019` and {cite:t}`understanding_barnichon_2021` push this agenda by showing that the effects of fiscal expansions and contractions need not be symmetric and may depend on labor-market slack, financing conditions, and the composition of shocks. In parallel, {cite:t}`heterogeneous_bernardini_2020` documents substantial variation in multiplier estimates around the Great Recession period, reinforcing the view that \"average\" multipliers can hide large cyclical and cross-country heterogeneity. Together, these papers moved the field away from point estimates toward conditional policy rules: when policymakers ask for a multiplier, the modern answer is \"conditional on regime, sign, and composition.\"\n",
"\n",
"A third direction, very close to the Dupor et al. research design, is local-to-aggregate mapping in spatial economies. Earlier regional IV designs established credible local treatment effects; the frontier is now to recover general-equilibrium propagation through production and trade linkages. {cite:t}`spatial_casas_2025` illustrates this trajectory with a spatial general-equilibrium treatment of local spending multipliers in EU regions, where cross-region demand and supply linkages are central rather than residual. This is conceptually aligned with the multi-region structure in {cite:t}`Dupor2023-bn`: local spending shocks generate spillovers that can make aggregate effects exceed local direct effects. The practical implication is that new empirical work increasingly combines regional quasi-experiments with explicit network or spatial structure, instead of treating spillovers as a nuisance. In the same spirit, {cite:t}`fiscal_sheremirov_2022` broadens external validity by comparing advanced and developing economies using military-spending variation, highlighting that institutions and macro regimes materially alter transmission.\n",
"\n",
"Related but distinct, another emerging branch studies household balance-sheet channels beyond standard transfer episodes. {cite:t}`macroeconomic_auclert_2019` (debt relief and bankruptcy protections) and {cite:t}`stock_chodorowreich_2021` (stock-market wealth and local labor markets) reinforce a broader post-2019 conclusion: fiscal and quasi-fiscal policies work through heterogeneous exposures to income, debt, and asset-price changes. This complements the consumption-MPC logic behind Dupor et al. and points to a richer notion of stimulus design where targeting depends not only on current income but also on liquidity, leverage, and wealth composition.\n",
"\n",
"Putting these strands together, the main research directions that emerged after 2019 are: (i) integrating fiscal and monetary policy in tractable HANK frameworks suitable for quantitative policy experiments; (ii) estimating regime-dependent, nonlinear multipliers rather than unconditional averages; (iii) connecting local causal estimates to national outcomes through explicit spatial and network propagation; and (iv) embedding household balance-sheet heterogeneity directly into fiscal design questions. The cutting edge now is not a single new estimate, but a synthesis: micro-founded heterogeneity plus credible quasi-experimental variation plus explicit aggregation structure. Under this view, the core contribution of {cite:t}`Dupor2023-bn` looks increasingly prescient, because it anticipated exactly this workflow by combining regional consumption evidence with a multi-region heterogeneous-agent model.\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Most important papers for where the field is heading (3-5)\n",
"\n",
"1. **Auclert, Rognlie, and Straub (2024), _Fiscal and Monetary Policy with Heterogeneous Agents_** ({cite:t}`fiscal_auclert_2024`) \n",
" Most important unifying framework for joint fiscal-monetary analysis in HANK settings; likely to be the benchmark platform for policy counterfactuals.\n",
"\n",
"2. **Hagedorn, Manovskii, and Mitman (2019), _The Fiscal Multiplier_** ({cite:t}`fiscal_hagedorn_2019`) \n",
" Foundational quantitative result showing how incomplete markets and policy financing/monetary rules jointly determine multiplier magnitudes.\n",
"\n",
"3. **Barnichon, Debortoli, and Matthes (2021), _Understanding the Size of the Government Spending Multiplier: It's in the Sign_** ({cite:t}`understanding_barnichon_2021`) \n",
" Influential for the nonlinear and asymmetry agenda; moves the literature toward sign- and regime-contingent fiscal effects.\n",
"\n",
"4. **Bernardini, De Schryder, and Peersman (2020), _Heterogeneous Government Spending Multipliers in the Era Surrounding the Great Recession_** ({cite:t}`heterogeneous_bernardini_2020`) \n",
" Key empirical evidence that multiplier heterogeneity across states and periods is substantial, sharpening the case against one-size-fits-all estimates.\n",
"\n",
"5. **Casas et al. (2025), _A Spatial General Equilibrium Analysis of Local Public Spending Multipliers in the European Union Regions_** ({cite:t}`spatial_casas_2025`) \n",
" Represents the frontier local-to-aggregate spatial GE approach, where spillovers and interregional linkages are modeled explicitly.\n",
"\n",
"If you want a \"top 3 only\" version, the strongest core set is: {cite:t}`fiscal_auclert_2024`, {cite:t}`fiscal_hagedorn_2019`, and {cite:t}`understanding_barnichon_2021`.\n"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"name": "python"
}
},
"nbformat": 4,
"nbformat_minor": 5
}
Loading