主观不确定性与边际消费倾向

Gizem Koşar|Davide Melcangi Subjective Uncertainty and the Marginal Propensity to ConsumeGizem Koşarand DavideMelcangi Federal Reserve Bank of New York Staff Reports, no.1148April2025https://doi.org/10.59576/sr.1148 Abstract Earnings uncertainty is central to most heterogeneous-household models.Yet, thereis surprisingly littleevidence on howsubjective uncertainty is related to consumptionbehavior.Using unique data from theSurvey of Consumer Expectations, we show thatthe marginal propensity to consume (MPC) is increasingand concave in individualspecificearnings growth uncertainty. In the workhorse consumption–savingsmodel,augmented with risk heterogeneity, MPCs decline with earnings uncertainty, contraryto theempirical evidence. We pinpoint which mechanisms, central to the model, createthis disconnect andshow how recently proposeddeviations from the full-informationrational expectations framework canreconcile theory with the empirical findings. JEL classification:D12, D84, E21Keywords:marginal propensity to consume, consumption, subjective uncertainty,heterogeneity Koşar: Federal Reserve Bank of New York, CESIfo(email:gizem.kosar@ny.frb.org).Melcangi:FederalReserve Bank of New York(email:davide.melcangi@ny.frb.org).The authorsthank Felix Aidala,AugustinBelin, and Sasha Thomas for excellent research assistance. For helpful comments andconversations,theyalsothank Orazio Attanasio, Marco Bassetto, Nicholas Bloom, Corina Boar, JardaBorovicka, Richard Crump,Marco del Negro, Keshav Dogra, Andres Drenik, Andreas Fuster, SimonGilchrist, Cosmin Ilut, KieranLarkin, Chen Lian, Elena Manresa, Virgiliu Midrigan, Luigi Pistaferri,Ayşegül Şahin, Chris Tonetti, JavierTuren, Rosen Valchev, Wilbert van der Klaauw, Joe Vavra, VenkyVenkateswaran,andBasit Zafar, as well asseminar participants at NY Fed, UCL, the SED 2024, UTAustin, Dartmouth, SEA 2024, SAEe 2024, NYU,Boston College, Bocconi University, and UVA. This paper presents preliminary findings and is being distributed to economists and other interestedreaders solely to stimulate discussion and elicit comments. The views expressed in this paper are those ofthe author(s) and do not necessarily reflect theposition of the Federal Reserve Bank of New York or theFederal Reserve System. Any errors or omissions are the responsibility of the author(s). To view the authors’ disclosure statements, visithttps://www.newyorkfed.org/research/staff_reports/sr1148.html. 1Introduction Labor income uncertainty lies at the core of most consumption models with household het-erogeneity. First, it is tightly connected with households’ precautionary savings; second, it iscrucially tied with ex-post household heterogeneity, implying that households display differ-ent marginal propensities to consume (MPC), even in the absence of ex-ante heterogeneity.1The MPC and its heterogeneity are crucial for understanding the effects of fiscal and mone-tary policy and have received a lot of attention in the literature over the past decade (e.g.,Kaplan and Violante (2014)). Despite the central role of idiosyncratic uncertainty, empir-ical evidence on these mechanisms remains surprisingly inconclusive, largely due to datalimitations. In this paper, we fill this gap and examine the relationship between individual-specificsubjective uncertainty and the MPC, contributing to the literature in three main ways.First, we use novel and unique data on subjective expectations to empirically documentthat MPCs follow a hump-shaped pattern with respect to individual-level earnings growthuncertainty. The relationship is increasing for most households and robust to a wide rangeof controls. Second, we find that this relationship is primarily driven by variation betweenhouseholds. We show that much of this variation remains unexplained by observable house-hold characteristics, suggesting a potentially important role for latent heterogeneity. Third,we show that most of our empirical findings are inconsistent with a standard incomplete mar-kets consumption-savings model augmented with heterogeneous earnings risk. We pinpointwhich mechanisms, central to the model, lie behind the disconnect with the data. Finally,we show how recently-proposed deviations from the full-information rational expectationsframework can reconcile theory with the empirical findings. Our empirical investigation is motivated by the central role of idiosyncratic uncertainty inshaping consumption decisions within the standard buffer-stock model, which has become thefoundation of most macroeconomic models with household heterogeneity.Two importantforces are at play in this model.On the one hand, for a given level of cash on hand,higher earnings uncertainty increases the precautionary savings motive, typically makes theconsumption function more concave, and thus raises the MPC (see Carroll and Kimball(1996) for a seminal contribution). On the other hand, higher savings lead to higher wealth,which is associated with a lower MPC. This model also has