Pith. sign in

REVIEW 1 major objections 7 minor 39 references

Cash-poor spend 70 cents of each windfall dollar; cash-rich spend 30

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

A Kalman-smoother-based estimator applied to Swedish administrative data reveals a steep, convex negative gradient in the marginal propensity to consume with respect to cash-on-hand, from 0.7 to 0.3 across deciles.

T0 review reviewed 2026-07-09 challenge →

load-bearing objection Letter to colleague on arXiv:2607.07055 the 1 major comments →

arxiv 2607.07055 v1 pith:IRJFRWRI submitted 2026-07-08 econ.GN q-fin.EC

Identifying the MPC-Liquidity Gradient in High-Quality Data

Mikael Carlsson , Marco D'Amico , Erik \"Oberg , Oskar N. Skans , Karl Walentin This is my paper
classification econ.GN q-fin.EC
keywords incomepass-throughcash-on-handconsumptionequationestimatorgradientadministrative
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper develops a new method for measuring how strongly households respond to income shocks depending on how much liquid wealth they hold. The method uses a Kalman smoother — a signal-extraction algorithm borrowed from engineering — to decompose each household's income history into permanent and transitory components, then regresses consumption growth on those recovered shocks. The key theoretical result is that, when income is measured accurately (as in administrative tax data), this procedure gives the most precise estimate possible among all methods that use income histories to identify spending responses. Applying it to Sweden's population-wide tax registers, the authors find that the marginal propensity to consume (MPC) out of transitory income falls sharply and convexly with cash-on-hand: from about 0.7 in the lowest liquidity decile to about 0.3 in the highest. The response to permanent income shocks is flat near one across the distribution. Traditional estimators, designed for noisy survey data, are too imprecise to detect this gradient.

Core claim

The central discovery is that the MPC-liquidity gradient — how much spending responses to income shocks vary with liquid wealth — is steep, negative, and convex, and that this gradient becomes visible only when one uses a more efficient estimator that exploits the full income history via Kalman smoothing rather than relying on distant future income moments as instruments. The Kalman-smoothed transitory shock is the best linear prediction of the latent shock from the household's entire income history, which gives minimum-variance identification of the pass-through coefficient. In Swedish administrative data, this reveals an MPC profile dropping from roughly 0.7 to 0.3 across cash-on-hand dec,

What carries the argument

The pass-through equation (Proposition 1) linearizes consumption growth from a buffer-stock model around current innovations, yielding state-dependent coefficients γ (transitory) and λ (permanent) that are functions of predetermined cash-on-hand. The Kalman smoother recovers latent permanent and transitory shocks as the best linear projection of the true shocks onto the household's full income history. The projection identity guarantees that the regression coefficient on the smoothed transitory shock identifies γ (Proposition 3), and that no other linear income-history signal can achieve lower asymptotic variance (Proposition 4). MPC bounds follow from converting γ using the consumption-to-

Load-bearing premise

The MPC bounds require assuming that news about future income never raises current consumption more than receiving that income today would. This is what lets the pass-through coefficient be converted into an MPC range rather than just an elasticity. The bounds tighten when transitory income is only weakly persistent — which the data support — but the assumption itself is not independently tested.

What would settle it

If the persistence of transitory income shocks were substantially higher than estimated (say θ₁ above 0.5), the MPC bounds would widen enough that the gradient might no longer be distinguishable from a flat profile. Alternatively, if the Kalman-smoothed transitory signal were contaminated by classical measurement error at magnitudes above about 5 percent of income growth variance, attenuation bias could flatten the estimated gradient.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • Fiscal stimulus targeted at low-liquidity households should produce substantially larger aggregate demand effects per dollar than broad-based transfers, because the MPC gradient means the spending multiplier varies by a factor of roughly 2.3 across the liquidity distribution.
  • Monetary policy transmission through the cash-flow channel depends on the liquidity distribution of exposed households; models that assume a uniform MPC will misstate both the level and the distributional incidence of policy effects.
  • The finding that liquid wealth rather than income drives the gradient suggests that policies affecting household liquidity (e.g., access to credit, mortgage refinancing) directly alter the sensitivity of consumption to income shocks.
  • The efficiency result implies that existing semi-structural MPC estimates from survey data using BPP-type estimators may be too imprecise to detect heterogeneity that is actually present, potentially explaining why prior literature found weaker or absent liquidity gradients.
  • The flat permanent-shock pass-through near one across the liquidity distribution provides a sharp empirical signature distinguishing buffer-stock models from alternative consumption theories where permanent shocks might be smoothed differently at different wealth levels.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

1 major / 7 minor

Summary. This paper develops a semi-structural estimator for state-dependent marginal propensities to consume (MPCs) using high-quality administrative income data. The authors derive a state-dependent pass-through equation from a canonical buffer-stock consumption model (Proposition 1), show that the transitory pass-through coefficient bounds the MPC conditional on household state variables (Proposition 2), and introduce a Kalman-smoother-based estimator that recovers latent permanent and transitory income shocks from household income histories. The key theoretical results are: (i) under negligible income measurement error, the Kalman-shock estimator identifies the pass-through coefficient (Proposition 3), and (ii) it attains minimum variance within the class of estimators linear in income histories, which includes BPP (Blundell et al. 2008) and BPP-C (Commault 2022) estimators (Proposition 4). Applied to Swedish administrative tax registers (2000–2007), the method reveals a sharp, convex negative MPC gradient with respect to normalized cash-on-hand: the annual MPC falls from approximately 0.7 in the lowest decile to 0.3 in the top decile. The BPP-C estimator, by contrast, is too imprecise to detect this gradient in the same data.

Significance. The paper makes a substantive methodological and empirical contribution to the literature on consumption-savings behavior and MPC estimation. The identification and efficiency results (Propositions 3–4) are cleanly derived from projection algebra and do not depend on the empirical findings, which is a notable structural strength. The efficiency result is particularly valuable: it formally establishes that the Kalman-smoothed signal is the minimum-variance linear income-history signal, subsuming BPP and BPP-C as special cases. The empirical finding of a steep, convex MPC-liquidity gradient is economically important for heterogeneous-agent macro models and fiscal policy design. The direct head-to-head comparison with BPP-C (Figure 8) convincingly demonstrates that the gradient is undetectable with less efficient estimators, motivating the new approach. The bootstrap procedure (clustered at household level, applied to the full pipeline) and robustness to MA(2) specifications (Appendix F) are commendable. The paper also provides a transparent discussion of the bias-variance trade-off when income measurement error is non-negligible (Section 3.6, Appendix D), including simulations for cả

major comments (1)
  1. Proposition 3 (Section 3.3, Appendix C.2): The identification proof requires that the residual ω_{i,t} in the pass-through equation be orthogonal to the income-history signal after conditioning on the state and permanent-shock component. The proof establishes Cov(ε̂_{i,t}, e_{i,t}) = 0 by asserting that e_{i,t} is orthogonal to all admissible income-history signals. However, the Kalman smoother uses the FULL income history (both past and future relative to t), and the pass-through equation (Proposition 1) is only a first-order linear approximation. Higher-order terms—interactions between shocks and state, curvature of the consumption policy—are absorbed into e_{i,t}. If the true consumption policy has curvature (∂²log c/∂m² ≠ 0, which the buffer-stock model implies, especially at low cash-on-hand where the MPC gradient is steepest), the linearization error is proportional to ε_{i,t}², or
minor comments (7)
  1. Section 4.1: The paper excludes observations with housing transactions when estimating the second-stage consumption function. This is well-motivated by the single-liquid-asset assumption, but the share of observations excluded and the potential for selection on unobservables correlated with cash-on-hand deciles is not discussed. Reporting the exclusion rate by decile would address this.
  2. Abstract/Figure 6: The abstract reports MPCs of 0.7 to 0.3, but Figure 6 shows lower bounds. The upper bounds (obtained by multiplying by ~1.22) would be 0.83 to 0.38. The abstract should clarify that these are lower bounds, or report both bounds.
  3. Section 5.1: The MA(1) baseline is selected because the second autocovariance is smaller in absolute value than the first. However, the second autocovariance (−0.0026) is statistically significant given the large sample. While the authors note that statistical significance is not a good criterion for lag selection in administrative data (footnote 7), the MA(2) robustness results in Appendix F show θ₂ is also significant. A brief discussion of why the MA(1) is the economically preferred specification, beyond the relative magnitude of autocovariances, would strengthen the justification.
  4. Section 3.7: The attenuation bias from measurement error in the state variable (normalized cash-on-hand) is bounded at approximately 11% worst case. The argument relies on the reliability ratio R²_ν of the recovered transitory component, but the actual value of R²_ν in the data is not reported. Providing this quantity would allow readers to assess how much the Kalman smoother reduces the worst-case bound.
  5. Assumption 2.1: The income process assumes common parameters across households. The paper notes this restriction is necessary for pooled estimation, but given the rich administrative data, some evidence on heterogeneity in income process parameters (e.g., by education or sector) and its potential impact on the recovered shocks would be informative, even if the baseline maintains the common-parameter assumption.
  6. Appendix D: The simulation exercises use 30,421 observations per sample. It would be useful to confirm that the bias-variance trade-off conclusions are not sensitive to this sample size, particularly given that the actual estimation samples are orders of magnitude larger (1.7M+ for Singles). A brief note on whether the qualitative conclusions extend to larger samples would address potential for small-sample artifacts.
  7. Figure 5: The permanent-shock pass-through is described as 'flat and centered around one,' but the figure appears to show some decline at very low cash-on-hand levels. A brief comment on whether this is statistically meaningful or an artifact of the binning would be helpful.

Circularity Check

0 steps flagged

No significant circularity: identification and efficiency proofs are self-contained projection algebra; MPC bounds rest on a domain assumption, not a circular one.

full rationale

The paper's core theoretical results (Propositions 1-4) are derived from first principles using standard projection algebra and the canonical buffer-stock model, without circular dependencies. Proposition 3 (identification) follows from the projection identity: the Kalman-smoothed shock is the best linear projection of the latent shock onto the income history, so the projection error is orthogonal to the signal by construction — this is a standard statistical fact, not a self-referential definition. Proposition 4 (efficiency) follows from Cauchy-Schwarz applied to the correlation between any linear income-history signal and the latent shock, showing the projection signal maximizes this correlation — again, a self-contained mathematical argument. The income process parameters are estimated from pooled income-growth moments independently of the consumption regression. The MPC bounds (Proposition 2) depend on Assumption 2.4 (bounded response to news), but this is a substantive domain restriction on the consumption policy's curvature, not a circular definition of the MPC in terms of the pass-through coefficient. The empirical application estimates the pass-through coefficient from the data and then converts it to MPC bounds using the estimated income process parameters and the consumption-to-income ratio — these are distinct objects. The only minor self-citation is to Blundell et al. (2013) for the linear approximation framework, but this is a standard building block used as a starting point, not a load-bearing uniqueness claim that forces the conclusion. The skeptic's concern about higher-order terms from the linearization being correlated with future income is a correctness/misspecification concern, not a circularity: the paper's identification claim is explicitly conditional on the linear pass-through equation being correctly specified after conditioning on the state, and the residual being orthogonal to income-history signals — this is stated as an assumption, not derived from the conclusion. No step in the derivation chain reduces to its own inputs by construction.

Axiom & Free-Parameter Ledger

4 free parameters · 4 axioms · 0 invented entities

The paper introduces no new entities or particles. All parameters are estimated from data. The key axioms are standard in the consumption-savings literature except Assumption 2.4, which is specific to this paper's MPC-bounding strategy and is the most fragile link in the chain from pass-through coefficients to MPC bounds.

free parameters (4)
  • θ₁ (MA(1) transitory persistence) = 0.2191 (Singles), 0.2213 (Households)
    Estimated from pooled income-growth autocovariances; determines MPC bound width and Kalman smoother weights.
  • σ²_ε (transitory shock variance) = 0.0123 (Singles), 0.0120 (Households)
    Estimated from income moments; enters Kalman smoother and pass-through regression.
  • σ²_η (permanent shock variance) = 0.0097 (Singles), 0.0095 (Households)
    Recovered from variance identity after estimating other parameters.
  • Consumption-to-income ratio C/Y = Not a single number; varies by decile
    Used as conversion factor between pass-through elasticity and MPC; computed from data.
axioms (4)
  • domain assumption Income follows a permanent-transitory MA(k) process with common parameters across households (Assumption 2.1)
    Standard in the semi-structural MPC literature; parameters estimated from pooled moments. The common-parameter restriction is necessary for household-level shock recovery.
  • domain assumption Consumption scales linearly with permanent income (Assumption 2.3)
    Standard implication of CRRA buffer-stock model; enables normalization by permanent income.
  • ad hoc to paper Bounded response to news: news about future income does not raise consumption more than equivalent current cash (Assumption 2.4)
    Used to bound the news channel and construct MPC bounds from pass-through coefficients. The authors argue it is weak when persistence is low, but it is not independently tested.
  • domain assumption Negligible measurement error in income
    Justified by third-party-reported tax data; central to identification (Proposition 3) and efficiency (Proposition 4). Simulations in Appendix D quantify bias under violation.

reviewed 2026-07-09 · how reviews work

0 comments
Cite this review

Pith. "Pith review of Identifying the MPC-Liquidity Gradient in High-Quality Data." pith.science (2026). https://pith.science/paper/IRJFRWRI

@misc{pith2026260707055,
  author       = {Pith},
  title        = {Pith review of: Identifying the MPC-Liquidity Gradient in High-Quality Data},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/IRJFRWRI}},
  note         = {Machine review of arXiv:2607.07055}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

We estimate the gradient of the Marginal Propensity to Consume (MPC) with respect to liquidity using a new estimator designed for administrative data with negligible measurement error in income. We derive a state-dependent consumption pass-through equation from the canonical buffer-stock model, and show that the pass-through coefficient of this equation can be used to construct tight bounds on the MPC conditional on the relevant state. We recover latent permanent and transitory income shocks with a Kalman smoother and use them as regressors in an empirical representation of the consumption equation. The Kalman-shock estimator identifies the theoretical pass-through coefficient under the assumption of negligible measurement error in income, and attains the minimum variance within the class of estimators that are linear in household income histories, including the canonical GMM estimator by Blundell et al (2008) and recent refinements thereof. Applying the method to Swedish administrative tax registers, we show that consumption responses to transitory shocks have a sharp negative and convex gradient with respect to cash-on-hand; the associated annual MPC falls from 0.7 in the lowest cash-on-hand decile to 0.3 in the top decile. The permanent-shock pass-through is close to one across the cash-on-hand distribution. These patterns are not visible when using traditional, less efficient, estimators.

Figures

Figures reproduced from arXiv: 2607.07055 by Erik \"Oberg, Karl Walentin, Marco D'Amico, Mikael Carlsson, Oskar N. Skans.

Figure 1
Figure 1. Figure 1: Dynamic footprints of income shocks Notes: The transitory shock is illustrated with a small MA(1) component, calibrated to the persistence estimated in our baseline income process. through equation hold within a state cell, or after partialling out state controls, and suppose that income is measured without error. Suppose also that the residual component ωi,t is or￾thogonal to the income-history signal aft… view at source ↗
Figure 2
Figure 2. Figure 2: Distribution of wealth components (a) Net total wealth, households (b) Net total wealth, singles (c) Net illiquid wealth, households (d) Net illiquid wealth, singles (e) Liquid wealth, households (f) Liquid wealth, singles Notes: Kernel density estimates pool household-year observations over 2000–2007. Variables are in thousand 2003 SEK. Peaks in the liquid-wealth panels are capped to show tail behavior. 2… view at source ↗
Figure 3
Figure 3. Figure 3: Distribution of consumption, disposable income, and cash-on-hand [PITH_FULL_IMAGE:figures/full_fig_p023_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Distribution of lagged normalized cash-on-hand [PITH_FULL_IMAGE:figures/full_fig_p025_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Pass-through coefficients by cash-on-hand [PITH_FULL_IMAGE:figures/full_fig_p026_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: MPC by cash-on-hand Notes: Estimates show lower bounds of the MPC, and are computed within decile bins of lagged normalized cash-on-hand; the horizontal axis plots the observation-count-weighted average normalized cash-on-hand in each bin. Shaded regions show 95 percent confidence intervals based on bootstrap standard errors clustered at the household level. However, our outcome is an expenditure measure r… view at source ↗
Figure 7
Figure 7. Figure 7: MPC by components of cash-on-hand Notes: Estimates show lower bounds of the MPC, and are computed within decile bins of the relevant permanent-income-normalized cash-on-hand component; the horizontal axis plots the observation￾count-weighted average component divided by permanent income in each bin. Shaded regions show 95 percent confidence intervals based on bootstrap standard errors clustered at the hous… view at source ↗
Figure 8
Figure 8. Figure 8: MPC by cash-on-hand: OLS-LP versus BPP Notes: Estimates show lower bounds of the MPC for Singles, and are computed within decile bins of lagged normalized cash-on-hand; the horizontal axis plots the observation-count-weighted average normalized cash-on-hand in each bin. OLS-LP uses the Kalman-smoothed transitory shock. BPP￾C uses the distant future-income-growth instrument. Shaded regions show 95 percent c… view at source ↗
Figure 9
Figure 9. Figure 9: Bias-variance trade-off with income measurement error [PITH_FULL_IMAGE:figures/full_fig_p042_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Registry consumption, national accounts, and household budget survey (HUT) [PITH_FULL_IMAGE:figures/full_fig_p048_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: MPC by cash-on-hand, MA(2) Notes: Estimates show lower bounds of the MPC, and are computed within decile bins of lagged nor￾malized cash-on-hand; the horizontal axis plots the observation-count-weighted average normalized cash-on-hand in each bin. The projection estimator uses the Kalman-smoothed transitory shock and reports 95 percent confidence intervals based on bootstrap standard errors clustered at t… view at source ↗
Figure 12
Figure 12. Figure 12: Pass-through by cash-on-hand, MA(2) Notes: Columns report Singles and Households; rows report permanent- and transitory-shock pass￾through coefficients from the OLS projection estimator. Estimates are computed within decile bins of lagged normalized cash-on-hand; the horizontal axis plots the observation-count-weighted average normalized cash-on-hand in each bin. Shaded regions show 95 percent confidence … view at source ↗
Figure 13
Figure 13. Figure 13: Near- versus distant-lead future-income IV estimates [PITH_FULL_IMAGE:figures/full_fig_p054_13.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

39 extracted references · 39 canonical work pages · 1 internal anchor

  1. [1]

    Earnings and consumption dynamics: A nonlinear panel data framework

    Manuel Arellano, Richard Blundell, and Stéphane Bonhomme. Earnings and consumption dynamics: A nonlinear panel data framework. Econometrica, 85 0 (3): 0 693--734, 2017. doi:https://doi.org/10.3982/ECTA13795. URL https://onlinelibrary.wiley.com/doi/abs/10.3982/ECTA13795

    work page doi:10.3982/ecta13795 2017

  2. [2]

    HANK : A new core of usable macroeconomics

    Adrien Auclert. HANK : A new core of usable macroeconomics. AEA Papers and Proceedings, 115: 0 153--157, 2025. doi:10.1257/pandp.20251114. URL https://www.aeaweb.org/articles?id=10.1257/pandp.20251114

    work page doi:10.1257/pandp.20251114 2025

  3. [3]

    Precautionary savings, illiquid assets, and the aggregate consequences of shocks to household income risk

    Christian Bayer, Ralph Luetticke, Lien Pham-Dao, and Volker Tjaden. Precautionary savings, illiquid assets, and the aggregate consequences of shocks to household income risk. Econometrica, 87 0 (1): 0 255--290, 2019. doi:10.3982/ECTA13601. URL https://doi.org/10.3982/ECTA13601

    work page doi:10.3982/ecta13601 2019

  4. [4]

    Bilbiie, Sigurd M lster Galaasen, Refet S

    Florin O. Bilbiie, Sigurd M lster Galaasen, Refet S. G \"u rkaynak, Mathis Maehlum, and Krisztina Molnar. HANKSSON . mimeo, University of Cambridge, 2025

    work page 2025

  5. [5]

    Consumption inequality and partial insurance

    Richard Blundell, Luigi Pistaferri, and Ian Preston. Consumption inequality and partial insurance. American Economic Review, 98 0 (5): 0 1887--1921, December 2008. doi:10.1257/aer.98.5.1887. URL https://www.aeaweb.org/articles?id=10.1257/aer.98.5.1887

    work page doi:10.1257/aer.98.5.1887 1921

  6. [6]

    Decomposing changes in income risk using consumption data

    Richard Blundell, Hamish Low, and Ian Preston. Decomposing changes in income risk using consumption data. Quantitative Economics, 4 0 (1): 0 1--37, 2013. doi:10.3982/QE44. URL https://onlinelibrary.wiley.com/doi/abs/10.3982/QE44

    work page doi:10.3982/qe44 2013

  7. [7]

    Five facts about MPCs : Evidence from a randomized experiment

    Johannes Boehm, Etienne Fize, and Xavier Jaravel. Five facts about MPCs : Evidence from a randomized experiment. American Economic Review, 115: 0 1--42, 2025

    work page 2025

  8. [8]

    Carter Braxton, Kyle Herkenhoff, Jonathan Rothbaum, and Lawrence Schmidt

    J. Carter Braxton, Kyle Herkenhoff, Jonathan Rothbaum, and Lawrence Schmidt. Changing income risk across the us skill distribution: Evidence from a generalized kalman filter. American Economic Review, 115 0 (12): 0 4438–75, December 2025. doi:10.1257/aer.20231656. URL https://www.aeaweb.org/articles?id=10.1257/aer.20231656

    work page doi:10.1257/aer.20231656 2025

  9. [9]

    Imputing consumption from income and wealth information

    Martin Browning and S ren Leth‐Petersen. Imputing consumption from income and wealth information. Economic Journal, 113 0 (488): 0 F282--F301, 2003

    work page 2003

  10. [10]

    Christopher D. Carroll. Buffer-stock saving and the life cycle/permanent income hypothesis. Quarterly Journal of Economics, 112 0 (1): 0 1--55, 1997

    work page 1997

  11. [11]

    Does consumption respond to transitory shocks? reconciling natural experiments and semistructural methods

    Jeanne Commault. Does consumption respond to transitory shocks? reconciling natural experiments and semistructural methods. American Economic Journal: Macroeconomics, 14: 0 96--122, 2022

    work page 2022

  12. [12]

    Consumption heterogeneity: Micro drivers and macro implications

    Edmund Crawley and Andreas Kuchler. Consumption heterogeneity: Micro drivers and macro implications. American Economic Journal: Macroeconomics, 15: 0 314--341, 2023

    work page 2023

  13. [13]

    Saving and liquidity constraints

    Angus Deaton. Saving and liquidity constraints. Econometrica, 59 0 (5): 0 1221--1248, 1991

    work page 1991

  14. [14]

    Stock market returns and consumption

    Marco Di Maggio , Amir Kermani, and Kaveh Majlesi. Stock market returns and consumption. Journal of Finance, 75 0 (6): 0 3175--3219, 2020. doi:10.1111/jofi.12968

    work page doi:10.1111/jofi.12968 2020

  15. [15]

    Holm, and Gisle J

    Andreas Fagereng, Martin B. Holm, and Gisle J. Natvik. Mpc heterogeneity and household balance sheets. American Economic Journal: Macroeconomics, 13 0 (4): 0 1--54, October 2021

    work page 2021

  16. [16]

    Torstensen

    Andreas Fagereng, Helene Onshuus, and Kjersti N. Torstensen. The consumption expenditure response to unemployment: Evidence from norwegian households. Journal of Monetary Economics, 146, 2024

    work page 2024

  17. [17]

    Household debt and monetary policy: Revealing the cash-flow channel

    Martin Flod \'e n, Matilda Kilstr \"o m, J \'o sef Sigurdsson, and Roine Vestman. Household debt and monetary policy: Revealing the cash-flow channel. Economic Journal, 131 0 (636): 0 1742--1771, 2021. doi:10.1093/ej/ueaa135

    work page doi:10.1093/ej/ueaa135 2021

  18. [18]

    A Theory of the Consumption Function

    Milton Friedman. A Theory of the Consumption Function. Princeton University Press, Princeton, NJ, 1957. URL https://www.nber.org/books-and-chapters/theory-consumption-function

    work page 1957

  19. [19]

    What would you do with \ 500? spending responses to gains, losses, news, and loans

    Andreas Fuster, Greg Kaplan, and Basit Zafar. What would you do with \ 500? spending responses to gains, losses, news, and loans. Review of Economic Studies, 88: 0 1760--1795, 2021. doi:10.1093/restud/rdaa076

    work page doi:10.1093/restud/rdaa076 2021

  20. [20]

    Liquid wealth and consumption smoothing of typical labor income shocks

    Peter Ganong, Damon Jones, Pascal Noel, Diana Farrell, Fiona Greig, and Chris Wheat. Liquid wealth and consumption smoothing of typical labor income shocks. Working paper, 2025

    work page 2025

  21. [21]

    How americans respond to idiosyncratic and exogenous changes in household wealth and unearned income

    Mikhail Golosov, Michael Graber, Magne Mogstad, and David Novgorodsky. How americans respond to idiosyncratic and exogenous changes in household wealth and unearned income. Quarterly Journal of Economics, 139 0 (2): 0 1321--1395, 2024. doi:10.1093/qje/qjad053. URL https://doi.org/10.1093/qje/qjad053

    work page doi:10.1093/qje/qjad053 2024

  22. [22]

    What do data on millions of u.s

    Fatih Guvenen, Fatih Karahan, Serdar Ozkan, and Jae Song. What do data on millions of u.s. workers reveal about lifecycle earnings dynamics? Econometrica, 89 0 (5): 0 2303--2339, 2021. doi:10.3982/ECTA14603. URL https://doi.org/10.3982/ECTA14603

    work page doi:10.3982/ecta14603 2021

  23. [23]

    Robert E. Hall. Stochastic implications of the life cycle-permanent income hypothesis: Theory and evidence. Journal of Political Economy, 86 0 (6): 0 971--987, 1978. doi:10.1086/260724. URL https://doi.org/10.1086/260724

    work page doi:10.1086/260724 1978

  24. [24]

    Johnson, Jonathan A

    David S. Johnson, Jonathan A. Parker, and Nicholas S. Souleles. Household expenditure and the income tax rebates of 2001. American Economic Review, 96 0 (5): 0 1589--1610, December 2006. doi:10.1257/aer.96.5.1589. URL https://www.aeaweb.org/articles?id=10.1257/aer.96.5.1589

    work page doi:10.1257/aer.96.5.1589 2001

  25. [26]

    Violante

    Greg Kaplan and Giovanni L. Violante. A model of the consumption response to fiscal stimulus payments. Econometrica, 82 0 (4): 0 1199--1239, 2014 b . doi:https://doi.org/10.3982/ECTA10528

    work page doi:10.3982/ecta10528 2014

  26. [27]

    Violante

    Greg Kaplan and Giovanni L. Violante. The marginal propensity to consume in heterogeneous agent models. Annual Review of Economics, 14: 0 747--775, 2022. doi:10.1146/annurev-economics-080217-053444. URL https://doi.org/10.1146/annurev-economics-080217-053444

    work page doi:10.1146/annurev-economics-080217-053444 2022

  27. [28]

    The value of registry data for consumption analysis: An application to health shocks

    Jonas Kolsrud, Camille Landais, and Johannes Spinnewijn. The value of registry data for consumption analysis: An application to health shocks. Journal of Public Economics, 189: 0 104088, 2020. ISSN 0047-2727

    work page 2020

  28. [29]

    Violante

    Dirk Krueger, Fabrizio Perri, Luigi Pistaferri, and Giovanni L. Violante. Cross-sectional facts for macroeconomists. Review of Economic Dynamics, 13 0 (1): 0 1--14, 2010. doi:10.1016/j.red.2009.12.001. URL https://www.sciencedirect.com/science/article/pii/S1094202509000726

    work page doi:10.1016/j.red.2009.12.001 2010

  29. [30]

    A simple framework for MPCs and MPXs

    Peter Maxted, David Laibson, and Benjamin Moll. A simple framework for MPCs and MPXs . Manuscript, December 2025, 2025

    work page 2025

  30. [31]

    Consumption, income changes, and heterogeneity: Evidence from two fiscal stimulus programs

    Kanishka Misra and Paolo Surico. Consumption, income changes, and heterogeneity: Evidence from two fiscal stimulus programs. American Economic Journal: Macroeconomics, 6: 0 84--106, 2014

    work page 2014

  31. [32]

    Utility analysis and the consumption function: An interpretation of cross-section data

    Franco Modigliani and Richard Brumberg. Utility analysis and the consumption function: An interpretation of cross-section data. In Kenneth K. Kurihara, editor, Post-Keynesian Economics, pages 388--436. Rutgers University Press, New Brunswick, NJ, 1954

    work page 1954

  32. [33]

    Parker, Nicholas S

    Jonathan A. Parker, Nicholas S. Souleles, David S. Johnson, and Robert McClelland. Consumer spending and the economic stimulus payments of 2008. American Economic Review, 103 0 (6): 0 2530--53, October 2013. doi:10.1257/aer.103.6.2530. URL https://www.aeaweb.org/articles?id=10.1257/aer.103.6.2530

    work page doi:10.1257/aer.103.6.2530 2008

  33. [34]

    The matching multiplier and the amplification of recessions

    Christina Patterson. The matching multiplier and the amplification of recessions. American Economic Review, 113 0 (4): 0 982--1012, 2023. doi:10.1257/aer.20210254. URL https://www.aeaweb.org/articles?id=10.1257/aer.20210254

    work page doi:10.1257/aer.20210254 2023

  34. [35]

    Rich pickings? risk, return, and skill in household wealth

    Laurent Bach, Laurent E Calvet, and Paolo Sodini. Rich pickings? risk, return, and skill in household wealth. American Economic Review, 110: 0 2703--2747, 2020. doi:10.2307/26966314. URL https://www.jstor.org/stable/10.2307/26966314

    work page doi:10.2307/26966314 2020

  35. [36]

    Evidence on the validity of cross-sectional and longitudinal labor market data

    John Bound, Charles Brown, Greg J Duncan, and Willard L Rodgers. Evidence on the validity of cross-sectional and longitudinal labor market data. Journal of Labor Economics, 12: 0 345--368, 1994

    work page 1994

  36. [37]

    Arie Kapteyn and Jelmer Y. Ypma. Measurement error and misclassification: A comparison of survey and administrative data. Journal of Labor Economics, 25 0 (3): 0 513--551, 2007. doi:10.1086/513298. URL https://doi.org/10.1086/513298

    work page doi:10.1086/513298 2007

  37. [38]

    A large-scale validation study of measurement errors in longitudinal survey data

    Nicolai Kristensen and Niels Westergaard-Nielsen. A large-scale validation study of measurement errors in longitudinal survey data. Journal of Economic and Social Measurement, 32: 0 65--92, 2007

    work page 2007

  38. [39]

    A Globally Convergent Proximal Newton-Type Method in Nonsmooth Convex Optimization

    Erik Meijer, Susann Rohwedder, and Tom Wansbeek. Measurement error in earnings data: Using a mixture model approach to combine survey and register data. Journal of Business & Economic Statistics, 30 0 (2): 0 191--201, 2012. doi:10.1198/jbes.2011.08166. URL https://pmc.ncbi.nlm.nih.gov/articles/PMC3604906/

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1198/jbes.2011.08166 2012

  39. [40]

    Measurement error and earnings dynamics: Some estimates from the psid validation study

    J \"o rn-Steffen Pischke. Measurement error and earnings dynamics: Some estimates from the psid validation study. Journal of Business & Economic Statistics, 13 0 (3): 0 305--314, 1995. doi:10.1080/07350015.1995.10524599. URL https://www.jstor.org/stable/1392190

    work page doi:10.1080/07350015.1995.10524599 1995

This paper was first reviewed by glm-5.2 on July 9, 2026.