Carroll, Crawley, Slacalek, Tokuoka and White (2020) resolved the conflict between micro and macro consumption evidence — the coexistence puzzle — by adding sticky expectations to a microfounded heterogeneous-agent model: each household updates its perception of aggregate permanent income only with probability Π per quarter (calibrated to 0.25), which generates aggregate excess smoothness while preserving realistic household-level behavior. The update probability, however, is exogenous and identical across households and over time, and the mechanism has not been connected directly to measured household expectations. The open problem, stated in the paper's own conclusion, is to enrich the model so that the degree of attention varies plausibly over time and across people (endogenous or state-dependent updating), and to validate the mechanism against available expectations data such as consumer sentiment, news-based uncertainty, and elicited macroeconomic expectations, including how behavior differs between normal and crisis times — while preserving the model's joint match of the micro and macro stylized facts. Later literature has advanced adjacent threads (behavioral New Keynesian general equilibrium, sticky-expectations heterogeneous-agent New Keynesian (HANK) models, measured-expectations policy counterfactuals) and has built endogenous-attention models for inflation expectations, but as of late 2025 the consumption setting with endogenous, validated attention remains open.

Background

Aggregate consumption in the data is smoother than aggregate income — the 'excess smoothness' finding of Campbell and Deaton (1989) — while household-level consumption is well explained by frictionless permanent-income logic. The representative-agent macroeconomics literature has matched the aggregate smoothness by assuming habit formation, in which utility depends on consumption relative to a habit stock of past consumption, making spending inertial. The meta-analysis of Havranek, Rusnak and Sokolova (2017) showed the resulting conflict: studies using macro data estimate large positive habit parameters, while micro studies find essentially none. This micro–macro conflict, sometimes called the coexistence puzzle (because aggregate consumption exhibits both excess smoothness and excess sensitivity to predictable income changes), is arguably the central puzzle in the microfoundations of aggregate consumption dynamics. Carroll, Crawley, Slacalek, Tokuoka and White (2020) propose a different resolution: each household solves a standard buffer-stock consumption-savings problem (a precautionary-savings model in which consumers hold wealth to smooth idiosyncratic income risk) and observes its own idiosyncratic circumstances perfectly, but updates its perception of the aggregate economy — the level and growth rate of aggregate permanent income — only occasionally, with probability Π per quarter. This 'sticky expectations' assumption descends from the sticky-information models of Mankiw and Reis (2002), in which agents' information sets are revised only sporadically, and from the rational-inattention framework of Sims (2003), in which agents with finite information-processing capacity allocate attention optimally; Gabaix (2014) proposed a sparsity-based model of bounded rationality in which agents simplify decision problems by neglecting their less important features. Under sticky expectations, aggregation over households whose information is of different ages makes aggregate consumption respond sluggishly to aggregate shocks, matching the macro facts, while household-level behavior stays close to the frictionless benchmark, matching the micro facts; the paper also shows the implied inattention has tiny utility costs (on the order of one two-thousandth of lifetime consumption, about 0.05 percent), because aggregate risk is a small share of the total uncertainty households face. The cost-of-stickiness analysis sketches a once-and-for-all optimal choice of Π but the model treats Π = 0.25 as exogenous, uniform across households and time, informed by estimates of how fast households' macroeconomic expectations adapt to news. The paper's conclusion states the unfinished agenda: the model should be enriched so that the degree of attention varies plausibly over time or across people, and the model should be connected to available expectations data — measures of consumer sentiment, news-based measures of uncertainty such as the Baker–Bloom–Davis index of policy-related economic uncertainty constructed from newspaper coverage, and direct elicitations of macroeconomic expectations — which would be especially useful for understanding how behavior differs between normal times and crisis times.

Problem Statement

Within the sticky-expectations framework of Carroll, Crawley, Slacalek, Tokuoka and White (2020) — a microfounded buffer-stock heterogeneous-agent consumption model in which households update their perceptions of aggregate permanent income with exogenous per-quarter probability Π (calibrated to 0.25) — the open problem, as stated in the paper's conclusion, is to enrich the model so that it has plausible implications for how the degree of attention varies over time and across people, and to connect the model to the available expectations data (for example, measures of consumer sentiment, news-based uncertainty measures, and direct elicitations of macroeconomic expectations). Concretely, the open problem is to (1) replace the exogenous, uniform update probability with an endogenous or state-dependent attention process — varying across household characteristics and with the macroeconomic environment, e.g. crisis versus normal times — motivated by plausible determinants of attention such as the costs and benefits of information, without prescribing a particular modeling approach, and solved within the model's quantitative consumption-savings environment; and (2) validate the mechanism and the attention process against measured household expectations data, testing whether the model's implied pattern of updating (who updates, when, and how this varies with conditions) matches what surveys of household macroeconomic expectations show, while preserving the framework's central result of jointly matching the micro evidence (no household-level excess smoothness, realistic spending responses) and the macro evidence (excess smoothness and excess sensitivity of aggregate consumption). These two components serve one objective: establishing the sticky-expectations resolution of the micro–macro consumption puzzle as an endogenous, empirically validated mechanism rather than a maintained calibration.

References

  1. Carroll, Christopher D., Edmund Crawley, Jiri Slacalek, Kiichi Tokuoka, and Matthew N. White (2020), 'Sticky Expectations and Consumption Dynamics', American Economic Journal: Macroeconomics 12(3), 40–76, DOI 10.1257/mac.20180286, https://doi.org/10.1257/mac.20180286. Open-access versions: ECB Working Paper No. 2152 (May 2018), https://www.ecb.europa.eu/pub/pdf/scpwps/ecb.wp2152.en.pdf, and the author-hosted near-final preprint, https://edmundcrawley.com/StickyE.pdf
  2. Campbell, John, and Angus Deaton (1989), 'Why is Consumption So Smooth?', Review of Economic Studies 56(3), 357–373, DOI 10.2307/2297552, https://doi.org/10.2307/2297552
  3. Havranek, Tomas, Marek Rusnak, and Anna Sokolova (2017), 'Habit formation in consumption: A meta-analysis', European Economic Review 95, 142–167, DOI 10.1016/j.euroecorev.2017.03.009, https://doi.org/10.1016/j.euroecorev.2017.03.009
  4. Mankiw, N. Gregory, and Ricardo Reis (2002), 'Sticky Information versus Sticky Prices: A Proposal to Replace the New Keynesian Phillips Curve', Quarterly Journal of Economics 117(4), 1295–1328, DOI 10.1162/003355302320935034, https://doi.org/10.1162/003355302320935034
  5. Sims, Christopher A. (2003), 'Implications of rational inattention', Journal of Monetary Economics 50(3), 665–690, DOI 10.1016/S0304-3932(03)00029-1, https://doi.org/10.1016/S0304-3932(03)00029-1
  6. Gabaix, Xavier (2014), 'A Sparsity-Based Model of Bounded Rationality', Quarterly Journal of Economics 129(4), 1661–1710, DOI 10.1093/qje/qju024, https://doi.org/10.1093/qje/qju024
  7. Gabaix, Xavier (2020), 'A Behavioral New Keynesian Model', American Economic Review 110(8), 2271–2327, DOI 10.1257/aer.20162005, https://doi.org/10.1257/aer.20162005
  8. Auclert, Adrien, Matthew Rognlie, and Ludwig Straub (2020), 'Micro Jumps, Macro Humps: Monetary Policy and Business Cycles in an Estimated HANK Model', NBER Working Paper 26647, DOI 10.3386/w26647, https://doi.org/10.3386/w26647
  9. Auclert, Adrien, Matthew Rognlie, and Ludwig Straub (2024), 'The Intertemporal Keynesian Cross', Journal of Political Economy 132(12), 4068–4121, DOI 10.1086/732531, https://doi.org/10.1086/732531
  10. Roth, Christopher, and Johannes Wohlfart (2020), 'How Do Expectations about the Macroeconomy Affect Personal Expectations and Behavior?', Review of Economics and Statistics 102(4), 731–748, DOI 10.1162/rest_a_00867, https://doi.org/10.1162/rest_a_00867
  11. Roth, Christopher, Mirko Wiederholt, and Johannes Wohlfart (2023), 'The Effects of Monetary Policy: Theory with Measured Expectations', SSRN working paper, DOI 10.2139/ssrn.4329475, https://doi.org/10.2139/ssrn.4329475
  12. Adams, Jonathan J., and Eugenio Rojas (2024), 'Household Consumption and Dispersed Information', Journal of Monetary Economics 147, 103592, DOI 10.1016/j.jmoneco.2024.103592, https://doi.org/10.1016/j.jmoneco.2024.103592
  13. Eva, Kenneth, Michael J. Lamla, and Damjan Pfajfar (2025), 'State-Dependent Sticky Expectations: Evidence and Theory', Federal Reserve Bank of Cleveland Working Paper No. 25-23 (October 2025), DOI 10.26509/frbc-wp-202523, https://doi.org/10.26509/frbc-wp-202523 (also SSRN DOI 10.2139/ssrn.5603532, https://doi.org/10.2139/ssrn.5603532)
  14. Boccanfuso, Jeremy (2022), 'Inattention region and consumption response to income windfalls', SSRN working paper, DOI 10.2139/ssrn.4138108, https://doi.org/10.2139/ssrn.4138108
  15. Carroll, Christopher D., Edmund Crawley, Jiri Slacalek, and Matthew N. White (2020), 'Modeling the Consumption Response to the CARES Act', Finance and Economics Discussion Series 2020-077, Board of Governors of the Federal Reserve System, DOI 10.17016/feds.2020.077, https://doi.org/10.17016/feds.2020.077